Doc. no. D????
Date: 2016-07-11
Project: Programming Language C++
Reply to: Marshall Clow <lwgchair@gmail.com>

C++ Standard Library Active Issues List (Revision R99)

Revised 2016-07-11 at 06:07:25 UTC

Reference ISO/IEC IS 14882:2014(E)

Also see:

The purpose of this document is to record the status of issues which have come before the Library Working Group (LWG) of the INCITS PL22.16 and ISO WG21 C++ Standards Committee. Issues represent potential defects in the ISO/IEC IS 14882:2014(E) document.

This document contains only library issues which are actively being considered by the Library Working Group, i.e., issues which have a status of New, Open, Ready, or Review. See Library Defect Reports List for issues considered defects and Library Closed Issues List for issues considered closed.

The issues in these lists are not necessarily formal ISO Defect Reports (DR's). While some issues will eventually be elevated to official Defect Report status, other issues will be disposed of in other ways. See Issue Status.

Prior to Revision 14, library issues lists existed in two slightly different versions; a Committee Version and a Public Version. Beginning with Revision 14 the two versions were combined into a single version.

This document includes [bracketed italicized notes] as a reminder to the LWG of current progress on issues. Such notes are strictly unofficial and should be read with caution as they may be incomplete or incorrect. Be aware that LWG support for a particular resolution can quickly change if new viewpoints or killer examples are presented in subsequent discussions.

For the most current official version of this document see http://www.open-std.org/jtc1/sc22/wg21/. Requests for further information about this document should include the document number above, reference ISO/IEC 14882:2014(E), and be submitted to Information Technology Industry Council (ITI), 1250 Eye Street NW, Washington, DC 20005.

Public information as to how to obtain a copy of the C++ Standard, join the standards committee, submit an issue, or comment on an issue can be found in the comp.std.c++ FAQ.

How to submit an issue

  1. Mail your issue to the author of this list.
  2. Specify a short descriptive title. If you fail to do so, the subject line of your mail will be used as the issue title.
  3. If the "From" on your email is not the name you wish to appear as issue submitter, then specify issue submitter.
  4. Provide a brief discussion of the problem you wish to correct. Refer to the latest working draft or standard using [section.tag] and paragraph numbers where appropriate.
  5. Provide proposed wording. This should indicate exactly how you want the standard to be changed. General solution statements belong in the discussion area. This area contains very clear and specific directions on how to modify the current draft. If you are not sure how to word a solution, you may omit this part. But your chances of a successful issue greatly increase if you attempt wording.
  6. It is not necessary for you to use html markup. However, if you want to, you can <ins>insert text like this</ins> and <del>delete text like this</del>. The only strict requirement is to communicate clearly to the list maintainer exactly how you want your issue to look.
  7. It is not necessary for you to specify other html font/formatting mark-up, but if you do the list maintainer will attempt to respect your formatting wishes (as described by html markup, or other common idioms).
  8. It is not necessary for you to specify open date or last modified date (the date of your mail will be used).
  9. It is not necessary for you to cross reference other issues, but you can if you like. You do not need to form the hyperlinks when you do, the list maintainer will take care of that.
  10. One issue per email is best.
  11. Between the time you submit the issue, and the next mailing deadline (date at the top of the Revision History), you own this issue. You control the content, the stuff that is right, the stuff that is wrong, the format, the misspellings, etc. You can even make the issue disappear if you want. Just let the list maintainer know how you want it to look, and he will try his best to accommodate you. After the issue appears in an official mailing, you no longer enjoy exclusive ownership of it.

Revision History

Issue Status

Issues reported to the LWG transition through a variety of statuses, indicating their progress towards a resolution. Typically, most issues will flow through the following stages.

New - The issue has not yet been reviewed by the LWG. Any Proposed Resolution is purely a suggestion from the issue submitter, and should not be construed as the view of LWG.

Open - The LWG has discussed the issue but is not yet ready to move the issue forward. There are several possible reasons for open status:

A Proposed Resolution for an open issue is still not be construed as the view of LWG. Comments on the current state of discussions are often given at the end of open issues in an italic font. Such comments are for information only and should not be given undue importance.

Review - Exact wording of a Proposed Resolution is now available for review on an issue for which the LWG previously reached informal consensus.

Ready - The LWG has reached consensus that the issue is a defect in the Standard, the Proposed Resolution is correct, and the issue is ready to forward to the full committee for further action as a Defect Report (DR).

Typically, an issue must have a proposed resolution in the currently published issues list, whose wording does not change during LWG review, to move to the Ready status.

Voting - This status should not be seen in a published issues list, but is a marker for use during meetings to indicate an issues was Ready in the pre-meeting mailing, the Proposed Resolution is correct, and the issue will be offered to the working group at the end of the current meeting to apply to the current working paper (WP) or to close in some other appropriate manner. This easily distinguishes such issues from those moving to Ready status during the meeting itself, that should not be forwarded until the next meeting. If the issue does not move forward, it should fall back to one of the other open states before the next list is published.

Immediate - This status should not be seen in a published issues list, but is a marker for use during meetings to indicate an issues was not Ready in the pre-meeting mailing, but the Proposed Resolution is correct, and the issue will be offered to the working group at the end of the current meeting to apply to the current working paper (WP) or to close in some other appropriate manner. This status is used only rarely, typically for fixes that are both small and obvious, and usually within a meeting of the expected publication of a revised standard. If the issue does not move forward, it should fall back to one of the other open states before the next list is published.

In addition, there are a few ways to categorise and issue that remains open to a resolution within the library, but is not actively being worked on.

Deferred - The LWG has discussed the issue, is not yet ready to move the issue forward, but neither does it deem the issue significant enough to delay publishing a standard or Technical Report. A typical deferred issue would be seeking to clarify wording that might be technically correct, but easily mis-read.

A Proposed Resolution for a deferred issue is still not be construed as the view of LWG. Comments on the current state of discussions are often given at the end of open issues in an italic font. Such comments are for information only and should not be given undue importance.

Core - The LWG has discussed the issue, and feels that some key part of resolving the issue is better handled by a cleanup of the language in the Core part of the standard. The issue is passed to the Core Working Group, which should ideally open a corresponding issue that can be linked from the library issue. Such issues will be revisitted after Core have made (or declined to make) any changes.

EWG - The LWG has discussed the issue, and wonder that some key part of resolving the issue is better handled by some (hopefully small) extension to the language. The issue is passed to the Evolution Working Group, which should ideally open a corresponding issue that can be linked from the library issue. Such issues will be revisitted after Evoltion have made (or declined to make) any recommendations. Positive recommendations from EWG will often mean the issue transition to Core status while we wait for some proposed new feature to land in the working paper.

LEWG - The LWG has discussed the issue, and deemd the issue is either an extension, however small, or changes the library design in some fundamental way, and so has delegated the initial work to the Library Evolution Working Group.

Ultimately, all issues should reach closure with one of the following statuses.

DR - (Defect Report) - The full WG21/PL22.16 committee has voted to forward the issue to the Project Editor to be processed as a Potential Defect Report. The Project Editor reviews the issue, and then forwards it to the WG21 Convenor, who returns it to the full committee for final disposition. This issues list accords the status of DR to all these Defect Reports regardless of where they are in that process.

WP - (Working Paper) - The proposed resolution has not been accepted as a Technical Corrigendum, but the full WG21/PL22.16 committee has voted to apply the Defect Report's Proposed Resolution to the working paper.

C++14 - (C++ Standard, as revised for 2014) - The full WG21/PL22.16 committee has voted to accept the Defect Report's Proposed Resolution into the published 2014 revision to the C++ standard, ISO/IEC IS 14882:2014(E).

C++11 - (C++ Standard, as revised for 2011) - The full WG21/PL22.16 committee has voted to accept the Defect Report's Proposed Resolution into the published 2011 revision to the C++ standard, ISO/IEC IS 14882:2011(E).

CD1 - (Committee Draft 2008) - The full WG21/PL22.16 committee has voted to accept the Defect Report's Proposed Resolution into the Fall 2008 Committee Draft.

TC1 - (Technical Corrigenda 1) - The full WG21/PL22.16 committee has voted to accept the Defect Report's Proposed Resolution as a Technical Corrigenda. Action on this issue is thus complete and no further action is possible under ISO rules.

TRDec - (Decimal TR defect) - The LWG has voted to accept the Defect Report's Proposed Resolution into the Decimal TR. Action on this issue is thus complete and no further action is expected.

Resolved - The LWG has reached consensus that the issue is a defect in the Standard, but the resolution adopted to resolve the issue came via some other mechanism than this issue in the list - typically by applying a formal paper, occasionally as a side effect of consolidating several interacting issue resolutions into a single issue.

Dup - The LWG has reached consensus that the issue is a duplicate of another issue, and will not be further dealt with. A Rationale identifies the duplicated issue's issue number.

NAD - The LWG has reached consensus that the issue is not a defect in the Standard.

NAD Editorial - The LWG has reached consensus that the issue can either be handled editorially, or is handled by a paper (usually linked to in the rationale).

Tentatively - This is a status qualifier. The issue has been reviewed online, or at an unofficial meeting, but not in an official meeting, and some support has been formed for the qualified status. Tentatively qualified issues may be moved to the unqualified status and forwarded to full committee (if Ready) within the same meeting. Unlike Ready issues, Tentatively Ready issues will be reviewed in subcommittee prior to forwarding to full committee. When a status is qualified with Tentatively, the issue is still considered active.

Pending - This is a status qualifier. When prepended to a status this indicates the issue has been processed by the committee, and a decision has been made to move the issue to the associated unqualified status. However for logistical reasons the indicated outcome of the issue has not yet appeared in the latest working paper.

The following statuses have been retired, but may show up on older issues lists.

NAD Future - In addition to the regular status, the LWG believes that this issue should be revisited at the next revision of the standard. That is now an ongoing task managed by the Library Evolution Working Group, and most issues in this status were reopended with the status LEWG.

NAD Concepts - This status reflects an evolution of the language during the development of C++11, where a new feature entered the language, called concepts, that fundamentally changed the way templates would be specified and written. While this language feature was removed towards the end of the C++11 project, there is a clear intent to revisit this part of the language design. During that development, a number of issues were opened against the updated library related to use of that feature, or requesting fixes that would require explicit use of the concepts feature. All such issues have been closed with this status, and may be revisitted should this or a similar language feature return for a future standard.

NAD Arrays - This status reflects an evolution of the language during the development of C++14/17, where work on a Technical Specification, called the Arrays TS was begun. In early 2016, this work was abandoned, and the work item was officially withdrawn. During development of the TS, a number of issues were opened the features in the TS. All such issues have been closed with this status, and may be revisitted should this or a similar language feature return for a future standard.

Issues are always given the status of New when they first appear on the issues list. They may progress to Open or Review while the LWG is actively working on them. When the LWG has reached consensus on the disposition of an issue, the status will then change to Dup, NAD, or Ready as appropriate. Once the full PL22.16 committee votes to forward Ready issues to the Project Editor, they are given the status of Defect Report (DR). These in turn may become the basis for Technical Corrigenda (TC1), an updated standard (C++11, C++14), or are closed without action other than a Record of Response (Resolved) where the desired effect has already been achieved by some other process. The intent of this LWG process is that only issues which are truly defects in the Standard move to the formal ISO DR status.

Active Issues


255. Why do basic_streambuf<>::pbump() and gbump() take an int?

Section: 27.6.3 [streambuf] Status: LEWG Submitter: Martin Sebor Opened: 2000-08-12 Last modified: 2016-02-12

Priority: Not Prioritized

View other active issues in [streambuf].

View all other issues in [streambuf].

View all issues with LEWG status.

Discussion:

The basic_streambuf members gbump() and pbump() are specified to take an int argument. This requirement prevents the functions from effectively manipulating buffers larger than std::numeric_limits<int>::max() characters. It also makes the common use case for these functions somewhat difficult as many compilers will issue a warning when an argument of type larger than int (such as ptrdiff_t on LLP64 architectures) is passed to either of the function. Since it's often the result of the subtraction of two pointers that is passed to the functions, a cast is necessary to silence such warnings. Finally, the usage of a native type in the functions signatures is inconsistent with other member functions (such as sgetn() and sputn()) that manipulate the underlying character buffer. Those functions take a streamsize argument.

[ 2009-07 Frankfurt ]

This is part of a bigger problem. If anyone cares enough, they should write a paper solving the bigger problem of offset types in iostreams.

This is related to the paper about large file sizes. Beman has already agreed to drop the section of that paper that deals with this.

int is big enough for reasonable buffers.

Move to NAD Future.

This is related to LWG 423.

Proposed resolution:

Change the signatures of these functions in the synopsis of template class basic_streambuf (27.5.2) and in their descriptions (27.5.2.3.1, p4 and 27.5.2.3.2, p4) to take a streamsize argument.

Although this change has the potential of changing the ABI of the library, the change will affect only platforms where int is different than the definition of streamsize. However, since both functions are typically inline (they are on all known implementations), even on such platforms the change will not affect any user code unless it explicitly relies on the existing type of the functions (e.g., by taking their address). Such a possibility is IMO quite remote.

Alternate Suggestion from Howard Hinnant, c++std-lib-7780:

This is something of a nit, but I'm wondering if streamoff wouldn't be a better choice than streamsize. The argument to pbump and gbump MUST be signed. But the standard has this to say about streamsize (27.4.1/2/Footnote):

[Footnote: streamsize is used in most places where ISO C would use size_t. Most of the uses of streamsize could use size_t, except for the strstreambuf constructors, which require negative values. It should probably be the signed type corresponding to size_t (which is what Posix.2 calls ssize_t). --- end footnote]

This seems a little weak for the argument to pbump and gbump. Should we ever really get rid of strstream, this footnote might go with it, along with the reason to make streamsize signed.

Rationale:

The LWG believes this change is too big for now. We may wish to reconsider this for a future revision of the standard. One possibility is overloading pbump, rather than changing the signature.

[ [2006-05-04: Reopened at the request of Chris (Krzysztof ?elechowski)] ]


423. effects of negative streamsize in iostreams

Section: 27 [input.output] Status: LEWG Submitter: Martin Sebor Opened: 2003-09-18 Last modified: 2016-02-12

Priority: Not Prioritized

View all other issues in [input.output].

View all issues with LEWG status.

Discussion:

A third party test suite tries to exercise istream::ignore(N) with a negative value of N and expects that the implementation will treat N as if it were 0. Our implementation asserts that (N >= 0) holds and aborts the test.

I can't find anything in section 27 that prohibits such values but I don't see what the effects of such calls should be, either (this applies to a number of unformatted input functions as well as some member functions of the basic_streambuf template).

[ 2009-07 Frankfurt ]

This is related to LWG 255.

Move to NAD Future.

Proposed resolution:

I propose that we add to each function in clause 27 that takes an argument, say N, of type streamsize a Requires clause saying that "N >= 0." The intent is to allow negative streamsize values in calls to precision() and width() but disallow it in calls to streambuf::sgetn(), istream::ignore(), or ostream::write().

[Kona: The LWG agreed that this is probably what we want. However, we need a review to find all places where functions in clause 27 take arguments of type streamsize that shouldn't be allowed to go negative. Martin will do that review.]


484. Convertible to T

Section: 24.2.3 [input.iterators] Status: LEWG Submitter: Chris Jefferson Opened: 2004-09-16 Last modified: 2016-02-12

Priority: Not Prioritized

View all other issues in [input.iterators].

View all issues with LEWG status.

Discussion:

From comp.std.c++:

I note that given an input iterator a for type T, then *a only has to be "convertable to T", not actually of type T.

Firstly, I can't seem to find an exact definition of "convertable to T". While I assume it is the obvious definition (an implicit conversion), I can't find an exact definition. Is there one?

Slightly more worryingly, there doesn't seem to be any restriction on the this type, other than it is "convertable to T". Consider two input iterators a and b. I would personally assume that most people would expect *a==*b would perform T(*a)==T(*b), however it doesn't seem that the standard requires that, and that whatever type *a is (call it U) could have == defined on it with totally different symantics and still be a valid inputer iterator.

Is this a correct reading? When using input iterators should I write T(*a) all over the place to be sure that the object I'm using is the class I expect?

This is especially a nuisance for operations that are defined to be "convertible to bool". (This is probably allowed so that implementations could return say an int and avoid an unnessary conversion. However all implementations I have seen simply return a bool anyway. Typical implemtations of STL algorithms just write things like while(a!=b && *a!=0). But strictly speaking, there are lots of types that are convertible to T but that also overload the appropriate operators so this doesn't behave as expected.

If we want to make code like this legal (which most people seem to expect), then we'll need to tighten up what we mean by "convertible to T".

[Lillehammer: The first part is NAD, since "convertible" is well-defined in core. The second part is basically about pathological overloads. It's a minor problem but a real one. So leave open for now, hope we solve it as part of iterator redesign.]

[ 2009-07-28 Reopened by Alisdair. No longer solved by concepts. ]

[ 2009-10 Santa Cruz: ]

Mark as NAD Future. We agree there's an issue, but there is no proposed solution at this time and this will be solved by concepts in the future.

Proposed resolution:

Rationale:

[ San Francisco: ]

Solved by N2758.


523. regex case-insensitive character ranges are unimplementable as specified

Section: 28 [re] Status: LEWG Submitter: Eric Niebler Opened: 2005-07-01 Last modified: 2016-02-12

Priority: Not Prioritized

View other active issues in [re].

View all other issues in [re].

View all issues with LEWG status.

Discussion:

A problem with TR1 regex is currently being discussed on the Boost developers list. It involves the handling of case-insensitive matching of character ranges such as [Z-a]. The proper behavior (according to the ECMAScript standard) is unimplementable given the current specification of the TR1 regex_traits<> class template. John Maddock, the author of the TR1 regex proposal, agrees there is a problem. The full discussion can be found at http://lists.boost.org/boost/2005/06/28850.php (first message copied below). We don't have any recommendations as yet.

-- Begin original message --

The situation of interest is described in the ECMAScript specification (ECMA-262), section 15.10.2.15:

"Even if the pattern ignores case, the case of the two ends of a range is significant in determining which characters belong to the range. Thus, for example, the pattern /[E-F]/i matches only the letters E, F, e, and f, while the pattern /[E-f]/i matches all upper and lower-case ASCII letters as well as the symbols [, \, ], ^, _, and `."

A more interesting case is what should happen when doing a case-insentitive match on a range such as [Z-a]. It should match z, Z, a, A and the symbols [, \, ], ^, _, and `. This is not what happens with Boost.Regex (it throws an exception from the regex constructor).

The tough pill to swallow is that, given the specification in TR1, I don't think there is any effective way to handle this situation. According to the spec, case-insensitivity is handled with regex_traits<>::translate_nocase(CharT) -- two characters are equivalent if they compare equal after both are sent through the translate_nocase function. But I don't see any way of using this translation function to make character ranges case-insensitive. Consider the difficulty of detecting whether "z" is in the range [Z-a]. Applying the transformation to "z" has no effect (it is essentially std::tolower). And we're not allowed to apply the transformation to the ends of the range, because as ECMA-262 says, "the case of the two ends of a range is significant."

So AFAICT, TR1 regex is just broken, as is Boost.Regex. One possible fix is to redefine translate_nocase to return a string_type containing all the characters that should compare equal to the specified character. But this function is hard to implement for Unicode, and it doesn't play nice with the existing ctype facet. What a mess!

-- End original message --

[ John Maddock adds: ]

One small correction, I have since found that ICU's regex package does implement this correctly, using a similar mechanism to the current TR1.Regex.

Given an expression [c1-c2] that is compiled as case insensitive it:

Enumerates every character in the range c1 to c2 and converts it to it's case folded equivalent. That case folded character is then used a key to a table of equivalence classes, and each member of the class is added to the list of possible matches supported by the character-class. This second step isn't possible with our current traits class design, but isn't necessary if the input text is also converted to a case-folded equivalent on the fly.

ICU applies similar brute force mechanisms to character classes such as [[:lower:]] and [[:word:]], however these are at least cached, so the impact is less noticeable in this case.

Quick and dirty performance comparisons show that expressions such as "[X-\\x{fff0}]+" are indeed very slow to compile with ICU (about 200 times slower than a "normal" expression). For an application that uses a lot of regexes this could have a noticeable performance impact. ICU also has an advantage in that it knows the range of valid characters codes: code points outside that range are assumed not to require enumeration, as they can not be part of any equivalence class. I presume that if we want the TR1.Regex to work with arbitrarily large character sets enumeration really does become impractical.

Finally note that Unicode has:

Three cases (upper, lower and title). One to many, and many to one case transformations. Character that have context sensitive case translations - for example an uppercase sigma has two different lowercase forms - the form chosen depends on context(is it end of a word or not), a caseless match for an upper case sigma should match either of the lower case forms, which is why case folding is often approximated by tolower(toupper(c)).

Probably we need some way to enumerate character equivalence classes, including digraphs (either as a result or an input), and some way to tell whether the next character pair is a valid digraph in the current locale.

Hoping this doesn't make this even more complex that it was already,

[ Portland: Alisdair: Detect as invalid, throw an exception. Pete: Possible general problem with case insensitive ranges. ]

[ 2009-07 Frankfurt ]

We agree that this is a problem, but we do not know the answer.

We are going to declare this NAD until existing practice leads us in some direction.

No objection to NAD Future.

Move to NAD Future.

Proposed resolution:


532. Tuple comparison

Section: 20.5.2.8 [tuple.rel], 99 [tr.tuple.rel] Status: LEWG Submitter: David Abrahams Opened: 2005-11-29 Last modified: 2016-02-12

Priority: Not Prioritized

View other active issues in [tuple.rel].

View all other issues in [tuple.rel].

View all issues with LEWG status.

Duplicate of: 348

Discussion:

Where possible, tuple comparison operators <,<=,=>, and > ought to be defined in terms of std::less rather than operator<, in order to support comparison of tuples of pointers.

[ 2009-07-28 Reopened by Alisdair. No longer solved by concepts. ]

[ 2009-10 Santa Cruz: ]

If we solve this for tuple we would have to solve it for pair algorithms, etc. It is too late to do that at this time. Move to NAD Future.

Proposed resolution:

change 6.1.3.5/5 from:

Returns: The result of a lexicographical comparison between t and u. The result is defined as: (bool)(get<0>(t) < get<0>(u)) || (!(bool)(get<0>(u) < get<0>(t)) && ttail < utail), where rtail for some tuple r is a tuple containing all but the first element of r. For any two zero-length tuples e and f, e < f returns false.

to:

Returns: The result of a lexicographical comparison between t and u. For any two zero-length tuples e and f, e < f returns false. Otherwise, the result is defined as: cmp( get<0>(t), get<0>(u)) || (!cmp(get<0>(u), get<0>(t)) && ttail < utail), where rtail for some tuple r is a tuple containing all but the first element of r, and cmp(x,y) is an unspecified function template defined as follows.

Where T is the type of x and U is the type of y:

if T and U are pointer types and T is convertible to U, returns less<U>()(x,y)

otherwise, if T and U are pointer types, returns less<T>()(x,y)

otherwise, returns (bool)(x < y)

[ Berlin: This issue is much bigger than just tuple (pair, containers, algorithms). Dietmar will survey and work up proposed wording. ]

Rationale:

Recommend NAD. This will be fixed with the next revision of concepts.

[ San Francisco: ]

Solved by N2770.


708. Locales need to be per thread and updated for POSIX changes

Section: 22 [localization] Status: LEWG Submitter: Peter Dimov Opened: 2007-07-28 Last modified: 2016-02-12

Priority: Not Prioritized

View all other issues in [localization].

View all issues with LEWG status.

Discussion:

The POSIX "Extended API Set Part 4,"

http://www.opengroup.org/sib/details.tpl?id=C065

introduces extensions to the C locale mechanism that allow multiple concurrent locales to be used in the same application by introducing a type locale_t that is very similar to std::locale, and a number of _l functions that make use of it.

The global locale (set by setlocale) is now specified to be per- process. If a thread does not call uselocale, the global locale is in effect for that thread. It can install a per-thread locale by using uselocale.

There is also a nice querylocale mechanism by which one can obtain the name (such as "de_DE") for a specific facet, even for combined locales, with no std::locale equivalent.

std::locale should be harmonized with the new POSIX locale_t mechanism and provide equivalents for uselocale and querylocale.

[ Kona (2007): Bill and Nick to provide wording. ]

[ San Francisco: Bill and Nick still intend to provide wording, but this is a part of the task to be addressed by the group that will look into issue 860. ]

[ 2009-07 Frankfurt: ]

It's our intention to stay in sync with WG14. If WG14 makes a decision that requires a change in WG21 the issue will be reopened.

Move to NAD Future.

Proposed resolution:


760. The emplace issue

Section: 23.2 [container.requirements] Status: Tentatively NAD Submitter: Paolo Carlini Opened: 2007-11-11 Last modified: 2016-02-12

Priority: 2

View other active issues in [container.requirements].

View all other issues in [container.requirements].

View all issues with Tentatively NAD status.

Discussion:

In an emplace member function the function parameter pack may be bound to a priori unlimited number of objects: some or all of them can be elements of the container itself. Apparently, in order to conform to the blanket statement 23.2 [container.requirements]/11, the implementation must check all of them for that possibility. A possible solution can involve extending the exception in 23.2 [container.requirements]/12 also to the emplace member. As a side note, the push_back and push_front member functions are luckily not affected by this problem, can be efficiently implemented anyway.

[ Related to 767 and to 2164 ]

[ Bellevue: ]

The proposed addition (13) is partially redundant with the existing paragraph 12. Why was the qualifier "rvalues" added to paragraph 12? Why does it not cover subelements and pointers?

Resolution: Alan Talbot to rework language, then set state to Review.

[ 2009-07 Frankfurt ]

The problem is broader than emplace. The LWG doesn't feel that it knows how to write wording that prohibits all of the problematic use cases at this time.

NAD Future.

[2015-02 Cologne]

LWG believes that 2164 addresses this issue and therefore considers 760 as NAD.

Proposed resolution:

Add after 23.2 [container.requirements]/12:

-12- Objects passed to member functions of a container as rvalue references shall not be elements of that container. No diagnostic required.

-13- Objects bound to the function parameter pack of the emplace member function shall not be elements or sub-objects of elements of the container. No diagnostic required.


839. Maps and sets missing splice operation

Section: 23.4 [associative], 23.5 [unord] Status: LEWG Submitter: Alan Talbot Opened: 2008-05-18 Last modified: 2016-02-12

Priority: Not Prioritized

View other active issues in [associative].

View all other issues in [associative].

View all issues with LEWG status.

Discussion:

Splice is a very useful feature of list. This functionality is also very useful for any other node based container, and I frequently wish it were available for maps and sets. It seems like an omission that these containers lack this capability. Although the complexity for a splice is the same as for an insert, the actual time can be much less since the objects need not be reallocated and copied. When the element objects are heavy and the compare operations are fast (say a map<int, huge_thingy>) this can be a big win.

Suggested resolution:

Add the following signatures to map, set, multimap, multiset, and the unordered associative containers:

 
void splice(list<T,Allocator>&& x);
void splice(list<T,Allocator>&& x, const_iterator i);
void splice(list<T,Allocator>&& x, const_iterator first, const_iterator last);

Hint versions of these are also useful to the extent hint is useful. (I'm looking for guidance about whether hints are in fact useful.)

 
void splice(const_iterator position, list<T,Allocator>&& x);
void splice(const_iterator position, list<T,Allocator>&& x, const_iterator i);
void splice(const_iterator position, list<T,Allocator>&& x, const_iterator first, const_iterator last);

[ Sophia Antipolis: ]

Don't try to splice "list" into the other containers, it should be container-type.

forward_list already has splice_after.

Would "splice" make sense for an unordered_map?

Jens, Robert: "splice" is not the right term, it implies maintaining ordering in lists.

Howard: adopt?

Jens: absorb?

Alan: subsume?

Robert: recycle?

Howard: transfer? (but no direction)

Jens: transfer_from. No.

Alisdair: Can we give a nothrow guarantee? If your compare() and hash() doesn't throw, yes.

Daniel: For unordered_map, we can't guarantee nothrow.

[ San Francisco: ]

Martin: this would possibly outlaw an implementation technique that is currently in use; caching nodes in containers.

Alan: if you cache in the allocator, rather than the individual container, this proposal doesn't interfere with that.

Martin: I'm not opposed to this, but I'd like to see an implementation that demonstrates that it works.

[ 2009-07 Frankfurt: ]

NAD Future.

[ 2009-09-19 Howard adds: ]

I'm not disagreeing with the NAD Future resolution. But when the future gets here, here is a possibility worth exploring:

Add to the "unique" associative containers:

typedef details      node_ptr;

node_ptr             remove(const_iterator p);
pair<iterator, bool> insert(node_ptr&& nd);
iterator             insert(const_iterator p, node_ptr&& nd);

And add to the "multi" associative containers:

typedef details node_ptr;

node_ptr remove(const_iterator p);
iterator insert(node_ptr&& nd);
iterator insert(const_iterator p, node_ptr&& nd);

Container::node_ptr is a smart pointer much like unique_ptr. It owns a node obtained from the container it was removed from. It maintains a reference to the allocator in the container so that it can properly deallocate the node if asked to, even if the allocator is stateful. This being said, the node_ptr can not outlive the container for this reason.

The node_ptr offers "const-free" access to the node's value_type.

With this interface, clients have a great deal of flexibility:

Here is how the customer might use this functionality:

The "node insertion" API maintains the API associated with inserting value_types so the customer can use familiar techniques for getting an iterator to the inserted node, or finding out whether it was inserted or not for the "unique" containers.

Lightly prototyped. No implementation problems. Appears to work great for the client.

Proposed resolution:


851. simplified array construction

Section: 23.3.7 [array] Status: LEWG Submitter: Benjamin Kosnik Opened: 2008-06-05 Last modified: 2016-02-12

Priority: Not Prioritized

View other active issues in [array].

View all other issues in [array].

View all issues with LEWG status.

Discussion:

This is an issue that came up on the libstdc++ list, where a discrepancy between "C" arrays and C++0x's std::array was pointed out.

In "C," this array usage is possible:

int ar[] = {1, 4, 6};

But for C++,

std::array<int> a = { 1, 4, 6 }; // error

Instead, the second parameter of the array template must be explicit, like so:

std::array<int, 3> a = { 1, 4, 6 };

Doug Gregor proposes the following solution, that assumes generalized initializer lists.

template<typename T, typename... Args>
inline array<T, sizeof...(Args)> 
make_array(Args&&... args) 
{ return { std::forward<Args>(args)... };  }

Then, the way to build an array from a list of unknown size is:

auto a = make_array<T>(1, 4, 6);

[ San Francisco: ]

Benjamin: Move to Ready?

Bjarne: I'm not convinced this is useful enough to add, so I'd like us to have time to reflect on it.

Alisdair: the constraints are wrong, they should be

template<ValueType T, ValueType... Args>
requires Convertible<Args, T>...
array<T, sizeof...(Args)> make_array(Args&&... args);

Alidair: this would be useful if we had a constexpr version.

Bjarne: this is probably useful for arrays with a small number of elements, but it's not clearly useful otherwise.

Consensus is to move to Open.

[ 2009-06-07 Daniel adds: ]

I suggest a fix and a simplification of the current proposal: Recent prototyping by Howard showed, that a fix is required because narrowing conversion 8.6.4 [dcl.init.list]/6 b.3 would severely limit the possible distribution of argument types, e.g. the expression make_array<double>(1, 2.0) is ill-formed, because the narrowing happens inside the function body where no constant expressions exist anymore. Furthermore given e.g.

int f();
double g();

we probably want to support

make_array<double>(f(), g());

as well. To make this feasible, the currently suggested expansion

{ std::forward<Args>(args)... }

needs to be replaced by

{ static_cast<T>(std::forward<Args>(args))... }

which is safe, because we already ensure convertibility via the element-wise Convertible<Args, T> requirement. Some other fixes are necessary: The ValueType requirement for the function parameters is invalid, because all lvalue arguments will deduce to an lvalue-reference, thereby no longer satisfying this requirement.

The suggested simplification is to provide a default-computed effective type for the result array based on common_type and decay, in unconstrained form:

template<typename... Args>
array<typename decay<typename common_type<Args...>::type>::type,
sizeof...(Args)>
make_array(Args&&... args);

The approach used below is similar to that of make_pair and make_tuple using a symbol C to represent the decayed common type [Note: Special handling of reference_wrapper types is intentionally not provided, because our target has so satisfy ValueType, thus under the revised proposal only an all-reference_wrapper-arguments would be well-formed and an array of reference_wrapper will be constructed]. I do currently not suggest to add new concepts reflecting decay and common_type, but an implementor will need something like this to succeed. Note that we use a similar fuzziness for make_pair and make_tuple currently. This fuzziness is not related to the currently missing Constructible<Vi, Ti&&> requirement for those functions. The following proposal fixes that miss for make_array. If the corresponding C type deduction is explicitly wanted for standardization, here the implementation

auto concept DC<typename... T> {
  typename type = typename decay<typename common_type<T...>::type>::type;
}

where C is identical to DC<Args...>::type in the proposed resolution below.

I intentionally added no further type relation between type and the concept template parameters, but instead added this requirement below to make the specification as transparent as possible. As written this concept is satisfied, if the corresponding associated type exists.

Suggested Resolution:

  1. Add to the array synopsis in 23.3 [sequences]:

    
    template<ReferentType... Args>
    requires ValueType<C> && IdentityOf<Args> && Constructible<C, Args&&>...
    array<C, sizeof...(Args)>
    make_array(Args&&... args);
    
    
  2. Append after 23.3.7.9 [array.tuple] Tuple interface to class template array the following new section:

    23.4.1.7 Array creation functions [array.creation]

    
    template<ReferentType... Args>
    requires ValueType<C> && IdentityOf<Args> && Constructible<C, Args&&>...
    array<C, sizeof...(Args)>
    make_array(Args&&... args);
    

    Let C be decay<common_type<Args...>::type>::type.

    Returns: an array<C, sizeof...(Args)> initialized with { static_cast<C>(std::forward<Args>(args))... }.

[ 2009-07 Frankfurt: ]

The proposed resolution uses concepts.

Daniel to rewrite the proposed resolution.

Leave Open.

[ 2009-07-25 Daniel provides rewritten proposed resolution. ]

[ 2009-10 Santa Cruz: ]

Argument for NAD future: everything about this could be added on. This does not require changes to the existing text.

[2015-11-29, Alisdair comments]

N4391 was adopted for Fundamentals 2 at the Lenexa meeting.

Proposed resolution:

  1. Add to the array synopsis in 23.3 [sequences]:

    template<class... Args>
      array<CT, sizeof...(Args)>
      make_array(Args&&... args);
    
  2. Append after 23.3.7.9 [array.tuple] "Tuple interface to class template array" the following new section:

    XX.X.X.X Array creation functions [array.creation]

    
    template<class... Args>
    array<CT, sizeof...(Args)>
    make_array(Args&&... args)
    

    Let CT be decay<common_type<Args...>::type>::type.

    Returns: An array<CT, sizeof...(Args)> initialized with { static_cast<CT>(std::forward<Args>(args))... }.

    [Example:

    
    int i = 0; int& ri = i;
    make_array(42u, i, 2.78, ri);
    

    returns an array of type

    
    array<double, 4>
    

    end example]


877. to throw() or to Throw: Nothing.

Section: 17 [library] Status: LEWG Submitter: Martin Sebor Opened: 2008-08-23 Last modified: 2016-02-12

Priority: Not Prioritized

View other active issues in [library].

View all other issues in [library].

View all issues with LEWG status.

Discussion:

Recent changes to the working draft have introduced a gratuitous inconsistency with the C++ 2003 version of the specification with respect to exception guarantees provided by standard functions. While the C++ 2003 standard consistenly uses the empty exception specification, throw(), to declare functions that are guaranteed not to throw exceptions, the current working draft contains a number of "Throws: Nothing." clause to specify essentially the same requirement. The difference between the two approaches is that the former specifies the behavior of programs that violate the requirement (std::unexpected() is called) while the latter leaves the behavior undefined.

A survey of the working draft reveals that there are a total of 209 occurrences of throw() in the library portion of the spec, the majority in clause 18, a couple (literally) in 19, a handful in 20, a bunch in 22, four in 24, one in 27, and about a dozen in D.9.

There are also 203 occurrences of "Throws: Nothing." scattered throughout the spec.

While sometimes there are good reasons to use the "Throws: Nothing." approach rather than making use of throw(), these reasons do not apply in most of the cases where this new clause has been introduced and the empty exception specification would be a better approach.

First, functions declared with the empty exception specification permit compilers to generate better code for calls to such functions. In some cases, the compiler might even be able to eliminate whole chunks of user-written code when instantiating a generic template on a type whose operations invoked from the template specialization are known not to throw. The prototypical example are the std::uninitialized_copy() and std::uninitialized_fill() algorithms where the entire catch(...) block can be optimized away.

For example, given the following definition of the std::uninitialized_copy function template and a user-defined type SomeType:

template <class InputIterator, class ForwardIterator>
ForwardIterator
uninitialized_copy (InputIterator first, InputIterator last, ForwardIterator res)
{
   typedef iterator_traits<ForwardIterator>::value_type ValueType;

   ForwardIterator start = res;

   try {
       for (; first != last; ++first, ++res)
           ::new (&*res) ValueType (*first);
   }
   catch (...) {
       for (; start != res; --start)
           (&*start)->~ValueType ();
       throw;
   }
   return res;
}

struct SomeType {
   SomeType (const SomeType&) throw ();
}

compilers are able to emit the following efficient specialization of std::uninitialized_copy<const SomeType*, SomeType*> (note that the catch block has been optimized away):

template <> SomeType*
uninitialized_copy (const SomeType *first, const SomeType *last, SomeType *res)
{
   for (; first != last; ++first, ++res)
       ::new (res) SomeType (*first);

   return res;
}

Another general example is default constructors which, when decorated with throw(), allow the compiler to eliminate the implicit try and catch blocks that it otherwise must emit around each the invocation of the constructor in new-expressions.

For example, given the following definitions of class MayThrow and WontThrow and the two statements below:

struct MayThrow {
   MayThrow ();
};

struct WontThrow {
   WontThrow () throw ();
};

MayThrow  *a = new MayThrow [N];
WontThrow *b = new WontThrow [N];

the compiler generates the following code for the first statement:

MayThrow *a;
{
   MayThrow *first = operator new[] (N * sizeof (*a));
   MayThrow *last  = first + N;
   MayThrow *next  = first;
   try {
       for ( ; next != last; ++next)
           new (next) MayThrow;
   }
   catch (...) {
       for ( ; first != first; --next)
           next->~MayThrow ();
       operator delete[] (first);
       throw;
   }
   a = first;
}

but it is can generate much more compact code for the second statement:

WontThrow *b    = operator new[] (N * sizeof (*b));
WontThrow *last = b + N;
for (WontThrow *next = b; next != last; ++next)
   new (next) WontThrow;

Second, in order for users to get the maximum benefit out of the new std::has_nothrow_xxx traits when using standard library types it will be important for implementations to decorate all non throwing copy constructors and assignment operators with throw(). Note that while an optimizer may be able to tell whether a function without an explicit exception specification can throw or not based on its definition, it can only do so when it can see the source code of the definition. When it can't it must assume that the function may throw. To prevent violating the One Definition Rule, the std::has_nothrow_xxx trait must return the most pessimistic guess across all translation units in the program, meaning that std::has_nothrow_xxx<T>::value must evaluate to false for any T whose xxx (where xxx is default or copy ctor, or assignment operator) is defined out-of-line.

Counterarguments:

During the discussion of this issue on c++std-lib@accu.org (starting with post c++std-lib-21950) the following arguments in favor of the "Throws: Nothing." style have been made.

  1. Decorating functions that cannot throw with the empty exception specification can cause the compiler to generate suboptimal code for the implementation of the function when it calls other functions that aren't known to the compiler not to throw (i.e., that aren't decorated with throw() even if they don't actually throw). This is a common situation when the called function is a C or POSIX function.
  2. Alternate, proprietary mechanisms exist (such as GCC __attribute__((nothrow)) or Visual C++ __declspec(nothrow)) that let implementers mark up non-throwing functions, often without the penalty mentioned in (1) above. The C++ standard shouldn't preclude the use of these potentially more efficient mechanisms.
  3. There are functions, especially function templates, that invoke user-defined functions that may or may not be declared throw(). Declaring such functions with the empty exception specification will cause compilers to generate suboptimal code when the user-defined function isn't also declared not to throw.

The answer to point (1) above is that implementers can (and some have) declare functions with throw() to indicate to the compiler that calls to the function can safely be assumed not to throw in order to allow it to generate efficient code at the call site without also having to define the functions the same way and causing the compiler to generate suboptimal code for the function definition. That is, the function is declared with throw() in a header but it's defined without it in the source file. The throw() declaration is suppressed when compiling the definition to avoid compiler errors. This technique, while strictly speaking no permitted by the language, is safe and has been employed in practice. For example, the GNU C library takes this approach. Microsoft Visual C++ takes a similar approach by simply assuming that no function with C language linkage can throw an exception unless it's explicitly declared to do so using the language extension throw(...).

Our answer to point (2) above is that there is no existing practice where C++ Standard Library implementers have opted to make use of the proprietary mechanisms to declare functions that don't throw. The language provides a mechanism specifically designed for this purpose. Avoiding its use in the specification itself in favor of proprietary mechanisms defeats the purpose of the feature. In addition, making use of the empty exception specification inconsistently, in some areas of the standard, while conspicuously avoiding it and making use of the "Throws: Nothing." form in others is confusing to users.

The answer to point (3) is simply to exercise caution when declaring functions and especially function templates with the empty exception specification. Functions that required not to throw but that may call back into user code are poor candidates for the empty exception specification and should instead be specified using "Throws: Nothing." clause.

[ 2009-07 Frankfurt ]

We need someone to do an extensive review.

NAD Future.

Proposed resolution:

We propose two possible solutions. Our recommendation is to adopt Option 1 below.

Option 1:

Except for functions or function templates that make calls back to user-defined functions that may not be declared throw() replace all occurrences of the "Throws: Nothing." clause with the empty exception specification. Functions that are required not to throw but that make calls back to user code should be specified to "Throw: Nothing."

Option 2:

For consistency, replace all occurrences of the empty exception specification with a "Throws: Nothing." clause.


933. Unique_ptr defect

Section: 20.11.1.2.5 [unique.ptr.single.modifiers] Status: LEWG Submitter: Alisdair Meredith Opened: 2008-11-27 Last modified: 2016-02-12

Priority: Not Prioritized

View all other issues in [unique.ptr.single.modifiers].

View all issues with LEWG status.

Discussion:

If we are supporting stateful deleters, we need an overload for reset that takes a deleter as well.

void reset( pointer p, deleter_type d);

We probably need two overloads to support move-only deleters, and this sounds uncomfortably like the two constructors I have been ignoring for now...

[ Batavia (2009-05): ]

Howard comments that we have the functionality via move-assigment.

Move to Open.

[ 2009-10 Santa Cruz: ]

Mark as NAD Future.

Proposed resolution:


935. clock error handling needs to be specified

Section: 20.17.7 [time.clock] Status: LEWG Submitter: Beman Dawes Opened: 2008-11-24 Last modified: 2016-02-12

Priority: Not Prioritized

View all issues with LEWG status.

Discussion:

Each of the three clocks specified in Clocks 20.17.7 [time.clock] provides the member function:

static time_point now();

The semantics specified by Clock requirements 20.17.3 [time.clock.req] make no mention of error handling. Thus the function may throw bad_alloc or an implementation-defined exception (17.6.5.12 [res.on.exception.handling] paragraph 4).

Some implementations of these functions on POSIX, Windows, and presumably on other operating systems, may fail in ways only detectable at runtime. Some failures on Windows are due to supporting chipset errata and can even occur after successful calls to a clock's now() function.

These functions are used in cases where exceptions are not appropriate or where the specifics of the exception or cause of error need to be available to the user. See N2828, Library Support for hybrid error handling (Rev 1), for more specific discussion of use cases. Thus some change in the interface of now is required.

The proposed resolution has been implemented in the Boost version of the chrono library. No problems were encountered.

[ Batavia (2009-05): ]

We recommend this issue be deferred until the next Committee Draft has been issued and the prerequisite paper has been accepted.

Move to Open.

[ 2009-10 Santa Cruz: ]

Mark as NAD future. Too late to make this change without having already accepted the hybrid error handling proposal.

Proposed resolution:

Accept the proposed wording of N2828, Library Support for hybrid error handling (Rev 1).

Change Clock requirements 20.17.3 [time.clock.req] as indicated:

-2- In Table 55 C1 and C2 denote clock types. t1 and t2 are values returned by C1::now() where the call returning t1 happens before (1.10) the call returning t2 and both of these calls happen before C1::time_point::max(). ec denotes an object of type error_code (19.5.3.1 [syserr.errcode.overview]).

Table 55 — Clock requirements
ExpressionReturn typeOperational semantics
... ... ...
C1::now() C1::time_point Returns a time_point object representing the current point in time.
C1::now(ec) C1::time_point Returns a time_point object representing the current point in time.

Change class system_clock 20.17.7.1 [time.clock.system] as indicated:

static time_point now(error_code& ec=throws());

Change class monotonic_clock 99 [time.clock.monotonic] as indicated:

static time_point now(error_code& ec=throws());

Change class high_resolution_clock 20.17.7.3 [time.clock.hires] as indicated:

static time_point now(error_code& ec=throws());

936. Mutex type overspecified

Section: 30.4.1 [thread.mutex.requirements] Status: LEWG Submitter: Pete Becker Opened: 2008-12-05 Last modified: 2016-02-12

Priority: Not Prioritized

View other active issues in [thread.mutex.requirements].

View all other issues in [thread.mutex.requirements].

View all issues with LEWG status.

Duplicate of: 961

Discussion:

30.4.1 [thread.mutex.requirements] describes the requirements for a type to be a "Mutex type". A Mutex type can be used as the template argument for the Lock type that's passed to condition_variable_any::wait (although Lock seems like the wrong name here, since Lock is given a different formal meaning in 30.4.2 [thread.lock]) and, although the WD doesn't quite say so, as the template argument for lock_guard and unique_lock.

The requirements for a Mutex type include:

Also, a Mutex type "shall not be copyable nor movable".

The latter requirement seems completely irrelevant, and the three requirements on return types are tighter than they need to be. For example, there's no reason that lock_guard can't be instantiated with a type that's copyable. The rule is, in fact, that lock_guard, etc. won't try to copy objects of that type. That's a constraint on locks, not on mutexes. Similarly, the requirements for void return types are unnecessary; the rule is, in fact, that lock_guard, etc. won't use any returned value. And with the return type of bool, the requirement should be that the return type is convertible to bool.

[ Summit: ]

Move to open. Related to conceptualization and should probably be tackled as part of that.

[ Post Summit Anthony adds: ]

Section 30.4.1 [thread.mutex.requirements] conflates the requirements on a generic Mutex type (including user-supplied mutexes) with the requirements placed on the standard-supplied mutex types in an attempt to group everything together and save space.

When applying concepts to chapter 30, I suggest that the concepts Lockable and TimedLockable embody the requirements for *use* of a mutex type as required by unique_lock/lock_guard/condition_variable_any. These should be relaxed as Pete describes in the issue. The existing words in 30.4.1 [thread.mutex.requirements] are requirements on all of std::mutex, std::timed_mutex, std::recursive_mutex and std::recursive_timed_mutex, and should be rephrased as such.

Proposed resolution:


961. Various threading bugs #11

Section: 30.4.1 [thread.mutex.requirements] Status: LEWG Submitter: Pete Becker Opened: 2009-01-07 Last modified: 2016-02-12

Priority: Not Prioritized

View other active issues in [thread.mutex.requirements].

View all other issues in [thread.mutex.requirements].

View all issues with LEWG status.

Duplicate of: 936

Discussion:

30.4.1 [thread.mutex.requirements] describes required member functions of mutex types, and requires that they throw exceptions under certain circumstances. This is overspecified. User-defined types can abort on such errors without affecting the operation of templates supplied by standard-library.

[ Summit: ]

Move to open. Related to conceptualization and should probably be tackled as part of that.

[ 2009-10 Santa Cruz: ]

Would be OK to leave it as is for time constraints, could loosen later.

Mark as NAD Future.

Proposed resolution:


1025. The library should provide more specializations for std::hash

Section: 20.14.14 [unord.hash] Status: LEWG Submitter: Alisdair Meredith Opened: 2009-03-11 Last modified: 2016-02-12

Priority: Not Prioritized

View other active issues in [unord.hash].

View all other issues in [unord.hash].

View all issues with LEWG status.

Discussion:

Addresses UK 208 [CD1]

std::hash should be implemented for much more of the standard library. In particular for pair, tuple and all the standard containers.

Proposed resolution:


1031. Need shared_ptr conversion to a unique_ptr

Section: 20.11.2.2 [util.smartptr.shared] Status: LEWG Submitter: Alisdair Meredith Opened: 2009-03-11 Last modified: 2016-02-12

Priority: Not Prioritized

View other active issues in [util.smartptr.shared].

View all other issues in [util.smartptr.shared].

View all issues with LEWG status.

Discussion:

Addresses US 78 [CD1]

There is presently no way to convert directly from a shared_ptr to a unique_ptr. Add an interface that performs the conversion.

[ Summit: ]

We look forward to a paper on this topic. We recommend no action until a paper is available. We believe that the shared pointer must use the default deleter for the conversion to succeed.

[ Peter Dimov adds: ]

This is basically a request for shared_ptr<>::release in disguise, with all the associated problems. Not a good idea.

[ 2009-07 post-Frankfurt: ]

The rationale for the omission of a release() member function from shared_ptr is given in: http://www.boost.org/doc/libs/1_39_0/libs/smart_ptr/shared_ptr.htm

The implementation of such a member is non-trivial (and maybe impossible), because it would need to account for the deleter.

[ 2009-07-26 Howard sets to Tentatively NAD Future. ]

I took an online poll and got 3 votes for NAD and 3 for NAD Future. Personally I prefer NAD Future as this does refer to an extension that could conceivably be considered beyond C++0X.

However such an extension would need to solve a couple of problems:

  1. What is the interface for such a conversion when the shared_ptr does not have unique ownership? Throw an exception? Create a null unique_ptr? Undefined behavior?
  2. How does one handle custom deleters given to the shared_ptr constructor?

    I do not believe it is possible to implement a general answer to this question. The shared_ptr deleter is a run time (or construction time) characteristic. The unique_ptr deleter is a compile time characteristic. In general one can not know to what type of unqiue_ptr you are converting to.

    One answer is for the user of the conversion to specify the deleter type and perhaps throw an exception if the specification turns out to be incorrect.

    Another answer is for the conversion to only be valid when the underlying deleter is default_delete. We would probalby need to specify that this is indeed the underlying deleter of a shared_ptr when a custom deleter is not given in the constructor.

At any rate, there are non-trivial design issues which would need to be implemented and tested in the field for usability prior to standardization.

[ 2009 Santa Cruz: ]

Moved to NAD Future.

Proposed resolution:


1041. Add associative/unordered container functions that allow to extract elements

Section: 23.2.4 [associative.reqmts] Status: LEWG Submitter: Alisdair Meredith Opened: 2009-03-12 Last modified: 2016-02-12

Priority: Not Prioritized

View other active issues in [associative.reqmts].

View all other issues in [associative.reqmts].

View all issues with LEWG status.

Discussion:

Addresses UK 239 [CD1]

It is not possible to take a move-only key out of an unordered container, such as (multi)set or (multi)map, or the new unordered containers.

Add below a.erase(q), a.extract(q), with the following notation:

a.extract(q)>, Return type pair<key, iterator> Extracts the element pointed to by q and erases it from the set. Returns a pair containing the value pointed to by q and an iterator pointing to the element immediately following q prior to the element being erased. If no such element exists,returns a.end().

[ Summit: ]

We look forward to a paper on this topic. We recommend no action until a paper is available. The paper would need to address exception safety.

[ Post Summit Alisdair adds: ]

Would value_type be a better return type than key_type?

[ 2009-07 post-Frankfurt: ]

Leave Open. Alisdair to contact Chris Jefferson about this.

[ 2009-09-20 Howard adds: ]

See the 2009-09-19 comment of 839 for an API which accomplishes this functionality and also addresses several other use cases which this proposal does not.

[ 2009-10 Santa Cruz: ]

Mark as NAD Future. No consensus to make the change at this time.

Proposed resolution:

In 23.2.4 [associative.reqmts] Table 85, add:

Table 85 -- Associative container requirements (in addition to container)
Expression Return type Assertion/note
pre-/post-condition
Complexity
a.erase(q) ... ... ...
a.extract(q) pair<key_type, iterator> Extracts the element pointed to by q and erases it from the set. Returns a pair containing the value pointed to by q and an iterator pointing to the element immediately following q prior to the element being erased. If no such element exists, returns a.end(). amortized constant

In 23.2.5 [unord.req] Table 87, add:

Table 87 -- Unordered associative container requirements (in addition to container)
Expression Return type Assertion/note
pre-/post-condition
Complexity
a.erase(q) ... ... ...
a.extract(q) pair<key_type, iterator> Extracts the element pointed to by q and erases it from the set. Returns a pair containing the value pointed to by q and an iterator pointing to the element immediately following q prior to the element being erased. If no such element exists, returns a.end(). amortized constant

1052. reverse_iterator::operator-> should also support smart pointers

Section: 24.5.1.3.5 [reverse.iter.opref] Status: LEWG Submitter: Alisdair Meredith Opened: 2009-03-12 Last modified: 2016-02-12

Priority: Not Prioritized

View all other issues in [reverse.iter.opref].

View all issues with LEWG status.

Discussion:

Addresses UK 281 [CD1]

The current specification for return value for reverse_iterator::operator-> will always be a true pointer type, but reverse_iterator supports proxy iterators where the pointer type may be some kind of 'smart pointer'.

[ Summit: ]

move_iterator avoids this problem by returning a value of the wrapped Iterator type. study group formed to come up with a suggested resolution.

move_iterator solution shown in proposed wording.

[ 2009-07 post-Frankfurt: ]

Howard to deconceptize. Move to Review after that happens.

[ 2009-08-01 Howard deconceptized: ]

[ 2009-10 Santa Cruz: ]

We can't think of any reason we can't just define reverse iterator's pointer types to be the same as the underlying iterator's pointer type, and get it by calling the right arrow directly.

Here is the proposed wording that was replaced:

template <class Iterator> 
class reverse_iterator { 
  ...
  typedef typename iterator_traits<Iterator>::pointer pointer;

Change 24.5.1.3.5 [reverse.iter.opref]:

pointer operator->() const;

Returns:

&(operator*());
this->tmp = current;
--this->tmp;
return this->tmp;

[ 2010-03-03 Daniel opens: ]

  1. There is a minor problem with the exposition-only declaration of the private member deref_tmp which is modified in a const member function (and the same problem occurs in the specification of operator*). The fix is to make it a mutable member.
  2. The more severe problem is that the resolution for some reasons does not explain in the rationale why it was decided to differ from the suggested fix (using deref_tmp instead of tmp) in the [ 2009-10 Santa Cruz] comment:

    this->deref_tmp = current;
    --this->deref_tmp;
    return this->deref_tmp;
    

    combined with the change of

    typedef typename iterator_traits<Iterator>::pointer pointer;
    

    to

    typedef Iterator pointer;
    

    The problem of the agreed on wording is that the following rather typical example, that compiled with the wording before 1052 had been applied, won't compile anymore:

    #include <iterator>
    #include <utility>
    
    int main() {
      typedef std::pair<int, double> P;
      P op;
      std::reverse_iterator<P*> ri(&op + 1);
      ri->first; // Error
    }
    

    Comeau online returns (if a correspondingly changed reverse_iterator is used):

    "error: expression must have class type
         return deref_tmp.operator->();
                ^
             detected during instantiation of "Iterator
                       reverse_iterator<Iterator>::operator->() const [with
                       Iterator=std::pair<int, double> *]""
    

    Thus the change will break valid, existing code based on std::reverse_iterator.

IMO the suggestion proposed in the comment is a necessary fix, which harmonizes with the similar specification of std::move_iterator and properly reflects the recursive nature of the evaluation of operator-> overloads.

Suggested resolution:

  1. In the class template reverse_iterator synopsis of 24.5.1.1 [reverse.iterator] change as indicated:

    namespace std {
    template <class Iterator>
    class reverse_iterator : public
                 iterator<typename iterator_traits<Iterator>::iterator_category,
                 typename iterator_traits<Iterator>::value_type,
                 typename iterator_traits<Iterator>::difference_type,
                 typename iterator_traits<Iterator>::pointer,
                 typename iterator_traits<Iterator>::reference> {
    public:
      [..]
      typedef typename iterator_traits<Iterator>::pointer pointer;
      [..]
    protected:
      Iterator current;
    private:
      mutable Iterator deref_tmp; // exposition only
    };
    
  2. Change 24.5.1.3.5 [reverse.iter.opref]/1 as indicated:
    pointer operator->() const;
    

    1 Returns Effects: &(operator*()).

    deref_tmp = current;
    --deref_tmp;
    return deref_tmp;
    

[ 2010 Pittsburgh: ]

We prefer to make to use a local variable instead of deref_tmp within operator->(). And although this means that the mutable change is no longer needed, we prefer to keep it because it is needed for operator*() anyway.

Here is the proposed wording that was replaced:

Change 24.5.1.3.5 [reverse.iter.opref]:

pointer operator->() const;

Returns:

&(operator*());
deref_tmp = current;
--deref_tmp;
return deref_tmp::operator->();

[ 2010-03-10 Howard adds: ]

Here are three tests that the current proposed wording passes, and no other solution I've seen passes all three:

  1. Proxy pointer support:

    #include <iterator>
    #include <cassert>
    
    struct X { int m; };
    
    X x;
    
    struct IterX {
        typedef std::bidirectional_iterator_tag iterator_category;
        typedef X& reference;
        struct pointer
        {
            pointer(X& v) : value(v) {}
            X& value;
            X* operator->() const {return &value;}
        };
        typedef std::ptrdiff_t difference_type;
        typedef X value_type;
        // additional iterator requirements not important for this issue
        
        reference operator*() const { return x; }
        pointer operator->() const { return pointer(x); }
        IterX& operator--() {return *this;}
    
    };
    
    int main()
    {
        std::reverse_iterator<IterX> ix;
        assert(&ix->m == &(*ix).m);
    }
    
  2. Raw pointer support:

    #include <iterator>
    #include <utility>
    
    int main() {
      typedef std::pair<int, double> P;
      P op;
      std::reverse_iterator<P*> ri(&op + 1);
      ri->first; // Error
    }
    
  3. Caching iterator support:

    #include <iterator>
    #include <cassert>
    
    struct X { int m; };
    
    struct IterX {
        typedef std::bidirectional_iterator_tag iterator_category;
        typedef X& reference;
        typedef X* pointer;
        typedef std::ptrdiff_t difference_type;
        typedef X value_type;
        // additional iterator requirements not important for this issue
        
        reference operator*() const { return value; }
        pointer operator->() const { return &value; }
        IterX& operator--() {return *this;}
    
    private:
        mutable X value;
    };
    
    int main()
    {
        std::reverse_iterator<IterX> ix;
        assert(&ix->m == &(*ix).m);
    }
    

[ 2010 Pittsburgh: ]

Moved to NAD Future, rationale added.

Rationale:

The LWG did not reach a consensus for a change to the WP.

Proposed resolution:

  1. In the class template reverse_iterator synopsis of 24.5.1.1 [reverse.iterator] change as indicated:

    namespace std {
    template <class Iterator>
    class reverse_iterator : public
                 iterator<typename iterator_traits<Iterator>::iterator_category,
                 typename iterator_traits<Iterator>::value_type,
                 typename iterator_traits<Iterator>::difference_type,
                 typename iterator_traits<Iterator&>::pointer,
                 typename iterator_traits<Iterator>::reference> {
    public:
      [..]
      typedef typename iterator_traits<Iterator&>::pointer pointer;
      [..]
    protected:
      Iterator current;
    private:
      mutable Iterator deref_tmp; // exposition only
    };
    
  2. Change 24.5.1.3.5 [reverse.iter.opref]/1 as indicated:
    pointer operator->() const;
    

    1 Returns Effects: &(operator*()).

    deref_tmp = current;
    --deref_tmp;
    return deref_tmp;
    

1053. Unify algorithms with operator and function object variants

Section: 25 [algorithms] Status: LEWG Submitter: Alisdair Meredith Opened: 2009-03-12 Last modified: 2016-02-12

Priority: Not Prioritized

View other active issues in [algorithms].

View all other issues in [algorithms].

View all issues with LEWG status.

Discussion:

Addresses UK 295 [CD1]

There is a level of redundancy in the library specification for many algorithms that can be eliminated with the combination of concepts and default parameters for function templates. Eliminating redundancy simplified specification and reduces the risk of introducing accidental inconsistencies.

Proposed resolution: Adopt N2743.

[ Summit: ]

NAD, this change would break code that takes the address of an algorithm.

[ Post Summit Alisdair adds: ]

Request 'Open'. The issues in the paper go beyond just reducing the number of signatures, but cover unifying the idea of the ordering operation used by algorithms, containers and other library components. At least, it takes a first pass at the problem.

For me (personally) that was the more important part of the paper, and not clearly addressed by the Summit resolution.

[ 2009-10 Santa Cruz: ]

Too inventive, too late, would really need a paper. Moved to NAD Future.

Proposed resolution:


1112. bitsets and new style for loop

Section: 20.9 [template.bitset] Status: LEWG Submitter: Alisdair Meredith Opened: 2009-05-06 Last modified: 2016-02-12

Priority: Not Prioritized

View other active issues in [template.bitset].

View all other issues in [template.bitset].

View all issues with LEWG status.

Discussion:

std::bitset is a homogeneous container-like sequence of bits, yet it does not model the Range concept so cannot be used with the new for-loop syntax. It is the only such type in the library that does NOT support the new for loop.

The obvious reason is that bitset does not support iterators.

At least two reasonable solutions are available:

  1. Add an iterator interface to bitset, bringing its interface close to that of std::array
  2. Provide an unspecified concept_map for Range<bitset>.

The latter will still need some kind of iterator-like adapter for bitset, but gives implementers greater freedom on the details. E.g. begin/end return some type that simply invokes operator[] on the object it wraps, and increments its index on operator++. A vendor can settle for InputIterator support, rather than wrapping up a full RandomAccessIterator.

I have a mild preference for option (ii) as I think it is less work to specify at this stage of the process, although (i) is probably more useful in the long run.

Hmm, my wording looks a little woolly, as it does not say what the element type of the range is. Do I get a range of bool, bitset<N>::reference, or something else entirely?

I guess most users will assume the behaviour of reference, but expect to work with bool. Bool is OK for read-only traversal, but you really need to take a reference to a bitset::reference if you want to write back.

[ Batavia (2009-05): ]

Move to Open. We further recommend this be deferred until after the next Committee Draft.

[ 2009-05-25 Alisdair adds: ]

I just stumbled over the Range concept_map for valarray and this should probably set the precedent on how to write the wording.

[ Howard: I've replaced the proposed wording with Alisdair's suggestion. ]

[ 2009-07-24 Daniel modifies the proposed wording for non-concepts. ]

[ 2009-10 post-Santa Cruz: ]

Mark as Tentatively NAD Future due to the loss of concepts.

Rationale:

All concepts-related text has been removed from the draft.

Proposed resolution:

  1. Modify the section 20.9 [template.bitset] <bitset> synopsis by adding the following at the end of the synopsis:

    
    // XX.X.X bitset range access [bitset.range]
    template<size_t N> unspecified-1 begin(bitset<N>&);
    template<size_t N> unspecified-2 begin(const bitset<N>&);
    template<size_t N> unspecified-1 end(bitset<N>&);
    template<size_t N> unspecified-2 end(const bitset<N>&);
    
    
  2. Add a new section "bitset range access" [bitset.range] after the current section 20.9.4 [bitset.operators] with the following series of paragraphs:

    1. In the begin and end function templates that follow, unspecified-1 is a type that meets the requirements of a mutable random access iterator (24.2.7 [random.access.iterators]) whose value_type is bool and whose reference type is bitset<N>::reference. unspecified-2 is a type that meets the requirements of a constant random access iterator (24.2.7 [random.access.iterators]) whose value_type is bool and whose reference type is bool.

    
    template<size_t N> unspecified-1 begin(bitset<N>&);
    template<size_t N> unspecified-2 begin(const bitset<N>&);
    
    

    2. Returns: an iterator referencing the first bit in the bitset.

    
    template<size_t N> unspecified-1 end(bitset<N>&);
    template<size_t N> unspecified-2 end(const bitset<N>&);
    

    3. Returns: an iterator referencing one past the last bit in the bitset.


1120. New type trait - remove_all

Section: 20.15 [meta] Status: LEWG Submitter: Alisdair Meredith Opened: 2009-05-23 Last modified: 2016-02-12

Priority: Not Prioritized

View other active issues in [meta].

View all other issues in [meta].

View all issues with LEWG status.

Discussion:

Sometimes it is necessary to remove all qualifiers from a type before passing on to a further API. A good example would be calling the tuple query APIs tuple_size or tuple_element with a deduced type inside a function template. If the deduced type is cv-qualified or a reference then the call will fail. The solution is to chain calls to remove_cv<remove_reference<T>::type>::type, and note that the order matters.

Suggest it would be helpful to add a new type trait, remove_all, that removes all top-level qualifiers from a type i.e. cv-qualification and any references. Define the term in such a way that if additional qualifiers are added to the language, then remove_all is defined as stripping those as well.

[ 2009-10-14 Daniel adds: ]

remove_all seems too generic, a possible alternative matching the current naming style could be remove_cv_reference or remove_reference_cv. It should also be considered whether this trait should also remove 'extents', or pointer 'decorations'. Especially if the latter situations are considered as well, it might be easier to chose the name not in terms of what it removes (which might be a lot), but in terms of it creates. In this case I could think of e.g. extract_value_type.

[ 2009-10 Santa Cruz: ]

NAD Future.

Proposed resolution:


1121. Support for multiple arguments

Section: 20.16.4 [ratio.arithmetic] Status: LEWG Submitter: Alisdair Meredith Opened: 2009-05-25 Last modified: 2016-02-12

Priority: Not Prioritized

View all other issues in [ratio.arithmetic].

View all issues with LEWG status.

Discussion:

Both add and multiply could sensibly be called with more than two arguments. The variadic template facility makes such declarations simple, and is likely to be frequently wrapped by end users if we do not supply the variant ourselves.

We deliberately ignore divide at this point as it is not transitive. Likewise, subtract places special meaning on the first argument so I do not suggest extending that immediately. Both could be supported with analogous wording to that for add/multiply below.

Note that the proposed resolution is potentially incompatible with that proposed for 921, although the addition of the typedef to ratio would be equally useful.

[ 2009-10-30 Alisdair adds: ]

The consensus of the group when we reviewed this in Santa Cruz was that 921 would proceed to Ready as planned, and the multi-paramater add/multiply templates should be renamed as ratio_sum and ratio_product to avoid the problem mixing template aliases with partial specializations.

It was also suggested to close this issue as NAD Future as it does not correspond directly to any NB comment. NBs are free to submit a specific comment (and re-open) in CD2 though.

Walter Brown also had concerns on better directing the order of evaluation to avoid overflows if we do proceed for 0x rather than TR1, so wording may not be complete yet.

[ Alisdair updates wording. ]

[ 2009-10-30 Howard: ]

Moved to Tentatively NAD Future after 5 positive votes on c++std-lib.

Rationale:

Does not have sufficient support at this time. May wish to reconsider for a future standard.

Proposed resolution:

Add the following type traits to p3 20.16 [ratio]

// ratio arithmetic
template <class R1, class R2> struct ratio_add;
template <class R1, class R2> struct ratio_subtract;
template <class R1, class R2> struct ratio_multiply;
template <class R1, class R2> struct ratio_divide;
template <class R1, class ... RList> struct ratio_sum;
template <class R1, class ... RList> struct ratio_product;

after 20.16.4 [ratio.arithmetic] p1: add

template <class R1, class ... RList> struct ratio_sum; // declared, never defined

template <class R1> struct ratio_sum<R1> : R1 {};

Requires: R1 is a specialization of class template ratio

template <class R1, class R2, class ... RList> 
 struct ratio_sum<R1, R2, RList...>
   : ratio_add< R1, ratio_sum<R2, RList...>> {
};

Requires: R1 and each element in parmater pack RList is a specialization of class template ratio

after 20.16.4 [ratio.arithmetic] p3: add

template <class R1, class ... RList> struct ratio_product; // declared, never defined

template <class R1> struct ratio_product<R1> : R1 {};

Requires: R1 is a specialization of class template ratio

template <class R1, class R2, class ... RList> 
 struct ratio_sum<R1, R2, RList...>
   : ratio_add< R1, ratio_product<R2, RList...>> {
};

Requires: R1 and each element in parmater pack RList is a specialization of class template ratio


1150. wchar_t, char16_t and char32_t filenames

Section: 27.9.5 [fstream] Status: LEWG Submitter: LWG Opened: 2009-06-28 Last modified: 2016-02-12

Priority: Not Prioritized

View all issues with LEWG status.

Discussion:

Addresses JP 73

Description

It is a problem from C++98, fstream cannot appoint a filename of wide character string(const wchar_t and const wstring&).

Suggestion

Add interface corresponding to wchar_t, char16_t and char32_t.

[ 2009-07-01 Alisdair notes that this is a duplicate of 454 which has more in-depth rationale. ]

[ 2009-09-21 Daniel adds: ]

I suggest to mark this issue as NAD Future with the intend to solve the issue with a single file path c'tor template assuming a provision of a TR2 filesystem library.

[ 2009 Santa Cruz: ]

NAD Future. This is a duplicate of 454.

Proposed resolution:


1154. complex should accept integral types

Section: 26.5 [complex.numbers] Status: LEWG Submitter: LWG Opened: 2009-06-28 Last modified: 2016-02-12

Priority: Not Prioritized

View other active issues in [complex.numbers].

View all other issues in [complex.numbers].

View all issues with LEWG status.

Discussion:

Addresses FR 35

Description

Instantiations of the class template complex<> have to be allowed for integral types, to reflect existing practice and ISO standards (LIA-III).

Suggestion

[ 2009-10-26 Proposed wording in N3002. ]

[ 2010 Pittsburgh: ]

Moved to NAD Future. Rationale added.

Rationale:

There is no consensus for making this change at this time.

Proposed resolution:

Adopt N3002.


1173. "Equivalence" wishy-washiness

Section: 17 [library] Status: Open Submitter: David Abrahams Opened: 2009-07-14 Last modified: 2016-02-12

Priority: 3

View other active issues in [library].

View all other issues in [library].

View all issues with Open status.

Discussion:

Issue: The CopyConstructible requirements are wishy-washy. It requires that the copy is "equivalent" to the original, but "equivalent" is never defined.

I believe this to be an example of a more general lack of rigor around copy and assignment, although I haven't done the research to dig up all the instances.

It's a problem because if you don't know what CopyConstructible means, you also don't know what it means to copy a pair of CopyConstructible types. It doesn't prevent us from writing code, but it is a hole in our ability to understand the meaning of copy.

Furthermore, I'm pretty sure that vector's copy constructor doesn't require the elements to be EqualityComparable, so that table is actually referring to some ill-defined notion of equivalence when it uses ==.

[ 2009 Santa Cruz: ]

Move to "Open". Dave is right that this is a big issue. Paper D2987 ("Defining Move Special Member Functions", Bjarne Stroustrup and Lawrence Crowl) touches on this but does not solve it. This issue is discussed in Elements of Programming.

[ 2010 Rapperswil: ]

This issue is quite vague, so it is difficult to know if and when it has been resolved. John Lakos wrote a paper covering this area a while back, and there is a real interest in providing some sort of clean-up in the future. We need a more clearly draughted issues with an addressable set of concerns, ideally with a paper proposing a resolution, but for a future revision of the standard. Move to Tentatively NAD Future.

[ Moved to NAD Future at 2010-11 Batavia ]

Proposed resolution:


1175. unordered complexity

Section: 23.2.5 [unord.req] Status: Open Submitter: Pablo Halpern Opened: 2009-07-17 Last modified: 2016-02-12

Priority: 3

View other active issues in [unord.req].

View all other issues in [unord.req].

View all issues with Open status.

Discussion:

When I look at the unordered_* constructors, I think the complexity is poorly described and does not follow the style of the rest of the standard.

The complexity for the default constructor is specified as constant. Actually, it is proportional to n, but there are no invocations of value_type constructors or other value_type operations.

For the iterator-based constructor the complexity should be:

Complexity: exactly n calls to construct value_type from InputIterator::value_type (where n = distance(f,l)). The number of calls to key_equal::operator() is proportional to n in the average case and n*n in the worst case.

[ 2010 Rapperswil: ]

Concern that the current wording may require O(1) where that cannot be delivered. We need to look at both the clause 23 requirements tables and the constructor description of each unordered container to be sure.

Howard suggests NAD Editorial as we updated the container requirement tables since this issue was written.

Daniel offers to look deeper, and hopefully produce wording addressing any outstanding concerns at the next meeting.

Move to Open.

[2011-02-26: Daniel provides wording]

I strongly suggest to clean-up the differences between requirement tables and individual specifications. In the usual way, the most specific specifications wins, which is in this case the wrong one. In regard to the concern expressed about missing DefaultConstructible requirements of the value type I disagree: The function argument n is no size-control parameter, but only some effective capacity parameter: No elements will be value-initialized by these constructors. The necessary requirement for the value type, EmplaceConstructible into *this, is already listed in Table 103 — Unordered associative container requirements. Another part of the proposed resolution is the fact that there is an inconsistency of the complexity counting when both a range and a bucket count is involved compared to constructions where only bucket counts are provided: E.g. the construction X a(n); has a complexity of n bucket allocations, but this part of the work is omitted for X a(i, j, n);, even though it is considerable larger (in the average case) for n ≫ distance(i, j).

[2011-03-24 Madrid meeting]

Move to deferred

[ 2011 Bloomington ]

The proposed wording looks good. Move to Review.

[2012, Kona]

Fix up some presentation issues with the wording, combining the big-O expressions into single expressions rather than the sum of two separate big-Os.

Strike "constant or linear", prefer "linear in the number of buckets". This allows for number of buckets being larger than requested n as well.

Default n to "unspecified" rather than "implementation-defined". It seems an un-necessary burden asking vendors to document a quantity that is easily determined through the public API of these classes.

Replace distance(f,l) with "number of elements in the range [f,l)"

Retain in Review with the updated wording

[2012, Portland: Move to Open]

The wording still does not call out Pablo's original concern, that the element constructor is called no more than N times, and that the N squared term applies to moves during rehash.

Inconsistent use of O(n)+O(N) vs. O(n+N), with a preference for the former.

AJM to update wording with a reference to "no more than N element constructor calls".

Matt concerned that calling out the O(n) requirements is noise, and dangerous noise in suggesting a precision we do not mean. The cost of constructing a bucket is very different to constructing an element of user-supplied type.

AJM notes that if there are multiple rehashes, the 'n' complexity is probably not linear.

Matt suggests back to Open, Pablo suggests potentially NAD if we keep revisitting without achieving a resolution.

Matt suggests complexity we are concerned with is the number of operations, such as constructing elements, moving nodes, and comparing/hashing keys. We are less concerned with constructing buckets, which are generally noise in this bigger picture.

[2015-01-29 Telecon]

AM: essentially correct, but do we want to complicate the spec?

HH: Pablo has given us permission to NAD it

JM: when I look at the first change in the P/R I find it mildly disturbing that the existing wording says you have a constant time constructor with a single element even if your n is 10^6, so I think adding this change makes people aware there might be a large cost in initializing the hash table, even though it doesn't show up in user-visible constructions.

HH: one way to avoid that problem is make the default ctor noexcept. Then the container isn't allowed to create an arbitrarily large hash table

AM: but this is the constructor where the user provides n

MC: happy with the changes, except I agree with the editorial recommendation to keep the two 𝒪s separate.

JW: yes, the constant 'k' is different in 𝒪(n) and 𝒪(N)

GR: do we want to talk about buckets at all

JM: yes, good to highlight that bucket construction might be a significant cost

HH: suggest we take the suggestion to split 𝒪(n+N) to 𝒪(n)+𝒪(N) and move to Tentatively Ready

GR: 23.2.1p2 says all complexity requirements are stated solely in terms of the number of operations on the contained object, so we shouldn't be stating complexity in terms of the hash table initialization

HH: channeling Pete, there's an implicit "unless otherwise specified" everywhere.

VV: seem to be requesting modifications that render this not Tentatively Ready

GR: I think it can't be T/R

AM: make the editorial recommendation, consider fixing 23.2.1/3 to give us permission to state complexity in terms of bucket initialization

HH: only set it to Review after we get new wording to review

[2015-02 Cologne]

Update wording, revisit later.

Proposed resolution:

  1. Modify the following rows in Table 103 — Unordered associative container requirements to add the explicit bucket allocation overhead of some constructions. As editorial recommendation it is suggested not to shorten the sum 𝒪(n) + 𝒪(N) to 𝒪(n + N), because two different work units are involved.

    Table 103 — Unordered associative container requirements (in addition to container)
    Expression Return type Assertion/note pre-/post-condition Complexity
    X(i, j, n, hf, eq)
    X a(i, j, n, hf, eq)
    X
    Effects: Constructs an empty container with at least n
    buckets, using hf as the hash function and eq as the key
    equality predicate, and inserts elements from [i, j) into it.
    Average case 𝒪(n + N) (N is distance(i, j)),
    worst case 𝒪(n) + 𝒪(N2)
    X(i, j, n, hf)
    X a(i, j, n, hf)
    X
    Effects: Constructs an empty container with at least n
    buckets, using hf as the hash function and key_equal() as the key
    equality predicate, and inserts elements from [i, j) into it.
    Average case 𝒪(n + N) (N is distance(i, j)),
    worst case 𝒪(n + N2)
    X(i, j, n)
    X a(i, j, n)
    X
    Effects: Constructs an empty container with at least n
    buckets, using hasher() as the hash function and key_equal() as the key
    equality predicate, and inserts elements from [i, j) into it.
    Average case 𝒪(n + N) (N is distance(i, j)),
    worst case 𝒪(n + N2)
  2. Modify 23.5.4.2 [unord.map.cnstr] p. 1-4 as indicated (The edits of p. 1 and p. 3 attempt to fix some editorial oversight.):

    explicit unordered_map(size_type n = see below,
                           const hasher& hf = hasher(),
                           const key_equal& eql = key_equal(),
                           const allocator_type& a = allocator_type());
    

    1 Effects: Constructs an empty unordered_map using the specified hash function, key equality function, and allocator, and using at least n buckets. If n is not provided, the number of buckets is unspecifiedimpldefdefault number of buckets in unordered_map. max_load_factor() returns 1.0.

    2 Complexity: ConstantLinear in the number of buckets.

    template <class InputIterator>
    unordered_map(InputIterator f, InputIterator l,
                  size_type n = see below,
                  const hasher& hf = hasher(),
                  const key_equal& eql = key_equal(),
                  const allocator_type& a = allocator_type());
    

    3 Effects: Constructs an empty unordered_map using the specified hash function, key equality function, and allocator, and using at least n buckets. If n is not provided, the number of buckets is unspecifiedimpldefdefault number of buckets in unordered_map. Then inserts elements from the range [f, l). max_load_factor() returns 1.0.

    4 Complexity: Average case linear, worst case quadraticLinear in the number of buckets. In the average case linear in N and in the worst case quadratic in N to insert the elements, where N is equal to number of elements in the range [f,l).

  3. Modify 23.5.5.2 [unord.multimap.cnstr] p. 1-4 as indicated (The edits of p. 1 and p. 3 attempt to fix some editorial oversight.):

    explicit unordered_multimap(size_type n = see below,
                                const hasher& hf = hasher(),
                                const key_equal& eql = key_equal(),
                                const allocator_type& a = allocator_type());
    

    1 Effects: Constructs an empty unordered_multimap using the specified hash function, key equality function, and allocator, and using at least n buckets. If n is not provided, the number of buckets is unspecifiedimpldefdefault number of buckets in unordered_multimap. max_load_factor() returns 1.0.

    2 Complexity: ConstantLinear in the number of buckets.

    template <class InputIterator>
    unordered_multimap(InputIterator f, InputIterator l,
                       size_type n = see below,
                       const hasher& hf = hasher(),
                       const key_equal& eql = key_equal(),
                       const allocator_type& a = allocator_type());
    

    3 Effects: Constructs an empty unordered_multimap using the specified hash function, key equality function, and allocator, and using at least n buckets. If n is not provided, the number of buckets is unspecifiedimpldefdefault number of buckets in unordered_multimap. Then inserts elements from the range [f, l). max_load_factor() returns 1.0.

    4 Complexity: Average case linear, worst case quadraticLinear in the number of buckets. In the average case linear in N and in the worst case quadratic in N to insert the elements, where N is equal to number of elements in the range [f,l).

  4. Modify 23.5.6.2 [unord.set.cnstr] p. 1-4 as indicated (The edits of p. 1 and p. 3 attempt to fix some editorial oversight.):

    explicit unordered_set(size_type n = see below,
                           const hasher& hf = hasher(),
                           const key_equal& eql = key_equal(),
                           const allocator_type& a = allocator_type());
    

    1 Effects: Constructs an empty unordered_set using the specified hash function, key equality function, and allocator, and using at least n buckets. If n is not provided, the number of buckets is unspecifiedimpldefdefault number of buckets in unordered_set. max_load_factor() returns 1.0.

    2 Complexity: ConstantLinear in the number of buckets.

    template <class InputIterator>
    unordered_set(InputIterator f, InputIterator l,
                  size_type n = see below,
                  const hasher& hf = hasher(),
                  const key_equal& eql = key_equal(),
                  const allocator_type& a = allocator_type());
    

    3 Effects: Constructs an empty unordered_set using the specified hash function, key equality function, and allocator, and using at least n buckets. If n is not provided, the number of buckets is unspecifiedimpldefdefault number of buckets in unordered_set. Then inserts elements from the range [f, l). max_load_factor() returns 1.0.

    4 Complexity: Average case linear, worst case quadraticLinear in the number of buckets. In the average case linear in N and in the worst case quadratic in N to insert the elements, where N is equal to number of elements in the range [f,l).

  5. Modify 23.5.7.2 [unord.multiset.cnstr] p. 1-4 as indicated (The edits of p. 1 and p. 3 attempt to fix some editorial oversight.):

    explicit unordered_multiset(size_type n = see below,
                                const hasher& hf = hasher(),
                                const key_equal& eql = key_equal(),
                                const allocator_type& a = allocator_type());
    

    1 Effects: Constructs an empty unordered_multiset using the specified hash function, key equality function, and allocator, and using at least n buckets. If n is not provided, the number of buckets is unspecifiedimpldefdefault number of buckets in unordered_multiset. max_load_factor() returns 1.0.

    2 Complexity: ConstantLinear in the number of buckets.

    template <class InputIterator>
    unordered_multiset(InputIterator f, InputIterator l,
                       size_type n = see below,
                       const hasher& hf = hasher(),
                       const key_equal& eql = key_equal(),
                       const allocator_type& a = allocator_type());
    

    3 Effects: Constructs an empty unordered_multiset using the specified hash function, key equality function, and allocator, and using at least n buckets. If n is not provided, the number of buckets is unspecifiedimpldefdefault number of buckets in unordered_multiset. Then inserts elements from the range [f, l). max_load_factor() returns 1.0.

    4 Complexity: Average case linear, worst case quadraticLinear in the number of buckets. In the average case linear in N and in the worst case quadratic in N to insert the elements, where N is equal to number of elements in the range [f,l).


1184. Feature request: dynamic bitset

Section: 23.3.11 [vector] Status: LEWG Submitter: Alisdair Meredith Opened: 2009-07-29 Last modified: 2016-02-12

Priority: Not Prioritized

View all other issues in [vector].

View all issues with LEWG status.

Discussion:

Opened at Alisdair's request, steming from 96. Alisdair recommends NAD Future.

[ 2009-10 Santa Cruz: ]

NAD Future. We want a heap allocated bitset, but we don't have one today and don't have time to add one.

Proposed resolution:


1188. Unordered containers should have a minimum load factor as well as a maximum

Section: 23.2.5 [unord.req], 23.5 [unord] Status: LEWG Submitter: Matt Austern Opened: 2009-08-10 Last modified: 2016-02-12

Priority: Not Prioritized

View other active issues in [unord.req].

View all other issues in [unord.req].

View all issues with LEWG status.

Discussion:

Unordered associative containers have a notion of a maximum load factor: when the number of elements grows large enough, the containers automatically perform a rehash so that the number of elements per bucket stays below a user-specified bound. This ensures that the hash table's performance characteristics don't change dramatically as the size increases.

For similar reasons, Google has found it useful to specify a minimum load factor: when the number of elements shrinks by a large enough, the containers automatically perform a rehash so that the number of elements per bucket stays above a user-specified bound. This is useful for two reasons. First, it prevents wasting a lot of memory when an unordered associative container grows temporarily. Second, it prevents amortized iteration time from being arbitrarily large; consider the case of a hash table with a billion buckets and only one element. (This was discussed even before TR1 was published; it was TR issue 6.13, which the LWG closed as NAD on the grounds that it was a known design feature. However, the LWG did not consider the approach of a minimum load factor.)

The only interesting question is when shrinking is allowed. In principle the cleanest solution would be shrinking on erase, just as we grow on insert. However, that would be a usability problem; it would break a number of common idioms involving erase. Instead, Google's hash tables only shrink on insert and rehash.

The proposed resolution allows, but does not require, shrinking in rehash, mostly because a postcondition for rehash that involves the minimum load factor would be fairly complicated. (It would probably have to involve a number of special cases and it would probably have to mention yet another parameter, a minimum bucket count.)

The current behavior is equivalent to a minimum load factor of 0. If we specify that 0 is the default, this change will have no impact on backward compatibility.

[ 2010 Rapperswil: ]

This seems to a useful extension, but is too late for 0x. Move to Tentatively NAD Future.

[ Moved to NAD Future at 2010-11 Batavia ]

Proposed resolution:

Add two new rows, and change rehash's postcondition in the unordered associative container requirements table in 23.2.5 [unord.req]:

Table 87 — Unordered associative container requirements (in addition to container)
ExpressionReturn typeAssertion/note pre-/post-condition Complexity
a.min_load_factor() float Returns a non-negative number that the container attempts to keep the load factor greater than or equal to. The container automatically decreases the number of buckets as necessary to keep the load factor above this number. constant
a.min_load_factor(z) void Pre: z shall be non-negative. Changes the container's minimum load factor, using z as a hint. [Footnote: the minimum load factor should be significantly smaller than the maximum. If z is too large, the implementation may reduce it to a more sensible value.] constant
a.rehash(n) void Post: a.bucket_count() >= n, and a.size() <= a.bucket_count() * a.max_load_factor(). [Footnote: It is intentional that the postcondition does not mention the minimum load factor. This member function is primarily intended for cases where the user knows that the container's size will increase soon, in which case the container's load factor will temporarily fall below a.min_load_factor().] a.bucket_cout > a.size() / a.max_load_factor() and a.bucket_count() >= n. Average case linear in a.size(), worst case quadratic.

Add a footnote to 23.2.5 [unord.req] p12:

The insert members shall not affect the validity of references to container elements, but may invalidate all iterators to the container. The erase members shall invalidate only iterators and references to the erased elements.

[A consequence of these requirements is that while insert may change the number of buckets, erase may not. The number of buckets may be reduced on calls to insert or rehash.]

Change paragraph 13:

The insert members shall not affect the validity of iterators if (N+n) < z * B zmin * B <= (N+n) <= zmax * B, where N is the number of elements in the container prior to the insert operation, n is the number of elements inserted, B is the container's bucket count, zmin is the container's minimum load factor, and zmax is the container's maximum load factor.

Add to the unordered_map class synopsis in section 23.5.4 [unord.map], the unordered_multimap class synopsis in 23.5.5 [unord.multimap], the unordered_set class synopsis in 23.5.6 [unord.set], and the unordered_multiset class synopsis in 23.5.7 [unord.multiset]:


float min_load_factor() const;
void min_load_factor(float z);

In 23.5.4.2 [unord.map.cnstr], 23.5.5.2 [unord.multimap.cnstr], 23.5.6.2 [unord.set.cnstr], and 23.5.7.2 [unord.multiset.cnstr], change:

... max_load_factor() returns 1.0 and min_load_factor() returns 0.


1201. Do we always want to unwrap ref-wrappers in make_tuple

Section: 20.5.2.4 [tuple.creation], 20.4 [pairs] Status: LEWG Submitter: Alisdair Meredith Opened: 2009-09-05 Last modified: 2016-02-12

Priority: Not Prioritized

View all other issues in [tuple.creation].

View all issues with LEWG status.

Discussion:

Spotting a recent thread on the boost lists regarding collapsing optional representations in optional<optional<T>> instances, I wonder if we have some of the same issues with make_tuple, and now make_pair?

Essentially, if my generic code in my own library is handed a reference_wrapper by a user, and my library in turn delegates some logic to make_pair or make_tuple, then I am going to end up with a pair/tuple holding a real reference rather than the intended reference wrapper.

There are two things as a library author I can do at this point:

  1. document my library also has the same reference-wrapper behaviour as std::make_tuple
  2. roll my own make_tuple that does not unwrap rereferences, a lost opportunity to re-use the standard library.

(There may be some metaprogramming approaches my library can use to wrap the make_tuple call, but all will be significantly more complex than simply implementing a simplified make_tuple.)

Now I don't propose we lose this library facility, I think unwrapping references will be the common behaviour. However, we might want to consider adding another overload that does nothing special with ref-wrappers. Note that we already have a second overload of make_tuple in the library, called tie.

[ 2009-09-30 Daniel adds: ]

I suggest to change the currently proposed paragraph for make_simple_pair

template<typename... Types>
  pair<typename decay<Types>::type...> make_simple_pair(Types&&... t);

Type requirements: sizeof...(Types) == 2. Remarks: The program shall be ill-formed, if sizeof...(Types) != 2.

...

or alternatively (but with a slightly different semantic):

Remarks: If sizeof...(Types) != 2, this function shall not participate in overload resolution.

to follow a currently introduced style and because the library does not have yet a specific "Type requirements" element. If such thing would be considered as useful this should be done as a separate issue. Given the increasing complexity of either of these wordings it might be preferable to use the normal two-argument-declaration style again in either of the following ways:

  1. template<class T1, class T2>
    pair<typename decay<T1>::type, typename decay<T2>::type>
    make_simple_pair(T1&& t1, T2&& t2);
    
  2. template<class T1, class T2>
    pair<V1, V2> make_simple_pair(T1&& t1, T2&& t2);
    

    Let V1 be typename decay<T1>::type and V2 be typename decay<T2>::type.

[ 2009-10 post-Santa Cruz: ]

Mark as Tentatively NAD Future.

Rationale:

Does not have sufficient support at this time. May wish to reconsider for a future standard.

Proposed resolution:

Add the following function to 20.4 [pairs] and signature in appropriate synopses:

template<typename... Types>
  pair<typename decay<Types>::type...> make_simple_pair(Types&&... t);

Type requirements: sizeof...(Types) == 2.

Returns: pair<typename decay<Types>::type...>(std::forward<Types>(t)...).

[ Draughting note: I chose a variadic representation similar to make_tuple rather than naming both types as it is easier to read through the clutter of metaprogramming this way. Given there are exactly two elements, the committee may prefer to draught with two explicit template type parameters instead ]

Add the following function to 20.5.2.4 [tuple.creation] and signature in appropriate synopses:

template<typename... Types>
  tuple<typename decay<Types>::type...> make_simple_tuple(Types&&... t);

Returns: tuple<typename decay<Types>::type...>(std::forward<Types>(t)...).


1203. More useful rvalue stream insertion

Section: 27.7.3.9 [ostream.rvalue], 27.7.2.6 [istream.rvalue] Status: LEWG Submitter: Howard Hinnant Opened: 2009-09-06 Last modified: 2016-02-12

Priority: Not Prioritized

View other active issues in [ostream.rvalue].

View all other issues in [ostream.rvalue].

View all issues with LEWG status.

Discussion:

27.7.3.9 [ostream.rvalue] was created to preserve the ability to insert into (and extract from 27.7.2.6 [istream.rvalue]) rvalue streams:

template <class charT, class traits, class T>
  basic_ostream<charT, traits>&
  operator<<(basic_ostream<charT, traits>&& os, const T& x);

1 Effects: os << x

2 Returns: os

This is good as it allows code that wants to (for example) open, write to, and close an ofstream all in one statement:

std::ofstream("log file") << "Some message\n";

However, I think we can easily make this "rvalue stream helper" even easier to use. Consider trying to quickly create a formatted string. With the current spec you have to write:

std::string s = static_cast<std::ostringstream&>(std::ostringstream() << "i = " << i).str();

This will store "i = 10" (for example) in the string s. Note the need to cast the stream back to ostringstream& prior to using the member .str(). This is necessary because the inserter has cast the ostringstream down to a more generic ostream during the insertion process.

I believe we can re-specify the rvalue-inserter so that this cast is unnecessary. Thus our customer now has to only type:

std::string s = (std::ostringstream() << "i = " << i).str();

This is accomplished by having the rvalue stream inserter return an rvalue of the same type, instead of casting it down to the base class. This is done by making the stream generic, and constraining it to be an rvalue of a type derived from ios_base.

The same argument and solution also applies to the inserter. This code has been implemented and tested.

[ 2009 Santa Cruz: ]

NAD Future. No concensus for change.

Proposed resolution:

Change 27.7.2.6 [istream.rvalue]:

template <class charT, class traits Istream, class T>
  basic_istream<charT, traits>& Istream&&
  operator>>(basic_istream<charT, traits> Istream&& is, T& x);

1 Effects: is >> x

2 Returns: std::move(is)

3 Remarks: This signature shall participate in overload resolution if and only if Istream is not an lvalue reference type and is derived from ios_base.

Change 27.7.3.9 [ostream.rvalue]:

template <class charT, class traits Ostream, class T>
  basic_ostream<charT, traits>& Ostream&&
  operator<<(basic_ostream<charT, traits> Ostream&& os, const T& x);

1 Effects: os << x

2 Returns: std::move(os)

3 Remarks: This signature shall participate in overload resolution if and only if Ostream is not an lvalue reference type and is derived from ios_base.


1213. Meaning of valid and singular iterator underspecified

Section: 24.2 [iterator.requirements] Status: Open Submitter: Daniel Krügler Opened: 2009-09-19 Last modified: 2016-02-12

Priority: 4

View other active issues in [iterator.requirements].

View all other issues in [iterator.requirements].

View all issues with Open status.

Discussion:

The terms valid iterator and singular aren't properly defined. The fuzziness of those terms became even worse after the resolution of 208 (including further updates by 278). In 24.2 [iterator.requirements] as of N2723 the standard says now:

5 - These values are called past-the-end values. Values of an iterator i for which the expression *i is defined are called dereferenceable. The library never assumes that past-the-end values are dereferenceable. Iterators can also have singular values that are not associated with any container. [...] Results of most expressions are undefined for singular values; the only exceptions are destroying an iterator that holds a singular value and the assignment of a non-singular value to an iterator that holds a singular value. [...] Dereferenceable values are always non-singular.

10 - An invalid iterator is an iterator that may be singular.

First, issue 208 intentionally removed the earlier constraint that past-the-end values are always non-singular. The reason for this was to support null pointers as past-the-end iterators of e.g. empty sequences. But there seem to exist different views on what a singular (iterator) value is. E.g. according to the SGI definition a null pointer is not a singular value:

Dereferenceable iterators are always nonsingular, but the converse is not true. For example, a null pointer is nonsingular (there are well defined operations involving null pointers) even thought it is not dereferenceable.

and proceeds:

An iterator is valid if it is dereferenceable or past-the-end.

Even if the standard prefers a different meaning of singular here, the change was incomplete, because by restricting feasible expressions of singular iterators to destruction and assignment isn't sufficient for a past-the-end iterator: Of-course it must still be equality-comparable and in general be a readable value.

Second, the standard doesn't clearly say whether a past-the-end value is a valid iterator or not. E.g. 20.10.10 [specialized.algorithms]/1 says:

In all of the following algorithms, the formal template parameter ForwardIterator is required to satisfy the requirements of a forward iterator (24.1.3) [..], and is required to have the property that no exceptions are thrown from [..], or dereference of valid iterators.

The standard should make better clear what "singular pointer" and "valid iterator" means. The fact that the meaning of a valid value has a core language meaning doesn't imply that for an iterator concept the term "valid iterator" has the same meaning.

Let me add a final example: In 99 [allocator.concepts.members] of N2914 we find:

pointer X::allocate(size_type n);

11 Returns: a pointer to the allocated memory. [Note: if n == 0, the return value is unspecified. —end note]

[..]

void X::deallocate(pointer p, size_type n);

Preconditions: p shall be a non-singular pointer value obtained from a call to allocate() on this allocator or one that compares equal to it.

If singular pointer value would include null pointers this make the preconditions unclear if the pointer value is a result of allocate(0): Since the return value is unspecified, it could be a null pointer. Does that mean that programmers need to check the pointer value for a null value before calling deallocate?

[ 2010-11-09 Daniel comments: ]

A later paper is in preparation.

[ 2010 Batavia: ]

Doesn't need to be resolved for Ox

[2014-02-20 Re-open Deferred issues as Priority 4]

Consider to await the paper.

Proposed resolution:


1217. Quaternion support

Section: 26.5 [complex.numbers] Status: LEWG Submitter: Ted Shaneyfelt Opened: 2009-09-26 Last modified: 2016-02-12

Priority: Not Prioritized

View other active issues in [complex.numbers].

View all other issues in [complex.numbers].

View all issues with LEWG status.

Discussion:

Concerning mathematically proper operation of the type:

complex<complex<T> >

Generally accepted mathematical semantics of such a construct correspond to quaternions through Cayly-Dickson construct

(w+xi) + (y+zi) j

The proper implementation seems straightforward by adding a few declarations like those below. I have included operator definition for combining real scalars and complex types, as well, which seems appropriate, as algebra of complex numbers allows mixing complex and real numbers with operators. It also allows for constructs such as complex<double> i=(0,1), x = 12.34 + 5*i;

Quaternions are often used in areas such as computer graphics, where, for example, they avoid the problem of Gimbal lock when rotating objects in 3D space, and can be more efficient than matrix multiplications, although I am applying them to a different field.

/////////////////////////ALLOW OPERATORS TO COMBINE REAL SCALARS AND COMPLEX VALUES /////////////////////////
template<typename T,typename S> complex<T> operator+(const complex<T> x,const S a) {
    complex<T> result(x.real()+a, x.imag());
    return result;
}
template<typename T,typename S> complex<T> operator+(const S a,const complex<T> x) {
    complex<T> result(a+x.real(), x.imag());
    return result;
}
template<typename T,typename S> complex<T> operator-(const complex<T> x,const S a) {
    complex<T> result(x.real()-a, x.imag());
    return result;
}
template<typename T,typename S> complex<T> operator-(const S a,const complex<T> x) {
    complex<T> result(a-x.real(), x.imag());
    return result;
}
template<typename T,typename S> complex<T> operator*(const complex<T> x,const S a) {
    complex<T> result(x.real()*a, x.imag()*a);
    return result;
}
template<typename T,typename S> complex<T> operator*(const S a,const complex<T> x) {
    complex<T> result(a*x.real(), a*x.imag());
    return result;
}

/////////////////////////PROPERLY IMPLEMENT QUATERNION SEMANTICS/////////////////////////
template<typename T> double normSq(const complex<complex<T> >q) {
    return q.real().real()*q.real().real()
         + q.real().imag()*q.real().imag()
         + q.imag().real()*q.imag().real()
         + q.imag().imag()*q.imag().imag();
}
template<typename T> double norm(const complex<complex<T> >q) {
    return sqrt(normSq(q));
}
/////// Cayley-Dickson Construction
template<typename T> complex<complex<T> > conj(const complex<complex<T> > x) {
    complex<complex<T> > result(conj(x.real()),-x.imag());
    return result;
}
template<typename T> complex<complex<T> > operator*(const complex<complex<T> > ab,const complex<complex<T> > cd) {
    complex<T> re(ab.real()*cd.real()-conj(cd.imag())*ab.imag());
    complex<T> im(cd.imag()*ab.real()+ab.imag()*conj(cd.real()));
    complex<complex<double> > q(re,im);
    return q;
}
//// Quaternion division
template<typename S,typename T> complex<complex<T> > operator/(const complex<complex<T> > q,const S a) {
    return q * (1/a);
}
template<typename S,typename T> complex<complex<T> > operator/(const S a,const complex<complex<T> > q) {
    return a*conj(q)/normSq(q);
}
template<typename T> complex<complex<T> > operator/(const complex<complex<T> > n, const complex<complex<T> > d) {
    return n * (conj(d)/normSq(d));
}

[ 2009-10 Santa Cruz: ]

NAD Future. There is no consensus or time to move this into C++0X.

Proposed resolution:


1235. Issue with C++0x random number proposal

Section: 99 [rand.concept.dist] Status: LEWG Submitter: Matthias Troyer Opened: 2009-10-12 Last modified: 2016-02-12

Priority: Not Prioritized

View all issues with LEWG status.

Discussion:

There exist optimized, vectorized vendor libraries for the creation of random number generators, such as Intel's MKL [1] and AMD's ACML [2]. In timing tests we have seen a performance gain of a factor of up to 80 (eighty) compared to a pure C++ implementation (in Boost.Random) when using these generator to generate a sequence of normally distributed random numbers. In codes dominated by the generation of random numbers (we have application codes where random number generation is more than 50% of the CPU time) this factor 80 is very significant.

To make use of these vectorized generators, we use a C++ class modeling the RandomNumberEngine concept and forwarding the generation of random numbers to those optimized generators. For example:

namespace mkl {
 class mt19937 {.... };
}

For the generation of random variates we also want to dispatch to optimized vectorized functions in the MKL or ACML libraries. See this example:

mkl::mt19937 eng;
std::normal_distribution<double> dist;

double n = dist(eng);

Since the variate generation is done through the operator() of the distribution there is no customization point to dispatch to Intel's or AMD's optimized functions to generate normally distributed numbers based on the mt19937 generator. Hence, the performance gain of 80 cannot be achieved.

Contrast this with TR1:

mkl::mt19937 eng;
std::tr1::normal_distribution<double> dist;
std::tr1::variate_generator<mkl::mt19937,std::tr1::normal_distribution<double> > rng(eng,dist);
double n = rng();

This - admittedly much uglier from an aestethic point of view - design allowed optimization by specializing the variate_generator template for mkl::mt19937:

namespace std { namespace tr1 {

template<>
class variate_generator<mkl::mt19937,std::tr1::normal_distribution<double> > { .... };

} }

A similar customization point is missing in the C++0x design and prevents the optimized vectorized version to be used.

Suggested resolution:

Add a customization point to the distribution concept. Instead of the variate_generator template this can be done through a call to a free function generate_variate found by ADL instead of operator() of the distribution:

template <RandomNumberDistribution, class RandomNumberEngine>
typename RandomNumberDistribution ::result_type
generate_variate(RandomNumberDistribution const& dist, RandomNumberEngine& eng);

This function can be overloaded for optimized enginges like mkl::mt19937.

[ 2009-10 Santa Cruz: ]

NAD Future. No time to add this feature for C++0X.

Proposed resolution:


1238. defining algorithms taking iterator for range

Section: 25 [algorithms] Status: LEWG Submitter: Alisdair Meredith Opened: 2009-10-15 Last modified: 2016-02-12

Priority: Not Prioritized

View other active issues in [algorithms].

View all other issues in [algorithms].

View all issues with LEWG status.

Discussion:

The library has many algorithms that take a source range represented by a pair of iterators, and the start of some second sequence given by a single iterator. Internally, these algorithms will produce undefined behaviour if the second 'range' is not as large as the input range, but none of the algorithms spell this out in Requires clauses, and there is no catch-all wording to cover this in clause 17 or the front matter of 25.

There was an attempt to provide such wording in paper n2944 but this seems incidental to the focus of the paper, and getting the wording of this issue right seems substantially more difficult than the simple approach taken in that paper. Such wording will be removed from an updated paper, and hopefully tracked via the LWG issues list instead.

It seems there are several classes of problems here and finding wording to solve all in one paragraph could be too much. I suspect we need several overlapping requirements that should cover the desired range of behaviours.

Motivating examples:

A good initial example is the swap_ranges algorithm. Here there is a clear requirement that first2 refers to the start of a valid range at least as long as the range [first1, last1). n2944 tries to solve this by positing a hypothetical last2 iterator that is implied by the signature, and requires distance(first2,last2) < distance(first1,last1). This mostly works, although I am uncomfortable assuming that last2 is clearly defined and well known without any description of how to obtain it (and I have no idea how to write that).

A second motivating example might be the copy algorithm. Specifically, let us image a call like:

copy(istream_iterator<int>(is),istream_iterator(),ostream_iterator<int>(os));

In this case, our input iterators are literally simple InputIterators, and the destination is a simple OutputIterator. In neither case am I happy referring to std::distance, in fact it is not possible for the ostream_iterator at all as it does not meet the requirements. However, any wording we provide must cover both cases. Perhaps we might deduce last2 == ostream_iterator<int>{}, but that might not always be valid for user-defined iterator types. I can well imagine an 'infinite range' that writes to /dev/null and has no meaningful last2.

The motivating example in n2944 is std::equal, and that seems to fall somewhere between the two.

Outlying examples might be partition_copy that takes two output iterators, and the _n algorithms where a range is specified by a specific number of iterations, rather than traditional iterator pair. We should also not accidentally apply inappropriate constraints to std::rotate which takes a third iterator that is not intended to be a separate range at all.

I suspect we want some wording similar to:

For algorithms that operate on ranges where the end iterator of the second range is not specified, the second range shall contain at least as many elements as the first.

I don't think this quite captures the intent yet though. I am not sure if 'range' is the right term here rather than sequence. More awkwardly, I am not convinced we can describe an Output sequence such as produce by an ostream_iterator as "containing elements", at least not as a precondition to the call before they have been written.

Another idea was to describe require that the trailing iterator support at least distance(input range) applications of operator++ and may be written through the same number of times if a mutable/output iterator.

We might also consider handling the case of an output range vs. an input range in separate paragraphs, if that simplifies how we describe some of these constraints.

[ 2009-11-03 Howard adds: ]

Moved to Tentatively NAD Future after 5 positive votes on c++std-lib.

Rationale:

Does not have sufficient support at this time. May wish to reconsider for a future standard.

Proposed resolution:


1242. Enable SCARY iterators

Section: 23 [containers] Status: LEWG Submitter: Herb Sutter Opened: 2009-10-21 Last modified: 2016-02-12

Priority: Not Prioritized

View other active issues in [containers].

View all other issues in [containers].

View all issues with LEWG status.

Discussion:

See N2980.

[ 2009-10 Santa Cruz ]

The paper was lengthy discussed but considerable concern remained to add this feature to C++0x. Strong consensus was found to consider it for C++1x, though.

Proposed resolution:

The LWG does not wish to make a change at this time.


1282. A proposal to add std::split algorithm

Section: 25 [algorithms] Status: LEWG Submitter: Igor Semenov Opened: 2009-12-07 Last modified: 2016-02-12

Priority: Not Prioritized

View other active issues in [algorithms].

View all other issues in [algorithms].

View all issues with LEWG status.

Discussion:

  1. Motivation and Scope

    Splitting strings into parts by some set of delimiters is an often task, but there is no simple and generalized solution in C++ Standard. Usually C++ developers use std::basic_stringstream<> to split string into parts, but there are several inconvenient restrictions:

  2. Impact on the Standard

    This algorithm doesn't interfere with any of current standard algorithms.

  3. Design Decisions

    This algorithm is implemented in terms of input/output iterators. Also, there is one additional wrapper for const CharType * specified delimiters.

  4. Example implementation

    template< class It, class DelimIt, class OutIt >
    void split( It begin, It end, DelimIt d_begin, DelimIt d_end, OutIt out )
    {
       while ( begin != end )
       {
           It it = std::find_first_of( begin, end, d_begin, d_end );
           *out++ = std::make_pair( begin, it );
           begin = std::find_first_of( it, end, d_begin, d_end,
               std::not2( std::equal_to< typename It::value_type >() ) );
       }
    }
    
    template< class It, class CharType, class OutIt >
    void split( It begin, It end, const CharType * delim, OutIt out )
    {
       split( begin, end, delim, delim + std::strlen( delim ), out );
    }
    
  5. Usage

    std::string ss( "word1 word2 word3" );
    std::vector< std::pair< std::string::const_iterator, std::string::const_iterator > > v;
    split( ss.begin(), ss.end(), " ", std::back_inserter( v ) );
    
    for ( int i = 0; i < v.size(); ++i )
    {
       std::cout << std::string( v[ i ].first, v[ i ].second ) << std::endl;
    }
    // word1
    // word2
    // word3
    

[ 2010-01-22 Moved to Tentatively NAD Future after 5 positive votes on c++std-lib. Rationale added below. ]

Rationale:

The LWG is not considering completely new features for standardization at this time. We would like to revisit this good suggestion for a future TR and/or standard.

Proposed resolution:

Add to the synopsis in 25.1 [algorithms.general]:

template< class ForwardIterator1, class ForwardIterator2, class OutputIterator >
  void split( ForwardIterator1 first, ForwardIterator1 last,
              ForwardIterator2 delimiter_first, ForwardIterator2 delimiter_last,
              OutputIterator result );

template< class ForwardIterator1, class CharType, class OutputIterator >
  void split( ForwardIterator1 first, ForwardIterator1 last,
              const CharType * delimiters, OutputIterator result );

Add a new section [alg.split]:

template< class ForwardIterator1, class ForwardIterator2, class OutputIterator >
  void split( ForwardIterator1 first, ForwardIterator1 last,
              ForwardIterator2 delimiter_first, ForwardIterator2 delimiter_last,
              OutputIterator result );

1. Effects: splits the range [first, last) into parts, using any element of [delimiter_first, delimiter_last) as a delimiter. Results are pushed to output iterator in the form of std::pair<ForwardIterator1, ForwardIterator1>. Each of these pairs specifies a maximal subrange of [first, last) which does not contain a delimiter.

2. Returns: nothing.

3. Complexity: Exactly last - first assignments.

template< class ForwardIterator1, class CharType, class OutputIterator >
  void split( ForwardIterator1 first, ForwardIterator1 last,
              const CharType * delimiters, OutputIterator result );

1. Effects: split the range [first, last) into parts, using any element of delimiters (interpreted as zero-terminated string) as a delimiter. Results are pushed to output iterator in the form of std::pair<ForwardIterator1, ForwardIterator1>. Each of these pairs specifies a maximal subrange of [first, last) which does not contain a delimiter.

2. Returns: nothing.

3. Complexity: Exactly last - first assignments.


1289. Generic casting requirements for smart pointers

Section: 20.2 [utility] Status: LEWG Submitter: Ion Gaztañaga Opened: 2009-12-14 Last modified: 2016-02-12

Priority: Not Prioritized

View all other issues in [utility].

View all issues with LEWG status.

Discussion:

In section 17.6.3.5 [allocator.requirements], Table 40 — Allocator requirements, the following expression is required for allocator pointers:

Table 40 — Allocator requirements
Expression Return type Assertion/note
pre-/post-condition
Default
static_cast<X::pointer>(w) X::pointer static_cast<X::pointer>(w) == p  

To achieve this expression, a smart pointer writer must introduce an explicit conversion operator from smart_ptr<void> to smart_ptr<T> so that static_cast<pointer>(void_ptr) is a valid expression. Unfortunately this explicit conversion weakens the safety of a smart pointer since the following expression (invalid for raw pointers) would become valid:

smart_ptr<void> smart_v = ...;
smart_ptr<T> smart_t(smart_v);

On the other hand, shared_ptr also defines its own casting functions in 20.11.2.2.9 [util.smartptr.shared.cast], and although it's unlikely that a programmer will use shared_ptr as allocator::pointer, having two different ways to do the same cast operation does not seem reasonable. A possible solution would be to replace static_cast<X::pointer>(w) expression with a user customizable (via ADL) static_pointer_cast<value_type>(w), and establish the xxx_pointer_cast functions introduced by shared_ptr as the recommended generic casting utilities of the standard.

Unfortunately, we've experienced problems in Boost when trying to establish xxx_pointer_cast as customization points for generic libraries (http://objectmix.com/c/40424-adl-lookup-explicit-template-parameters.html) because these casting functions are called with explicit template parameters and the standard says in 14.8.1 [temp.arg.explicit] p.8 "Explicit template argument specification":

8 ...But when a function template with explicit template arguments is used, the call does not have the correct syntactic form unless there is a function template with that name visible at the point of the call. If no such name is visible, the call is not syntactically well-formed and argument-dependent lookup does not apply.

So we can do this:

template<class BasePtr>
void generic_ptr_swap(BasePtr p)
{
  //ADL customization point
  swap(p, p);
  //...
}

but not the following:

template<class BasePtr>
void generic_ptr_algo(BasePtr p)
{
  typedef std::pointer_traits<BasePtr>::template
     rebind<Derived> DerivedPtr;
  DerivedPtr dp = static_pointer_cast<Derived>(p);
}

The solution to make static_pointer_cast a customization point is to add a generic declaration (no definition) of static_pointer_cast in a namespace (like std) and apply "using std::static_pointer_cast" declaration to activate ADL:

namespace std{

template<typename U, typename T>
unspecified
static_pointer_cast(T&&) = delete;

}

template<class BasePtr>
void generic_ptr_algo(BasePtr p)
{
  typedef std::pointer_traits<BasePtr>::template
     rebind<Derived> DerivedPtr;

  //ADL applies because static_pointer_cast is made
  //  visible according to [temp.arg.explicit]/8
  using std::static_pointer_cast;

  DerivedPtr dp = static_pointer_cast<Derived>(p);

  //...
}

A complete solution will need also the definition of static_pointer_cast for raw pointers, and this definition has been present in Boost (http://www.boost.org/boost/ pointer_cast.hpp) for years.

[ 2010-03-26 Daniel made editorial adjustments to the proposed wording. ]

[ Moved to NAD Future at 2010-11 Batavia ]

This is a new feature rather than a defect. It can be added later: "this is such a hairy area that people will put up with changes"

Proposed resolution:

Add to section 20.2 [utility] Utility components, Header <utility> synopsis:

// 20.3.X, generic pointer cast functions

template<typename U, typename T>
unspecified
static_pointer_cast(T&&) = delete;

template<typename U, typename T>
unspecified
dynamic_pointer_cast(T&&) = delete;

template<typename U, typename T>
unspecified
const_pointer_cast(T&&) = delete;

//Overloads for raw pointers
template<typename U, typename T>
auto static_pointer_cast(T* t) -> decltype(static_cast<U*>(t));

template<typename U, typename T>
auto dynamic_pointer_cast(T* t) -> decltype(dynamic_cast<U*>(t));

template<typename U, typename T>
auto const_pointer_cast(T* t) -> decltype(const_cast<U*>(t));

Add to section 20.2 [utility] Utility components, a new subclause 20.3.X Pointer cast utilities [pointer.cast]:

20.3.X Pointer cast utilities [pointer.cast]

1 The library defines generic pointer casting function templates so that template code can explicitly make these names visible and activate argument-dependent lookup for pointer cast calls.

//Generic declarations
template<typename U, typename T>
unspecified
static_pointer_cast(T&&) = delete;

template<typename U, typename T>
unspecified
dynamic_pointer_cast(T&&) = delete;

template<typename U, typename T>
unspecified
const_pointer_cast(T&&) = delete;

2 The library also defines overloads of these functions for raw pointers.

//Overloads for raw pointers
template<typename U, typename T>
auto static_pointer_cast(T* t) -> decltype(static_cast<U*>(t));

Returns: static_cast<U*>(t)

template<typename U, typename T>
auto dynamic_pointer_cast(T* t) -> decltype(dynamic_cast<U*>(t));

Returns: dynamic_cast<U*>(t)

template<typename U, typename T>
auto const_pointer_cast(T* t) -> decltype(const_cast<U*>(t));

Returns: const_cast<U*>(t)

[Example:

#include <utility> //static_pointer_cast
#include <memory>  //pointer_traits

class Base{};
class Derived : public Base{};

template<class BasePtr>
void generic_pointer_code(BasePtr b)
{
   typedef std::pointer_traits<BasePtr>::template
      rebind<Derived> DerivedPtr;

   using std::static_pointer_cast;
   //ADL applies now that static_pointer_cast is visible
   DerivedPtr d = static_pointer_cast<Derived>(b);
}

end example]

Replace in section 17.6.3.5 [allocator.requirements] Table 40 — Allocator requirements, the following table entries for allocator pointers:

Table 40 — Allocator requirements
Expression Return type Assertion/note
pre-/post-condition
Default
static_pointer_cast<X::pointerT>(w) X::pointer static_pointer_cast<X::pointerT>(w) == p  
static_pointer_cast<X::const_pointerconst T>(w) X::const_pointer static_pointer_cast<X::const_pointerconst T>(z) == q  

1317. make_hash

Section: 20.14.14 [unord.hash] Status: LEWG Submitter: Nicolai M. Josuttis Opened: 2010-02-10 Last modified: 2016-02-12

Priority: Not Prioritized

View other active issues in [unord.hash].

View all other issues in [unord.hash].

View all issues with LEWG status.

Discussion:

Currently, the library lacks a convenient way to provide a hash function that can be used with the provided unordered containers to allow the usage of non trivial element types.

While we can easily declare an

std::unordered_set<int>

or

std::unordered_set<std::string>

we have no easy way to declare an unordered_set for a user defined type. IMO, this is a big obstacle to use unordered containers in practice. Note that in Java, the wide usage of HashMap is based on the fact that there is always a default hash function provided.

Of course, a default hash function implies the risk to provide poor hash functions. But often even poor hash functions are good enough.

While I really would like to see a default hash function, I don't propose it here because this would probably introduce a discussion that's too big for this state of C++0x.

However, I strongly suggest at least to provide a convenience variadic template function make_hash<>() to allow an easy definition of a (possibly poor) hash function.

As a consequence for a user-defined type such as

class Customer {
   friend class CustomerHash;
   private:
     string firstname;
     string lastname;
     long   no;
   ...
 };

would allow to specify:

class CustomerHash : public std::unary_function<Customer, std::size_t>
{
  public:
    std::size_t operator() (const Customer& c) const  {
       return make_hash(c.firstname,c.lastname,c.no);
    }
};

instead of:

class CustomerHash : public std::unary_function<Customer, std::size_t>
{
  public:
    std::size_t operator() (const Customer& c) const  {
       return std::hash<std::string>()(c.firstname) +
              std::hash<std::string>()(c.lastname) +
              std::hash<long>()(c.no);
    }
};

Note that, in principle, we can either specify that

make_hash returns the sum of a call of std::hash<T>()(x) for each argument x of type T

or we can specify that

make_hash provides a hash value for each argument, for which a std::hash() function is provided

with the possible note that the hash value may be poor or only a good hash value if the ranges of all passed arguments is equally distributed.

For my convenience, I propose wording that describes the concrete implementation.

[ 2010 Pittsburgh: Moved to NAD Editorial, rationale added below. ]

Rationale:

There is no consensus to make this change at this time.

Proposed resolution:

In Function objects 20.14 [function.objects] in paragraph 2 at the end of the Header <functional> synopsis insert:

// convenience functions
template <class T>
  size_t make_hash (const T&);
template <class T, class... Types>
  size_t make_hash (const T&, const Types&...);

In Class template hash 20.14.14 [unord.hash] add:

20.7.16.1 Hash creation functions [hash.creation]

template <class T>
  size_t make_hash (const T& val);

Returns: hash<T>()(val);

template <class T, class... Types>
  size_t make_hash (const T& val, const Types&... args);

Returns: hash<T>()(val) + std::make_hash(args...)


1320. Header for iter_swap

Section: 24.3 [iterator.synopsis] Status: LEWG Submitter: Alisdair Meredith Opened: 2010-02-16 Last modified: 2016-02-12

Priority: Not Prioritized

View all other issues in [iterator.synopsis].

View all issues with LEWG status.

Discussion:

The iter_swap function template appears in the <algorithm> header, yet its main use is in building further algorithms, not calling existing ones. The main clients are implementers of data structures and their iterators, so it seems most appropriate to place the template in the <iterator> header instead.

Note that this is not an issue for implementers of the standard library, as they rarely use the standard headers directly, designing a more fine-grained set of headers for their own internal use. This option is not available to customers of the standard library.

Note that we cannot remove iter_swap from <algorithm> without breaking code, but there is no reason we cannot offer the same declaration via two standard headers. Alternatively, require <algorithm> to #include <iterator>, but introducing the dependency on the iterator adaptors seems un-necessary.

[ ]

Discussed possibly moving to <utility> but don't like that. Some not seeing this as a defect, and want to keep it in <algorithm>. No one seems to feel strongly about moving to <iterator>.

Proposed resolution:

Add the declaration of iter_swap to the <iterator> header synopsis (24.3 [iterator.synopsis]), with a note that it is documented in clause 25 [algorithms].

...
template <class T, size_t N> T* end(T (&array)[N]);

// documented in 25 [algorithms]
template<class ForwardIterator1, class ForwardIterator2>
  void iter_swap(ForwardIterator1 a, ForwardIterator2 b);

1396. regex should support allocators

Section: 28.8 [re.regex] Status: LEWG Submitter: INCITS Opened: 2010-08-25 Last modified: 2016-02-12

Priority: Not Prioritized

View all other issues in [re.regex].

View all issues with LEWG status.

Duplicate of: 1451

Discussion:

Addresses US-104, US-141

std::basic_regex should have an allocator for all the reasons that a std::string does. For example, I can use boost::interprocess to put a string or vector in shared memory, but not a regex.

[ Resolution proposed by ballot comment ]

Add allocators to regexes

[ 2010-10-24 Daniel adds: ]

Accepting n3171 would solve this issue.

[2011-03-22 Madrid]

Close 1396 as NAD Future.

Rationale:

No consensus for a change at this time

Proposed resolution:


1406. Support hashing smart-pointers based on owner

Section: 20.11.2.2 [util.smartptr.shared] Status: LEWG Submitter: Japan Opened: 2010-08-25 Last modified: 2016-02-12

Priority: Not Prioritized

View other active issues in [util.smartptr.shared].

View all other issues in [util.smartptr.shared].

View all issues with LEWG status.

Discussion:

Addresses JP-5

Hash support based on ownership sharing should be supplied for shared_ptr and weak_ptr. For two shared_ptr objects p and q, two distinct equivalence relations can be defined. One is based on equivalence of pointer values, which is derived from the expression p.get() == q.get() (hereafter called address based equivalence relation), the other is based on equivalence of ownership sharing, which is derived from the expression !p.owner_before(q) && !q.owner_before(p) (hereafter called ownership-based equivalence relation). These two equivalence relations are independent in general. For example, a shared_ptr object created by the constructor of the signature shared_ptr(shared_ptr<U> const &, T *) could reveal a difference between these two relations. Therefore, hash support based on each equivalence relation should be supplied for shared_ptr. However, while the standard library provides the hash support for address-based one (20.9.11.6 paragraph 2), it lacks the hash support for ownership-based one. In addition, associative containers work well in combination with the shared_ptr's ownership-based comparison but unordered associative containers don't. This is inconsistent.

For the case of weak_ptr, hash support for the ownership based equivalence relation can be safely defined on weak_ptrs, and even on expired ones. The absence of hash support for the ownership-based equivalence relation is fatal, especially for expired weak_ptrs. And the absence of such hash support precludes some quite effective use-cases, e.g. erasing the unordered_map entry of an expired weak_ptr key from a customized deleter supplied to shared_ptrs.

Hash support for the ownership-based equivalence relation cannot be provided by any user-defined manner because information about ownership sharing is not available to users at all. Therefore, the only way to provide ownership-based hash support is to offer it intrusively by the standard library.

As far as we know, such hash support is implementable. Typical implementation of such hash function could return the hash value of the pointer of the counter object that is internally managed by shared_ptr and weak_ptr.

[2010 Rapperswil:]

No consensus to make this change at this time.

Proposed resolution:

Add the following non-static member functions to shared_ptr and weak_ptr class template;

Update [util.smartptr.shared], 20.9.11.2 paragraph 1

namespace std{
template<class T> class shared_ptr {
public:
...
  size_t owner_hash() const;
...
};
}

Update [util.smartptr.weak], 20.9.11.3 paragraph 1

namespace std{
template<class T> class weak_ptr {
public:
...
  size_t owner_hash() const;
...
};
}

These functions satisfy the following requirements. Let p and q be objects of either shared_ptr or weak_ptr, H be a hypothetical function object type that satisfies the hash requirements ([hash.requirements], 20.2.4) and h be an object of the type H. The expression p.owner_hash() behaves as if it were equivalent to the expression h(p). In addition, h(p) == h(q) must become true if p and q share ownership.


1422. vector<bool> iterators are not random access

Section: 23.3.12 [vector.bool] Status: LEWG Submitter: BSI Opened: 2010-08-25 Last modified: 2016-02-12

Priority: Not Prioritized

View all other issues in [vector.bool].

View all issues with LEWG status.

Discussion:

Addresses GB-118

vector<bool> iterators are not random access iterators because their reference type is a special class, and not bool &. All standard libary operations taking iterators should treat this iterator as if it was a random access iterator, rather than a simple input iterator.

[ Resolution proposed in ballot comment ]

Either revise the iterator requirements to support proxy iterators (restoring functionality that was lost when the Concept facility was removed) or add an extra paragraph to the vector<bool> specification requiring the library to treat vector<bool> iterators as-if they were random access iterators, despite having the wrong reference type.

[ Rapperswil Review ]

The consensus at Rapperswil is that it is too late for full support for proxy iterators, but requiring the library to respect vector<bool> iterators as-if they were random access would be preferable to flagging this container as deliberately incompatible with standard library algorithms.

Alisdair to write the note, which may become normative Remark depending on the preferences of the project editor.

[ Post-Rapperswil Alisdair provides wording ]

Initial wording is supplied, deliberately using Note in preference to Remark although the author notes his preference for Remark. The issue of whether iterator_traits<vector<bool>>::iterator_category is permitted to report random_access_iterator_tag or must report input_iterator_tag is not addressed.

[ Old Proposed Resolution: ]

Insert a new paragraph into 23.3.12 [vector.bool] between p4 and p5:

[Note All functions in the library that take a pair of iterators to denote a range shall treat vector<bool> iterators as-if they were random access iterators, even though the reference type is not a true reference.-- end note]

[ 2010-11 Batavia: ]

Closed as NAD Future, because the current iterator categories cannot correctly describe vector<bool>::iterator. But saying that they are Random Access Iterators is also incorrect, because it is not too hard to create a corresponding test that fails. We should deal with the more general proxy iterator problem in the future, and see no benefit to take a partial workaround specific to vector<bool> now.

Proposed resolution:

Rationale:

No consensus to make this change at this time.


1459. Overlapping evaluations are allowed

Section: 29.3 [atomics.order] Status: LEWG Submitter: Canada Opened: 2010-08-25 Last modified: 2016-02-12

Priority: Not Prioritized

View other active issues in [atomics.order].

View all other issues in [atomics.order].

View all issues with LEWG status.

Duplicate of: 1458

Discussion:

Addresses CA-21, GB-131

29.4 [atomics.lockfree] p.8 states:

An atomic store shall only store a value that has been computed from constants and program input values by a finite sequence of program evaluations, such that each evaluation observes the values of variables as computed by the last prior assignment in the sequence.

... but 1.9 [intro.execution] p.13 states:

If A is not sequenced before B and B is not sequenced before A, then A and B are unsequenced. [ Note: The execution of unsequenced evaluations can overlap. — end note ]

Overlapping executions can make it impossible to construct the sequence described in 29.4 [atomics.lockfree] p.8. We are not sure of the intention here and do not offer a suggestion for change, but note that 29.4 [atomics.lockfree] p.8 is the condition that prevents out-of-thin-air reads.

For an example, suppose we have a function invocation f(e1,e2). The evaluations of e1 and e2 can overlap. Suppose that the evaluation of e1 writes y and reads x whereas the evaluation of e2 reads y and writes x, with reads-from edges as below (all this is within a single thread).

 e1           e2
Wrlx y--   --Wrlx x
      rf\ /rf
         X
        / \
Rrlx x<-   ->Rrlx y

This seems like it should be allowed, but there seems to be no way to produce a sequence of evaluations with the property above.

In more detail, here the two evaluations, e1 and e2, are being executed as the arguments of a function and are consequently not sequenced-before each other. In practice we'd expect that they could overlap (as allowed by 1.9 [intro.execution] p.13), with the two writes taking effect before the two reads. However, if we have to construct a linear order of evaluations, as in 29.4 [atomics.lockfree] p.8, then the execution above is not permited. Is that really intended?

[ Resolution proposed by ballot comment ]

Please clarify.

[2011-03-09 Hans comments:]

I'm not proud of 29.3 [atomics.order] p9 (formerly p8), and I agree with the comments that this isn't entirely satisfactory. 29.3 [atomics.order] p9 was designed to preclude out-of-thin-air results for races among memory_order_relaxed atomics, in spite of the fact that Java experience has shown we don't really know how to do that adequately. In the long run, we probably want to revisit this.

However, in the short term, I'm still inclined to declare this NAD, for two separate reasons:

  1. 1.9 [intro.execution] p15 states: "If a side effect on a scalar object is unsequenced relative to either another side effect on the same scalar object or a value computation using the value of the same scalar object, the behavior is undefined." I think the examples presented here have undefined behavior as a result. It's not completely clear to me whether examples can be constructed that exhibit this problem, and don't have undefined behavior.

  2. This comment seems to be using a different meaning of "evaluation" from what is used elsewhere in the standard. The sequence of evaluations here doesn't have to consist of full expression evaluations. They can be evaluations of operations like lvalue to rvalue conversion, or individual assignments. In particular, the reads and writes executed by e1 and e2 in the example could be treated as separate evaluations for purposes of producing the sequence. The definition of "sequenced before" in 1.9 [intro.execution] makes little sense if the term "evaluation" is restricted to any notion of complete expression. Perhaps we should add yet another note to clarify this? 29.3 [atomics.order] p10 probably leads to the wrong impression here.

    An alternative resolution would be to simply delete our flakey attempt at preventing out-of-thin-air reads, by removing 29.3 [atomics.order] p9-11, possibly adding a note that explains that we technically allow, but strongly discourage them. If we were starting this from scratch now, that would probably be my preference. But it seems like too drastic a resolution at this stage.

[2011-03-24 Madrid]

Moved to NAD Future

Proposed resolution:


1484. Need a way to join a thread with a timeout

Section: 30.3.1 [thread.thread.class] Status: LEWG Submitter: INCITS Opened: 2010-08-25 Last modified: 2016-02-12

Priority: Not Prioritized

View all issues with LEWG status.

Discussion:

Addresses US-183

There is no way to join a thread with a timeout.

[ Resolution proposed by ballot comment: ]

Add join_for and join_until. Or decide one should never join a thread with a timeout since pthread_join doesn't have a timeout version.

[ 2010 Batavia ]

The concurrency working group deemed this an extension beyond the scope of C++0x.

Rationale:

The LWG does not wish to make a change at this time.

Proposed resolution:


1488. Improve interoperability between the C++0x and C1x threads APIs

Section: 30.4 [thread.mutex] Status: LEWG Submitter: INCITS Opened: 2010-08-25 Last modified: 2016-02-12

Priority: Not Prioritized

View all other issues in [thread.mutex].

View all issues with LEWG status.

Discussion:

Addresses US-185

Cooperate with WG14 to improve interoperability between the C++0x and C1x threads APIs. In particular, C1x mutexes should be conveniently usable with a C++0x lock_guard. Performance overheads for this combination should be considered.

[ Resolution proposed by ballot comment: ]

Remove C++0x timed_mutex and timed_recursive_mutex if that facilitates development of more compatible APIs.

[ 2010 Batavia ]

The concurrency sub-group reviewed the options, and decided that closer harmony should wait until both standards are published.

Rationale:

The LWG does not wish to make any change at this time.

Proposed resolution:


1493. Add mutex, recursive_mutex, is_locked function

Section: 30.4.1 [thread.mutex.requirements] Status: LEWG Submitter: INCITS Opened: 2010-08-25 Last modified: 2016-02-12

Priority: Not Prioritized

View other active issues in [thread.mutex.requirements].

View all other issues in [thread.mutex.requirements].

View all issues with LEWG status.

Discussion:

Addresses US-189

mutex and recursive_mutex should have an is_locked() member function. is_locked allows a user to test a lock without acquiring it and can be used to implement a lightweight try_try_lock.

[ Resolution proposed by ballot comment: ]

Add a member function:

bool is_locked() const;

to std::mutex and std::recursive_mutex. These functions return true if the current thread would not be able to obtain a mutex. These functions do not synchronize with anything (and, thus, can avoid a memory fence).

[ 2010 Batavia ]

The Concurrency subgroup reviewed this issue and deemed it to be an extension to be handled after publishing C++0x.

Rationale:

The LWG does not wish to make a change at this time.

Proposed resolution:


1499. Condition variables preclude wakeup optimization

Section: 30.5 [thread.condition] Status: LEWG Submitter: INCITS Opened: 2010-08-25 Last modified: 2016-02-12

Priority: Not Prioritized

View all other issues in [thread.condition].

View all issues with LEWG status.

Discussion:

Addresses US-193

Condition variables preclude a wakeup optimization.

[ Resolution proposed by ballot comment: ]

Change condition_variable to allow such optimization. See Appendix 1 - Additional Details

[ 2010 Batavia ]

The Concurrency subgroup reviewed the issue, and deemed it an extension to be handled after C++0x.

Rationale:

The LWG does not wish to make the change at this time.

Proposed resolution:


1521. Requirements on internal pointer representations in containers

Section: 23.2.1 [container.requirements.general] Status: LEWG Submitter: Mike Spertus Opened: 2010-10-16 Last modified: 2016-02-12

Priority: Not Prioritized

View other active issues in [container.requirements.general].

View all other issues in [container.requirements.general].

View all issues with LEWG status.

Discussion:

Addresses US-104, US-141

The standard doesn't say that containers should use abstract pointer types internally. Both Howard and Pablo agree that this is the intent. Further, it is necessary for containers to be stored, for example, in shared memory with an interprocess allocator (the type of scenario that allocators are intended to support).

In spite of the (possible) agreement on intent, it is necessary to make this explicit:

An implementations may like to store the result of dereferencing the pointer (which is a raw reference) as an optimization, but that prevents the data structure from being put in shared memory, etc. In fact, a container could store raw references to the allocator, which would be a little weird but conforming as long as it has one by-value copy. Furthermore, pointers to locales, ctypes, etc. may be there, which also prevents the data structure from being put in shared memory, so we should make explicit that a container does not store raw pointers or references at all.

[ Pre-batavia ]

This issue is being opened as part of the response to NB comments US-104/141. See paper N3171 in the pre-Batavia mailing.

[2011-03-23 Madrid meeting]

Deferred

[ 2011 Batavia ]

This may be an issue, but it is not clear. We want to gain a few years experience with the C++11 allocator model to see if this is already implied by the existing specification.

Proposed resolution:

Add to the end of 23.2.1 [container.requirements.general] p. 8:

[..] In all container types defined in this Clause, the member get_allocator() returns a copy of the allocator used to construct the container or, if that allocator has been replaced, a copy of the most recent replacement. The container may not store internal objects whose types are of the form T * or T & except insofar as they are part of the item type or members.


2035. Output iterator requirements are broken

Section: 24.2.4 [output.iterators] Status: Open Submitter: Daniel Krügler Opened: 2011-02-27 Last modified: 2016-02-12

Priority: 3

View other active issues in [output.iterators].

View all other issues in [output.iterators].

View all issues with Open status.

Discussion:

During the Pittsburgh meeting the proposal N3066 became accepted because it fixed several severe issues related to the iterator specification. But the current working draft (N3225) does not reflect all these changes. Since I'm unaware whether every correction can be done editorial, this issue is submitted to take care of that. To give one example: All expressions of Table 108 — "Output iterator requirements" have a post-condition that the iterator is incrementable. This is impossible, because it would exclude any finite sequence that is accessed by an output iterator, such as a pointer to a C array. The N3066 wording changes did not have these effects.

[2011-03-01: Daniel comments:]

This issue has some overlap with the issue 2038 and I would prefer if we could solve both at one location. I suggest the following approach:

  1. The terms dereferencable and incrementable could be defined in a more general way not restricted to iterators (similar to the concepts HasDereference and HasPreincrement from working draft N2914). But on the other hand, all current usages of dereferencable and incrementable are involved with types that satisfy iterator requirements. Thus, I believe that it is sufficient for C++0x to add corresponding definitions to 24.2.1 [iterator.requirements.general] and to let all previous usages of these terms refer to this sub-clause. Since the same problem occurs with the past-the-end iterator, this proposal suggest providing similar references to usages that precede its definition as well.

  2. We also need to ensure that all iterator expressions get either an operational semantics in terms of others or we need to add missing pre- and post-conditions. E.g. we have the following ones without semantics:

    *r++ = o // output iterator
    *r--     // bidirectional iterator
    

    According to the SGI specification these correspond to

    { *r = o; ++r; }                         // output iterator
    { reference tmp = *r; --r; return tmp; } // bidirectional iterator
    

    respectively. Please note especially the latter expression for bidirectional iterator. It fixes a problem that we have for forward iterator as well: Both these iterator categories provide stronger guarantees than input iterator, because the result of the dereference operation is reference, and not only convertible to the value type (The exact form from the SGI documentation does not correctly refer to reference).

[2011-03-14: Daniel comments and updates the suggested wording]

In addition to the before mentioned necessary changes there is another one need, which became obvious due to issue 2042: forward_list<>::before_begin() returns an iterator value which is not dereferencable, but obviously the intention is that it should be incrementable. This leads to the conclusion that imposing dereferencable as a requirement for the expressions ++r is wrong: We only need the iterator to be incrementable. A similar conclusion applies to the expression --r of bidirectional iterators.

[ 2011 Bloomington ]

Consensus this is the correct direction, but there are (potentially) missing incrementable preconditions on some table rows, and the Remarks on when an output iterator becomes dereferencable are probably better handled outside the table, in a manner similar to the way we word for input iterators.

There was some concern about redundant pre-conditions when the operational semantic is defined in terms of operations that have preconditions, and a similar level of concern over dropping such redundancies vs. applying a consistent level of redundant specification in all the iterator tables. Wording clean-up in either direction would be welcome.

[2011-08-18: Daniel adapts the proposed resolution to honor the Bloomington request]

There is only a small number of further changes suggested to get rid of superfluous requirements and essentially non-normative assertions. Operations should not have extra pre-conditions, if defined by "in-terms-of" semantics, see e.g. a != b or a->m for Table 107. Further, some remarks, that do not impose anything or say nothing new have been removed, because I could not find anything helpful they provide. E.g. consider the remarks for Table 108 for the operations dereference-assignment and preincrement: They don't provide additional information say nothing surprising. With the new pre-conditions and post-conditions it is implied what the remarks intend to say.

[ 2011-11-03: Some observations from Alexander Stepanov via c++std-lib-31405 ]

The following sentence is dropped from the standard section on OutputIterators:

"In particular, the following two conditions should hold: first, any iterator value should be assigned through before it is incremented (this is, for an output iterator i, i++; i++; is not a valid code sequence); second, any value of an output iterator may have at most one active copy at any given time (for example, i = j; *++i = a; *j = b; is not a valid code sequence)."

[ 2011-11-04: Daniel comments and improves the wording ]

In regard to the first part of the comment, the intention of the newly proposed wording was to make clear that for the expression

*r = o

we have the precondition dereferenceable and the post-condition incrementable. And for the expression

++r

we have the precondition incrementable and the post-condition dereferenceable or past-the-end. This should not allow for a sequence like i++; i++; but I agree that it doesn't exactly say that.

In regard to the second point: To make this point clearer, I suggest to add a similar additional wording as we already have for input iterator to the "Assertion/note" column of the expression ++r:

"Post: any copies of the previous value of r are no longer required to be dereferenceable or incrementable."

The proposed has been updated to honor the observations of Alexander Stepanov.

[2015-02 Cologne]

The matter is complicated, Daniel volunteers to write a paper.

Proposed resolution:

  1. Add a reference to 24.2.1 [iterator.requirements.general] to the following parts of the library preceding Clause 24 Iterators library: (I stopped from 23.2.5 [unord.req] on, because the remaining references are the concrete containers)

    1. 17.6.3.2 [swappable.requirements] p5:

      -5- A type X satisfying any of the iterator requirements (24.2) is ValueSwappable if, for any dereferenceable (24.2.1 [iterator.requirements.general]) object x of type X, *x is swappable.

    2. 17.6.3.5 [allocator.requirements], Table 27 — "Descriptive variable definitions", row with the expression c:

      a dereferenceable (24.2.1 [iterator.requirements.general]) pointer of type C*

    3. 20.10.3.2 [pointer.traits.functions]:

      Returns: The first template function returns a dereferenceable (24.2.1 [iterator.requirements.general]) pointer to r obtained by calling Ptr::pointer_to(r); […]

    4. 21.3.1.3 [string.iterators] p. 2:

      Returns: An iterator which is the past-the-end value (24.2.1 [iterator.requirements.general]).

    5. 22.4.5.1.2 [locale.time.get.virtuals] p. 11:

      iter_type do_get(iter_type s, iter_type end, ios_base& f,
        ios_base::iostate& err, tm *t, char format, char modifier) const;
      

      Requires: t shall be dereferenceable (24.2.1 [iterator.requirements.general]).

    6. 23.2.1 [container.requirements.general] p. 6:

      […] end() returns an iterator which is the past-the-end (24.2.1 [iterator.requirements.general]) value for the container. […]

    7. 23.2.3 [sequence.reqmts] p. 3:

      […] q denotes a valid dereferenceable (24.2.1 [iterator.requirements.general]) const iterator to a, […]

    8. 23.2.4 [associative.reqmts] p. 8 (I omit intentionally one further reference in the same sub-clause):

      […] q denotes a valid dereferenceable (24.2.1 [iterator.requirements.general]) const iterator to a, […]

    9. 23.2.5 [unord.req] p. 10 (I omit intentionally one further reference in the same sub-clause):

      […] q and q1 are valid dereferenceable (24.2.1 [iterator.requirements.general]) const iterators to a, […]

  2. Edit 24.2.1 [iterator.requirements.general] p. 5 as indicated (The intent is to properly define incrementable and to ensure some further library guarantee related to past-the-end iterator values):

    -5- Just as a regular pointer to an array guarantees that there is a pointer value pointing past the last element of the array, so for any iterator type there is an iterator value that points past the last element of a corresponding sequence. These values are called past-the-end values. Values of an iterator i for which the expression *i is defined are called dereferenceable. Values of an iterator i for which the expression ++i is defined are called incrementable. The library never assumes that past-the-end values are dereferenceable or incrementable. Iterators can also have singular values that are not associated with any sequence. […]

  3. Modify the column contents of Table 106 — "Iterator requirements", 24.2.2 [iterator.iterators], as indicated:

    Table 106 — Iterator requirements
    Expression Return type Operational semantics Assertion/note
    pre-/post-condition
    *r reference   pre: r is dereferenceable.
    ++r X&   pre: r is incrementable.
  4. Modify the column contents of Table 107 — "Input iterator requirements", 24.2.3 [input.iterators], as indicated [Rationale: The wording changes attempt to define a minimal "independent" set of operations, namely *a and ++r, and to specify the semantics of the remaining ones. This approach seems to be in agreement with the original SGI specificationend rationale]:

    Table 107 — Input iterator requirements (in addition to Iterator)
    Expression Return type Operational semantics Assertion/note
    pre-/post-condition
    a != b contextually
    convertible to bool
    !(a == b) pre: (a, b) is in the domain
    of ==.
    *a convertible to T   pre: a is dereferenceable.
    The expression
    (void)*a, *a is equivalent
    to *a.
    If a == b and (a,b) is in
    the domain of == then *a is
    equivalent to *b.
    a->m   (*a).m pre: a is dereferenceable.
    ++r X&   pre: r is dereferenceableincrementable.
    post: r is dereferenceable or
    r is past-the-end.
    post: any copies of the
    previous value of r are no
    longer required either to be
    dereferenceable, incrementable,
    or to be in the domain of ==.
    (void)r++   (void)++r equivalent to (void)++r
    *r++ convertible to T { T tmp = *r;
    ++r;
    return tmp; }
     
  5. Modify the column contents of Table 108 — "Output iterator requirements", 24.2.4 [output.iterators], as indicated [Rationale: The wording changes attempt to define a minimal "independent" set of operations, namely *r = o and ++r, and to specify the semantics of the remaining ones. This approach seems to be in agreement with the original SGI specificationend rationale]:

    Table 108 — Output iterator requirements (in addition to Iterator)
    Expression Return type Operational semantics Assertion/note
    pre-/post-condition
    *r = o result is not used   pre: r is dereferenceable.
    Remark: After this operation
    r is not required to be
    dereferenceable and any copies of
    the previous value of r are no
    longer required to be dereferenceable
    or incrementable.

    post: r is incrementable.
    ++r X&   pre: r is incrementable.
    &r == &++r.
    Remark: After this operation
    r is not required to be
    dereferenceable.
    Remark: After this operation
    r is not required to be
    incrementable and any copies of
    the previous value of r are no
    longer required to be dereferenceable
    or incrementable.

    post: r is dereferenceable
    or r is past-the-end
    incrementable.
    r++ convertible to const X& { X tmp = r;
    ++r;
    return tmp; }
    Remark: After this operation
    r is not required to be
    dereferenceable.
    post: r is incrementable.
    *r++ = o result is not used { *r = o; ++r; } Remark: After this operation
    r is not required to be
    dereferenceable.
    post: r is incrementable.
  6. Modify the column contents of Table 109 — "Forward iterator requirements", 24.2.5 [forward.iterators], as indicated [Rationale: Since the return type of the expression *r++ is now guaranteed to be type reference, the implied operational semantics from input iterator based on value copies is wrong — end rationale]

    Table 109 — Forward iterator requirements (in addition to input iterator)
    Expression Return type Operational semantics Assertion/note
    pre-/post-condition
    r++ convertible to const X& { X tmp = r;
    ++r;
    return tmp; }
     
    *r++ reference { reference tmp = *r;
    ++r;
    return tmp; }
     
  7. Modify the column contents of Table 110 — "Bidirectional iterator requirements", 24.2.6 [bidirectional.iterators], as indicated:

    Table 110 — Bidirectional iterator requirements (in addition to forward iterator)
    Expression Return type Operational semantics Assertion/note
    pre-/post-condition
    --r X&   pre: there exists s such that
    r == ++s.
    post: r is dereferenceableincrementable.
    --(++r) == r.
    --r == --s implies r == s.
    &r == &--r.
    r-- convertible to const X& { X tmp = r;
    --r;
    return tmp; }
     
    *r-- reference { reference tmp = *r;
    --r;
    return tmp; }
     

2038. Missing definition for incrementable iterator

Section: 24.2.4 [output.iterators] Status: Open Submitter: Pete Becker Opened: 2011-02-27 Last modified: 2016-02-12

Priority: 3

View other active issues in [output.iterators].

View all other issues in [output.iterators].

View all issues with Open status.

Discussion:

In comp.lang.c++, Vicente Botet raises the following questions:

"In "24.2.4 Output iterators" there are 3 uses of incrementable. I've not found the definition. Could some one point me where it is defined?

Something similar occurs with dereferenceable. While the definition is given in "24.2.1 In general" it is used several times before.

Shouldn't these definitions be moved to some previous section?"

He's right: both terms are used without being properly defined.

There is no definition of "incrementable".

While there is a definition of "dereferenceable", it is, in fact, a definition of "dereferenceable iterator". "dereferenceable" is used throughout Clause 23 (Containers) before its definition in Clause 24. In almost all cases it's referring to iterators, but in 17.6.3.2 [swappable.requirements] there is a mention of "dereferenceable object"; in 17.6.3.5 [allocator.requirements] the table of Descriptive variable definitions refers to a "dereferenceable pointer"; 20.10.3.2 [pointer.traits.functions] refers to a "dereferenceable pointer"; in 22.4.5.1.2 [locale.time.get.virtuals]/11 (do_get) there is a requirement that a pointer "shall be dereferenceable". In those specific cases it is not defined.

[2011-03-02: Daniel comments:]

I believe that the currently proposed resolution of issue 2035 solves this issue as well.

[ 2011 Bloomington ]

Agree with Daniel, this will be handled by the resolution of 2035.

Proposed resolution:


2040. Missing type traits related to is_convertible

Section: 20.15 [meta] Status: LEWG Submitter: Daniel Krügler Opened: 2011-03-03 Last modified: 2016-02-12

Priority: Not Prioritized

View other active issues in [meta].

View all other issues in [meta].

View all issues with LEWG status.

Discussion:

When n3142 was suggested, it concentrated on constructions, assignments, and destructions, but overlooked to complement the single remaining compiler-support trait

template <class From, class To> struct is_convertible;

with the no-throw and triviality related aspects as it had been done with the other expression-based traits. Specifically, the current specification misses to add the following traits:

template <class From, class To> struct is_nothrow_convertible;
template <class From, class To> struct is_trivially_convertible;

In particular the lack of is_nothrow_convertible is severly restricting. This was recently recognized when the proposal for decay_copy was prepared by n3255. There does not exist a portable means to define the correct conditional noexcept specification for the decay_copy function template, which is declared as:

template <class T> 
typename decay<T>::type decay_copy(T&& v) noexcept(???);

The semantics of decay_copy bases on an implicit conversion which again influences the overload set of functions that are viable here. In most circumstances this will have the same effect as comparing against the trait std::is_nothrow_move_constructible, but there is no guarantee for that being the right answer. It is possible to construct examples, where this would lead to the false result, e.g.

struct S {
  S(const S&) noexcept(false);
 
  template<class T>
  explicit S(T&&) noexcept(true);
};

std::is_nothrow_move_constructible will properly honor the explicit template constructor because of the direct-initialization context which is part of the std::is_constructible definition and will in this case select it, such that std::is_nothrow_move_constructible<S>::value == true, but if we had the traits is_nothrow_convertible, is_nothrow_convertible<S, S>::value would evaluate to false, because it would use the copy-initialization context that is part of the is_convertible definition, excluding any explicit constructors and giving the opposite result.

The decay_copy example is surely not one of the most convincing examples, but is_nothrow_convertible has several use-cases, and can e.g. be used to express whether calling the following implicit conversion function could throw an exception or not:

template<class T, class U>
T implicit_cast(U&& u) noexcept(is_nothrow_convertible<U, T>::value) 
{
  return std::forward<U>(u);
}

Therefore I suggest to add the missing trait is_nothrow_convertible and for completeness also the missing trait is_trivially_convertible to 20.15 [meta].

[2011-03-24 Madrid meeting]

Daniel K: This is a new feature so out of scope.

Pablo: Any objections to moving 2040 to Open?

No objections.

[Bloomington, 2011]

Move to NAD Future, this would be an extension to existing functionality.

Proposed resolution:

  1. Ammend the following declarations to the header <type_traits> synopsis in 20.15.2 [meta.type.synop]:

    namespace std {
      …
      // 20.9.6, type relations:
      template <class T, class U> struct is_same;
      template <class Base, class Derived> struct is_base_of;
      template <class From, class To> struct is_convertible;
      template <class From, class To> struct is_trivially_convertible;
      template <class From, class To> struct is_nothrow_convertible;
    
      …
    }
    
  2. Modify Table 51 — "Type relationship predicates" as indicated. The removal of the remaining traces of the trait is_explicitly_convertible is an editorial step, it was removed by n3047:

    Table 51 — Type relationship predicates
    Template Condition Comments
    template <class From, class To>
    struct is_convertible;
    see below From and To shall be complete
    types, arrays of unknown bound, or
    (possibly cv-qualified) void
    types.
    template <class From, class To>
    struct is_explicitly_convertible;
    is_constructible<To, From>::value a synonym for a two-argument
    version of is_constructible.
    An implementation may define it
    as an alias template.
    template <class From, class To>
    struct is_trivially_convertible;
    is_convertible<From,
    To>::value
    is true and the
    conversion, as defined by
    is_convertible, is known
    to call no operation that is
    not trivial ([basic.types], [special]).
    From and To shall be complete
    types, arrays of unknown bound,
    or (possibly cv-qualified) void
    types.
    template <class From, class To>
    struct is_nothrow_convertible;
    is_convertible<From,
    To>::value
    is true and the
    conversion, as defined by
    is_convertible, is known
    not to throw any
    exceptions ([expr.unary.noexcept]).
    From and To shall be complete
    types, arrays of unknown bound,
    or (possibly cv-qualified) void
    types.

2055. std::move in std::accumulate and other algorithms

Section: 26.8 [numeric.ops] Status: LEWG Submitter: Chris Jefferson Opened: 2011-01-01 Last modified: 2016-02-12

Priority: Not Prioritized

View all other issues in [numeric.ops].

View all issues with LEWG status.

Discussion:

The C++0x draft says std::accumulate uses: acc = binary_op(acc, *i).

Eelis van der Weegen has pointed out, on the libstdc++ mailing list, that using acc = binary_op(std::move(acc), *i) can lead to massive improvements (particularly, it means accumulating strings is linear rather than quadratic).

Consider the simple case, accumulating a bunch of strings of length 1 (the same argument holds for other length buffers). For strings s and t, s+t takes time length(s)+length(t), as you have to copy both s and t into a new buffer.

So in accumulating n strings, step i adds a string of length i-1 to a string of length 1, so takes time i.

Therefore the total time taken is: 1+2+3+...+n = O(n2)

std::move(s)+t, for a "good" implementation, is amortized time length(t), like vector, just copy t onto the end of the buffer. So the total time taken is:

1+1+1+...+1 (n times) = O(n). This is the same as push_back on a vector.

I'm trying to decide if this implementation might already be allowed. I suspect it might not be (although I can't imagine any sensible code it would break). There are other algorithms which could benefit similarly (inner_product, partial_sum and adjacent_difference are the most obvious).

Is there any general wording for "you can use rvalues of temporaries"?

The reflector discussion starting with message c++std-lib-29763 came to the conclusion that above example is not covered by the "as-if" rules and that enabling this behaviour would seem quite useful.

[ 2011 Bloomington ]

Moved to NAD Future. This would be a larger change than we would consider for a simple TC.

Proposed resolution:


2062. Effect contradictions w/o no-throw guarantee of std::function swaps

Section: 20.14.12.2 [func.wrap.func], 20.14.12.2.2 [func.wrap.func.mod] Status: Open Submitter: Daniel Krügler Opened: 2011-05-28 Last modified: 2016-02-12

Priority: 2

View other active issues in [func.wrap.func].

View all other issues in [func.wrap.func].

View all issues with Open status.

Discussion:

Howard Hinnant observed in reflector message c++std-lib-30841 that 20.14.12.2 [func.wrap.func] makes the member swap noexcept, even though the non-member swap is not noexcept.

The latter was an outcome of the discussions during the Batavia meeting and the Madrid meeting involving LWG 1349, which seems to indicate that the remaining noexcept specifier at the member swap is incorrect and should be removed.

But if we allow for a potentially throwing member swap of std::function, this causes another conflict with the exception specification for the following member function:

template<class F> function& operator=(reference_wrapper<F> f) noexcept;

Effects: function(f).swap(*this);

Note that in this example the sub-expression function(f) does not cause any problems, because of the nothrow-guarantee given in 20.14.12.2.1 [func.wrap.func.con] p. 10. The problem is located in the usage of the swap which could potentially throw given the general latitude.

So, either the Madrid meeting decision need to be revised (and both member and free swap of std::function should be noexcept), or this function needs to be adapted as well, e.g. by taking the exception-specification away or by changing the semantics.

One argument for "swap-may-throw" would be to allow for small-object optimization techniques where the copy of the target may throw. But given the fact that the swap function has been guaranteed to be "Throws: Nothing" from TR1 on, it seems to me that that there would still be opportunities to perform small-object optimizations just restricted to the set of target copies that cannot throw.

In my opinion member swap of std::function has always been intended to be no-throw, because otherwise there would be no good technical reason to specify the effects of several member functions in terms of the "construct-swap" idiom (There are three functions that are defined this way), which provides the strong exception safety in this case. I suggest to enforce that both member swap and non-member swap of std::function are nothrow functions as it had been guaranteed since TR1 on.

[ 2011 Bloomington ]

Dietmar: May not be swappable in the first place.

Alisdair: This is wide contact. Then we should be taking noexcept off instead of putting it on. This is preferred resolution.

Pablo: This is bigger issue. Specification of assignment in terms of swap is suspect to begin with. It is over specification. How this was applied to string is a better example to work from.

Pablo: Two problems: inconsistency that should be fixed (neither should have noexcept), the other issues is that assignment should not be specified in terms of swap. There are cases where assignment should succeed where swap would fail. This is easier with string as it should follow container rules.

Action Item (Alisdair): There are a few more issues found to file.

Dave: This is because of allocators? The allocator makes this not work.

Howard: There is a type erased allocator in shared_ptr. There is a noexcept allocator in shared_ptr.

Pablo: shared_ptr is a different case. There are shared semantics and the allocator does move around. A function does not have shared semantics.

Alisdair: Function objects think they have unique ownership.

Howard: In function we specify semantics with copy construction and swap.

Action Item (Pablo): Write this up better (why assignment should not be defined in terms of swap)

Howard: Not having trouble making function constructor no throw.

Dietmar: Function must allocate memory.

Howard: Does not put stuff that will throw on copy or swap in small object optimization. Put those on heap. Storing allocator, but has to be no throw copy constructable.

Pablo: Are you allowed to or required to swap or move allocators in case or swap or move.

Dave: An allocator that is type erased should be different...

Pablo: it is

Dave: Do you need to know something about allocator types? But only at construction time.

Pablo: You could have allocators that are different types.

Dave: Swap is two ended operation.

Pablo: Opinion is that both have to say propagate on swap for them to swap.

John: It is not arbitrary. If one person says no. No is no.

Howard: Find noexcept swap to be very useful. Would like to move in that direction and bring container design along.

Dave: If you have something were allocator must not propagate you can detect that at construction time.

...

Pablo: Need to leave this open and discuss in smaller group.

Alisdair: Tried to add boost::any as TR2 proposal and ran into this issue. Only the first place where we run into issues with type erased allocators. Suggest we move it to open.

Action Item: Move to open.

Action Item (Pablo works with Howard and Daniel): Address the more fundamental issue (which may be multiple issues) and write up findings.

[ Original resolution: ]

This wording is relative to the FDIS.

  1. Modify the header <functional> synopsis in 20.14 [function.objects] as indicated:

    namespace std {
      […]
    
      template<class R, class... ArgTypes>
      void swap(function<R(ArgTypes...)>&, function<R(ArgTypes...)>&) noexcept;
    
      […]
    }
    
  2. Modify the class template function synopsis in 20.14.12.2 [func.wrap.func] as indicated:

    namespace std {
      […]
    
      // [func.wrap.func.alg], specialized algorithms:
      template<class R, class... ArgTypes>
      void swap(function<R(ArgTypes...)>&, function<R(ArgTypes...)>&) noexcept;
    
      […]
    }
    
  3. Modify 20.14.12.2.7 [func.wrap.func.alg] as indicated:

    template<class R, class... ArgTypes>
    void swap(function<R(ArgTypes...)>& f1, function<R(ArgTypes...)>& f2) noexcept;
    

    -1- Effects: f1.swap(f2);

[2014-02-28 (Post Issaquah), Pablo provides more information]

For cross-referencing purposes: The resolution of this issue should be harmonized with any resolution to LWG 2370, which addresses inappropriate noexcepts in some function constructors.

We have the following choices:

  1. swap() does not throw

    Discussion: This definition is desirable, and allows assignment to be implemented with the strong exception guarantee, but it does have consequences: The implementation cannot use the small-object optimization for a function-object F unless F is NothrowMovable (nothrow-swappable is unimportant because F is not swapped with another F). Note that many functors written before C++11 will not have move constructors decorated with noexcept, so this limitation could affect a lot of code.

    It is not clear what other implementation restrictions might be needed. Allocators are required not to throw on move or copy. Is that sufficient?

  2. swap() can throw

    Discussion: This definition gives maximum latitude to implementation to use small-object optimization. However, the strong guarantee on assignment is difficult to achieve. Should we consider giving up on the strong guarantee? How much are we willing to pessimize code for exceptions?

  3. swap() will not throw if both functions have NoThrowMoveable functors

    Discussion: This definition is similar to option 2, but gives slightly stronger guarantees. Here, swap() can throw, but the programmer can theoretically prevent that from happening. This should be straight-forward to implement and gives the implementation a lot of latitude for optimization. However, because this is a dynamic decision, the program is not as easy to reason about. Also, the strong guarantee for assignment is compromized as in option 2.

Proposed resolution:


2070. allocate_shared should use allocator_traits<A>::construct

Section: 20.11.2.2.6 [util.smartptr.shared.create] Status: Open Submitter: Jonathan Wakely Opened: 2011-07-11 Last modified: 2016-03-08

Priority: 2

View other active issues in [util.smartptr.shared.create].

View all other issues in [util.smartptr.shared.create].

View all issues with Open status.

Discussion:

20.11.2.2.6 [util.smartptr.shared.create] says:

-2- Effects: Allocates memory suitable for an object of type T and constructs an object in that memory via the placement new expression ::new (pv) T(std::forward<Args>(args)...). The template allocate_shared uses a copy of a to allocate memory. If an exception is thrown, the functions have no effect.

This explicitly requires placement new rather than using allocator_traits<A>::construct(a, (T*)pv, std::forward<Args>(args)...) In most cases that would result in the same placement new expression, but would allow more control over how the object is constructed e.g. using scoped_allocator_adaptor to do uses-allocator construction, or using an allocator declared as a friend to construct objects with no public constructors.

[2011-08-16 Bloomington:]

Agreed to fix in principle, but believe that make_shared and allocate_shared have now diverged enough that their descriptions should be separated. Pablo and Stefanus to provide revised wording.

Daniel's (old) proposed resolution:

This wording is relative to the FDIS.

  1. Change the following paragraphs of 20.11.2.2.6 [util.smartptr.shared.create] as indicated (The suggested removal of the last sentence of p1 is not strictly required to resolve this issue, but is still recommended, because it does not say anything new but may give the impression that it says something new):

    template<class T, class... Args> shared_ptr<T> make_shared(Args&&... args);
    template<class T, class A, class... Args>
      shared_ptr<T> allocate_shared(const A& a, Args&&... args);
    

    -1- Requires: For the template make_shared, tThe expression ::new (pv) T(std::forward<Args>(args)...), where pv has type void* and points to storage suitable to hold an object of type T, shall be well formed. For the template allocate_shared, the expression allocator_traits<A>::construct(a, pt, std::forward<Args>(args)...), where pt has type T* and points to storage suitable to hold an object of type T, shall be well formed. A shall be an allocator ([allocator.requirements]). The copy constructor and destructor of A shall not throw exceptions.

    -2- Effects: Allocates memory suitable for an object of type T and constructs an object in that memory. The template make_shared constructs the object via the placement new expression ::new (pv) T(std::forward<Args>(args)...). The template allocate_shared uses a copy of a to allocate memory and constructs the object by calling allocator_traits<A>::construct(a, pt, std::forward<Args>(args)...). If an exception is thrown, the functions have no effect.

    -3- Returns: A shared_ptr instance that stores and owns the address of the newly constructed object of type T.

    -4- Postconditions: get() != 0 && use_count() == 1

    -5- Throws: bad_alloc, or, for the template make_shared, an exception thrown from the constructor of T, or, for the template allocate_shared, an exception thrown from A::allocate or from allocator_traits<A>::constructfrom the constructor of T.

    -6- Remarks: Implementations are encouraged, but not required, to perform no more than one memory allocation. [ Note: This provides efficiency equivalent to an intrusive smart pointer. — end note ]

    -7- [ Note: These functions will typically allocate more memory than sizeof(T) to allow for internal bookkeeping structures such as the reference counts. — end note ]

[2011-12-04: Jonathan and Daniel improve wording]

See also c++std-lib-31796

[2013-10-13, Ville]

This issue is related to 2089.

[2014-02-15 post-Issaquah session : move to Tentatively NAD]

STL: This takes an allocator, but then ignores its construct. That's squirrely.

Alisdair: The convention is when you take an allocator, you use its construct.

STL: 23.2.1 [container.requirements.general]/3, argh! This fills me with despair, but I understand it now.

STL: Ok, this is some cleanup.

STL: You're requiring b to be of type A and not being rebound, is that an overspecification?

Pablo: Good point. Hmm, that's only a requirement on what must be well-formed.

STL: If it's just a well-formed requirement, then why not just use a directly?

Pablo: Yeah, the well-formed requirement is overly complex. It's not a real call, we could just use a directly. It makes it harder to read.

Alisdair: b should be an allocator in the same family as a.

Pablo: This is a well-formed requirement, I wonder if it's the capital A that's the problem here. It doesn't matter here, this is way too much wording.

Alisdair: It's trying to tie the constructor arguments into the allocator requirements.

Pablo: b could be struck, that's a runtime quality. The construct will work with anything that's in the family of A.

Alisdair: The important part is the forward of Args.

Pablo: A must be an allocator, and forward Args must work with that.

Alisdair: First let's nail down A.

Pablo: Then replace b with a, and strike the rest.

STL: You need pt's type, at least.

Pablo: There's nothing to be said about runtime constraints here, this function doesn't even take a pt.

STL: Looking at the Effects, I believe b is similarly messed up, we can use a2 to construct an object.

Alisdair: Or any allocator in the family of a.

STL: We say this stuff for the deallocate too, it should be lifted up.

STL: "owns the address" is weird.

Alisdair: shared_ptr owns pointers, although it does sound funky.

Walter: "to destruct" is ungrammatical.

STL: "When ownership is given up" is not what we usually say.

Alisdair: I think the Returns clause is the right place to say this.

STL: The right place to say this is shared_ptr's dtor, we don't want to use Core's "come from" convention.

Alisdair: I'm on the hook to draft cleaner wording.

[2015-10, Kona Saturday afternoon]

AM: I was going to clean up the wording, but haven't done it yet.
Defer until we have new wording.

[2016-03, Jacksonville]

Alisdair: we need to figure out whether we should call construct or not; major implementation divergence
STL: this does not grant friendship, does it?
Jonathan: some people want it.
Thomas: scoped allocator adapter should be supported, so placement new doesn't work
Alisdair: this makes the make_ functions impossible
Thomas: you don't want to use those though.
Alisdair: but people use that today, at Bloomberg
Alisdair: and what do we do about fancy pointers?
Jonathan: we constrain it to only non-fancy pointers.
STL: shared_ptr has never attempted to support fancy pointers; seems like a paper is needed.
Poll: call construct:6 operator new: 0 don't care: 4
Poll: should we support fancy pointers? Yes: 1 No: 4 don't care: 4
STL: 20.8.2.2.6p2: 'and pv->~T()' is bogus for void
STL: 20.8.2.2.6p4: is this true even if we're going to allocate a bit more?
Alisdair: yes
Alisdair: coming up with new wording

Proposed resolution:

This wording is relative to the FDIS.

  1. Change the following paragraphs of 20.11.2.2.6 [util.smartptr.shared.create] as indicated:

    template<class T, class... Args> shared_ptr<T> make_shared(Args&&... args);
    template<class T, class A, class... Args>
      shared_ptr<T> allocate_shared(const A& a, Args&&... args);
    

    -1- Requires: The expression ::new (pv) T(std::forward<Args>(args)...), where pv has type void* and points to storage suitable to hold an object of type T, shall be well formed. A shall be an allocator (17.6.3.5 [allocator.requirements]). The copy constructor and destructor of A shall not throw exceptions.

    -2- Effects: Equivalent to

     
    return allocate_shared<T>(allocator<T>(), std::forward<Args>(args)...);
    

    Allocates memory suitable for an object of type T and constructs an object in that memory via the placement new expression ::new (pv) T(std::forward<Args>(args)...). The template allocate_shared uses a copy of a to allocate memory. If an exception is thrown, the functions have no effect.

    -?- Remarks: An implementation may meet the effects (and the implied guarantees) without creating the allocator object [Note: That is, user-provided specializations of std::allocator may not be instantiated, the expressions ::new (pv) T(std::forward<Args>(args)...) and pv->~T() may be evaluated directly — end note].

    -3- Returns: A shared_ptr instance that stores and owns the address of the newly constructed object of type T.

    -4- Postconditions: get() != 0 && use_count() == 1

    -5- Throws: bad_alloc, or an exception thrown from A::allocate or from the constructor of T.

    -6- Remarks: Implementations are encouraged, but not required, to perform no more than one memory allocation. [Note: This provides efficiency equivalent to an intrusive smart pointer. — end note]

    -7- [Note: These functions will typically allocate more memory than sizeof(T) to allow for internal bookkeeping structures such as the reference counts. — end note]

  2. Add the following set of new paragraphs immediately following the previous paragraph 7 of 20.11.2.2.6 [util.smartptr.shared.create]:

    template<class T, class A, class... Args>
      shared_ptr<T> allocate_shared(const A& a, Args&&... args);
    

    -?- Requires: The expressions allocator_traits<A>::construct(b, pt, std::forward<Args>(args)...) and allocator_traits<A>::destroy(b, pt) shall be well-formed and well-defined, where b has type A and is a copy of a and where pt has type T* and points to storage suitable to hold an object of type T. A shall meet the allocator requirements (17.6.3.5 [allocator.requirements]).

    -?- Effects: Uses an object a2 of type allocator_traits<A>::rebind_alloc<unspecified> that compares equal to a to allocate memory suitable for an object of type T. Uses a copy b of type A from a to construct an object of type T in that memory by calling allocator_traits<A>::construct(b, pt, std::forward<Args>(args)...). If an exception is thrown, the function has no effect.

    -?- Returns: A shared_ptr instance that stores and owns the address of the newly constructed object of type T. When ownership is given up, the effects are as follows: Uses a copy b2 of type A from a to destruct an object of type T by calling allocator_traits<A>::destroy(b2, pt2) where pt2 has type T* and refers to the newly constructed object. Then uses an object of type allocator_traits<A>::rebind_alloc<unspecified> that compares equal to a to deallocate the allocated memory.

    -?- Postconditions: get() != 0 && use_count() == 1

    -?- Throws: Nothing unless memory allocation or allocator_traits<A>::construct throws an exception.

    -?- Remarks: Implementations are encouraged, but not required, to perform no more than one memory allocation. [Note: Such an implementation provides efficiency equivalent to an intrusive smart pointer. — end note]

    -?- [Note: This function will typically allocate more memory than sizeof(T) to allow for internal bookkeeping structures such as the reference counts. — end note]


2077. Further incomplete constraints for type traits

Section: 20.15.4.3 [meta.unary.prop] Status: Open Submitter: Daniel Krügler Opened: 2011-08-20 Last modified: 2016-02-12

Priority: 3

View other active issues in [meta.unary.prop].

View all other issues in [meta.unary.prop].

View all issues with Open status.

Discussion:

The currently agreed on proposed wording for 2015 using remove_all_extents<T>::type instead of the "an array of unknown bound" terminology in the precondition should be extended to some further entries especially in Table 49, notably the is_*constructible, is_*assignable, and is_*destructible entries. To prevent ODR violations, incomplete element types of arrays must be excluded for value-initialization and destruction for example. Construction and assignment has to be honored, when we have array-to-pointer conversions or pointer conversions of incomplete pointees in effect.

[2012, Kona]

The issue is that in three type traits, we are accidentally saying that in certain circumstances the type must give a specified answer when given an incomplete type. (Specifically: an array of unknown bound of incomplete type.) The issue asserts that there's an ODR violation, since the trait returns false in that case but might return a different version when the trait is completed.

Howard argues: no, there is no risk of an ODR violation. is_constructible<A[]> must return false regardless of whether A is complete, so there's no reason to forbid an array of unknown bound of incomplete types. Same argument applies to is_assignable. General agreement with Howard's reasoning.

There may be a real issue for is_destructible. None of us are sure what is_destructible is supposed to mean for an array of unknown bound (regardless of whether its type is complete), and the standard doesn't make it clear. The middle column doesn't say what it's supposed to do for incomplete types.

In at least one implementation, is_destructible<A[]> does return true if A is complete, which would result in ODR violation unless we forbid it for incomplete types.

Move to open. We believe there is no issue for is_constructible or is_assignable, but that there is a real issue for is_destructible.

Proposed resolution:


2088. std::terminate problem

Section: 18.8.4 [exception.terminate] Status: Open Submitter: Daniel Krügler Opened: 2011-09-25 Last modified: 2016-02-12

Priority: 3

View all issues with Open status.

Discussion:

Andrzej Krzemienski reported the following on comp.std.c++:

In N3290, which is to become the official standard, in 18.8.4.4 [terminate], paragraph 1 reads

Remarks: Called by the implementation when exception handling must be abandoned for any of several reasons (15.5.1), in effect immediately after evaluating the throw-expression (18.8.3.1). May also be called directly by the program.

It is not clear what is "in effect". It was clear in previous drafts where paragraphs 1 and 2 read:

Called by the implementation when exception handling must be abandoned for any of several reasons (15.5.1). May also be called directly by the program.

Effects: Calls the terminate_handler function in effect immediately after evaluating the throw-expression (18.8.3.1), if called by the implementation, or calls the current terminate_handler function, if called by the program.

It was changed by N3189. The same applies to function unexpected (D. 11.4, paragraph 1).

Assuming the previous wording is still intended, the wording can be read "unless std::terminate is called by the program, we will use the handler that was in effect immediately after evaluating the throw-expression".

This assumes that there is some throw-expression connected to every situation that triggers the call to std::terminate. But this is not the case:

Which one is referred to?

In case std::nested_exception::rethrow_nested is called for an object that has captured no exception, there is no throw-expression involved directly (and may no throw be involved even indirectly).

Next, 18.8.4.1 [terminate.handler], paragraph 2 says

Required behavior: A terminate_handler shall terminate execution of the program without returning to the caller.

This seems to allow that the function may exit by throwing an exception (because word "return" implies a normal return).

One could argue that words "terminate execution of the program" are sufficient, but then why "without returning to the caller" would be mentioned. In case such handler throws, noexcept specification in function std::terminate is violated, and std::terminate would be called recursively - should std::abort not be called in case of recursive std::terminate call? On the other hand some controlled recursion could be useful, like in the following technique.

The here mentioned wording changes by N3189 in regard to 18.8.4.4 [terminate] p1 were done for a better separation of effects (Effects element) and additional normative wording explanations (Remarks element), there was no meaning change intended. Further, there was already a defect existing in the previous wording, which was not updated when further situations where defined, when std::terminate where supposed to be called by the implementation.

The part

"in effect immediately after evaluating the throw-expression"

should be removed and the quoted reference to 18.8.4.1 [terminate.handler] need to be part of the effects element where it refers to the current terminate_handler function, so should be moved just after

"Effects: Calls the current terminate_handler function."

It seems ok to allow a termination handler to exit via an exception, but the suggested idiom should better be replaced by a more simpler one based on evaluating the current exception pointer in the terminate handler, e.g.

void our_terminate (void) {
  std::exception_ptr p = std::current_exception();
  if (p) {
    ... // OK to rethrow and to determine it's nature
  } else {
    ... // Do something else
  }
}

[2011-12-09: Daniel comments]

A related issue is 2111.

[2012, Kona]

Move to Open.

There is an interaction with Core issues in this area that Jens is already supplying wording for. Review this issue again once Jens wording is available.

Alisdair to review clause 15.5 (per Jens suggestion) and recommend any changes, then integrate Jens wording into this issue.

Proposed resolution:


2089. std::allocator::construct should use uniform initialization

Section: 20.10.9.1 [allocator.members] Status: EWG Submitter: David Krauss Opened: 2011-10-07 Last modified: 2016-02-12

Priority: 2

View all other issues in [allocator.members].

View all issues with EWG status.

Discussion:

When the EmplaceConstructible (23.2.1 [container.requirements.general]/13) requirement is used to initialize an object, direct-initialization occurs. Initializing an aggregate or using a std::initializer_list constructor with emplace requires naming the initialized type and moving a temporary. This is a result of std::allocator::construct using direct-initialization, not list-initialization (sometimes called "uniform initialization") syntax.

Altering std::allocator<T>::construct to use list-initialization would, among other things, give preference to std::initializer_list constructor overloads, breaking valid code in an unintuitive and unfixable way — there would be no way for emplace_back to access a constructor preempted by std::initializer_list without essentially reimplementing push_back.

std::vector<std::vector<int>> v;
v.emplace_back(3, 4); // v[0] == {4, 4, 4}, not {3, 4} as in list-initialization

The proposed compromise is to use SFINAE with std::is_constructible, which tests whether direct-initialization is well formed. If is_constructible is false, then an alternative std::allocator::construct overload is chosen which uses list-initialization. Since list-initialization always falls back on direct-initialization, the user will see diagnostic messages as if list-initialization (uniform-initialization) were always being used, because the direct-initialization overload cannot fail.

I can see two corner cases that expose gaps in this scheme. One occurs when arguments intended for std::initializer_list satisfy a constructor, such as trying to emplace-insert a value of {3, 4} in the above example. The workaround is to explicitly specify the std::initializer_list type, as in v.emplace_back(std::initializer_list<int>(3, 4)). Since this matches the semantics as if std::initializer_list were deduced, there seems to be no real problem here.

The other case is when arguments intended for aggregate initialization satisfy a constructor. Since aggregates cannot have user-defined constructors, this requires that the first nonstatic data member of the aggregate be implicitly convertible from the aggregate type, and that the initializer list have one element. The workaround is to supply an initializer for the second member. It remains impossible to in-place construct an aggregate with only one nonstatic data member by conversion from a type convertible to the aggregate's own type. This seems like an acceptably small hole.

The change is quite small because EmplaceConstructible is defined in terms of whatever allocator is specified, and there is no need to explicitly mention SFINAE in the normative text.

[2012, Kona]

Move to Open.

There appears to be a real concern with initializing aggregates, that can be performed only using brace-initialization. There is little interest in the rest of the issue, given the existence of 'emplace' methods in C++11.

Move to Open, to find an acceptable solution for intializing aggregates. There is the potential that EWG may have an interest in this area of language consistency as well.

[2013-10-13, Ville]

This issue is related to 2070.

[2015-02 Cologne]

Move to EWG, Ville to write a paper.

[2015-09, Telecom]

Ville: N4462 reviewed in Lenexa. EWG discussion to continue in Kona.

Proposed resolution:

This wording is relative to the FDIS.

Change 20.10.9.1 [allocator.members] p12 as indicated:

template <class U, class... Args>
  void construct(U* p, Args&&... args);

12 Effects: ::new((void *)p) U(std::forward<Args>(args)...) if is_constructible<U, Args...>::value is true, else ::new((void *)p) U{std::forward<Args>(args)...}


2095. promise and packaged_task missing constructors needed for uses-allocator construction

Section: 30.6.5 [futures.promise], 30.6.9 [futures.task] Status: Open Submitter: Jonathan Wakely Opened: 2011-11-01 Last modified: 2016-02-12

Priority: 4

View other active issues in [futures.promise].

View all other issues in [futures.promise].

View all issues with Open status.

Discussion:

This example is ill-formed according to C++11 because uses_allocator<promise<R>, A>::value is true, but is_constructible<promise<R>, A, promise<R>&&>::value is false. Similarly for packaged_task.

#include <future>
#include <memory>
#include <tuple>

using namespace std;

typedef packaged_task<void()> task;
typedef promise<void> prom;
allocator<task> a;

tuple<task, prom> t1{ allocator_arg, a };
tuple<task, prom> t2{ allocator_arg, a, task{}, prom{} };

[2012, Portland]

This is an allocator issue, and should be dealt with directly by LWG.

[2013-03-06]

Jonathan suggests to make the new constructors non-explicit and makes some representational improvements.

[2013-09 Chicago]

Move to deferred.

This issue has much in common with similar problems with std::function that are being addressed by the polymorphic allocators proposal currently under evaluation in LEWG. Defer further discussion on this topic until the final outcome of that paper and its proposed resolution is known.

[2014-02-20 Re-open Deferred issues as Priority 4]

Proposed resolution:

[This wording is relative to the FDIS.]

  1. Add to 30.6.5 [futures.promise], class template promise synopsis, as indicated:

    namespace std {
      template <class R>
      class promise {
      public:
        promise();
        template <class Allocator>
        promise(allocator_arg_t, const Allocator& a);
        template <class Allocator>
        promise(allocator_arg_t, const Allocator& a, promise&& rhs) noexcept;
        promise(promise&& rhs) noexcept;
        promise(const promise& rhs) = delete;
        ~promise();	
        […]
      };
      […]
    }
    
  2. Change 30.6.5 [futures.promise] as indicated:

    promise(promise&& rhs) noexcept;
    template <class Allocator>
    promise(allocator_arg_t, const Allocator& a, promise&& rhs) noexcept;
    

    -5- Effects: constructs a new promise object and transfers ownership of the shared state of rhs (if any) to the newly-constructed object.

    -6- Postcondition: rhs has no shared state.

    -?- [Note: a is not used — end note]

  3. Add to 30.6.9 [futures.task], class template packaged_task synopsis, as indicated:

    namespace std {
      template<class> class packaged_task; // undefined
    
      template<class R, class... ArgTypes>
      class packaged_task<R(ArgTypes...)> {
      public:
        // construction and destruction
        packaged_task() noexcept;
        template <class Allocator>
          packaged_task(allocator_arg_t, const Allocator& a) noexcept;
        template <class F>
          explicit packaged_task(F&& f);
        template <class F, class Allocator>
          explicit packaged_task(allocator_arg_t, const Allocator& a, F&& f);
        ~packaged_task();
    	
        // no copy
        packaged_task(const packaged_task&) = delete;
        template<class Allocator>
          packaged_task(allocator_arg_t, const Allocator& a, const packaged_task&) = delete;
        packaged_task& operator=(const packaged_task&) = delete;
        
        // move support
        packaged_task(packaged_task&& rhs) noexcept;
        template <class Allocator>
          packaged_task(allocator_arg_t, const Allocator& a, packaged_task&& rhs) noexcept;
        packaged_task& operator=(packaged_task&& rhs) noexcept;
        void swap(packaged_task& other) noexcept;
        […]
      };
      […]
    }
    
  4. Change 30.6.9.1 [futures.task.members] as indicated:

    packaged_task() noexcept;
    template <class Allocator>
      packaged_task(allocator_arg_t, const Allocator& a) noexcept;
    

    -1- Effects: constructs a packaged_task object with no shared state and no stored task.

    -?- [Note: a is not used — end note]

    […]

    packaged_task(packaged_task&& rhs) noexcept;
    template <class Allocator>
      packaged_task(allocator_arg_t, const Allocator& a, packaged_task&& rhs) noexcept;
    

    -5- Effects: constructs a new packaged_task object and transfers ownership of rhs's shared state to *this, leaving rhs with no shared state. Moves the stored task from rhs to *this.

    -6- Postcondition: rhs has no shared state.

    -?- [Note: a is not used — end note]


2114. Incorrect "contextually convertible to bool" requirements

Section: 17.6.3.3 [nullablepointer.requirements], 24.2.3 [input.iterators], 24.2.7 [random.access.iterators], 25.1 [algorithms.general], 25.5 [alg.sorting], 30.2.1 [thread.req.paramname] Status: Open Submitter: Daniel Krügler Opened: 2011-12-09 Last modified: 2016-02-29

Priority: 3

View all issues with Open status.

Discussion:

As of 17.6.3.1 [utility.arg.requirements] Table 17/18, the return types of the expressions

a == b

or

a < b

for types satisfying the EqualityComparable or LessThanComparable types, respectively, are required to be "convertible to bool" which corresponds to a copy-initialization context. But several newer parts of the library that refer to such contexts have lowered the requirements taking advantage of the new terminology of "contextually convertible to bool" instead, which corresponds to a direct-initialization context (In addition to "normal" direct-initialization constructions, operands of logical operations as well as if or switch conditions also belong to this special context).

One example for these new requirements are input iterators which satisfy EqualityComparable but also specify that the expression

a != b

shall be just "contextually convertible to bool". The same discrepancy exists for requirement set NullablePointer in regard to several equality-related expressions.

For random access iterators we have

a < b contextually convertible to bool

as well as for all derived comparison functions, so strictly speaking we could have a random access iterator that does not satisfy the LessThanComparable requirements, which looks like an artifact to me.

Even if we keep with the existing requirements based on LessThanComparable or EqualityComparable we still would have the problem that some current specifications are actually based on the assumption of implicit convertibility instead of "explicit convertibility", e.g. 20.11.1.5 [unique.ptr.special] p3:

template <class T1, class D1, class T2, class D2>
bool operator!=(const unique_ptr<T1, D1>& x, const unique_ptr<T2, D2>& y);

-3- Returns: x.get() != y.get().

Similar examples exist in 20.11.1.2.2 [unique.ptr.single.dtor] p2, 20.11.1.2.3 [unique.ptr.single.asgn] p9, 20.11.1.2.4 [unique.ptr.single.observers] p1+3+8, etc.

In all these places the expressions involving comparison functions (but not those of the conversion of a NullablePointer to bool!) assume to be "convertible to bool". I think this is a very natural assumption and all delegations of the comparison functions of some type X to some other API type Y in third-party code does so assuming that copy-initialization semantics will just work.

The actual reason for using the newer terminology can be rooted back to LWG 556. My hypotheses is that the resolution of that issue also needs a slight correction. Why so?

The reason for opening that issue were worries based on the previous "convertible to bool" wording. An expressions like "!pred(a, b)" might not be well-formed in those situations, because operator! might not be accessible or might have an unusual semantics (and similarly for other logical operations). This can indeed happen with unusual proxy return types, so the idea was that the evaluation of Predicate, BinaryPredicate (25.1 [algorithms.general] p8+9), and Compare (25.5 [alg.sorting] p2) should be defined based on contextual conversion to bool. Unfortunately this alone is not sufficient: In addition, I think, we also want the predicates to be (implicitly) convertible to bool! Without this wording, several conditions are plain wrong, e.g. 25.3.5 [alg.find] p2, which talks about "pred(*i) != false" (find_if) and "pred(*i) == false" (find_if_not). These expressions are not within a boolean context!

While we could simply fix all these places by proper wording to be considered in a "contextual conversion to bool", I think that this is not the correct solution: Many third-party libraries already refer to the previous C++03 Predicate definition — it actually predates C++98 and is as old as the SGI specification. It seems to be a high price to pay to switch to direct initialization here instead of fixing a completely different specification problem.

A final observation is that we have another definition for a Predicate in 30.2.1 [thread.req.paramname] p2:

If a parameter is Predicate, operator() applied to the actual template argument shall return a value that is convertible to bool.

The problem here is not that we have two different definitions of Predicate in the standard — this is confusing, but this fact alone is not a defect. The first (minor) problem is that this definition does not properly apply to function objects that are function pointers, because operator() is not defined in a strict sense. But the actually worse second problem is that this wording has the very same problem that has originally lead to LWG 556! We only need to look at 30.5.1 [thread.condition.condvar] p15 to recognice this:

while (!pred())
  wait(lock);

The negation expression here looks very familiar to the example provided in LWG 556 and is sensitive to the same "unusual proxy" problem. Changing the 30.2.1 [thread.req.paramname] wording to a corresponding "contextual conversion to bool" wouldn't work either, because existing specifications rely on "convertible to bool", e.g. 30.5.1 [thread.condition.condvar] p32+33+42 or 30.5.2 [thread.condition.condvarany] p25+26+32+33.

To summarize: I believe that LWG 556 was not completely resolved. A pessimistic interpretation is, that even with the current wording based on "contextually convertible to bool" the actual problem of that issue has not been fixed. What actually needs to be required here is some normative wording that basically expresses something along the lines of:

The semantics of any contextual conversion to bool shall be equivalent to the semantics of any implicit conversion to bool.

This is still not complete without having concepts, but it seems to be a better approximation. Another way of solving this issue would be to define a minimum requirements table with equivalent semantics. The proposed wording is a bit simpler but attempts to express the same thing.

[2012, Kona]

Agree with Daniel that we potentially broke some C++03 user code, accept the changes striking "contextually" from tables. Stefan to provide revised wording for section 25, and figure out changes to section 30.

Move to open, and then to Review when updated wording from Stefan is available.

[2012-10-12, STL comments]

  1. The current proposed resolution still isn't completely satisfying. It would certainly be possible for the Standard to require these various expressions to be implicitly and contextually convertible to bool, but that would have a subtle consequence (which, I will argue, is undesirable - regardless of the fact that it dates all the way back to C++98/03). It would allow users to provide really wacky types to the Standard Library, with one of two effects:

    1. Standard Library implementations would have to go to great lengths to respect such wacky types, essentially using static_cast<bool> when invoking any predicates or comparators.

    2. Otherwise, such wacky types would be de facto nonportable, because they would make Standard Library implementations explode.

    Effect B is the status quo we're living with today. What Standard Library implementations want to do with pred(args) goes beyond "if (pred(args))" (C++03), contextually converting pred(args) to bool (C++11), or implicitly and contextually converting pred(args) to bool (the current proposed resolution). Implementations want to say things like:

    if (pred(args))
    if (!pred(args))
    if (cond && pred(args))
    if (cond && !pred(args))
    

    These are real examples taken from Dinkumware's implementation. There are others that would be realistic ("pred(args) && cond", "cond || pred(args)", etc.)

    Although negation was mentioned in this issue's Discussion section, and in LWG 556's, the current proposed resolution doesn't fix this problem. Requiring pred(args) to be implicitly and contextually convertible to bool doesn't prevent operator!() from being overloaded and returning std::string (as a wacky example). More ominously, it doesn't prevent operator&&() and operator||() from being overloaded and destroying short-circuiting.

  2. I would like LWG input before working on Standardese for a new proposed resolution. Here's an outline of what I'd like to do:

    1. Introduce a new "concept" in 17.6.3 [utility.requirements], which I would call BooleanTestable in the absence of better ideas.

    2. Centralize things and reduce verbosity by having everything simply refer to BooleanTestable when necessary. I believe that the tables could say "Return type: BooleanTestable", while Predicate/BinaryPredicate/Compare would need the incantation "shall satisfy the requirements of BooleanTestable".

    3. Resolve the tug-of-war between users (who occasionally want to do weird things) and implementers (who don't want to have to contort their code) by requiring that:

      1. Given a BooleanTestable x, x is both implicitly and contextually convertible to bool.

      2. Given a BooleanTestable x, !x is BooleanTestable. (This is intentionally "recursive".)

      3. Given a BooleanTestable x, bool t = x, t2(x), f = !x; has the postcondition t == t2 && t != f.

      4. Given a BooleanTestable x and a BooleanTestable y of possibly different types, "x && y" and "x || y" invoke the built-in operator&&() and operator||(), triggering short-circuiting.

      5. bool is BooleanTestable.

    I believe that this simultaneously gives users great latitude to use types other than bool, while allowing implementers to write reasonable code in order to get their jobs done. (If I'm forgetting anything that implementers would want to say, please let me know.)

  3. About requirement (I): As Daniel patiently explained to me, we need to talk about both implicit conversions and contextual conversions, because it's possible for a devious type to have both "explicit operator bool()" and "operator int()", which might behave differently (or be deleted, etc.).

  4. About requirement (IV): This is kind of tricky. What we'd like to say is, "BooleanTestable can't ever trigger an overloaded logical operator". However, given a perfectly reasonable type Nice - perhaps even bool itself! - other code (perhaps a third-party library) could overload operator&&(Nice, Evil). Therefore, I believe that the requirement should be "no first use" - the Standard Library will ask for various BooleanTestable types from users (for example, the result of "first != last" and the result of "pred(args)"), and as long as they don't trigger overloaded logical operators with each other, everything is awesome.

  5. About requirement (V): This is possibly redundant, but it's trivial to specify, makes it easier for users to understand what they need to do ("oh, I can always achieve this with bool"), and provides a "base case" for requirement (IV) that may or may not be necessary. Since bool is BooleanTestable, overloading operator&&(bool, Other) (etc.) clearly makes the Other type non-BooleanTestable.

Previous resolution from Daniel [SUPERSEDED]:

This wording is relative to the FDIS.

  1. Change Table 25 — "NullablePointer requirements" in 17.6.3.3 [nullablepointer.requirements] as indicated:

    Table 25 — NullablePointer requirements
    Expression Return type Operational semantics
    […]
    a != b contextually convertible to bool !(a == b)
    a == np
    np == a
    contextually convertible to bool a == P()
    a != np
    np != a
    contextually convertible to bool !(a == np)
  2. Change Table 107 — "Input iterator requirements" in 24.2.3 [input.iterators] as indicated:

    Table 107 — Input iterator requirements (in addition to Iterator)
    Expression Return type Operational semantics Assertion/note
    pre-/post-condition
    a != b contextually convertible to bool !(a == b) pre: (a, b) is in the domain of ==.
    […]
  3. Change Table 111 — "Random access iterator requirements" in 24.2.7 [random.access.iterators] as indicated:

    Table 111 — Random access iterator requirements (in addition to bidirectional iterator)
    Expression Return type Operational semantics Assertion/note
    pre-/post-condition
    […]
    a < b contextually convertible to bool b - a > 0 < is a total ordering relation
    a > b contextually convertible to bool b < a > is a total ordering relation opposite to <.
    a >= b contextually convertible to bool !(a < b)
    a <= b contextually convertible to bool !(a > b)
  4. Change 25.1 [algorithms.general] p8+9 as indicated:

    -8- The Predicate parameter is used whenever an algorithm expects a function object (20.14 [function.objects]) that, when applied to the result of dereferencing the corresponding iterator, returns a value testable as true. In other words, if an algorithm takes Predicate pred as its argument and first as its iterator argument, it should work correctly in the construct pred(*first) implicitly or contextually converted to bool (Clause 4 [conv]). The function object pred shall not apply any non-constant function through the dereferenced iterator.

    -9- The BinaryPredicate parameter is used whenever an algorithm expects a function object that when applied to the result of dereferencing two corresponding iterators or to dereferencing an iterator and type T when T is part of the signature returns a value testable as true. In other words, if an algorithm takes BinaryPredicate binary_pred as its argument and first1 and first2 as its iterator arguments, it should work correctly in the construct binary_pred(*first1, *first2) implicitly or contextually converted to bool (Clause 4 [conv]). BinaryPredicate always takes the first iterator's value_type as its first argument, that is, in those cases when T value is part of the signature, it should work correctly in the construct binary_pred(*first1, value) implicitly or contextually converted to bool (Clause 4 [conv]). binary_pred shall not apply any non-constant function through the dereferenced iterators.

  5. Change 25.5 [alg.sorting] p2 as indicated:

    -2- Compare is a function object type (20.14 [function.objects]). The return value of the function call operation applied to an object of type Compare, when implicitly or contextually converted to bool (4 [conv]), yields true if the first argument of the call is less than the second, and false otherwise. Compare comp is used throughout for algorithms assuming an ordering relation. It is assumed that comp will not apply any non-constant function through the dereferenced iterator.

  6. Change 30.2.1 [thread.req.paramname] p2 as indicated:

    -2- If a parameter is Predicate, operator() applied to the actual template argument shall return a value that is convertible to boolPredicate is a function object type (20.14 [function.objects]). The return value of the function call operation applied to an object of type Predicate, when implicitly or contextually converted to bool (4 [conv]), yields true if the corresponding test condition is satisfied, and false otherwise.

[2014-05-20, Daniel suggests concrete wording based on STL's proposal]

The presented wording follows relatively closely STL's outline with the following notable exceptions:

  1. A reference to BooleanTestable in table "Return Type" specifications seemed very unusual to me and I found no "prior art" for this in the Standard. Instead I decided to follow the usual style to add a symbol with a specific meaning to a specific paragraph that specifies symbols and their meanings.

  2. STL's requirement IV suggested to directly refer to built-in operators && and ||. In my opinion this concrete requirement isn't needed if we simply require that two BooleanTestable operands behave equivalently to two those operands after conversion to bool (each of them).

  3. I couldn't find a good reason to require normatively that type bool meets the requirements of BooleanTestable: My assertion is that after having defined them, the result simply falls out of this. But to make this a bit clearer, I added also a non-normative note to these effects.

[2014-06-10, STL comments]

In the current wording I would like to see changed the suggested changes described by bullet #6:

  1. In 23.2.1 [container.requirements.general] p4 undo the suggested change

  2. Then change the 7 occurrences of "convertible to bool" in the denoted tables to "bool".

[2015-05-05 Lenexa]

STL: Alisdair wanted to do something here, but Daniel gave us updated wording.

[2015-07 Telecom]

Alisdair: Should specify we don't break short circuiting.
Ville: Looks already specified because that's the way it works for bool.
Geoffrey: Maybe add a note about the short circuiting.
B2/P2 is somewhat ambiguous. It implies that B has to be both implicitly convertible to bool and contextually convertible to bool.
We like this, just have nits.
Status stays Open.
Marshall to ping Daniel with feedback.

[2016-02-27, Daniel updates wording]

  1. The revised wording has been updated from N3936 to N4567.

  2. To satisfy the Kona 2015 committee comments, the wording in [booleantestable.requirements] has been improved to better separate the two different requirements of "can be contextually converted to bool" and "can be implicitly converted to bool. Both are necessary because it is possible to define a type that has the latter property but not the former, such as the following one:

    using Bool = int;
    
    struct OddBoolean 
    {
      explicit OddBoolean(bool) = delete;
      OddBoolean(Bool){}
    };
    
    OddBoolean b2 = true; // OK
    OddBoolean b1(true);  // Error
    
  3. In [booleantestable.requirements] a note has been added to ensure that an implementation is not allowed to break any short-circuiting semantics.

  4. I decided to separate LWG 2587/2588 from this issue. Both these issues aren't exactly the same but depending on the committee's position, their resolution might benefit from the new vocabulary introduced here.

Proposed resolution:

This wording is relative to N4567.

  1. Change 17.6.3.1 [utility.arg.requirements] p1, Table 17 — "EqualityComparable requirements", and Table 18 — "LessThanComparable requirements" as indicated:

    -1- […] In these tables, T is an object or reference type to be supplied by a C++ program instantiating a template; a, b, and c are values of type (possibly const) T; s and t are modifiable lvalues of type T; u denotes an identifier; rv is an rvalue of type T; and v is an lvalue of type (possibly const) T or an rvalue of type const T; and BT denotes a type that meets the BooleanTestable requirements ([booleantestable.requirements]).

    […]

    Table 17 — EqualityComparable requirements [equalitycomparable]
    Expression Return type Requirement
    a == b convertible to
    bool
    BT
    == is an equivalence relation, that is, it has the following properties: […]

    […]

    Table 18 — LessThanComparable requirements [lessthancomparable]
    Expression Return type Requirement
    a < b convertible to
    bool
    BT
    < is a strict weak ordering relation (25.5 [alg.sorting])
  2. Between 17.6.3.2 [swappable.requirements] and 17.6.3.3 [nullablepointer.requirements] insert a new sub-clause as indicated:

    ?.?.?.? BooleanTestable requirements [booleantestable.requirements]

    -?- A BooleanTestable type is a boolean-like type that also supports conversions to bool. A type B meets the BooleanTestable requirements if the expressions described in Table ?? are valid and have the indicated semantics, and if B also satisfies all the other requirements of this sub-clause [booleantestable.requirements].

    An object b of type B can be implicitly converted to bool and in addition can be contextually converted to bool (Clause 4). The result values of both kinds of conversions shall be equivalent.

    [Example: The types bool, std::true_type, and std::bitset<>::reference are BooleanTestable types. — end example]

    For the purpose of Table ??, let B2 and Bn denote types (possibly both equal to B or to each other) that meet the BooleanTestable requirements, let b1 denote a (possibly const) value of B, let b2 denote a (possibly const) value of B2, and let t1 denote a value of type bool.

    [Note: These rules ensure what an implementation can rely on but doesn't grant it license to break short-circuiting behavior of a BooleanTestable type. — end note]

  3. Somewhere within the new sub-clause [booleantestable.requirements] insert the following new Table (?? denotes the assigned table number):

    Table ?? — BooleanTestable requirements [booleantestable]
    Expression Return type Operational semantics
    bool(b1) bool Remarks: bool(b1) == t1 for every value
    b1 implicitly converted to t1.
    !b1 Bn Remarks: bool(b1) == !bool(!b1) for
    every value b1.
    b1 && b2 bool bool(b1) && bool(b2)
    b1 || b2 bool bool(b1) || bool(b2)
  4. Change 17.6.3.3 [nullablepointer.requirements] p5 and Table 25 — "NullablePointer requirements" as indicated:

    […]

    -5- In Table 25, u denotes an identifier, t denotes a non-const lvalue of type P, a and b denote values of type (possibly const) P, and np denotes a value of type (possibly const) std::nullptr_t, and BT denotes a type that meets the BooleanTestable requirements ([booleantestable.requirements]).

    […]

    Table 25 — NullablePointer requirements [nullablepointer]
    Expression Return type Operational semantics
    a != b contextually convertible to boolBT […]
    a == np
    np == a
    contextually convertible to boolBT […]
    a != np
    np != a
    contextually convertible to boolBT […]
  5. Change 20.5.2.8 [tuple.rel] as indicated;

    template<class... TTypes, class... UTypes>
    constexpr bool operator==(const tuple<TTypes...>& t, const tuple<UTypes...>& u);
    

    -1- Requires: For all i, where 0 <= i and i < sizeof...(TTypes), get<i>(t) == get<i>(u) is a valid expression returning a type that is convertible to boolmeets the BooleanTestable requirements ([booleantestable.requirements]). sizeof...(TTypes) == sizeof...(UTypes).

    […]

    template<class... TTypes, class... UTypes>
    constexpr bool operator<(const tuple<TTypes...>& t, const tuple<UTypes...>& u);
    

    -4- Requires: For all i, where 0 <= i and i < sizeof...(TTypes), get<i>(t) < get<i>(u) and get<i>(u) < get<i>(t) are valid expressions returning types that are convertible to boolmeet the BooleanTestable requirements ([booleantestable.requirements]). sizeof...(TTypes) == sizeof...(UTypes).

    […]

  6. Change 23.2.1 [container.requirements.general], Table 95 — "Container requirements", and Table 97 — "Optional container operations" as indicated:

    -4- In Tables 95, 96, and 97 X denotes a container class containing objects of type T, a and b denote values of type X, u denotes an identifier, r denotes a non-const value of type X, and rv denotes a non-const rvalue of type X, and BT denotes a type that meets the BooleanTestable requirements ([booleantestable.requirements]).

    Table 95 — Container requirements
    Expression Return type […]
    a == b convertible to
    bool
    BT
    […]
    a != b convertible to
    bool
    BT
    […]
    a.empty() convertible to
    bool
    BT
    […]

    […]

    Table 97 — Optional container requirements
    Expression Return type […]
    a < b convertible to
    bool
    BT
    […]
    a > b convertible to
    bool
    BT
    […]
    a <= b convertible to
    bool
    BT
    […]
    a >= b convertible to
    bool
    BT
    […]
  7. Change 24.2.1 [iterator.requirements.general], Table 106 — "Input iterator requirements", and Table 110 — "Random access iterator requirements" as indicated:

    -12- In the following sections, a and b denote values of type X or const X, difference_type and reference refer to the types iterator_traits<X>::difference_type and iterator_traits<X>::reference, respectively, n denotes a value of difference_type, u, tmp, and m denote identifiers, r denotes a value of X&, t denotes a value of value type T, o denotes a value of some type that is writable to the output iterator, and BT denotes a type that meets the BooleanTestable requirements ([booleantestable.requirements]).

    Table 106 — Input iterator requirements
    Expression Return type […]
    a != b contextually convertible to
    bool
    BT
    […]

    […]

    Table 110 — Random access iterator requirements
    Expression Return type […]
    a < b contextually convertible to
    bool
    BT
    […]
    a > b contextually convertible to
    bool
    BT
    […]
    a >= b contextually convertible to
    bool
    BT
    […]
    a <= b contextually convertible to
    bool
    BT
    […]
  8. Change 25.1 [algorithms.general] p8+p9 as indicated:

    [Drafting note: The wording changes below also fix (a) unusual wording forms used ("should work") which are unclear in which sense they are imposing normative requirements and (b) the problem, that the current wording seems to allow that the predicate may mutate a call argument, if that is not a dereferenced iterator. Upon applying the new wording it became obvious that the both the previous and the new wording has the effect that currently algorithms such as adjacent_find, search_n, unique, and unique_copy are not correctly described (because they have no iterator argument named first1), which could give raise to a new library issue. — end drafting note]

    -8- The Predicate parameter is used whenever an algorithm expects a function object (20.9) that, when applied to the result of dereferencing the corresponding iterator, returns a value testable as true. In other words, iIf an algorithm takes Predicate pred as its argument and first as its iterator argument, it should work correctly in the construct pred(*first) contextually converted to bool (Clause 4)the expression pred(*first) shall have a type that meets the BooleanTestable requirements ( [booleantestable.requirements]). The function object pred shall not apply any non-constant function through the dereferenced iteratorits argument.

    -9- The BinaryPredicate parameter is used whenever an algorithm expects a function object that when applied to the result of dereferencing two corresponding iterators or to dereferencing an iterator and type T when T is part of the signature returns a value testable as true. In other words, iIf an algorithm takes BinaryPredicate binary_pred as its argument and first1 and first2 as its iterator arguments, it should work correctly in the construct binary_pred(*first1, *first2) contextually converted to bool (Clause 4)the expression binary_pred(*first1, *first2) shall have a type that meets the BooleanTestable requirements ( [booleantestable.requirements]). BinaryPredicate always takes the first iterator's value_type as its first argument, that is, in those cases when T value is part of the signature, it should work correctly in the construct binary_pred(*first1, value) contextually converted to bool (Clause 4)the expression binary_pred(*first1, value) shall have a type that meets the BooleanTestable requirements ( [booleantestable.requirements]). binary_pred shall not apply any non-constant function through the dereferenced iteratorsany of its arguments.

  9. Change 25.5 [alg.sorting] p2 as indicated:

    […]

    -2- Compare is a function object type (20.9). The return value of the function call operation applied to an object of type Compare, when contextually converted to bool(Clause 4), yields true if the first argument of the call is less than the second, and false otherwise. Compare comp is used throughout for algorithms assuming an ordering relation. Let a and b denote two argument values whose types depend on the corresponding algorithm. Then the expression comp(a, b) shall have a type that meets the BooleanTestable requirements ( [booleantestable.requirements]). The return value of comp(a, b), converted to bool, yields true if the first argument a is less than the second argument b, and false otherwise. It is assumed that comp will not apply any non-constant function through the dereferenced iteratorany of its arguments.

    […]

  10. Change 27.5.4.2 [fpos.operations] and Table 126 — "Position type requirements" as indicated:

    -1- Operations specified in Table 126 are permitted. In that table,

    • P refers to an instance of fpos,

    • […]

    • o refers to a value of type streamoff,

    • BT refers to a type that meets the BooleanTestable requirements ([booleantestable.requirements]),

    • […]

    Table 126 — Position type requirements
    Expression Return type […]
    p == q convertible to boolBT […]
    p != q convertible to boolBT […]
  11. Change 30.2.1 [thread.req.paramname] p1 as indicated:

    -1- Throughout this Clause, the names of template parameters are used to express type requirements. If a template parameter is named Predicate, operator() applied to the template argument shall return a value that is convertible to boolPredicate is a function object type (20.14 [function.objects]). Let pred denote an lvalue of type Predicate. Then the expression pred() shall have a type that meets the BooleanTestable requirements ( [booleantestable.requirements]). The return value of pred(), converted to bool, yields true if the corresponding test condition is satisfied, and false otherwise.


2115. Undefined behaviour for valarray assignments with mask_array index?

Section: 26.7.8 [template.mask.array] Status: Open Submitter: Thomas Plum Opened: 2011-12-10 Last modified: 2016-02-12

Priority: 4

View all issues with Open status.

Discussion:

Recently I received a Service Request (SR) alleging that one of our testcases causes an undefined behavior. The complaint is that 26.7.8 [template.mask.array] in C++11 (and the corresponding subclause in C++03) are interpreted by some people to require that in an assignment "a[mask] = b", the subscript mask and the rhs b must have the same number of elements.

IMHO, if that is the intended requirement, it should be stated explicitly.

In any event, there is a tiny editorial cleanup that could be made:

In C++11, 26.7.8.1 [template.mask.array.overview] para 2 mentions

"the expression a[mask] = b;"

but the semicolon cannot be part of an expression. The correction could omit the semicolon, or change the word "expression" to "assignment" or "statement".

Here is the text of the SR, slightly modified for publication:

Subject: SR01174 LVS _26322Y31 has undefined behavior [open]

[Client:]
The test case t263.dir/_26322Y31.cpp seems to be illegal as it has an undefined behaviour. I searched into the SRs but found SRs were not related to the topic explained in this mail (SR00324, SR00595, SR00838).

const char vl[] = {"abcdefghijklmnopqrstuvwxyz"};
const char vu[] = {"ABCDEFGHIJKLMNOPQRSTUVWXYZ"};
const std::valarray<char> v0(vl, 27), vm5(vu, 5), vm6(vu, 6);
std::valarray<char> x = v0;
[…]
const bool vb[] = {false, false, true, true, false, true};
const std::valarray<bool> vmask(vb, 6);
x = v0;
x[vmask] = vm5;      // ***** HERE....
steq(&x[0], "abABeCghijklmnopqrstuvwxyz");
x2 = x[vmask];       // ***** ....AND HERE
[…]

This problem has already been discussed between [experts]: See thread http://gcc.gnu.org/ml/libstdc++/2009-11/threads.html#00051 Conclusion http://gcc.gnu.org/ml/libstdc++/2009-11/msg00099.html

[Plum Hall:]
Before I log this as an SR, I need to check one detail with you.

I did read the email thread you mentioned, and I did find a citation (see INCITS ISO/IEC 14882-2003 Section 26.3.2.6 on valarray computed assignments):

Quote: "If the array and the argument array do not have the same length, the behavior is undefined",

But this applies to computed assignment (*=, +=, etc), not to simple assignment. Here is the C++03 citation re simple assignment:

26.3.2.2 valarray assignment [lib.valarray.assign]

valarray<T>& operator=(const valarray<T>&);

1 Each element of the *this array is assigned the value of the corresponding element of the argument array. The resulting behavior is undefined if the length of the argument array is not equal to the length of the *this array.

In the new C++11 (N3291), we find ...

26.6.2.3 valarray assignment [valarray.assign]

valarray<T>& operator=(const valarray<T>& v);

1 Each element of the *this array is assigned the value of the corresponding element of the argument array. If the length of v is not equal to the length of *this, resizes *this to make the two arrays the same length, as if by calling resize(v.size()), before performing the assignment.

So it looks like the testcase might be valid for C++11 but not for C++03; what do you think?

[Client:]
I quite agree with you but the two problems I mentioned:

x[vmask] = vm5;      // ***** HERE....
[…]
x2 = x[vmask];       // ***** ....AND HERE

refer to mask_array assignment hence target the C++03 26.3.8 paragraph. Correct?

[Plum Hall:]
I mentioned the contrast between C++03 26.3.2.2 para 1 versus C++11 26.6.2.3 para 1.

But in C++03 26.3.8, I don't find any corresponding restriction. Could you quote the specific requirement you're writing about?

[Client:]
I do notice the difference between c++03 26.3.2.2 and c++11 26.6.2.3 about assignments between different sized valarray and I perfectly agree with you.

But, as already stated, this is not a simple valarray assignment but a mask_array assignment (c++03 26.3.8 / c++11 26.6.8). See c++11 quote below:

26.6.8 Class template mask_array
26.6.8.1 Class template mask_array overview
[....]

  1. This template is a helper template used by the mask subscript operator: mask_array<T> valarray<T>::operator[](const valarray<bool>&).

  2. It has reference semantics to a subset of an array specified by a boolean mask. Thus, the expression a[mask] = b; has the effect of assigning the elements of b to the masked elements in a (those for which the corresponding element in mask is true.)

26.6.8.2 mask_array assignment

void operator=(const valarray<T>&) const;
const mask_array& operator=(const mask_array&) const;

1 These assignment operators have reference semantics, assigning the values of the argument array elements to selected elements of the valarray<T> object to which it refers.

In particular, [one of the WG21 experts] insisted on the piece "the elements of b".

That is why I reported the test t263.dir/_26322Y31.cpp having an undefined behaviour.

[Plum Hall:]
OK, I can see that I will have to ask WG21; I will file an appropriate issue with the Library subgroup. In the meantime, I will mark this testcase as "DISPUTED" so that it is not required for conformance testing, until we get a definitive opinion.

[2012, Kona]

Moved to Open.

There appears to be a real need for clarification in the standard, and implementations differ in their current interpretation. This will need some research by implementers and a proposed resolution before further discussion is likely to be fruitful.

Proposed resolution:


2116. std::swap noexcept(what?)

Section: 20.15.4.3 [meta.unary.prop] Status: Open Submitter: Dave Abrahams Opened: 2011-12-09 Last modified: 2016-02-12

Priority: 3

View other active issues in [meta.unary.prop].

View all other issues in [meta.unary.prop].

View all issues with Open status.

Discussion:

IMO if we specified is_[nothrow_]constructible in terms of a variable declaration whose validity requires destructibility, it is clearly a bug in our specification and a failure to realize the actual original intent. The specification should have been in terms of placement-new.

Daniel:
At the time of the specification this was intended and the solution is not done by removing the destruction semantics of is_constructible.

The design of is_constructible was also impacted by the previous Constructible concept that explicitly contained destruction semantics, because during conceptification of the library it turned out to simplify the constraints in the library because you did not need to add Destructible all the time. It often was implied but never spoken out in C++03.

Pure construction semantics was considered as useful as well, so HasConstructor did also exist and would surely be useful as trait as well.

Another example that is often overlooked: This also affects wrapper types like pair, tuple, array that contain potentially more than one type: This is easy to understand if you think of T1 having a deleted destructor and T2 having a constructor that may throw: Obviously the compiler has potentially need to use the destructor of T1 in the constructor of std::pair<T1, T2> to ensure that the core language requirements are satisfied (All previous fully constructed sub-objects must be destructed).

The core language also honors this fact in 12.8 [class.copy] p11:

A defaulted copy/move constructor for a class X is defined as deleted (8.4.3 [dcl.fct.def.delete]) if X has:
[…]
— any direct or virtual base class or non-static data member of a type with a destructor that is deleted or inaccessible from the defaulted constructor,
[…]

Dave:
This is about is_nothrow_constructible in particular. The fact that it is foiled by not having a noexcept dtor is a defect.

[2012, Kona]

Move to Open.

is_nothrow_constructible is defined in terms of is_constructible, which is defined by looking at a hypothetical variable and asking whether the variable definition is known not to throw exceptions. The issue claims that this also examines the type's destructor, given the context, and thus will return false if the destructor can potentially throw. At least one implementation (Howard's) does return false if the constructor is noexcept(true) and the destructor is noexcept(false). So that's not a strained interpretation. The issue is asking for this to be defined in terms of placement new, instead of in terms of a temporary object, to make it clearer that is_nothrow_constructible looks at the noexcept status of only the constructor, and not the destructor.

Sketch of what the wording would look like:

require is_constructible, and then also require that a placement new operation does not throw. (Remembering the title of this issue... What does this imply for swap?

If we accept this resolution, do we need any changes to swap?

STL argues: no, because you are already forbidden from passing anything with a throwing desturctor to swap.

Dietmar argues: no, not true. Maybe statically the destructor can conceivably throw for some values, but maybe there are some values known not to throw. In that case, it's correct to pass those values to swap.

Proposed resolution:


2117. ios_base manipulators should have showgrouping/noshowgrouping

Section: 22.4.2.2.2 [facet.num.put.virtuals], 27.5.3.1.2 [ios::fmtflags], 27.5.6.1 [fmtflags.manip] Status: Open Submitter: Benjamin Kosnik Opened: 2011-12-15 Last modified: 2016-04-15

Priority: 3

View other active issues in [facet.num.put.virtuals].

View all other issues in [facet.num.put.virtuals].

View all issues with Open status.

Discussion:

Iostreams should include a manipulator to toggle grouping on/off for locales that support grouped digits. This has come up repeatedly and been deferred. See LWG 826 for the previous attempt.

If one is using a locale that supports grouped digits, then output will always include the generated grouping characters. However, very plausible scenarios exist where one might want to output the number, un-grouped. This is similar to existing manipulators that toggle on/off the decimal point, numeric base, or positive sign.

See some user commentary here.

[21012, Kona]

Move to Open.

This is a feature request.

Walter is slightly uncomfortable with processing feature requests through the issues lists.

Alisdair says this is far from the first feature request that has come in from the issues list.

STL: The fact that you can turn off grouping on hex output is compelling.

Marshall: if we add this flag, we'll need to update tables 87-91 as well.

STL: If it has been implemented somewhere, and it works, we'd be glad to add it.

Howard: We need to say what the default is.

Alisdair sumarizes:

(1) We want clear wording that says what the effect is of turning the flag off;

(2) what the default values are, and

(3) how this fits into tables 87-90. (and 128)

[Issaquah 2014-02-10-12: Move to LEWG]

Since this issue was filed, we have grown a new working group that is better placed to handle feature requests.

We will track such issues with an LEWG status until we get feedback from the Library Evolution Working Group.

[Issaquah 2014-02-12: LEWG discussion]

Do we think this feature should exist?
SFFNASA
2 4100

Think about the ABI break for adding a flag. But this could be mitigated by putting the data into an iword instead of a flag.

This needs to change Stage 2 in [facet.num.put.virtuals].

Previous resolution, which needs the above corrections:

This wording is relative to the FDIS.

  1. Insert in 22.4.2.2.2 [facet.num.put.virtuals] paragraph 5:

    Stage 1: The first action of stage 1 is to determine a conversion specifier. The tables that describe this determination use the following local variables

    fmtflags flags = str.flags() ;
    fmtflags basefield = (flags & (ios_base::basefield));
    fmtflags uppercase = (flags & (ios_base::uppercase));
    fmtflags floatfield = (flags & (ios_base::floatfield));
    fmtflags showpos = (flags & (ios_base::showpos));
    fmtflags showbase = (flags & (ios_base::showbase));
    fmtflags showgrouping = (flags & (ios_base::showgrouping));
    
  2. Change header <ios> synopsis, 27.5.1 [iostreams.base.overview] as indicated:

    #include <iosfwd>
    
    namespace std {
      […]
      // 27.5.6, manipulators:
      […]
      ios_base& showpoint     (ios_base& str);
      ios_base& noshowpoint   (ios_base& str);
      ios_base& showgrouping  (ios_base& str);
      ios_base& noshowgrouping(ios_base& str);
      ios_base& showpos       (ios_base& str);
      ios_base& noshowpos     (ios_base& str);
      […]
    }
    
  3. Change class ios_base synopsis, 27.5.3 [ios.base] as indicated:

    namespace std {
      class ios_base {
      public:
      class failure;
        // 27.5.3.1.2 fmtflags
        typedef T1 fmtflags;
        […]
        static constexpr fmtflags showpoint = unspecified ;
        static constexpr fmtflags showgrouping = unspecified ;
        static constexpr fmtflags showpos = unspecified ;
        […]
      };
    }
    
  4. Add a new entry to Table 122 — "fmtflags effects" as indicated:

    Table 122 — fmtflags effects
    Element Effect(s) if set
    […]
    showpoint generates a decimal-point character unconditionally in generated floatingpoint output
    showgrouping generates grouping characters unconditionally in generated output
    […]
  5. After 27.5.3.1.2 [ios::fmtflags] p12 insert the following:

    ios_base& showgrouping(ios_base& str);
    

    -?- Effects: Calls str.setf(ios_base::showgrouping).

    -?- Returns: str.

    ios_base& noshowgrouping(ios_base& str);
    

    -?- Effects: Calls str.unsetf(ios_base::showgrouping).

    -?- Returns: str.

Proposed resolution:


2121. app for string streams

Section: 27.8.5.1 [stringstream.cons] Status: New Submitter: Nicolai Josuttis Opened: 2012-01-15 Last modified: 2016-02-12

Priority: 3

View all issues with New status.

Discussion:

This issue was raised while discussing issue 1448.

Note the following program:

string s("s1: 123456789");
ostringstream s1(s, ios_base::out|ios_base::app);
s1 << "hello";
cout << s1.str() << endl;

With g++4.x it prints:

s1: 123456789hello

With VisualC++10 it prints:

hello23456789

From my intuitive understanding the flag "app" should result in the output of g++4.x. I also would read that from 27.5.3.1.4 [ios::openmode] claiming:

app   seek to end before each write

However in issue 1448 P.J.Plauger comments:

I think we should say nothing special about app at construction time (thus leaving the write pointer at the beginning of the buffer). Leave implementers wiggle room to ensure subsequent append writes as they see fit, but don't change existing rules for initial seek position.

Note that the flag ate on both platforms appends "hello" to s.

Proposed resolution:


2136. Postconditions vs. exceptions

Section: 17.5.1 [structure] Status: Open Submitter: Jens Maurer Opened: 2012-03-08 Last modified: 2016-02-12

Priority: 3

View all issues with Open status.

Discussion:

The front matter in clause 17 should clarify that postconditions will not hold if a standard library function exits via an exception. Postconditions or guarantees that apply when an exception is thrown (beyond the basic guarantee) are described in an "Exception safety" section.

[ 2012-10 Portland: Move to Open ]

Consensus that we do not clearly say this, and that we probably should. A likely location to describe the guarantees of postconditions could well be a new sub-clause following 17.6.4.11 [res.on.required] which serves the same purpose for requires clauses. However, we need such wording before we can make progress.

Also, see 2137 for a suggestion that we want to see a paper resolving both issues together.

[2015-05-06 Lenexa: EirkWF to write paper addressing 2136 and 2137]

MC: Idea is to replace all such "If no exception" postconditions with "Exception safety" sections.

Proposed resolution:


2137. Misleadingly constrained post-condition in the presence of exceptions

Section: 28.8.3 [re.regex.assign] Status: Open Submitter: Jonathan Wakely Opened: 2012-03-08 Last modified: 2016-02-12

Priority: 3

View all other issues in [re.regex.assign].

View all issues with Open status.

Discussion:

The post-conditions of basic_regex<>::assign 28.8.3 [re.regex.assign] p16 say:

If no exception is thrown, flags() returns f and mark_count() returns the number of marked sub-expressions within the expression.

The default expectation in the library is that post-conditions only hold, if there is no failure (see also 2136), therefore the initial condition should be removed to prevent any misunderstanding.

[ 2012-10 Portland: Move to Open ]

A favorable resolution clearly depends on a favorable resolution to 2136. There is also a concern that this is just one example of where we would want to apply such a wording clean-up, and which is really needed to resolve both this issue and 2136 is a paper providing the clause 17 wording that gives the guarantee for postcondition paragaraphs, and then reviews clauses 18-30 to apply that guarantee consistently. We do not want to pick up these issues piecemeal, as we risk openning many issues in an ongoing process.

[2015-05-06 Lenexa: EirkWF to write paper addressing 2136 and 2137]

Proposed resolution:

This wording is relative to N3376.

template <class string_traits, class A>
  basic_regex& assign(const basic_string<charT, string_traits, A>& s,
    flag_type f = regex_constants::ECMAScript);

[…]

-15- Effects: Assigns the regular expression contained in the string s, interpreted according the flags specified in f. If an exception is thrown, *this is unchanged.

-16- Postconditions: If no exception is thrown, flags() returns f and mark_count() returns the number of marked sub-expressions within the expression.


2139. What is a user-defined type?

Section: 17.6.4.2.1 [namespace.std], 19.5 [syserr], 20.10.7.1 [allocator.uses.trait], 20.14.10.1 [func.bind.isbind], 20.14.10.2 [func.bind.isplace], 20.14.14 [unord.hash], 20.15.7.6 [meta.trans.other], 22.3.1 [locale], 22.4.1.4 [locale.codecvt], 28.12.1.4 [re.regiter.incr] Status: Open Submitter: Loïc Joly Opened: 2012-03-08 Last modified: 2016-02-12

Priority: 4

View all other issues in [namespace.std].

View all issues with Open status.

Discussion:

The expression "user-defined type" is used in several places in the standard, but I'm not sure what it means. More specifically, is a type defined in the standard library a user-defined type?

From my understanding of English, it is not. From most of the uses of this term in the standard, it seem to be considered as user defined. In some places, I'm hesitant, e.g. 17.6.4.2.1 [namespace.std] p1:

A program may add a template specialization for any standard library template to namespace std only if the declaration depends on a user-defined type and the specialization meets the standard library requirements for the original template and is not explicitly prohibited.

Does it mean we are allowed to add in the namespace std a specialization for std::vector<std::pair<T, U>>, for instance?

Additional remarks from the reflector discussion: The traditional meaning of user-defined types refers to class types and enum types, but the library actually means here user-defined types that are not (purely) library-provided. Presumably a new term - like user-provided type - should be introduced and properly defined.

[ 2012-10 Portland: Move to Deferred ]

The issue is real, in that we never define this term and rely on a "know it when I see it" intuition. However, there is a fear that any attempt to pin down a definition is more likely to introduce bugs than solve them - getting the wording for this precisely correct is likely far more work than we are able to give it.

There is unease at simple closing as NAD, but not real enthusiasm to provide wording either. Move to Deferred as we are not opposed to some motivated individual coming back with full wording to review, but do not want to go out of our way to encourage someone to work on this in preference to other issues.

[2014-02-20 Re-open Deferred issues as Priority 4]

[2015-03-05 Jonathan suggests wording]

I dislike the suggestion to change to "user-provided" type because I already find the difference between user-declared / user-provided confusing for special member functions, so I think it would be better to use a completely different term. The core language uses "user-defined conversion sequence" and "user-defined literal" and similar terms for things which the library provides, so I think we should not refer to "user" at all to distinguish entities defined outside the implementation from things provided by the implementation.

I propose "program-defined type" (and "program-defined specialization"), defined below. The P/R below demonstrates the scope of the changes required, even if this name isn't adopted. I haven't proposed a change for "User-defined facets" in [locale].

[Lenexa 2015-05-06]

RS, HT: The core language uses "user-defined" in a specific way, including library things but excluding core language things, let's use a different term.

MC: Agree.

RS: "which" should be "that", x2

RS: Is std::vector<MyType> a "program-defined type"?

MC: I think it should be.

TK: std::vector<int> seems to take the same path.

JW: std::vector<MyType> isn't program-defined, we don't need it to be, anything that depends on that also depends on =MyType.

TK: The type defined by an "explicit template specialization" should be a program-defined type.

RS: An implicit instantiation of a "program-defined partial specialization" should also be a program-defined type.

JY: This definition formatting is horrible and ugly, can we do better?

RS: Checking ISO directives.

RS: Define "program-defined type" and "program-defined specialization" instead, to get rid of the angle brackets.

JW redrafting.

Proposed resolution:

This wording is relative to N4296.

  1. Add a new sub-clause to 17.3 [definitions]:

    17.3.? [defns.program.defined]

    program-defined

    <type> a class type or enumeration type which is not part of the C++ standard library and not defined by the implementation. [Note: Types defined by the implementation include extensions (1.4 [intro.compliance]) and internal types used by the library. — end note]

    program-defined

    <specialization> an explicit template specialization or partial specialization which is not part of the C++ standard library and not defined by the implementation.

  2. Change 17.6.4.2.1 [namespace.std] paragraph 1+2:

    -1- The behavior of a C++ program is undefined if it adds declarations or definitions to namespace std or to a namespace within namespace std unless otherwise specified. A program may add a template specialization for any standard library template to namespace std only if the declaration depends on a userprogram-defined type and the specialization meets the standard library requirements for the original template and is not explicitly prohibited.

    -2- The behavior of a C++ program is undefined if it declares

    […]

    A program may explicitly instantiate a template defined in the standard library only if the declaration depends on the name of a userprogram-defined type and the instantiation meets the standard library requirements for the original template.

  3. Change 19.5 [syserr] paragraph 4:

    -4- The is_error_code_enum and is_error_condition_enum may be specialized for userprogram-defined types to indicate that such types are eligible for class error_code and class error_condition automatic conversions, respectively.

  4. Change 20.10.7.1 [allocator.uses.trait] paragraph 1:

    -1- Remarks: automatically detects […]. A program may specialize this template to derive from true_type for a userprogram-defined type T that does not have a nested allocator_type but nonetheless can be constructed with an allocator where either: […]

  5. Change 20.14.10.1 [func.bind.isbind] paragraph 2:

    -2- Instantiations of the is_bind_expression template […]. A program may specialize this template for a userprogram-defined type T to have a BaseCharacteristic of true_type to indicate that T should be treated as a subexpression in a bind call.

  6. Change 20.14.10.2 [func.bind.isplace] paragraph 2:

    -2- Instantiations of the is_placeholder template […]. A program may specialize this template for a userprogram-defined type T to have a BaseCharacteristic of integral_constant<int, N> with N > 0 to indicate that T should be treated as a placeholder type.

  7. Change 20.14.14 [unord.hash] paragraph 1:

    The unordered associative containers defined in 23.5 use specializations of the class template hash […], the instantiation hash<Key> shall:

  8. Change 20.15.7.5 [meta.trans.ptr] Table 57 (common_type row):

    Table 57 — Other transformations
    Template Condition Comments
    template <class... T>
    struct common_type;
      The member typedef type shall be
    defined or omitted as specified below.
    […]. A program may
    specialize this trait if at least one
    template parameter in the
    specialization is a userprogram-defined type.
    […]
  9. Change 22.4.1.4 [locale.codecvt] paragraph 3:

    -3- The specializations required in Table 81 (22.3.1.1.1) […]. Other encodings can be converted by specializing on a userprogram-defined stateT type.[…]

  10. Change 28.12.1.4 [re.regiter.incr] paragraph 8:

    -8- [Note: This means that a compiler may call an implementation-specific search function, in which case a userprogram-defined specialization of regex_search will not be called. — end note]


2146. Are reference types Copy/Move-Constructible/Assignable or Destructible?

Section: 17.6.3.1 [utility.arg.requirements] Status: Open Submitter: Nikolay Ivchenkov Opened: 2012-03-23 Last modified: 2016-06-21

Priority: 2

View all other issues in [utility.arg.requirements].

View all issues with Open status.

Discussion:

According to 17.6.3.1 [utility.arg.requirements] p1

The template definitions in the C++ standard library refer to various named requirements whose details are set out in tables 17-24. In these tables, T is an object or reference type to be supplied by a C++ program instantiating a template; a, b, and c are values of type (possibly const) T; s and t are modifiable lvalues of type T; u denotes an identifier; rv is an rvalue of type T; and v is an lvalue of type (possibly const) T or an rvalue of type const T.

Is it really intended that T may be a reference type? If so, what should a, b, c, s, t, u, rv, and v mean? For example, are "int &" and "int &&" MoveConstructible?

As far as I understand, we can explicitly specify template arguments for std::swap and std::for_each. Can we use reference types there?

  1. #include <iostream>
    #include <utility>
    
    int main()
    {
      int x = 1;
      int y = 2;
      std::swap<int &&>(x, y); // undefined?
      std::cout << x << " " << y << std::endl;
    }
    
  2. #include <algorithm>
    #include <iostream>
    #include <iterator>
    #include <utility>
    
    struct F
    {
      void operator()(int n)
      {
        std::cout << n << std::endl;
        ++count;
      }
      int count;
    } f;
    
    int main()
    {
      int arr[] = { 1, 2, 3 };
      auto&& result = std::for_each<int *, F &&>( // undefined?
        std::begin(arr),
        std::end(arr),
        std::move(f));
      std::cout << "count: " << result.count << std::endl;
    }
    

Are these forms of usage well-defined?

Let's also consider the following constructor of std::thread:

template <class F, class ...Args>
explicit thread(F&& f, Args&&... args);

Requires: F and each Ti in Args shall satisfy the MoveConstructible requirements.

When the first argument of this constructor is an lvalue (e.g. a name of a global function), template argument for F is deduced to be lvalue reference type. What should "MoveConstructible" mean with regard to an lvalue reference type? Maybe the wording should say that std::decay<F>::type and each std::decay<Ti>::type (where Ti is an arbitrary item in Args) shall satisfy the MoveConstructible requirements?

[2013-03-15 Issues Teleconference]

Moved to Open.

The questions raised by the issue are real, and should have a clear answer.

[2015-10, Kona Saturday afternoon]

STL: std::thread needs to be fixed, and anything behaving like it needs to be fixed, rather than reference types. std::bind gets this right. We need to survey this. GR: That doesn't sound small to me. STL: Seach for CopyConstructible etc. It may be a long change, but not a hard one.

MC: It seems that we don't have a PR. Does anyone have one? Is anyone interested in doing a survey?

[2016-03, Jacksonville]

Casey volunteers to make a survey

[2016-06, Oulu]

During an independent survey performed by Daniel as part of the analysis of LWG 2716, some overlap was found between these two issues. Daniel suggested to take responsibility for surveying LWG 2146 and opined that the P/R of LWG 2716 should restrict to forwarding references, where the deduction to lvalue references can happen without providing an explicit template argument just by providing an lvalue function argument.

Proposed resolution:


2151. basic_string<>::swap semantics ignore allocators

Section: 21.3.1.1 [string.require] Status: Open Submitter: Robert Shearer Opened: 2012-04-13 Last modified: 2016-02-12

Priority: 3

View all other issues in [string.require].

View all issues with Open status.

Discussion:

In C++11, basic_string is not described as a "container", and is not governed by the allocator-aware container semantics described in sub-clause 23.2 [container.requirements]; as a result, and requirements or contracts for the basic_string interface must be documented in Clause 21 [strings].

Sub-clause 21.3.1.6.8 [string::swap] defines the swap member function with no requirements, and with guarantees to execute in constant time without throwing. Fulfilling such a contract is not reasonable in the presence of unequal non-propagating allocators.

In contrast, 23.2.1 [container.requirements.general] p7 declares the behavior of member swap for containers with unequal non-propagating allocators to be undefined.

Resolution proposal:

Additional language from Clause 23 [containers] should probably be copied to Clause 21 [strings]. I will refrain from an exactly recommendation, however, as I am raising further issues related to the language in Clause 23 [containers].

[2013-03-15 Issues Teleconference]

Moved to Open.

Alisdair has offered to provide wording.

Telecon notes that 23.2.1 [container.requirements.general]p13 says that string is an allocator-aware container.

Proposed resolution:


2152. Instances of standard container types are not swappable

Section: 17.6.3.2 [swappable.requirements], 23.2.1 [container.requirements.general] Status: Open Submitter: Robert Shearer Opened: 2012-04-13 Last modified: 2016-03-08

Priority: 2

View all other issues in [swappable.requirements].

View all issues with Open status.

Discussion:

Sub-clause 17.6.3.2 [swappable.requirements] defines two notions of swappability: a binary version defining when two objects are swappable with one another, and a unary notion defining whether an object is swappable (without qualification), with the latter definition requiring that the object satisfy the former with respect to all values of the same type.

Let T be a container type based on a non-propagating allocator whose instances do not necessarily compare equal. Then sub-clause 23.2.1 [container.requirements.general] p7 implies that no object t of type T is swappable (by the unary definition).

Throughout the standard it is the unary definition of "swappable" that is listed as a requirement (with the exceptions of 20.2.2 [utility.swap] p4, 20.4.2 [pairs.pair] p31, 20.5.2.3 [tuple.swap] p2, 25.4.3 [alg.swap] p2, and 25.4.3 [alg.swap] p6, which use the binary definition). This renders many of the mutating sequence algorithms of sub-clause 25.4 [alg.modifying.operations], for example, inapplicable to sequences of standard container types, even where every element of the sequence is swappable with every other.

Note that this concern extends beyond standard containers to all future allocator-based types.

Resolution proposal:

I see two distinct straightforward solutions:

  1. Modify the requirements of algorithms from sub-clause 25.4 [alg.modifying.operations], and all other places that reference the unary "swappable" definition, to instead use the binary "swappable with" definition (over a domain appropriate to the context). The unary definition of "swappable" could then be removed from the standard.
  2. Modify sub-clause 23.2.1 [container.requirements.general] such that objects of standard container types are "swappable" by the unary definition.

I favor the latter solution, for reasons detailed in the following issue.

[ 2012-10 Portland: Move to Open ]

The issue is broader than containers with stateful allocotors, although they are the most obvious example contained within the standard itself. The basic problem is that once you have a stateful allocator, that does not propagate_on_swap, then whether two objects of this type can be swapped with well defined behavior is a run-time property (the allocators compare equal) rather than a simple compile-time property that can be deduced from the type. Strictly speaking, any type where the nature of swap is a runtime property does not meet the swappable requirements of C++11, although typical sequences of such types are going to have elements that are all swappable with any other element in the sequence (using our other term of art for specifying requirements) as the common case is a container of elements who all share the same allocator.

The heart of the problem is that the swappable requirments demand that any two objects of the same type be swappable with each other, so if any two such objects would not be swappable with each other, then the whole type is never swappable. Many algorithms in clause 25 are specified in terms of swappable which is essentially an overspecification as all they actually need is that any element in the sequence is swappable with any other element in the sequence.

At this point Howard joins the discussion and points out that the intent of introducing the two swap-related terms was to support vector<bool>::reference types, and we are reading something into the wording that was never intended. Consuses is that regardless of the intent, that is what the words today say.

There is some support to see a paper reviewing the whole of clause 25 for this issue, and other select clauses as may be necessary.

There was some consideration to introducing a note into the front of clause 25 to indicate swappable requirements in the clause should be interpreted to allow such awkward types, but ultimately no real enthusiasm for introducing a swappable for clause 25 requirement term, especially if it confusingly had the same name as a term used with a subtly different meaning through the rest of the standard.

There was no enthusiasm for the alternate resolution of requiring containers with unequal allocators that do not propagate provide a well-defined swap behavior, as it is not believed to be possible without giving swap linear complexity for such values, and even then would require adding the constraint that the container element types are CopyConstructible.

Final conclusion: move to open pending a paper from a party with a strong interest in stateful allocators.

[2016-03 Jacksonville]

Alisdair says that his paper P0178 addresses this.

Proposed resolution:


2153. Narrowing of the non-member swap contract

Section: 20.2.2 [utility.swap], 17.6.3.2 [swappable.requirements], 23.2.1 [container.requirements.general] Status: Open Submitter: Robert Shearer Opened: 2012-04-13 Last modified: 2016-03-08

Priority: 2

View all other issues in [utility.swap].

View all issues with Open status.

Discussion:

Sub-clause 20.2.2 [utility.swap] defines a non-member 'swap' function with defined behavior for all MoveConstructible and MoveAssignable types. It does not guarantee constant-time complexity or noexcept in general, however this definition does render all objects of MoveConstructible and MoveAssignable type swappable (by the unary definition of sub-clause 17.6.3.2 [swappable.requirements]) in the absence of specializations or overloads.

The overload of the non-member swap function defined in Table 96, however, defines semantics incompatible with the generic non-member swap function, since it is defined to call a member swap function whose semantics are undefined for some values of MoveConstructible and MoveAssignable types.

The obvious (perhaps naive) interpretation of sub-clause 17.6.3.2 [swappable.requirements] is as a guide to the "right" semantics to provide for a non-member swap function (called in the context defined by 17.6.3.2 [swappable.requirements] p3) in order to provide interoperable user-defined types for generic programming. The standard container types don't follow these guidelines.

More generally, the design in the standard represents a classic example of "contract narrowing". It is entirely reasonable for the contract of a particular swap overload to provide more guarantees, such as constant-time execution and noexcept, than are provided by the swap that is provided for any MoveConstructible and MoveAssignable types, but it is not reasonable for such an overload to fail to live up to the guarantees it provides for general types when it is applied to more specific types. Such an overload or specialization in generic programming is akin to an override of an inherited virtual function in OO programming: violating a superclass contract in a subclass may be legal from the point of view of the language, but it is poor design and can easily lead to errors. While we cannot prevent user code from providing overloads that violate the more general swap contract, we can avoid doing so within the library itself.

My proposed resolution is to draw a sharp distinction between member swap functions, which provide optimal performance but idiosyncratic contracts, and non-member swap functions, which should always fulfill at least the contract of 20.2.2 [utility.swap] and thus render objects swappable. The member swap for containers with non-propagating allocators, for example, would offer constant-time guarantees and noexcept but would only offer defined behavior for values with allocators that compare equal; non-member swap would test allocator equality and then dispatch to either member swap or std::swap depending on the result, providing defined behavior for all values (and rendering the type "swappable"), but offering neither the constant-time nor the noexcept guarantees.

[2013-03-15 Issues Teleconference]

Moved to Open.

This topic deserves more attention than can be given in the telocon, and there is no proposed resolution.

[2013-03-15 Issues Teleconference]

Moved to Open.

This topic deserves more attention than can be given in the telocon, and there is no proposed resolution.

[2016-03 Jacksonville]

Alisdair says that his paper P0178 addresses this.

Proposed resolution:


2154. What exactly does compile-time complexity imply?

Section: 26.6.1.3 [rand.req.urng] Status: New Submitter: John Salmon Opened: 2012-04-26 Last modified: 2016-02-12

Priority: 4

View all other issues in [rand.req.urng].

View all issues with New status.

Discussion:

The expressions G::min() and G::max() in Table 116 in 26.6.1.3 [rand.req.urng] are specified as having "compile-time" complexity.

It is not clear what, exactly, this requirement implies. If a URNG has a method:

static int min();

then is the method required to have a constexpr qualifier? I believe the standard would benefit from clarification of this point.

Proposed resolution:


2155. Macro __bool_true_false_are_defined should be removed

Section: 18.10 [support.runtime] Status: Open Submitter: Thomas Plum Opened: 2012-04-30 Last modified: 2016-02-12

Priority: 4

View other active issues in [support.runtime].

View all other issues in [support.runtime].

View all issues with Open status.

Discussion:

Since C99, the C standard describes a macro named __bool_true_false_are_defined.

In the process of harmonizing C++11 with C99, this name became part of the C++ standard.

I propose that all mention of this name should be removed from the C and C++ standards.

Here's the problem: The name was originally proposed as a transition tool, so that the headers for a project could contain lines like the following.

#if !defined(__bool_true_false_are_defined)
#define bool int /* or whatever */
#define true 1
#define false 0
#endif

Then when the project was compiled by a "new" compiler that implemented bool as defined by the evolving C++98 or C99 standards, those lines would be skipped; but when compiled by an "old" compiler that didn't yet provide bool, true, and false, then the #define's would provide a simulation that worked for most purposes.

It turns out that there is an unfortunate ambiguity in the name. One interpretation is as shown above, but a different reading says "bool, true, and false are #define'd", i.e. that the meaning of the macro is to assert that these names are macros (not built-in) ... which is true in C, but not in C++.

In C++11, the name appears in parentheses followed by a stray period, so some editorial change is needed in any event:

18.10 [support.runtime] para 1:

Headers <csetjmp> (nonlocal jumps), <csignal> (signal handling), <cstdalign> (alignment), <cstdarg> (variable arguments), <cstdbool> (__bool_true_false_are_defined). <cstdlib> (runtime environment getenv(), system()), and <ctime> (system clock clock(), time()) provide further compatibility with C code.

However, para 2 says

"The contents of these headers are the same as the Standard C library headers <setjmp.h>, <signal.h>, <stdalign.h>, <stdarg.h>, <stdbool.h>, <stdlib.h>, and <time.h>, respectively, with the following changes:",

and para 8 says

"The header <cstdbool> and the header <stdbool.h> shall not define macros named bool, true, or false."

Thus para 8 doesn't exempt the C++ implementation from the arguably clear requirement of the C standard, to provide a macro named __bool_true_false_are_defined defined to be 1.

Real implementations of the C++ library differ, so the user cannot count upon any consistency; furthermore, the usefulness of the transition tool has faded long ago.

That's why my suggestion is that both C and C++ standards should eliminate any mention of __bool_true_false_are_defined. In that case, the name belongs to implementers to provide, or not, as they choose.

[2013-03-15 Issues Teleconference]

Moved to Open.

While not strictly necessary, the clean-up look good.

We would like to hear from our C liaison before moving on this issue though.

[2015-05 Lenexa]

LWG agrees. Jonathan provides wording.

Proposed resolution:

This wording is relative to N4296.

  1. Edit the footnote on 17.6.1.2 [headers] p7:

    176) In particular, including any of the standard headers <stdbool.h>, <cstdbool>, <iso646.h> or <ciso646> has no effect.

  2. Edit 18.10 [support.runtime] p1 as indicated (and remove the index entry for __bool_true_false_are_defined):

    -1- Headers <csetjmp> (nonlocal jumps), <csignal> (signal handling), <cstdalign> (alignment), <cstdarg> (variable arguments), <cstdbool>, (__bool_true_false_are_defined). <cstdlib> (runtime environment getenv(), system()), and <ctime> (system clock clock(), time()) provide further compatibility with C code.

  3. Remove Table 38 — Header <cstdbool> synopsis [tab:support.hdr.cstdbool] from 18.10 [support.runtime]

    Table 38 — Header <cstdbool> synopsis
    TypeName(s)
    Macro:__bool_true_false_are_defined

2157. How does std::array<T,0> initialization work when T is not default-constructible?

Section: 23.3.7.8 [array.zero] Status: Open Submitter: Daryle Walker Opened: 2012-05-08 Last modified: 2016-02-12

Priority: 3

View all other issues in [array.zero].

View all issues with Open status.

Discussion:

Objects of std::array<T,N> are supposed to be initialized with aggregate initialization (when not the destination of a copy or move). This clearly works when N is positive. What happens when N is zero? To continue using an (inner) set of braces for initialization, a std::array<T,0> implementation must have an array member of at least one element, and let default initialization take care of those secret elements. This cannot work when T has a set of constructors and the default constructor is deleted from that set. Solution: Add a new paragraph in 23.3.7.8 [array.zero]:

The unspecified internal structure of array for this case shall allow initializations like:

array<T, 0> a = { };

and said initializations must be valid even when T is not default-constructible.

[2012, Portland: Move to Open]

Some discussion to understand the issue, which is that implementations currently have freedom to implement an empty array by holding a dummy element, and so might not support value initialization, which is surprising when trying to construct an empty container. However, this is not mandated, it is an unspecified implementation detail.

Jeffrey points out that the implication of 23.3.7.1 [array.overview] is that this initialization syntax must be supported by empty array objects already. This is a surprising inference that was not obvious to the room, but consensus is that the reading is accurate, so the proposed resolution is not necessary, although the increased clarity may be useful.

Further observation is that the same clause effectively implies that T must always be DefaultConstructible, regardless of N for the same reasons - as an initializer-list may not supply enough values, and the remaining elements must all be value initialized.

Concern that we are dancing angels on the head of pin, and that relying on such subtle implications in wording is not helpful. We need a clarification of the text in this area, and await wording.

[2015-02 Cologne]

DK: What was the outcome of Portland? AM: Initially we thought we already had the intended behaviour. We concluded that T must always be DefaultConstructible, but I'm not sure why. GR: It's p2 in std::array, "up to N". AM: That wording already implies that "{}" has to work when N is zero. But the wording of p2 needs to be fixed to make clear that it does not imply that T must be DefaultConstructible.

Conclusion: Update wording, revisit later.

[2015-10, Kona Saturday afternoon]

MC: How important is this? Can you not just use default construction for empty arrays?

TK: It needs to degenerate properly from a pack. STL agrees.

JW: Yes, this is important, and we have to make it work.

MC: I hate the words "initialization like".

JW: I'll reword this.

WEB: Can I ask that once JW has reworded this we move it to Review rather than Open?

MC: We'll try to review it in a telecon and hopefully get it to tentatively ready.

STL: Double braces must also work: array<T, 0> a = {{}};.

Jonathan to reword.

Proposed resolution:

This wording is relative to N3376.

Add the following new paragraph between the current 23.3.7.8 [array.zero] p1 and p2:

-1- array shall provide support for the special case N == 0.

-?- The unspecified internal structure of array for this case shall allow initializations like:

array<T, 0> a = { };

and said initializations must be valid even when T is not default-constructible.

-2- In the case that N == 0, begin() == end() == unique value. The return value of data() is unspecified.

-3- The effect of calling front() or back() for a zero-sized array is undefined.

-4- Member function swap() shall have a noexcept-specification which is equivalent to noexcept(true).


2158. Conditional copy/move in std::vector

Section: 23.3.11.3 [vector.capacity] Status: New Submitter: Nikolay Ivchenkov Opened: 2012-05-08 Last modified: 2016-02-12

Priority: 2

View other active issues in [vector.capacity].

View all other issues in [vector.capacity].

View all issues with New status.

Discussion:

There are various operations on std::vector that can cause elements of the vector to be moved from one location to another. A move operation can use either rvalue or const lvalue as argument; the choice depends on the value of !is_nothrow_move_constructible<T>::value && is_copy_constructible<T>::value, where T is the element type. Thus, some operations on std::vector (e.g. 'resize' with single parameter, 'reserve', 'emplace_back') should have conditional requirements. For example, let's consider the requirement for 'reserve' in N3376 – 23.3.11.3 [vector.capacity]/2:

Requires: T shall be MoveInsertable into *this.

This requirement is not sufficient if an implementation is free to select copy constructor when !is_nothrow_move_constructible<T>::value && is_copy_constructible<T>::value evaluates to true. Unfortunately, is_copy_constructible cannot reliably determine whether T is really copy-constructible. A class may contain public non-deleted copy constructor whose definition does not exist or cannot be instantiated successfully (e.g., std::vector<std::unique_ptr<int>> has copy constructor, but this type is not copy-constructible). Thus, the actual requirements should be:

Maybe it would be useful to introduce a new name for such conditional requirement (in addition to "CopyInsertable" and "MoveInsertable").

Proposed resolution:


2161. const equivalence of std::map

Section: 23.4 [associative], 23.5 [unord] Status: New Submitter: Bjarne Stroustrup Opened: 2012-06-18 Last modified: 2016-02-12

Priority: 2

View other active issues in [associative].

View all other issues in [associative].

View all issues with New status.

Discussion:

As described in the reflector discussion c++std-core-21860 consider the following example:

map<const int, int> mci{};
map<int, int> mi = mci; // ??
mci[1] = 2;
mi[1] = 2;

Should it be required that the marked initialization is well-formed? As a possible solution this could be realized by an alias template:

template <class K, class T>
struct OriginalMap { […] };

template <class K, class T>
using ImprovedMap = OriginalMap<const K, T>;

Proposed resolution:


2164. What are the semantics of vector.emplace(vector.begin(), vector.back())?

Section: 23.3.11.5 [vector.modifiers], 23.2 [container.requirements] Status: Open Submitter: Howard Hinnant Opened: 2012-07-07 Last modified: 2016-02-12

Priority: 2

View other active issues in [vector.modifiers].

View all other issues in [vector.modifiers].

View all issues with Open status.

Discussion:

Nikolay Ivchenkov recently brought the following example on the std-discussion newsgroup, asking whether the following program well-defined:

#include <iostream>
#include <vector>

int main()
{
  std::vector<int> v;
  v.reserve(4);
  v = { 1, 2, 3 };
  v.emplace(v.begin(), v.back());
  for (int x : v)
    std::cout << x << std::endl;
}

Nikolay Ivchenkov:

I think that an implementation of vector's 'emplace' should initialize an intermediate object with v.back() before any shifts take place, then perform all necessary shifts and finally replace the value pointed to by v.begin() with the value of the intermediate object. So, I would expect the following output:

3
1
2
3

GNU C++ 4.7.1 and GNU C++ 4.8.0 produce other results:

2
1
2
3

Howard Hinnant:

I believe Nikolay is correct that vector should initialize an intermediate object with v.back() before any shifts take place. As Nikolay pointed out in another email, this appears to be the only way to satisfy the strong exception guarantee when an exception is not thrown by T's copy constructor, move constructor, copy assignment operator, or move assignment operator as specified by 23.3.11.5 [vector.modifiers]/p1. I.e. if the emplace construction throws, the vector must remain unaltered.

That leads to an implementation that tolerates objects bound to the function parameter pack of the emplace member function may be elements or sub-objects of elements of the container.

My position is that the standard is correct as written, but needs a clarification in this area. Self-referencing emplace should be legal and give the result Nikolay expects. The proposed resolution of LWG 760 is not correct.

[2015-02 Cologne]

LWG agrees with the analysis including the assessment of LWG 760 and would appreciate a concrete wording proposal.

[2015-04-07 dyp comments]

The Standard currently does not require that creation of such intermediate objects is legal. 23.2.3 [sequence.reqmts] Table 100 — "Sequence container requirements" currently specifies:

Table 100 — Sequence container requirements
Expression Return type Assertion/note
pre-/post-condition
a.emplace(p, args); iterator Requires: T is EmplaceConstructible into X from args. For vector and deque, T is also MoveInsertable into X and MoveAssignable. […]

The EmplaceConstructible concept is defined via allocator_traits<A>::construct in 23.2.1 [container.requirements.general] p15.5 That's surprising to me since the related concepts use the suffix Insertable if they refer to the allocator. An additional requirement such as std::is_constructible<T, Args...> is necessary to allow creation of intermediate objects.

The creation of intermediate objects also affects other functions, such as vector.insert. Since aliasing the vector is only allowed for the single-element forms of insert and emplace (see 526), the range-forms are not affected. Similarly, aliasing is not allowed for the rvalue-reference overload. See also LWG 2266.

There might be a problem with a requirement of std::is_constructible<T, Args...> related to the issues described in LWG 2461. For example, a scoped allocator adapter passes additional arguments to the constructor of the value type. This is currently not done in recent implementations of libstdc++ and libc++ when creating the intermediate objects, they simply create the intermediate object by perfectly forwarding the arguments. If such an intermediate object is then moved to its final destination in the vector, a change of the allocator instance might be required — potentially leading to an expensive copy. One can also imagine worse problems, such as run-time errors (allocators not comparing equal at run-time) or compile-time errors (if the value type cannot be created without the additional arguments). I have not looked in detail into this issue, but I'd be reluctant adding a requirement such as std::is_constructible<T, Args...> without further investigation.

It should be noted that the creation of intermediate objects currently is inconsistent in libstdc++ vs libc++. For example, libstdc++ creates an intermediate object for vector.insert, but not vector.emplace, whereas libc++ does the exact opposite in this respect.

A live demo of the inconsistent creation of intermediate objects can be found here.

[2015-10, Kona Saturday afternoon]

HH: If it were easy, it'd have wording. Over the decades I have flipped 180 degrees on this. My current position is that it should work even if the element is in the same container.

TK: What's the implentation status? JW: Broken in GCC. STL: Broken in MSVS. Users complain about this every year.

MC: 526 says push_back has to work.

HH: I think you have to make a copy of the element anyway for reasons of exception safety. [Discussion of exception guarantees]

STL: vector has strong exception guarantees. Could we not just provide the Basic guarantee here.

HH: It would terrify me to relax that guarantee. It'd be an ugly, imperceptible runtime error.

HH: I agree if we had a clean slate that strong exception safety is costing us here, and we shouldn't provide it if it costs us.

STL: I have a mail here, "how can vector provide the strong guarantee when inserting in the middle".

HH: The crucial point is that you only get the strong guarantee if the exception is thrown by something other than the copy and move operations that are used to make the hole.

STL: I think we need to clean up the wording. But it does mandate currently that the self-emplacement must work, because nothings says that you can't do it. TK clarifies that a) self-emplacement must work, and b) you get the strong guarantee only if the operations for making the hole don't throw, otherwise basic. HH agrees. STL wants this to be clear in the Standard.

STL: Should it work for deque, too? HH: Yes.

HH: I will attempt wording for this.

TK: Maybe mail this to the reflector, and maybe someone has a good idea?

JW: I will definitely not come up with anything better, but I can critique wording.

Moved to Open; Howard to provide wording, with feedback from Jonathan.

Proposed resolution:


2166. Heap property underspecified?

Section: 25.5.6 [alg.heap.operations] Status: New Submitter: Peter Sommerlad Opened: 2012-07-09 Last modified: 2016-02-12

Priority: 3

View all other issues in [alg.heap.operations].

View all issues with New status.

Discussion:

Another similar issue to the operator< vs greater in nth_element but not as direct occurs in 25.5.6 [alg.heap.operations]:

-1- A heap is a particular organization of elements in a range between two random access iterators [a,b). Its two key properties are:

  1. There is no element greater than *a in the range and
  2. *a may be removed by pop_heap(), or a new element added by push_heap(), in O(log(N)) time.

As noted by Richard Smith, it seems that the first bullet should read:

*a is not less than any element in the range

Even better the heap condition could be stated here directly, instead of leaving it unspecified, i.e.,

Each element at (a+2*i+1) and (a+2*i+2) is less than the element at (a+i), if those elements exist, for i>=0.

But may be that was may be intentional to allow other heap organizations?

See also follow-up discussion of c++std-lib-32780.

Proposed resolution:


2173. The meaning of operator + in the description of the algorithms

Section: 25 [algorithms] Status: New Submitter: Nikolay Ivchenkov Opened: 2012-08-01 Last modified: 2016-02-12

Priority: 4

View other active issues in [algorithms].

View all other issues in [algorithms].

View all issues with New status.

Discussion:

According to 25.1 [algorithms.general]/12,

In the description of the algorithms operators + and - are used for some of the iterator categories for which they do not have to be defined. In these cases the semantics of a+n is the same as that of

X tmp = a;
advance(tmp, n);
return tmp;

There are several places where such operator + is applied to an output iterator — for example, see the description of std::copy:

template<class InputIterator, class OutputIterator>
OutputIterator copy(InputIterator first, InputIterator last,
                    OutputIterator result);

-1- Effects: Copies elements in the range [first,last) into the range [result,result + (last - first)) starting from first and proceeding to last. For each non-negative integer n < (last - first), performs *(result + n) = *(first + n).

std::advance is not supposed to be applicable to output iterators, so we need a different method of description.

See also message c++std-lib-32908.

[2014-06-07 Daniel comments and provides wording]

The specification for output iterators is somewhat tricky, because here a sequence of increments is required to be combined with intervening assignments to the dereferenced iterator. I tried to respect this fact by using a conceptual assignment operation as part of the specification.

Another problem in the provided as-if-code is the question which requirements are imposed on n. Unfortunately, the corresponding function advance is completely underspecified in this regard, so I couldn't borrow wording from it. We cannot even assume here that n is the difference type of the iterator, because for output iterators there is no requirements for this associated type to be defined. The presented wording attempts to minimize assumptions, but still can be considered as controversial.

Proposed resolution:

This wording is relative to N3936.

  1. Change 25.1 [algorithms.general] around p12 as indicated:

    -12- In the description of the algorithms operators + and - are used for some of the iterator categories for which they do not have to be defined. In these cases the semantics of a+n is the same as that of

    X tmp = a;
    advance(tmp, n);
    return tmp;
    

    when X meets the input iterator requirements (24.2.3 [input.iterators]), otherwise it is the same as that of

    X tmp = a;
    for (auto i = n; i; ++tmp, (void) --i) 
      *tmp = Expr(i); 
    return tmp;
    

    where Expr(i) denotes the n-i-th expression that is assigned to for the corresponding algorithm; and that of b-a is the same as of

    return distance(a, b);
    

2178. Allocator requirement changes not mentioned Annex C

Section: 17.6.3.5 [allocator.requirements], C.5 [diff.library] Status: Open Submitter: Nevin Liber Opened: 2012-08-14 Last modified: 2016-02-12

Priority: 3

View other active issues in [allocator.requirements].

View all other issues in [allocator.requirements].

View all issues with Open status.

Discussion:

Given that a number of things were removed from the allocator requirements (reference, const_reference, address() in 17.6.3.5 [allocator.requirements]), it seems that these incompatible changes should be mentioned in Annex C.5 [diff.library], more specifically in [diff.cpp03].

[ 2012-10 Portland: Move to Open ]

It was clearly pointed out by Bill during the C++11 process that our change to allocator requirements potentially broke 3rd party user containers written to expect C++03 allocators, or rather, an allocator written to the minimal requirements of C++11 might not be guaranteed to work with a container written to the previous rules. This was a trade-off in making allocaters easier to write by use of the allocator_traits framework.

This probably does merit a write-up in Annex C, and we look forward to seeing wording. Until then, the best we can do is move the issue to Open.

Proposed resolution:


2179. enable_shared_from_this and construction from raw pointers

Section: 20.11.2.5 [util.smartptr.enab], 20.11.2.2.1 [util.smartptr.shared.const] Status: Tentatively Resolved Submitter: Daniel Krügler Opened: 2012-08-16 Last modified: 2016-05-17

Priority: 3

View other active issues in [util.smartptr.enab].

View all other issues in [util.smartptr.enab].

View all issues with Tentatively Resolved status.

Discussion:

On reflector message c++std-lib-32927, Matt Austern asked whether the following example should be well-defined or not

struct X : public enable_shared_from_this<X> { };
auto xraw = new X;
shared_ptr<X> xp1(xraw);
shared_ptr<X> xp2(xraw);

pointing out that 20.11.2.2.1 [util.smartptr.shared.const] does not seem to allow it, since xp1 and xp2 aren't allowed to share ownership, because each of them is required to have use_count() == 1. Despite this wording it might be reasonable (and technical possible) to implement that request.

On the other hand, there is the non-normative note in 20.11.2.5 [util.smartptr.enab] p11 (already part of TR1):

The shared_ptr constructors that create unique pointers can detect the presence of an enable_shared_from_this base and assign the newly created shared_ptr to its __weak_this member.

Now according to the specification in 20.11.2.2.1 [util.smartptr.shared.const] p3-7:

template<class Y> explicit shared_ptr(Y* p);

the notion of creating unique pointers can be read to be included by this note, because the post-condition of this constructor is unique() == true. Evidence for this interpretation seems to be weak, though.

Howard Hinnant presented the counter argument, that actually the following is an "anti-idiom" and it seems questionable to teach it to be well-defined in any case:

auto xraw = new X;
shared_ptr<X> xp1(xraw);
shared_ptr<X> xp2(xraw);

He also pointed out that the current post-conditions of the affected shared_ptr constructor would need to be reworded.

It needs to be decided, which direction to follow. If this idiom seems too much broken to be supported, the note could be improved. If it should be supported, the constructors in 20.11.2.2.1 [util.smartptr.shared.const] need a careful analysis to ensure that post-conditions are correct.

Several library implementations currently do not support this example, instead they typically cause a crash. Matt points out that there are currently no explicit requirements imposed on shared_ptr objects to prevent them from owning the same underlying object without sharing the ownership. It might be useful to add such a requirement.

[2013-03-15 Issues Teleconference]

Moved to Open.

More discussion is needed to pick a direction to guide a proposed resolution.

[2013-05-09 Jonathan comments]

The note says the newly created shared_ptr is assigned to the weak_ptr member. It doesn't say before doing that the shared_ptr should check if the weak_ptr is non-empty and possibly share ownership with some other pre-existing shared_ptr.

[2015-08-26 Daniel comments]

LWG issue 2529 is independent but related to this issue.

[2016-03-16, Alisdair comments]

This issues should be closed as Resolved by paper p0033r1 at Jacksonville.

Proposed resolution:


2183. Muddled allocator requirements for match_results constructors

Section: 28.10.1 [re.results.const], 28.10.6 [re.results.all] Status: New Submitter: Pete Becker Opened: 2012-08-29 Last modified: 2016-02-12

Priority: 3

View other active issues in [re.results.const].

View all other issues in [re.results.const].

View all issues with New status.

Discussion:

28.10.1 [re.results.const] p1 says:

In all match_results constructors, a copy of the Allocator argument shall be used for any memory allocation performed by the constructor or member functions during the lifetime of the object.

There are three constructors:

match_results(const Allocator& = Allocator());
match_results(const match_results& m);
match_results(match_results&& m) noexcept;

The second and third constructors do no have an Allocator argument, so despite the "all match_results constructors", it is not possible to use "the Allocator argument" for the second and third constructors.

The requirements for those two constructors also does not give any guidance. The second constructor has no language about allocators, and the third states that the stored Allocator value is move constructed from m.get_allocator(), but doesn't require using that allocator to allocate memory.

The same basic problem recurs in 28.10.6 [re.results.all], which gives the required return value for get_allocator():

Returns: A copy of the Allocator that was passed to the object's constructor or, if that allocator has been replaced, a copy of the most recent replacement.

Again, the second and third constructors do not take an Allocator, so there is nothing that meets this requirement when those constructors are used.

Proposed resolution:


2184. Muddled allocator requirements for match_results assignments

Section: 28.10.1 [re.results.const], 28.10.6 [re.results.all] Status: New Submitter: Pete Becker Opened: 2012-08-29 Last modified: 2016-02-12

Priority: 3

View other active issues in [re.results.const].

View all other issues in [re.results.const].

View all issues with New status.

Discussion:

The effects of the two assignment operators are specified in Table 141. Table 141 makes no mention of allocators, so, presumably, they don't touch the target object's allocator. That's okay, but it leaves the question: match_results::get_allocator() is supposed to return "A copy of the Allocator that was passed to the object's constructor or, if that allocator has been replaced, a copy of the most recent replacement"; if assignment doesn't replace the allocator, how can the allocator be replaced?

Proposed resolution:


2189. Throwing swap breaks unordered containers' state

Section: 23.2.5.1 [unord.req.except] Status: Open Submitter: Alisdair Meredith Opened: 2012-09-23 Last modified: 2016-02-12

Priority: 3

View all issues with Open status.

Discussion:

The hash functor and key-comparison functor of unordered containers are allowed to throw on swap.

23.2.5.1 [unord.req.except]p3 "For unordered associative containers, no swap function throws an exception unless that exception is thrown by the swap of the container's Hash or Pred object (if any)."

In such a case we must offer the basic exception safety guarantee, where both objects are left in valid but unspecified states, and no resources are leaked. This yields a corrupt, un-usable container if the first swap succeeds, but the second fails by throwing, as the functors form a matched pair.

So our basic scenario is first, swap the allocators if the allocators propagate on swap, according to allocator_traits. Next we swap the pointers to our internal hash table data structures, so that they match the allocators that allocated them. (Typically, this operation cannot throw). Now our containers are back in a safely destructible state if an exception follows.

Next, let's say we swap the hash functor, and that throws. We have a corrupt data structure, in that the buckets are not correctly indexed by the correct functors, lookups will give unpredicatable results etc. We can safely restore a usable state by forcibly clearing each container - which does not leak resources and leaves us with two (empty but) usable containers.

Now let us assume that the hasher swap succeeds. Next we swap the equality comparator functor, and this too could throw. The important point to bear in mind is that these two functors form an important pairing - two objects that compare equal by the equality functor must also hash to the same value. If we swap one without the other, we most likely leave the container in an unusable state, even if we clear out all elements.

1. A colleague pointed out that the solution for this is to dynamically allocate the two functors, and then we need only swap pointers, which is not a throwing operation. And if we don't want to allocate on default construction (a common QoI request), we might consider moving to a dynamically allocated functors whenever swap is called, or on first insertion. Of course, allocating memory in swap is a whole new can of worms, but this does not really sound like the design we had intended.

2. The simplest option is to say that we do not support hasher or equality functors that throw on ADL swap. Note that the requirement is simply to not throw, rather than to be explicitly marked as noexcept. Throwing functors are allowed, so long as we never use values that would actually manifest a throw when used in an unordered container.

Pablo went on to give me several more options, to be sure we have a full set to consider:

3. Disallow one or the other functor from throwing. In that case, the possibly-throwing functor must be swapped first, then the other functor, the allocator, and the data pointer(s) afterwards (in any order -- there was a TC that allocator assignment and swap may not throw if the corresponding propagation trait is true.). Of course, the question becomes: which functor is allowed to throw and which one is not?

4. Require that any successful functor swap be reliably reversible. This is very inventive. I know of no other place in the standard where such a requirement is stated, though I have occasionally wanted such a guarantee.

5. Allow a failed swap to leave the containers in a state where future insertions may fail for reasons other than is currently allowed. Specifically, if the hash and equality functors are out of sync, all insertions will fail. Presumably some "incompletely swapped" exception would be thrown. This is "slightly" inventive, although people have been discussing "radioactive" states for a while.

[2013-03-15 Issues Teleconference]

Moved to Open.

Proposed resolution:


2191. Incorrect specification of match_results(match_results&&)

Section: 28.10.1 [re.results.const] Status: New Submitter: Pete Becker Opened: 2012-10-02 Last modified: 2016-02-12

Priority: 4

View other active issues in [re.results.const].

View all other issues in [re.results.const].

View all issues with New status.

Discussion:

28.10.1 [re.results.const]/3: "Move-constructs an object of class match_results satisfying the same postconditions as Table 141."

Table 141 lists various member functions and says that their results should be the results of the corresponding member function calls on m. But m has been moved from, so the actual requirement ought to be based on the value that m had before the move construction, not on m itself.

In addition to that, the requirements for the copy constructor should refer to Table 141.

Ganesh:

Also, the requirements for move-assignment should refer to Table 141. Further it seems as if in Table 141 all phrases of "for all integers n < m.size()" should be replaced by "for all unsigned integers n < m.size()".

Proposed resolution:


2195. Missing constructors for match_results

Section: 28.10 [re.results] Status: Open Submitter: Daniel Krügler Opened: 2012-10-06 Last modified: 2016-02-12

Priority: 3

View other active issues in [re.results].

View all other issues in [re.results].

View all issues with Open status.

Discussion:

The requirement expressed in 28.10 [re.results] p2

The class template match_results shall satisfy the requirements of an allocator-aware container and of a sequence container, as specified in 23.2.3 [sequence.reqmts], except that only operations defined for const-qualified sequence containers are supported.

can be read to require the existence of the described constructors from as well, but they do not exist in the synopsis.

The missing sequence constructors are:

match_results(initializer_list<value_type>);
match_results(size_type, const value_type&);
template<class InputIterator> match_results(InputIterator, InputIterator);

The missing allocator-aware container constructors are:

match_results(const match_results&, const Allocator&);
match_results(match_results&&, const Allocator&);

It should be clarified, whether (a) constructors are an exception of above mentioned operations or (b) whether at least some of them (like those accepting a match_results value and an allocator) should be added.

As visible in several places of the standard (including the core language), constructors seem usually to be considered as "operations" and they certainly can be invoked for const-qualified objects.

The below given proposed resolution applies only the minimum necessary fix, i.e. it excludes constructors from above requirement.

[2013-04-20, Bristol]

Check current implementations to see what they do and, possibly, write a paper.

[2013-09 Chicago]

Ask Daniel to update the proposed wording to include the allocator copy and move constructors.

[2014-01-18 Daniel changes proposed resolution]

Previous resolution from Daniel [SUPERSEDED]:

  1. Change 28.10 [re.results] p2 as indicated:

    The class template match_results shall satisfy the requirements of an allocator-aware container and of a sequence container, as specified in 23.2.3 [sequence.reqmts], except that only operations defined for const-qualified sequence containers that are not constructors are supported.

[2015-05-06 Lenexa]

MC passes important knowledge to EF.

VV, RP: Looks good.

TK: Second form should be conditionally noexcept

JY: Sequence constructors are not here, but mentioned in the issue writeup. Why?

TK: That would have been fixed by the superseded wording.

JW: How does this interact with Mike Spertus' allocator-aware regexes? [...] Perhaps it doesn't.

JW: Can't create match_results, want both old and new resolution.

JY: It's problematic that users can't create these, but not this issue.

VV: Why conditional noexcept?

MC: Allocator move might throw.

JW: Update superseded wording to "only non-constructor operations that are"?

MC: Only keep superseded, but append "and the means of constructing match_results are limited to [...]"?

JY: Bullet 4 paragraph 2 needs to address the allocator constructor.

Assigned to JW for drafting.

[2015-10, Kona Saturday afternoon]

STL: I want Mike Spertus to be aware of this issue.

Proposed resolution:

This wording is relative to N3936.

  1. Change 28.10 [re.results] p4, class template match_results synopsis, as indicated:

    […]
    // 28.10.1, construct/copy/destroy:
    explicit match_results(const Allocator& a = Allocator());
    match_results(const match_results& m);
    match_results(const match_results& m, const Allocator& a);
    match_results(match_results&& m) noexcept;
    match_results(match_results&& m, const Allocator& a) noexcept;
    […]
    
  2. Change 28.10.1 [re.results.const] as indicated: [Drafting note: Paragraph 6 as currently written, makes not much sense, because the noexcept does not allow any exception to propagate. Further-on, the allocator requirements do not allow for throwing move constructors. Deleting it seems to be near to editorial — end drafting note]

    match_results(const match_results& m);
    match_results(const match_results& m, const Allocator& a);
    

    -4- Effects: Constructs an object of class match_results, as a copy of m.

    match_results(match_results&& m) noexcept;
    match_results(match_results&& m, const Allocator& a) noexcept;
    

    -5- Effects: Move-constructs an object of class match_results from m satisfying the same postconditions as Table 142. AdditionallyFor the first form, the stored Allocator value is move constructed from m.get_allocator().

    -6- Throws: Nothing if the allocator's move constructor throws nothing.


2198. max_load_factor(z) makes no strong guarantees, but bans useful behavior

Section: 23.2.5 [unord.req] Status: Open Submitter: Alisdair Meredith Opened: 2012-10-09 Last modified: 2016-02-12

Priority: 3

View other active issues in [unord.req].

View all other issues in [unord.req].

View all issues with Open status.

Discussion:

The user cannot specify a max_load_factor for their unordered container at construction, it must be supplied after the event, when the container is potentially not empty. The contract for this method is deliberately vague, not guaranteeing to use the value supplied by the user, and any value actually used will be used as a ceiling that the container will attempt to respect.

The only guarantee we have is that, if user requests a max_load_factor that is less than the current load_factor, then the operation will take constant time, thus outlawing an implementation that chooses to rehash and so preserve as a class invariant that load_factor < max_load_factor.

Reasonable options conforming to the standard include ignoring the user's request if the requested value is too low, or deferring the rehash to the next insert operation and allowing the container to have a strange state (wrt max_load_factor) until then - and there is still the question of rehashing if the next insert is for a duplicate key in a unique container.

Given the deliberate vagueness of the current wording, to support a range of reasonable (but not perfect) behaviors, it is not clear why the equally reasonable rehash to restore the constraint should be outlawed. It is not thought that this is a performance critical operation, where users will be repeatedly setting low load factors on populated containers, in a tight or (less unlikely) an instant response scenario.

[2013-03-15 Issues Teleconference]

Moved to Open.

Alisdair to provide wording.

Proposed resolution:


2199. unordered containers are required to have an initial max load factor of 1.0

Section: 23.2.5 [unord.req] Status: Open Submitter: Alisdair Meredith Opened: 2012-10-09 Last modified: 2016-02-12

Priority: 3

View other active issues in [unord.req].

View all other issues in [unord.req].

View all issues with Open status.

Discussion:

The default constructor, allocator-aware constructor, and range-based constructors for the unordered containers do not offer a means to control the initial max_load_factor, so the standard mandates the value 1.0. This seems overly restrictive, as there is plenty of research suggesting a value between 0.5 and 1.0 is more often optimal for unique-key containers, and perhaps a slightly higher value might be appropriate for multi-containers.

Rather than guess at the appropriate max_load_factor, it seems reasonable that the standard should allow vendors to pick a value at their discretion, with perhaps a note of advice. It is less clear whether the default value should be implementation-defined or unspecified, given the ease of a user determining this by querying this attribute immediately after construction.

[2013-03-15 Issues Teleconference]

Moved to Open.

Alisdair to provide wording.

Marshall: It seems to me that what you really want is to be able to pass a max load factor in the constructor, but that's a different issue.

Alisdair agrees in principle, but concerned with adding yet more constructors to these classes.

Proposed resolution:


2201. Missing macro entries from C standard library

Section: C.5 [diff.library] Status: New Submitter: Kevin McCarty Opened: 2012-02-03 Last modified: 2016-02-12

Priority: 2

View all other issues in [diff.library].

View all issues with New status.

Discussion:

It seems that in C.5 [diff.library], Table 150 the following macros from 18.3.3 [c.limits], Table 31 are missing:

LLONG_MIN 
LLONG_MAX
ULLONG_MAX

In addition in C.5 [diff.library], Table 150 the following macros from 18.3.3 [c.limits], Table 32 are missing:

DECIMAL_DIG 
FLT_EVAL_METHOD

Furtheron it seems that in C.5 [diff.library], Table 149/150 further macros are missing as well, e.g. HUGE_VALF, INFINITY, etc.

[2014-02 Issaquah:]

This is an issue, all of C has not been updated for C99, C99 functions are missing, whole section needs to be overhauled.

The issue needs to be updated for functions and other missing items and when that happens the issue title is wrong and needs to be adapted.

Proposed resolution:


2202. Missing allocator support by async

Section: 30.6.8 [futures.async] Status: Deferred Submitter: Detlef Vollmann Opened: 2012-10-19 Last modified: 2016-02-12

Priority: 4

View all other issues in [futures.async].

Discussion:

promise, packaged_task, and async are the only places where a shared state is actually supposed to be allocated. Accordingly, promise and packaged_task are "allocator-aware". But function template async provides no way to provide an allocator.

[2013-09 Chicago]

Matt: deprecate async

Nico: read my paper

Alisdair: defer issues to wait for polymorphic allocators

Alisdair: defer, active topic of research Deferred

[2014-02-20 Re-open Deferred issues as Priority 4]

[2015-05 Lenexa, SG1 response]

We want whatever status approximates: "will not fix; we're working on a replacement facility and don't want to add features to a broken one"

Proposed resolution:


2206. Inaccuracy in initializer_list constructor requirements

Section: 23.2.3 [sequence.reqmts], 23.2.4 [associative.reqmts], 23.2.5 [unord.req], 26.6.1.2 [rand.req.seedseq] Status: Open Submitter: Jeffrey Yasskin Opened: 2012-10-21 Last modified: 2016-02-12

Priority: 3

View other active issues in [sequence.reqmts].

View all other issues in [sequence.reqmts].

View all issues with Open status.

Discussion:

In 23.2.3 [sequence.reqmts] p3, we have "il designates an object of type initializer_list<value_type>", and then several functions that take 'il' as an argument. However, an expression like {1, 2, 'a'} is not an object of type initializer_list<int> unless it's used to initialize an explicitly-typed variable of that type. I believe we want:

std::vector<int> v;
v = {1, 2, 'a'};

to compile portably, so we should say something different when defining 'il'. The same phrasing happens in 23.2.4 [associative.reqmts], 23.2.5 [unord.req], and 26.6.1.2 [rand.req.seedseq].

This may just be an editorial issue because the actual class synopses declare the functions to take initializer_list<exact_type>.

[2013-03-15 Issues Teleconference]

Moved to Open.

This is definitely not NAD

Should copy the suggested wording as the proposed resolution.

Proposed resolution:


2208. std::reverse_iterator should be a literal type

Section: 24.5.1 [reverse.iterators] Status: Tentatively Resolved Submitter: Jeffrey Yasskin Opened: 2012-10-30 Last modified: 2016-05-17

Priority: 3

View all other issues in [reverse.iterators].

View all issues with Tentatively Resolved status.

Discussion:

std::reverse_iterator::reverse_iterator(Iterator) should be constexpr so that other constexpr functions can return reverse_iterators. Of the other methods, the other constructors, base(), operator+, operator-, operator[], and the non-member operators can probably also be constexpr.

operator* cannot be constexpr because it involves an assignment to a member variable. Discussion starting with c++std-lib-33282 indicated that it would be useful to make reverse_iterator a literal type despite this restriction on its use at compile time.

Proposed resolution:

This issue was Resolved by paper P0031R0 adopted at Jacksonville, 2016.

2214. Clarify basic_ios::init call restrictions

Section: 27.5.5.2 [basic.ios.cons] Status: Open Submitter: Andrey Semashev Opened: 2012-11-09 Last modified: 2016-02-12

Priority: 4

View all other issues in [basic.ios.cons].

View all issues with Open status.

Discussion:

There is an ambiguity in how std::basic_ios::init method (27.5.5.2 [basic.ios.cons]) can be used in the derived class. The Standard only specify the state of the basic_ios object after the call completes. However, in basic_ios default constructor description (27.5.5.2 [basic.ios.cons]) there is this sentence:

Effects: Constructs an object of class basic_ios (27.5.3.7 [ios.base.cons]) leaving its member objects uninitialized. The object shall be initialized by calling basic_ios::init before its first use or before it is destroyed, whichever comes first; otherwise the behavior is undefined.

This restriction hints that basic_ios::init should be called exactly once before the object can be used or destroyed, because basic_ios::init may not know whether it was called before or not (i.e. whether its members are actually uninitialized or are initialized by the previous call to basic_ios::init). There is no such restriction in the basic_ios::init preconditions so it is not clear whether it is allowed to call basic_ios::init multiple times or not.

This problem has already affected publicly available implementations. For example, Microsoft Visual C++ STL introduces a memory leak if basic_ios::init is called multiple times, while GCC 4.7 and STLPort reinitialize the basic_ios object correctly without memory leak or any other undesired effects. There was a discussion of this issue on Boost developers mailing list, and there is a test case that reproduces the problem. The test case is actually a bug report for my Boost.Log library, which attempts to cache basic_ostream-derived objects internally to avoid expensive construction and destruction. My stream objects allowed resetting the stream buffer pointers the stream is attached to, without requiring to destroy and construct the stream.

My personal view of the problem and proposed resolution follows.

While apparently the intent of basic_ios::init is to provide a way to initialize basic_ios after default construction, I see no reason to forbid it from being called multiple times to reinitialize the stream. Furthermore, it is possible to implement a conforming basic_ios that does not have this restriction.

The quoted above section of the Standard that describes the effects of the default constructor is misleading. The Standard does not mandate any data members of basic_ios or ios_base (27.5.3 [ios.base]), which it derives from. This means that the implementation is allowed to use non-POD data members with default constructors that initialize the members with particular default values. For example, in the case of Microsoft Visual C++ STL the leaked memory is an std::locale instance that is dynamically allocated during basic_ios::init, a raw pointer to which is stored within ios_base. It is possible to store e.g. an unique_ptr instead of a raw pointer as a member of ios_base, the smart pointer will default initialize the underlying raw pointer on default construction and automatically destroy the allocated object upon being reset or destroyed, which would eliminate the leak and allow basic_ios::init to be called multiple times. This leads to conclusion that the default constructor of basic_ios cannot leave "its member objects uninitialized" but instead performs default initialization of the member objects, which would mean the same thing in case of POD types.

However, I feel that restricting ios_base and basic_ios members to non-POD types is not acceptable. Since multiple calls to basic_ios::init are not forbidden by the Standard, I propose to correct the basic_ios default constructor description so that it is allowed to destroy basic_ios object without calling basic_ios::init. This would imply that any raw members of basic_ios and ios_base should be initialized to values suitable for destruction (essentially, this means only initializing raw pointers to NULL). The new wording could look like this:

Effects: Constructs an object of class basic_ios (27.5.3.7 [ios.base.cons]) initializing its member objects to unspecified state, only suitable for basic_ios destruction. The object shall be initialized by calling basic_ios::init before its first use; otherwise the behavior is undefined.

This would remove the hint that basic_ios::init must be called exactly once. Also, this would remove the requirement for basic_ios::init to be called at all before the destruction. This is also an important issue because the derived stream constructor may throw an exception before it manages to call basic_ios::init (for example, if the streambuf constructor throws), and in this case the basic_ios destructor has undefined behavior.

To my mind, the described modification is sufficient to resolve the issue. But to emphasize the possibility to call basic_ios::init multiple times, a remark or a footnote for basic_ios::init postconditions could be added to explicitly state the semantics of calling it multiple times. The note could read as follows:

The function can be called multiple times during the object lifetime. Each subsequent call reinitializes the object to the described in postconditions initial state.

[2013-04-20, Bristol]

Alisdair: The current wording is unclear but the proposed resolution is wrong

Solution: Clarify that init must be called once and only once. Move then to review.

Proposed resolution:

This wording is relative to N3485.

  1. Edit 27.5.5.2 [basic.ios.cons] as indicated:

    basic_ios();
    

    -2- Effects: Constructs an object of class basic_ios (27.5.3.7 [ios.base.cons]) leaving its member objects uninitializedinitializing its member objects to unspecified state, only suitable for basic_ios destruction. The object shall be initialized by calling basic_ios::init before its first use or before it is destroyed, whichever comes first; otherwise the behavior is undefined.

    void init(basic_streambuf<charT,traits>* sb);
    

    Postconditions: The postconditions of this function are indicated in Table 128.

    -?- Remarks: The function can be called multiple times during the object lifetime. Each subsequent call reinitializes the object to the described in postconditions initial state.


2215. (unordered) associative container functors should be CopyConstructible

Section: 23.2.4 [associative.reqmts], 23.2.5 [unord.req] Status: Open Submitter: Alisdair Meredith Opened: 2012-11-14 Last modified: 2015-10-23

Priority: 3

View other active issues in [associative.reqmts].

View all other issues in [associative.reqmts].

View all issues with Open status.

Discussion:

The requirements on the functors used to arrange elements in the various associative and unordered containers are given by a set of expressions in tables 102 — Associative container requirements, and 103 — Unordered associative container requirements. In keeping with Library convention these expressions make the minimal requirements necessary on their types. For example, we have the following 3 row extracts for the unordered containers:

Expression Assertion/note pre-/post-condition
X(n, hf, eq)
X a(n, hf, eq)
Requires: hasher and key_equal are CopyConstructible.
X(n, hf)
X a(n, hf)
Requires: hasher is CopyConstructible and key_equal is DefaultConstructible.
X(n)
X a(n)
Requires: hasher and key_equal are DefaultConstructible.

However, the signature for each class template requires that the functors must effectively be CopyConstructible for each of these expressions:

template <class Key,
          class T,
          class Hash  = hash<Key>,
          class Pred  = std::equal_to<Key>,
          class Allocator = std::allocator<std::pair<const Key, T> > >
class unordered_map
{
  ...

  // construct/destroy/copy
  explicit unordered_map(size_type n = see below,
                         const hasher& hf = hasher(),
                         const key_equal& eql = key_equal(),
                         const allocator_type& a = allocator_type());

  ...
}

The letter of the standard can be honored as long as implementors recognize their freedom to split this one signature into multiple overloads, so that the documented default arguments (requiring a CopyConstructible functor) are not actually passed as default arguments.

As we look into the requirements for the copy constructor and copy-assignment operator, the requirements are even more vague, as the explicit requirements on the functors are not called out, other than saying that the functors are copied.

Must the functors be CopyAssignable? Or is CopyConstructible sufficient in this case? Do we require that the functors be Swappable so that the copy-swap idiom can be deployed here? Note that a type that is both CopyConstructible and CopyAssignable is still not guaranteed to be Swappable as the user may delete the swap function for their type in their own namespace, which would be found via ADL.

Some clean-up of the requirements table looks necessary, to at least document the assignment behavior. In addition, we should have clear guidance on whether these functors should always be CopyConstructible, as suggested by the class template definitions, or if the requirement tables are correct and we should explicitly split up the constructors in the (unordered) associative containers to no longer use default (function) arguments to obtain their defaulted functors.

I recommend the simplest solution would be to always require that the functors for (unordered) associative containers be CopyConstructible, above the requirements tables themselves, so that the issue need not be addressed within the tables. I suggest that the assignment operators for these containers add the requirement that the functors be Swappable, rather than forwarding the corresponding Assignable requirement.

[2013-03-15 Issues Teleconference]

Moved to Open.

Alisdair to propose wording.

[2014-06-08, Daniel comments]

The area of this issue partially overlaps what LWG 2227 addresses.

[2015-10-20, Daniel comments]

The revised resolution of LWG 2227 should resolve this issue as well. It follows the recommendations of the submitter to require CopyConstructible requirements for the function objects owned by containers, but it does not impose any further fundamental requirements.

Proposed resolution:

See the resolution of LWG 2227.


2216. regex_replace(basic_string) allocator handling

Section: 28.11.4 [re.alg.replace] Status: New Submitter: Jeffrey Yasskin Opened: 2012-11-26 Last modified: 2016-02-12

Priority: 3

View all other issues in [re.alg.replace].

View all issues with New status.

Discussion:

template <class traits, class charT, class ST, class SA>
  basic_string<charT, ST, SA>
  regex_replace(const basic_string<charT, ST, SA>& s,
      const basic_regex<charT, traits>& e,
      const charT* fmt,
      regex_constants::match_flag_type flags = 
	    regex_constants::match_default);

and friends are documented as

Constructs an empty string result of type basic_string<charT, ST, SA> and calls regex_replace(back_inserter(result), s.begin(), s.end(), e, fmt, flags).

This appears to require the result to have a default-constructed allocator, which isn't even possible for all allocator types. I suspect the allocator should be copied from 's' instead. Possibly there should be an additional defaulted argument to override the allocator of the result.

Proposed resolution:


2220. Under-specification of operator== for regex_token_iterator

Section: 28.12.2.2 [re.tokiter.comp] Status: New Submitter: Pete Becker Opened: 2012-11-21 Last modified: 2016-02-12

Priority: 2

View all issues with New status.

Discussion:

Consider the following example:

std::string str0("x");
std::regex rg0("a");
std::regex_token_iterator it0(str0.begin(), str0.end(), rg0, -1); // points at "x" in str0
std::string str1("x");
std::regex rg1("b");
std::regex_token_iterator it1(str1.begin(), str1.end(), rg1, -1); // points at "x" in str1

28.12.2.2 [re.tokiter.comp] p1 says that it0.operator==(it1) returns true "if *this and right are both suffix iterators and suffix == right.suffix"; both conditions are satisfied in this example. It does not say that they must both be iterators into the same sequence, nor does it say (as general iterator requirements do) that they must both be in the domain of == in order for the comparison to be meaningful. It's a simple statement: they're equal if the strings they point at compare equal. Given this being a valid comparison, the obtained result of "true" looks odd.

The problem is that for iterator values prior to the suffix iterator, equality means the same regular expression and the same matched sequence (both uses of "same" refer to identity, not equality); for the suffix iterator, equality means that the matched sequences compare equal.

Proposed resolution:


2221. No formatted output operator for nullptr

Section: 27.7.3 [output.streams] Status: New Submitter: Matt Austern Opened: 2012-12-07 Last modified: 2016-02-12

Priority: 3

View all issues with New status.

Discussion:

When I write

std::cout << nullptr << std::endl;

I get a compilation error, "ambiguous overload for 'operator<<' in 'std::cout << nullptr'". As far as I can tell, the compiler is right to issue that error. There are inserters for const void*, const char*, const signed char*, and const unsigned char*, and none for nullptr_t, so the expression really is ambiguous.

Proposed wording:

The obvious library solution is to add a nullptr_t overload, which would be defined something like

template<class C, class T>
basic_ostream<C, T>& operator<<(basic_ostream<C, T>& os, nullptr_t) 
{ 
  return os << (void*) nullptr; 
}

We might also consider addressing this at a core level: add a special-case language rule that addresses all cases where you write f(nullptr) and f is overloaded on multiple pointer types. (Perhaps a tiebreaker saying that void* is preferred in such cases.)

[2016-01-18, comments from Mike and Ville collected by Walter Brown]

Mike Miller: "Changing overload resolution sounds like something that should be considered by EWG before CWG […]"

Ville: "Agreed, such a change would be Evolutionary. Personally, I think it would also be wrong, because I don't see how void* is the right choice to prefer in the case of code that is currently ambiguous. Sure, it would solve this particular library issue, but it seemingly has wider repercussions. If LWG really wants to, EWG can certainly discuss this issue, but I would recommend solving it on the LWG side (which doesn't mean that the standard necessarily needs to change, I wouldn't call it far-fetched to NAD it)."

Proposed resolution:


2223. shrink_to_fit effect on iterator validity

Section: 23.3.11.3 [vector.capacity] Status: Open Submitter: Juan Soulie Opened: 2012-12-17 Last modified: 2016-02-12

Priority: 2

View other active issues in [vector.capacity].

View all other issues in [vector.capacity].

View all issues with Open status.

Discussion:

After the additions by 2033, it appears clear that the intended effect includes a reallocation and thus the potential effect on iterators should be explicitly added to the text in order to not contradict 23.2.1 [container.requirements.general]/11, or at the very least, explicitly state that a reallocation may happen.

Taking consistency with "reserve" into consideration, I propose:

BTW, while we are at it, I believe the effect on iterators should also be explicitly stated in the other instance a reallocation may happen: 23.3.11.5 [vector.modifiers]/1 — even if obvious, it only contradicts 23.2.1 [container.requirements.general]/11 implicitly.

I propose to also insert "Reallocation invalidates all the references, pointers, and iterators referring to the elements in the sequence." at the appropriate location in its "Remarks".

[2012-12-19: Jonathan Wakely comments]

The described problem also affects std::basic_string and std::deque.

[2013-03-15 Issues Teleconference]

Moved to Review.

[2013-04-18, Bristol]

Daniel extends the P/R.

Rationale:

The wording in 21.3.1.4 [string.capacity] combined with 21.3.1.1 [string.require] seems to say the necessary things. We cannot impose all requirements as we do for vector, because we want to allow the short-string-optimization.

[2014-02-15 post-Issaquah session]

STL: I think that shrink_to_fit should be a no-op when called twice.

STL: Do we ever define reallocation for deque? Nope, all mentions of "reallocation" are in vector. We define what it means in vector::reserve(), but not for deque.

STL: Oh duh, they define reallocate in the PR. But I think we can do better here.

STL: Optimally, deque shrinking just allocates a new map of pointers, and drops empty blocks, but preserves pointers/references to elements.

Alisdair: That's like unordered containers, invalidating only iterators.

Pablo: It doesn't make sense to reduce capacity() to size(), because deque doesn't have capacity!

STL: For vector, "effectively reduces the capacity" is unnecessary, the capacity there is observable.

STL: There is a strong reason to provide an optimal shrink to fit for deque, since only the library implementer can do this.

STL: The other thing I don't like the repeated definition of reallocation for vector, we define it once and use it in a bunch of places. At most we can lift it up to the vector synopsis.

STL: I'll write new wording.

[2014-10-01, STL adds discussion and provides new wording]

Compared to the previous proposed resolution:

My wording doesn't directly say that shrink_to_fit() should be a no-op when called twice in a row. (Indirectly, if the first call reduces capacity() to size(), the second call must preserve iterators/etc.) I considered rewording the complexity to say "linear if reallocation happens", but that's potentially problematic (what if we copy almost all N elements, then one throws and we have to unwind? There are no effects, so reallocation didn't happen, yet we took longer than constant time). Implementers can always do better than the stated complexity bounds.

I chose not to modify deque's requirements, so implementations remain free to reallocate the elements themselves.

I didn't attempt to centralize vector's reallocation wording. That can be done editorially, if someone is sufficiently motivated.

Previous resolution from Juan Soulie/Daniel [SUPERSEDED]:

This wording is relative to N3485.

  1. Keep 21.3.1.4 [string.capacity] around p14 unchanged, because we don't speak about reallocations and we give the strong exception guarantee in 21.3.1.1 [string.require] (Invalidation specification also at that place):

    void shrink_to_fit();
    

    -14- Remarks: shrink_to_fit is a non-binding request to reduce capacity() to size(). [Note: The request is non-binding to allow latitude for implementation-specific optimizations. — end note ].

  2. Edit 23.3.8.3 [deque.capacity] around p7 as indicated:

    void shrink_to_fit();
    

    -5- Requires: T shall be MoveInsertable into *this.

    -?- Effects: shrink_to_fit is a non-binding request to reduce capacity() to size(). [Note: The request is non-binding to allow latitude for implementation-specific optimizations. — end note ] Reallocation happens at this point if and only if the function effectively reduces the capacity. If an exception is thrown other than by the move constructor of a non-CopyInsertable T there are no effects.

    -6- Complexity: Linear in the size of the sequence.

    -7- Remarks: shrink_to_fit is a non-binding request to reduce capacity() to size(). [Note: The request is non-binding to allow latitude for implementation-specific optimizations. — end note ] If an exception is thrown other than by the move constructor of a non-CopyInsertable T there are no effects.Reallocation invalidates all the references, pointers, and iterators referring to the elements in the sequence.

  3. Edit 23.3.11.3 [vector.capacity] around p7 as indicated:

    void shrink_to_fit();
    

    -7- Requires: T shall be MoveInsertable into *this.

    -?- Effects: shrink_to_fit is a non-binding request to reduce capacity() to size(). [Note: The request is non-binding to allow latitude for implementation-specific optimizations. — end note ] Reallocation happens at this point if and only if the function effectively reduces the capacity. If an exception is thrown other than by the move constructor of a non-CopyInsertable T there are no effects.

    -8- Complexity: Linear in the size of the sequence.

    -9- Remarks: shrink_to_fit is a non-binding request to reduce capacity() to size(). [Note: The request is non-binding to allow latitude for implementation-specific optimizations. — end note ] If an exception is thrown other than by the move constructor of a non-CopyInsertable T there are no effects.Reallocation invalidates all the references, pointers, and iterators referring to the elements in the sequence.

  4. Edit 23.3.11.5 [vector.modifiers] p1 as indicated:

    iterator insert(const_iterator position, const T& x);
    iterator insert(const_iterator position, T&& x);
    iterator insert(const_iterator position, size_type n, const T& x);
    template <class InputIterator>
    iterator insert(const_iterator position, InputIterator first, InputIterator last);
    iterator insert(const_iterator position, initializer_list<T>);
    template <class... Args> void emplace_back(Args&&... args);
    template <class... Args> iterator emplace(const_iterator position, Args&&... args);
    void push_back(const T& x);
    void push_back(T&& x);
    

    -1- Remarks: Causes reallocation if the new size is greater than the old capacity. Reallocation invalidates all the references, pointers, and iterators referring to the elements in the sequence. If no reallocation happens, all the iterators and references before the insertion point remain valid. If an exception is thrown other than by the copy constructor, move constructor, assignment operator, or move assignment operator of T or by any InputIterator operation there are no effects. If an exception is thrown by the move constructor of a non-CopyInsertable T, the effects are unspecified.

[2015-02, Cologne]

GR: I'm concerned that shrink_to_fit may cause reallocation without changing the capacity. […] It's about correctness. The statement about invalidation is useless if I cannot detect whether reallocation has happened?

AM: It seems like the logic goes the other way round: It's the capacity change that causes reallocation, so if there's no capacity change, there's no reallocation. But that's not quite how I'd like to say it... maybe this, : "If capacity does not change, no reallocation occurs."

GR: Where does it actually say that reserve() invalidates? AM: It should say that in the container requirements. VV: vector specifies in reserve that there's reallocation if and only if the capacity changes. GR: I can't find anything in the container requirements about reserve. DK: No, it's specified for every container separately. GR: It isn't specified for string.

GR: I'm noticing that the issue touches on shrink_to_fit for a bunch of containers. Anyway, I think the reserve issue [re string] is in scope for this issue. This change is touching on a lot of members.

AM: Landing this change will provide clarity for what we should do with basic_string. GR: We're already asking for changes; we should fix string as well. AM: If one of the changes is ready before the other, I'd like to land the finished part first, but if both are ready for Lenexa, I'm equally happy to fix them in one go.

DK will reword this.

Conclusion: Update wording, revisit in Lenexa.

Proposed resolution:

This wording is relative to N3936.

  1. Change 21.3.1.4 [string.capacity] p14 as depicted:

    void shrink_to_fit();
    

    -14- RemarksEffects: shrink_to_fit is a non-binding request to reduce capacity() to size(). [Note: The request is non-binding to allow latitude for implementation-specific optimizations. — end note] It does not increase capacity(), but may reduce capacity() by causing reallocation.

    -?- Complexity: Linear in the size of the sequence.

    -?- Remarks: Reallocation invalidates all the references, pointers, and iterators referring to the elements in the sequence. If no reallocation happens, they remain valid.

  2. Change 23.3.8.3 [deque.capacity] p5-p7 as depicted:

    void shrink_to_fit();
    

    -5- Requires: T shall be MoveInsertable into *this.

    -?- Effects: shrink_to_fit is a non-binding request to reduce memory use but does not change the size of the sequence. [Note: The request is non-binding to allow latitude for implementation-specific optimizations. — end note] If an exception is thrown other than by the move constructor of a non-CopyInsertable T there are no effects.

    -6- Complexity: Linear in the size of the sequence.

    -7- Remarks: shrink_to_fit is a non-binding request to reduce memory use but does not change the size of the sequence. [Note: The request is non-binding to allow latitude for implementation-specific optimizations. — end note]shrink_to_fit invalidates all the references, pointers, and iterators referring to the elements in the sequence.

  3. Change 23.3.11.3 [vector.capacity] p7-p9 as depicted:

    void shrink_to_fit();
    

    -7- Requires: T shall be MoveInsertable into *this.

    -?- Effects: shrink_to_fit is a non-binding request to reduce capacity() to size(). [Note: The request is non-binding to allow latitude for implementation-specific optimizations. — end note] It does not increase capacity(), but may reduce capacity() by causing reallocation. If an exception is thrown other than by the move constructor of a non-CopyInsertable T there are no effects.

    -8- Complexity: Linear in the size of the sequence.

    -9- Remarks: shrink_to_fit is a non-binding request to reduce capacity() to size(). [Note: The request is non-binding to allow latitude for implementation-specific optimizations. — end note] If an exception is thrown other than by the move constructor of a non-CopyInsertable T there are no effects.Reallocation invalidates all the references, pointers, and iterators referring to the elements in the sequence. If no reallocation happens, they remain valid.

  4. Change 23.3.11.5 [vector.modifiers] p1 as depicted:

    -1- Remarks: Causes reallocation if the new size is greater than the old capacity. Reallocation invalidates all the references, pointers, and iterators referring to the elements in the sequence. If no reallocation happens, all the iterators and references before the insertion point remain valid. […]


2226. wstring_convert methods do not take allocator instance

Section: 22.3.3.2.2 [conversions.string] Status: LEWG Submitter: Glen Fernandes Opened: 2012-12-14 Last modified: 2016-02-12

Priority: Not Prioritized

View other active issues in [conversions.string].

View all other issues in [conversions.string].

View all issues with LEWG status.

Discussion:

The wstring_convert class template, described in 22.3.3.2.2 [conversions.string], does not support custom stateful allocators. It only supports custom stateless allocators.

The to_bytes member function returns basic_string<char, char_traits<char>, Byte_alloc> but it does not take an instance of Byte_alloc to pass to the constructor of the basic_string.

Similarly the from_bytes member function returns basic_string<Elem, char_traits<Elem>, Wide_alloc> but it does not take an instance of Wide_alloc to pass to the constructor of the basic_string.

This makes these two member functions and the wstring_convert class template not usable when Wide_alloc or Byte_alloc are stateful allocators.

[2013-01-22, Glen provides wording]

[2013-03-15 Issues Teleconference]

Moved to NAD Future.

This is clearly an extension that the LEWG may want to take a look at, once we have more experience with appropriate use of allocators with the C++11 model.

Proposed resolution:

This wording is relative to N3485.

  1. In 22.3.3.2.2 [conversions.string]/2 and /6 "Class template wstring_convert synopsis" change the overloads of the member function from_bytes() so that all four overloads take an additional parameter which is an instance of Wide_alloc:

    wide_string from_bytes(char byte, const Wide_alloc& alloc = Wide_alloc());
    wide_string from_bytes(const char *ptr, const Wide_alloc& alloc = Wide_alloc());
    wide_string from_bytes(const byte_string& str, const Wide_alloc& alloc = Wide_alloc());
    wide_string from_bytes(const char *first, const char *last, const Wide_alloc& alloc = Wide_alloc());
    
  2. In 22.3.3.2.2 [conversions.string] /8 specify that this Wide_alloc allocator parameter is used to construct the wide_string object returned from the function:

    -7- Effects: The first member function shall convert the single-element sequence byte to a wide string. The second member function shall convert the null-terminated sequence beginning at ptr to a wide string. The third member function shall convert the sequence stored in str to a wide string. The fourth member function shall convert the sequence defined by the range [first, last) to a wide string.

    -8- In all cases:

  3. In 22.3.3.2.2 [conversions.string]/2 and /12 "Class template wstring_convert synopsis" change the overloads of the member function to_bytes() so that all four overloads take an additional parameter which is an instance of Byte_alloc:

    byte_string to_bytes(Elem wchar, const Byte_alloc& alloc = Byte_alloc());
    byte_string to_bytes(const Elem *wptr, const Byte_alloc& alloc = Byte_alloc());
    byte_string to_bytes(const wide_string& wstr, const Byte_alloc& alloc = Byte_alloc());
    byte_string to_bytes(const Elem *first, const Elem *last, const Byte_alloc& alloc = Byte_alloc());
    
  4. In 22.3.3.2.2 [conversions.string] /13 specify that this Byte_alloc allocator parameter is used to construct the byte_string object returned from the function:

    -12- Effects: The first member function shall convert the single-element sequence wchar to a byte string. The second member function shall convert the null-terminated sequence beginning at wptr to a byte string. The third member function shall convert the sequence stored in wstr to a byte string. The fourth member function shall convert the sequence defined by the range [first, last) to a byte string.

    -13- In all cases:


2227. Stateful comparison objects in associative containers

Section: 23.2.4 [associative.reqmts] Status: Open Submitter: Juan Soulie Opened: 2012-12-19 Last modified: 2015-10-23

Priority: 3

View other active issues in [associative.reqmts].

View all other issues in [associative.reqmts].

View all issues with Open status.

Discussion:

Table 102 in 23.2.4 [associative.reqmts]/8 states on expression a.key_comp() that it "returns the comparison object out of which a was constructed". At the same time, 23.2.1 [container.requirements.general]/8 states (starting in the third line) that "...Any Compare, Pred, or Hash objects belonging to a and b shall be swappable and shall be exchanged by unqualified calls to non-member swap...". This is problematic for any compliant implementation, since once swapped the container cannot return the comparison object out of which it was constructed unless incurring in storing an otherwise needless object.

The simple solution is to correct that statement in Table 102, but I believe this is part of a larger problem of underspecified behavior: The new standard has made an effort in regards to allocators and now fully specifies what happens to stateful allocator objects. It has even specified what happens to stateful hasher and key_equal members of unordered containers (they propagate), but it says nothing about stateful comparison objects of (ordered) associative containers, except for the statement in 23.2.1 [container.requirements.general]/8 referred above and only related to swap.

For example, it is unclear to me what is specified to happen on an assignment: should the comparison object be copied/moved along with the elements, or should the left-hand side object keep its own? Maybe this has been intentionally left unspecified with the purpose of compatibility with C++98, which I understand it specified that comparison objects were kept for the entire life of the container (like allocators) — an unfortunate choice. But anyway, the segment of 23.2.1 [container.requirements.general] quoted above seems to break any possible backwards compatibility with C++98 in this regard.

Therefore, taking into consideration consistency with how this is dealed with for unordered associative containers, I propose that Table 102 is modified as follows:

[2013-03-15 Issues Teleconference]

Moved to Review.

[2013-04-18, Bristol]

STL: can't believe we don't specify this already. this is totally necessary

Alisdair: how does it do this? copy construction? assignment?

Also need it for move.

STL: we already specify this for constructing from a comparator, not during copy construction though.

Jonathan: don't like wording, should say "key_compare is CopyConstructible. Uses b.key_comp() as a comparison object."

STL: we get it right for unordered!

Jonathan: can't wordsmith this now, but I think implementations do the right thing.

Alisdair: not sure what right thing is for moves. Also we say nothing about propagating allocators to functors.

Moved to Open.

[2015-02 Cologne]

TK: There's no need for fine-grained propagate/not-propagate control. If you don't want to propagate the predicate, you can simply construct or insert from an iterator range.

VV: libstdc++ already implements the resolution of this issue.

GR: There are a couple of other problems. We don't specify move constructor and move assignment for maps. Those are just general.

TK: General container requirements already describe the semantics for {copy,move}-{construction,assignment}, so it doesn't seem that there's room for choice in std::map assignments. unordered_map is different, though.

[Note: Check what general container requirements say about container equality.]

DK will draft wording. The decision is to unambiguously make all {copy,move}-{construction,assignment} operations endow the LHS with the exact state of the RHS, including all predicates and hash function states.

Conclusion: Update wording, revisit later.

[2015-05-06 Lenexa: Waiting for updated wording]

Previous resolution [SUPERSEDED]:

This wording is relative to N3485.

  1. Change Table 102 as indicated:

    Table 102 — Associative container requirements (in addition to container)
    Expression Return type Assertion/note pre-/post-condition Complexity
    X(il) Same as X(il.begin(), il.end()). same as X(il.begin(), il.end()).
    X(b)
    X a(b)
    Copy constructor. In addition to
    the requirements of Table 96, copies
    the comparison object.
    Linear in b.size()
    a = b X& Copy assignment operator. In addition to
    the requirements of Table 96, copies the
    comparison object.
    Linear in a.size() and b.size()
    a.key_comp() X::key_compare rReturns thea's comparison object
    out of which a was constructed.
    constant

[2015-10-19 Daniel comments and provides alternative wording]

The current standard is especially unclear in regard to what effects move operations of unordered/associative containers should have. We have one example that is standardized exactly in this way by looking at 23.6.5.2 [priqueue.cons.alloc] p7:

template <class Alloc> priority_queue(priority_queue&& q, const Alloc& a);

-7- Effects: Initializes c with std::move(q.c) as the first argument and a as the second argument, and initializes comp with std::move(q.comp)

A similarly comparable example are the move-operations of std::unique_ptr in regard to the deleter (when this is no a reference), which also respect move-capabilities of that function object.

We have wording from C++98 for associative containers (but not for unordered containers!) that was never adjusted to C++11 move-semantics in 23.2.4 [associative.reqmts] p12:

When an associative container is constructed by passing a comparison object the container shall not store a pointer or reference to the passed object, even if that object is passed by reference. When an associative container is copied, either through a copy constructor or an assignment operator, the target container shall then use the comparison object from the container being copied, as if that comparison object had been passed to the target container in its constructor.

The second sentence of this wording is problematic for several reasons:

  1. It only talks about copy operations, not about move operations, except that the term "assignment" without leading "copy" is a bit ambigious (albeit it seems clear in the complete context).

  2. It is not really clear how to interpret "as if that comparison object had been passed to the target container in its constructor" for an assignment operation. A possible but not conclusive interpretation could be that this is wording supporting a "copy-via-swap" idiom.

  3. There does not exist similar wording for unordered containers, except that Table 102 provides entries for copy construction and copy assignment of the containers whose wording just talks of "copies" in either case.

Existing implementations differ already:

  1. Visual Studio 2015 uses copy construction and copy assignment for the two copy operations but uses swap operations for the move operations.

  2. GCC's libstdc++ performs copy construction and copy assignment for the two copy operations and for the two move operations, respectively

  3. clang++'s libc++ performs copy/move construction and copy/move assignment for the corresponding four copy/move operations

The alternative wording provided below attempts to clarify that container copy/move operations perform the corresponding copy/move operations on the owned function objects.

In addition the wording also resolves LWG 2215: I believe that the current wording should require that container function objects should meet the CopyConstructible requirements. Adding this general requirement also fixes the underspecified requirements of the accessor functions key_comp() and value_comp().

I don't think that a general requirement for Swappable is needed, only the member swap function currently requires this. Nonetheless the wording below does support stateful functors that are also moveable or move-assignable, therefore the specified semantics in terms of move operations.

I should add the following warning, though: If this proposed wording would be accepted, there is a little chance of code breakage, because the current wording can be read that in general there is no requirement that the container functors are CopyConstructible. The following code example is accepted by gcc + libstd++:

#include <map>
#include <utility>
#include <iostream>

struct Cmp {
  Cmp() = default;
  Cmp(const Cmp&) = delete;
  Cmp(Cmp&&) = delete;
  Cmp& operator=(const Cmp&) = delete;
  Cmp& operator=(Cmp&&) = delete;
  template<class T>
  bool operator()(const T& x, const T& y) const
  {
    return x < y;
  }
};

typedef std::map<int, int, Cmp> MyMap;

int main() {
  MyMap m;
  std::cout << (m.find(12) == m.end()) << std::endl;
}

Proposed resolution:

This wording is relative to N4527.

  1. Change 23.2.4 [associative.reqmts] p8 as indicated:

    -8- In Table 101, X denotes an associative container class, a denotes a value of type X, b denotes a possibly const value of type X, rv denotes a non-const rvalue of type X, u denotes the name of a variable being declared, […]

  2. Change Table 101 as indicated:

    Table 101 — Associative container requirements (in addition to container)
    Expression Return type Assertion/note pre-/post-condition Complexity
    X::key_compare Compare Requires: Compare is CopyConstructible.
    defaults to less<key_type>
    compile time
    X(c)
    X u(c);
    Requires: key_compare is CopyConstructible.
    Effects: Constructs an empty container.
    Uses a copy of c as a comparison object.
    […]
    X(i,j,c)
    X u(i,j,c);
    Requires: key_compare is CopyConstructible.
    value_type is EmplaceConstructible into X from *i.
    Effects: Constructs an empty container and inserts elements
    from the range [i, j) into it; uses c as a comparison object.
    […]
    X(il) Same as X(il.begin(), il.end()). same as X(il.begin(), il.end()).
    X(b)
    X a(b)
    (In addition to the requirements of Table 95)
    Effects: Copy constructs the comparison object of a from
    the comparison object of b.
    Linear in b.size()
    X(rv)
    X a(rv)
    (In addition to the requirements of Table 95 and Table 98)
    Effects: Move constructs the comparison object of a from
    the comparison object of rv.
    constant
    a = b X& (In addition to the requirements of Table 95 and Table 98)
    Requires: key_compare is CopyAssignable.
    Effects: Copy assigns the comparison object of b
    to the comparison object of a.
    Linear in a.size() and b.size()
    a = rv X& (In addition to the requirements of Table 95 and Table 98)
    Requires: key_compare is MoveAssignable.
    Effects: Move assigns from the comparison object of rv
    to the comparison object of a.
    Linear
    a.key_comp() X::key_compare rReturns thea's comparison object
    out of which a was constructed.
    constant
  3. Change 23.2.4 [associative.reqmts] p12 as indicated:

    -12- When an associative container is constructed by passing a comparison object the container shall not store a pointer or reference to the passed object, even if that object is passed by reference. When an associative container is copied, either through a copy constructor or an assignment operator, the target container shall then use the comparison object from the container being copied, as if that comparison object had been passed to the target container in its constructor.

  4. Change 23.2.5 [unord.req] p11 as indicated:

    -11- In Table 102: X denotes an unordered associative container class, a denotes a value of type X, b denotes a possibly const value of type X, rv denotes a non-const rvalue of type X, […]

  5. Change Table 102 as indicated:

    Table 102 — Unordered associative container requirements (in addition to container)
    Expression Return type Assertion/note pre-/post-condition Complexity
    X::hasher Hash Requires: Hash is CopyConstructible.
    Hash shall be a unary function object type
    such that the expression hf(k) has type std::size_t.
    compile time
    X::key_equal Pred Requires: Pred is CopyConstructible.
    Pred shall be a binary predicate that takes
    two arguments of type Key.
    Pred is an equivalence relation.
    compile time
    X(n, hf, eq)
    X a(n, hf, eq)
    X Requires: hasher and key_equal are CopyConstructible.
    Effects: […]
    […]
    X(n, hf)
    X a(n, hf)
    X Requires: hasher is CopyConstructible and
    key_equal is DefaultConstructible.
    Effects: […]
    […]
    X(i, j, n, hf, eq)
    X a(i, j, n, hf, eq)
    X Requires: hasher and key_equal are CopyConstructible.
    value_type is EmplaceConstructible into X from *i.
    Effects: […]
    […]
    X(i, j, n, hf)
    X a(i, j, n, hf)
    X Requires: hasher is CopyConstructible and
    key_equal is DefaultConstructible.
    value_type is EmplaceConstructible into X from *i.
    Effects: […]
    […]
    X(b)
    X a(b)
    X Copy constructor. In addition
    to the requirements of Table 95,
    copies the hash function,
    predicate, and maximum load
    factor.
    (In addition to the requirements of Table 95)
    Effects: Copy constructs the hash function, predicate, and maximum load factor
    of a from the corresponding objects of b.
    Average case linear in
    b.size(),
    worst case quadratic.
    X(rv)
    X a(rv)
    X (In addition to the requirements of Table 95 and Table 98)
    Effects: Move constructs the hash function, predicate, and maximum load factor
    of a from the corresponding objects of rv.
    constant
    a = b X& Copy assignment operator. In
    addition to the requirements of
    Table 95, copies the hash
    function, predicate, and
    maximum load factor.
    (In addition to the requirements of Table 95 and Table 98)
    Requires: hasher and key_equal are CopyAssignable.
    Effects: Copy assigns the hash function, predicate, and maximum load factor
    of b to the corresponding objects of a.
    Average case linear in
    b.size(),
    worst case quadratic.
    a = rv X& (In addition to the requirements of Table 95 and Table 98)
    Requires: hasher and key_equal are MoveAssignable.
    Effects: Move assigns the hash function, predicate, and maximum load factor
    from rv to the corresponding objects of a.
    Linear

2232. [CD] The char_traits specializations should declare their length(), compare(), and find() members constexpr

Section: 21.2.3 [char.traits.specializations] Status: LEWG Submitter: Jeffrey Yasskin Opened: 2012-12-24 Last modified: 2016-02-12

Priority: Not Prioritized

View all other issues in [char.traits.specializations].

View all issues with LEWG status.

Discussion:

Addresses ES 14, US 19

These functions have easy recursive constexpr implementations that, unfortunately, aren't efficient at runtime. EWG is still figuring out how to solve this problem in general (e.g., N3444 isn't sufficient to avoid stack overflows in debug builds or to get the optimal assembly-based implementations at runtime), so users can't portably solve this problem for themselves, but implementations can use compiler-specific techniques to choose the right implementation inside their standard libraries.

The LWG is still undecided about whether individual implementations can add constexpr to these functions, so we need to add constexpr to the standard here for implementations to be able to improve this.

[2013-03-15 Issues Teleconference]

Moved to Open.

There are a number of people who have a strong interest in this issue not available for the telecon.

It also plays at the heart of a discussion about library freedoms for constexpr and specifying a library that may depend on unspecified compiler intrinsics to be implementable.

[2013-09 Chicago]

Moved to NAD Future.

While it is clear that this feature can be implemented using only C++14 constexpr features, there is real concern that we cannot call the efficient, highly optimized, C implementations of these functions under a C++14 constexpr implementation, nor implement similar ourselves as this typically involves use of inline asm instructions.

Clang and libc++ have some experience of using intrinsics to try to address the performance issue, but the current intrinsics are not general enough to support char_traits. The intrinsics support only operations on character string literals, and the string literal is no longer visible as a literal after passing as a const char * to the char_traits functions.

Additional concern was raised that these operations are unlikely to be useful anyway, as the only client is basic_string which relies on dynamic memory allocation, and so cannot effectively be made a literal type. Jeffrey then pointed out the pending string_view library that will also use char_traits and would most certainly benefit from being a literal type.

Given the choice of giving up performance on a critical library component, or requiring a compiler intrinsic with only unsuccessful implementation experience, the consensus is to not reject this, unless compelling implementation experience is demonstrated. NAD Future seems the appropriate resolution.

Proposed resolution:

This wording is relative to N3691.

  1. In 21.2.3.1 [char.traits.specializations.char], 21.2.3.2 [char.traits.specializations.char16_t], 21.2.3.3 [char.traits.specializations.char32_t], and 21.2.3.4 [char.traits.specializations.wchar.t]:

    static constexpr int compare(const char_type* s1, const char_type* s2, size_t n);
    static constexpr size_t length(const char_type* s);
    static constexpr const char_type* find(const char_type* s, size_t n, const char_type& a);
    

2236. kill_dependency unconditionally noexcept

Section: 29.2 [atomics.syn], 29.3 [atomics.order] Status: SG1 Submitter: Daniel Krügler Opened: 2013-01-19 Last modified: 2016-02-12

Priority: Not Prioritized

View all other issues in [atomics.syn].

View all issues with SG1 status.

Discussion:

The "magic" kill_dependency function is a function without any constraints on the template parameter T and is specified as

template <class T>
T kill_dependency(T y) noexcept;

-14- Effects: The argument does not carry a dependency to the return value (1.10).

-15- Returns: y.

I wonder whether the unconditional noexcept is really intended here: Assume we have some type U that has a potentially throwing move constructor (or it has a potentially throwing copy constructor and no move constructor), for any "normal" function template with the same signature and the same effects (modulo the dependency magic) this would mean that it cannot safely be declared noexcept because of the return statement being part of the complete function call affected by noexcept (The by-value function argument is irrelevant in this context). In other words it seems that a function call such as

struct S {
  ...
  S(const S& r) { if(some condition) throw Something(); }
  ...
};

int main() {
  S s1 = ...;
  S s2 = std::kill_dependency(s1);
}

would be required to call std::terminate if the copy constructor of S throws during the return of std::kill_dependency.

To require copy elision for this already magic function would look like a low-hanging fruit to solve this problem, but this case is not covered by current copy elision rules see 12.8 p31 b1:

"— in a return statement in a function with a class return type, when the expression is the name of a non-volatile automatic object (other than a function or catch-clause parameter) with the same cv-unqualified type as the function return type, the copy/move operation can be omitted by constructing the automatic object directly into the function's return value".

Some options come into my mind:

  1. Make the exception-specification a constrained one in regard via std::is_nothrow_move_constructible:

    template <class T>
    T kill_dependency(T y) noexcept(see below);
    

    This is similar to the approach taken for function templates such as std::swap.

  2. Use perfect forwarding (This needs further wording to correct the effects):

    template <class T>
    T&& kill_dependency(T&& y) noexcept;
    
  3. Impose constraints on the template arguments in regard to throwing exceptions while copying/moving.

  4. Keep the state as it is but possibly add a note about a call of std::terminate in above scenario.

A second problem is that the current wording is not clear whether it is well-defined to call the function with types that are reference types, such as in the following example:

#include <atomic>

int main()
{
  int a = 12;
  int& b = std::kill_dependency<int&>(a);
}

It is unclear what kind of dependency is killed here. This is presumably a core language problem, but could affect the possible resolutions of the problem.

[2014-11 Urbana]

Recommend using a revised example:

int lookup(class D* p) 
{
  class E* q = p->a.load(memory_order_consume);
  int y = std::kill_dependency(q->y);
}

[2015-02 Cologne]

Handed over to SG1.

Proposed resolution:


2237. <cuchar> macros

Section: 21.5 [c.strings] Status: New Submitter: Jason Merrill Opened: 2013-01-29 Last modified: 2016-02-12

Priority: 4

View other active issues in [c.strings].

View all other issues in [c.strings].

View all issues with New status.

Discussion:

Apparently C1X changes __STDC_UTF_16__ and __STDC_UTF_32__ from macros defined in uchar.h (and reflected in C++ by Table 79) to be predefined by the compiler. Do we want to do the same?

Proposed resolution:


2238. Problematic iterator-pair constructor of containers

Section: 21.5 [c.strings] Status: Open Submitter: Johannes Schaub Opened: 2013-02-02 Last modified: 2016-02-12

Priority: 3

View other active issues in [c.strings].

View all other issues in [c.strings].

View all issues with Open status.

Discussion:

The non-explicit nature of the iterator-pair constructor of containers, such a

template <class InputIterator>
vector(InputIterator first, InputIterator last, const Allocator& = Allocator());

can be selected in unexpected situations, leading to a hard runtime error, as demonstrated by the following example:

#include <vector>

void f(std::vector<char> v){ /* ... */}

int main() {
  f({"A", "B"});
}

The actually intended initializer-list constructor isn't feasible here, so the best match is the constructor template

template <class InputIterator>
vector(InputIterator first, InputIterator last, const Allocator& = Allocator());

This compiles, but will result in code running amok. The potential trap (that cannot be easily detected by the library implementation) could be reduced by making this constructor explicit. It would still have the effect to be selected here, but the code would be ill-formed, so the programmer gets a clear meassage here.

[2014-06 Rapperswil]

JW: can't fix this, don't want to touch this, Do The Right Thing clause has been a source of tricky issues. only really happens with string literals, that's the only way to create an array that isn't obviously an array

GR: want to see paper

AM: is it only string literals, or also UDLs?

STL: maybe, but we don't need to deal with that. This is only a problem in a very specific case

Leave as Open.

Proposed resolution:


2241. <cstdalign> and #define of alignof

Section: 18.10 [support.runtime] Status: Tentatively Resolved Submitter: Richard Smith Opened: 2013-02-14 Last modified: 2016-06-28

Priority: 2

View other active issues in [support.runtime].

View all other issues in [support.runtime].

View all issues with Tentatively Resolved status.

Discussion:

According to 18.10 [support.runtime] p2:

The contents of these headers are the same as the Standard C library headers [..], <stdalign.h>, [..]

Since our base C standard is C99, which doesn't have a <stdalign.h>, the reference to a non-existing C header is irritating (In this context <stdalign.h> doesn't refer to the deprecated C++ header <stdalign.h> described in D.4 [depr.c.headers]).

Furthermore, it would be also important that it doesn not define a macro named alignof, which C11 also defines in this header.

Currently we only have the following guarantee as part of 18.10 [support.runtime] p7:

The header <cstdalign> and the header <stdalign.h> shall not define a macro named alignas.

It is unclear what the better strategy is: Striking the reference to <stdalign.h> in 18.10 [support.runtime] p2 or upgrading to C11 as new base C standard.

[2014-02-15 Issaquah]

STL: related to earlier issue on C4, 2201, and now we get a C11 header
JY: find _Alignof as keyword C11 FDIS has four defines in stdalign.h
AM: need paper for C11 as base library we should really do that
STL: really need vendor input
STL: don't think we need to do anything right now not P1
AM: any objections to downscale to P2 (no objections)

[2016-03 Jacksonville]

Walter: this is on track to go away if we adopt Clark's paper to rebase to C11
Room: tentatively resolved; revisit after C11 paper: P0063

[2016-03 Oulu]

P0063 was adopted.

Change status to Tentatively Resolved

Proposed resolution:


2242. [uninitialized_]copy_n() defect

Section: 25.4.1 [alg.copy], 20.10.10.4 [uninitialized.copy] Status: New Submitter: Sean Parent Opened: 2013-02-14 Last modified: 2016-02-12

Priority: 2

View other active issues in [alg.copy].

View all other issues in [alg.copy].

View all issues with New status.

Discussion:

copy_n() and uninitialized_copy_n() only return the output iterator, and not the input iterator. Likely the interface was simply copied from the original STL. Unfortunately the interface in the original STL contains a bug.

copy_n() and uninitialized_copy_n() must return the resulting input iterator as well as the output iterator (I would suggest returning a pair). Without this, there is no way to continue reading from an actual input iterator — and if it is really a forward iterator, it will cost n increments to get back to where you were.

Proposed resolution:


2243. istream::putback problem

Section: 27.7.2.3 [istream.unformatted] Status: New Submitter: Juan Soulie Opened: 2013-03-01 Last modified: 2016-02-12

Priority: 3

View all other issues in [istream.unformatted].

View all issues with New status.

Discussion:

In 27.7.2.3 [istream.unformatted] / 34, when describing putback, it says that "rdbuf->sputbackc()" is called. The problem are not the obvious typos in the expression, but the fact that it may lead to different interpretations, since nowhere is specified what the required argument to sputbackc is.

It can be guessed to be "rdbuf()->sputbackc(c)", but "rdbuf()->sputbackc(char_type())" or just anything would be as conforming (or non-confoming) as the first guess.

Proposed resolution:


2245. packaged_task::reset() memory allocation

Section: 30.6.9.1 [futures.task.members] Status: Open Submitter: Jonathan Wakely Opened: 2013-03-05 Last modified: 2016-02-12

Priority: 3

View all other issues in [futures.task.members].

View all issues with Open status.

Discussion:

The effects of packaged_task::reset() result in memory allocation, but don't allow a user to provide an allocator.

packaged_task::reset() needs to be overloaded like so:

template<class Alloc>  
void reset(const Alloc&);

Alternatively, the effects of reset() need to require the same allocator is used as at construction, which would require the constructor to store the allocator for later use.

I like to remark that GCC at the moment uses the second option, i.e. the allocator passed to the constructor (if any) is used to create the new shared state, because this didn't require any change to the interface.

[2015-02 Cologne]

Handed over to SG1.

[2015-05 Lenexa, SG1 response]

No strong opinions in SG1, and this is really an LWG issue. Back to you.

Proposed resolution:


2248. numeric_limits::is_iec559 misnamed

Section: 18.3.2 [limits] Status: New Submitter: Pete Becker Opened: 2013-03-08 Last modified: 2016-02-12

Priority: 4

View all other issues in [limits].

View all issues with New status.

Discussion:

This member should probably be named "is_ieee754". Or at least the standard should explain that IEC-559 no longer exists, and that it's been superseded by IEEE-754.

Proposed resolution:


2256. On vector iterator invalidation

Section: 23.3.11.5 [vector.modifiers] Status: New Submitter: Howard Hinnant Opened: 2013-04-29 Last modified: 2016-05-22

Priority: 3

View other active issues in [vector.modifiers].

View all other issues in [vector.modifiers].

View all issues with New status.

Discussion:

23.3.11.5 [vector.modifiers]/p3 says:

iterator erase(const_iterator position);
iterator erase(const_iterator first, const_iterator last);

Effects: Invalidates iterators and references at or after the point of the erase.

Consider this example:

#include <vector>
#include <cassert>

int main()
{
  typedef std::vector<int> C;
  C c = {1, 2, 3, 4};
  C::iterator i = c.begin() + 1;
  C::iterator j = c.end() - 1;
  assert(*i == 2);
  assert(*j == 4);
  c.erase(c.begin());
  assert(*i == 3); // Why is this not perfectly fine?!
}

Why has the iterator i been invalidated? It still refers to a perfectly reasonable, fully constructed object. If vector::iterator were to be implemented as a pointer (which is legal), it is not possible for that last line to do anything but run fine.

The iterator j on the other hand now points at end, and any iterators that may now point beyond end(), into uninitialized memory, are clearly invalid.

But why do we say that an iterator that must point to a valid object is invalid? This looks to me like we simply got sloppy in our specification.

[2016-05 Issues Telecom]

This is related to 2698

Proposed resolution:


2260. Missing requirement for Allocator::pointer

Section: 17.6.3.5 [allocator.requirements] Status: Open Submitter: Jonathan Wakely Opened: 2013-05-14 Last modified: 2016-02-12

Priority: 3

View other active issues in [allocator.requirements].

View all other issues in [allocator.requirements].

View all issues with Open status.

Discussion:

For an allocator A<T> which defines A<T>::pointer to a class type, i.e. not T*, I see no requirement that A<T>::pointer is convertible to A<U>::pointer, even if T* is convertible to U*. Such conversions are needed in containers to convert from e.g. ListNodeBase* to ListNode<T>*.

The obvious way to do such conversions appears to be pointer_traits::pointer_to(), but that's ill-formed if the static member function A<T>::pointer::pointer_to() doesn't exist and the allocator requirements don't mention that function, so you need to cast A<T>::pointer to A<T>::void_pointer then cast that to A<U>::pointer.

Is converting via void_pointer really intended, or are we missing a requirement that pointer_traits<A<T>::pointer>::pointer_to() be well-formed?

Proposed resolution:

Add to the Allocator requirements table the following requirement:

The expression pointer_traits<XX::pointer>::pointer_to(r) is well-defined.

[2013-09 Chicago]

Pablo to come back with proposed wording

[2015-07 Telecom]

Marshall to ping Pablo for proposed wording and disable current wording.

Previous resolution [SUPERSEDED]:
  1. Edit Table 28 as indicated:

    Table 28 — Allocator requirements (continued)
    Expression Return type Assertion/note pre-/post-condition Default
    static_cast<X::const_pointer>(z) X::const_pointer static_cast<X::const_pointer>(z) == q  
    pointer_traits<X::pointer>::pointer_to(r) X::pointer    

Proposed resolution:


2261. Are containers required to use their 'pointer' type internally?

Section: 23.2 [container.requirements] Status: New Submitter: Jonathan Wakely Opened: 2013-05-14 Last modified: 2016-02-12

Priority: 2

View other active issues in [container.requirements].

View all other issues in [container.requirements].

View all issues with New status.

Discussion:

Is a container C only supposed to refer to allocated memory (blocks of contiguous storage, nodes, etc.) through objects of type C::pointer rather than C::value_type*?

I don't see anything explicitly requiring this, so a container could immediately convert the result of get_allocator().allocate(1) to a built-in pointer of type value_type* and only deal with the built-in pointer until it needs to deallocate it again, but that removes most of the benefit of allowing allocators to use custom pointer types.

[2014-06-12, Jonathan comments]

This issue is basically the same issue as LWG 1521, which agrees it's an issue, to be dealt with in the future, so I request that 2261 not be closed as a dup unless we reopen 1521.

Proposed resolution:


2262. Requirement for unique_ptr<T>::get_deleter()(p) to be able to destroy the unique_ptr

Section: 20.11.1.2 [unique.ptr.single] Status: Open Submitter: Rob Desbois Opened: 2013-05-15 Last modified: 2016-02-12

Priority: 3

View all other issues in [unique.ptr.single].

View all issues with Open status.

Discussion:

N3337 20.11.1.2.5 [unique.ptr.single.modifiers] contains 2 non-normative notes stating:

[para 4]: "The order of these operations is significant because the call to get_deleter() may destroy *this."

[para 5]: "The postcondition does not hold if the call to get_deleter() destroys *this since this->get() is no longer a valid expression."

It seems this wording was created to resolve 998 due to the possibility that a unique_ptr may be destroyed through deletion of its stored pointer where that directly or indirectly refers to the same unique_ptr. If unique_ptr is required to support circular references then it seems this must be normative text: an implementation is currently allowed to operate on *this after the assignment and deletion specified in para 4, since this is only 'disallowed' by the non-normative note.

I propose the following draft rewording:

[para 4]: Effects: assigns p to the stored pointer, and then if the old value of the stored pointer, old_p, was not equal to nullptr, calls get_deleter()(old_p). No operation shall be performed after the call to get_deleter()(old_p) that requires *this to be valid, because the deletion may destroy *this if it is referred to directly or indirectly by the stored pointer. [Note: The order of these operations is significant because the call to get_deleter() may destroy *this. — end note]

[para 5]: Postconditions: If the call get_deleter()(old_p) destroyed *this, none. Otherwise, get() == p. [Note: The postcondition does not hold if the call to get_deleter() destroys *this since this->get() is no longer a valid expression. — end note]

I expect it will also be necessary to amend the requirements for a deleter, so in addition:

20.11.1.2 [unique.ptr.single] [para 1]: The default type for the template parameter D is default_delete. A client-supplied template argument D shall be a function object type (20.10), lvalue-reference to function, or lvalue-reference to function object type for which, given a value d of type D and a value ptr of type unique_ptr<T, D>::pointer, the expression d(ptr) is valid and has the effect of disposing of the pointer as appropriate for that deleter. Where D is not an lvalue reference type, d(ptr) shall be valid if ptr refers directly or indirectly to the invoking unique_ptr object.

[2013-10-05, Stephan T. Lavavej comments and provides alternative wording]

In Chicago, we determined that the original proposed change to 20.11.1.2 [unique.ptr.single]/1 was insufficient, because d might be a reference to a deleter functor that's destroyed during self-destruction.

We believed that 20.11.1.2.5 [unique.ptr.single.modifiers]/4 was already sufficiently clear. The Standard occasionally prevents implementations of X from doing various things, through the principle of "nothing allows X to fail in that situation". For example, v.push_back(v[0]) is required to work for non-empty vectors because nothing allows that to fail. In this case, the intent to allow self-destruction is already clear.

Additionally, we did not believe that 20.11.1.2.5 [unique.ptr.single.modifiers]/5 had to be changed. The current note is slightly squirrely but it does not lead to confusion for implementers or users.

Previous resolution from Rob Desbois:

  1. Edit 20.11.1.2 [unique.ptr.single] p1 as indicated:

    The default type for the template parameter D is default_delete. A client-supplied template argument D shall be a function object type (20.10), lvalue-reference to function, or lvalue-reference to function object type for which, given a value d of type D and a value ptr of type unique_ptr<T, D>::pointer, the expression d(ptr) is valid and has the effect of disposing of the pointer as appropriate for that deleter. Where D is not an lvalue reference type, d(ptr) shall be valid if ptr refers directly or indirectly to the invoking unique_ptr object.

  2. Edit 20.11.1.2.5 [unique.ptr.single.modifiers] p4+5 as indicated:

    void reset(pointer p = pointer()) noexcept;
    

    -3- Requires: The expression get_deleter()(get()) shall be well formed, shall have well-defined behavior, and shall not throw exceptions.

    -4- Effects: assigns p to the stored pointer, and then if the old value of the stored pointer, old_p, was not equal to nullptr, calls get_deleter()(old_p). No operation shall be performed after the call to get_deleter()(old_p) that requires *this to be valid, because the deletion may destroy *this if it is referred to directly or indirectly by the stored pointer. [Note: The order of these operations is significant because the call to get_deleter() may destroy *this. — end note]

    -5- Postconditions: If the call get_deleter()(old_p) destroyed *this, none. Otherwise, get() == p. [Note: The postcondition does not hold if the call to get_deleter() destroys *this since this->get() is no longer a valid expression. — end note]

Previous resolution [SUPERSEDED]:

This wording is relative to N3691.

  1. Edit 20.11.1.2 [unique.ptr.single] p1 as indicated:

    The default type for the template parameter D is default_delete. A client-supplied template argument D shall be a function object type (20.10), lvalue-reference to function, or lvalue-reference to function object type for which, given a value d of type D and a value ptr of type unique_ptr<T, D>::pointer, the expression d(ptr) is valid and has the effect of disposing of the pointer as appropriate for that deleter. d(ptr) shall be valid even if it triggers the destruction of d or (if D is an lvalue reference to function object type) the function object that d refers to.

[2015-05, Lenexa]

After some discussion in Lenexa there was some wavering on if the added sentence is necessary. Here is example code that demonstrates why the extra sentence is necessary. In this example the call to d(ptr) is valid, however the deleter references *this after destructing its element:

#include <cassert>
#include <memory>
#include <iostream>

class Deleter
{
    int state_ = 0;

    enum
    {
        destructed            = -4,
        self_move_assigned    = -3,
        move_assigned_from    = -2,
        move_constructed_from = -1
    };
public:
    ~Deleter() {state_ = destructed;}

    Deleter() = default;
    Deleter(Deleter const&) = default;
    Deleter& operator=(Deleter const&) = default;

    Deleter(Deleter&& a) noexcept
        : state_(a.state_)
    {a.state_ = move_constructed_from;}

    Deleter& operator=(Deleter&& a) noexcept
    {
        if (this == &a)
            state_ = self_move_assigned;
        else
        {
            state_ = a.state_;
            a.state_ = move_assigned_from;
        }
        return *this;
    }

    Deleter(int state)
        : state_(state)
    {
        assert(state >= 0);
    }

    template <class T>
    void
    operator()(T* t) const
    {
        std::cout << "Deleter beginning operator()(T*)\n";
        std::cout << "The deleter = " << *this << '\n';
        std::cout << "Deleter about to destruct the X.\n";
        delete t;
        std::cout << "Deleter has destructed the X.\n";
        std::cout << "The deleter = " << *this << '\n';
        std::cout << "Deleter ending operator()(T*)\n";
    }

    friend
    std::ostream&
    operator<<(std::ostream& os, const Deleter& a)
    {
        switch (a.state_)
        {
        case destructed:
            os << "**destructed**";
            break;
        case self_move_assigned:
            os << "self_move_assigned";
            break;
        case move_assigned_from:
            os << "move_assigned_from";
            break;
        case move_constructed_from:
            os << "move_constructed_from";
            break;
        default:
            os << a.state_;
            break;
        }
        return os;
    }
};

struct X
{
    Deleter deleter_{1};
};

int main()
{
    auto xp = new X;
    {
        std::unique_ptr<X, Deleter&> p(xp, xp->deleter_);
        std::cout << "unique_ptr is constructed.\n";
        std::cout << "The deleter = " << p.get_deleter() << '\n';
        std::cout << "Destructing unique_ptr...\n";
    }
    std::cout << "unique_ptr is destructed.\n";
}

Which outputs:

unique_ptr is constructed.
The deleter = 1
Destructing unique_ptr...
Deleter beginning operator()(T*)
The deleter = 1
Deleter about to destruct the X.
Deleter has destructed the X.
The deleter = **destructed**
Deleter ending operator()(T*)
unique_ptr is destructed.

The line "The deleter = **destructed**" represents the deleter referencing itself after it has been destructed by the d(ptr) expression, but prior to that call returning.

Suggested alternative to the current proposed wording:

The expression d(ptr) shall not refer to the object d after it executes ptr->~T().

[2015-07, Telecom]

Geoffrey: Deleter may or may not execute ~T().
Alisdair: After the destructor after the element has run. Say it in words instead of code.
Howard will provide updated wording. Perhaps need both normative and non-normative wording.

[2015-08-03, Howard updates P/R per telecon discussion.]

Proposed resolution:

This wording is relative to N4431.

  1. Edit 20.11.1.2 [unique.ptr.single] p1 as indicated:

    The default type for the template parameter D is default_delete. A client-supplied template argument D shall be a function object type (20.9), lvalue-reference to function, or lvalue-reference to function object type for which, given a value d of type D and a value ptr of type unique_ptr<T, D>::pointer, the expression d(ptr) is valid and has the effect of disposing of the pointer as appropriate for that deleter. The expression d(ptr), if it destructs the object referred to by ptr, shall not refer to the object d after it destructs *ptr. [Note: The object being destructed may control the lifetime of d. — end note]


2265. 29.3p9 appears to rule out some acceptable executions

Section: 29.3 [atomics.order] Status: Open Submitter: Brian Demsky Opened: 2013-06-17 Last modified: 2016-02-12

Priority: 4

View other active issues in [atomics.order].

View all other issues in [atomics.order].

View all issues with Open status.

Discussion:

I believe that the following variation on IRIW should admit executions in which c1 = d1 = 5 and c2 = d2 = 0. If this is allowed, then what is sequence of program evaluations for 29.3 [atomics.order] p9 that justifies the store to z? It seems that 29.3 [atomics.order] p9 should not allow this execution because one of the stores to x or y has to appear earlier in the sequence, each of the fetch_adds reads the previous load in the thread (and thus must appear later in the sequence), and 29.3 [atomics.order] p9 states that each load must read from the last prior assignment in the sequence.

atomic_int x;
atomic_int y;
atomic_int z;
int c1, c2, d1, d2;

static void a(void* obj)
{
  atomic_store_explicit(&x, 5, memory_order_relaxed); 
}

static void b(void* obj)
{
  atomic_store_explicit(&y, 5, memory_order_relaxed); 
}

static void c(void* obj)
{
  c1 = atomic_load_explicit(&x, memory_order_relaxed);
  // this could also be an atomic load if the address depends on c1:
  c2 = atomic_fetch_add_explicit(&y, c1, memory_order_relaxed);  
}

static void d(void* obj)
{
  d1 = atomic_load_explicit(&y, memory_order_relaxed);
  d2 = atomic_fetch_add_explicit(&x, d1, memory_order_relaxed); 
}

int user_main(int argc, char** argv)
{
  thrd_t t1, t2, t3, t4;

  atomic_init(&x, 0);
  atomic_init(&y, 0);

  printf("Main thread: creating 4 threads\n");
  thrd_create(&t1, (thrd_start_t)&a, NULL);
  thrd_create(&t2, (thrd_start_t)&b, NULL);
  thrd_create(&t3, (thrd_start_t)&c, NULL);
  thrd_create(&t4, (thrd_start_t)&d, NULL);

  thrd_join(t1);
  thrd_join(t2);
  thrd_join(t3);
  thrd_join(t4);
  printf("c1=%d c2=%d\n",c1,c2);
  printf("d1=%d d2=%d\n",d1,d2);

  // Can this store write 1000 (i.e., c1=d1=5, c2=d2=0)?
  atomic_store(&z, (c1+d1)*100+c2+d2);

  printf("Main thread is finished\n");

  return 0;
}

It seems that the easiest fix is to allow a load in 29.3 [atomics.order] p9 to read from any prior store in the evaluation order.

That said, I would personally advocate the following: It seems to me that C/C++ atomics are in a bit of different situation than Java because:

  1. People are expected to use relaxed C++ atomics in potentially racy situations, so it isn't clear that semantics as complicated as the JMM's causality would be sane.

  2. People who use C/C++ atomics are likely to be experts and use them in a very controlled fashion. I would be really surprised if compilers would find any real wins by optimizing the use of atomics.

Why not do something like:

There is satisfaction DAG of all program evaluations. Each evaluation observes the values of variables as computed by some prior assignment in the DAG.

There is an edge x->y between two evaluations x and y if:

  1. the evaluation y observes a value computed by the evaluation x or

  2. the evaluation y is an atomic store, the evaluation x is an atomic load, and there is a condition branch c that may depend (intrathread dependence) on x and x-sb->c and c-sb->y.

This seems to allow reordering of relaxed atomics that processors do without extra fence instructions, allows most reorderings by the compiler, and gets rid of satisfaction cycles.

[2015-02 Cologne]

Handed over to SG1.

[2015-05 Lenexa, SG1 response]

This was partially addressed (weasel-worded) in C++14 (See N3786). The remainder is an open research problem. N3710 outlines a "solution" that doesn't have a consensus behind it because it costs performance. We have no better solution at the moment.

Proposed resolution:


2267. partial_sort_copy underspecified for ranges of two different types

Section: 25.5.1.4 [partial.sort.copy] Status: New Submitter: Matt Austern Opened: 2013-06-26 Last modified: 2016-02-12

Priority: 3

View all issues with New status.

Discussion:

The signature of this function is:

template<class InputIterator, class RandomAccessIterator>
RandomAccessIterator
partial_sort_copy(InputIterator first, InputIterator last,
                  RandomAccessIterator result_first,
                  RandomAccessIterator result_last);

(and the usual overload for an explicitly provided comparison function). The standard says nothing about requirements in the case where the input type (iterator_traits<InputIterator>::value_type) and the output type (iterator_traits<RandomAccessIterator>::value_type) are different.

Presumably the input type must be convertible to the output type. What's less clear is what the requirements are on the comparison operator. Does the algorithm only perform comparisons on two values of the output type, or does it also perform comparisons on values of the input type, or might it even perform heterogeneous comparisons?

Proposed resolution:


2269. Container iterators and argument-dependent lookup

Section: 23.2.1 [container.requirements.general] Status: New Submitter: Matt Austern Opened: 2013-06-26 Last modified: 2016-02-12

Priority: 4

View other active issues in [container.requirements.general].

View all other issues in [container.requirements.general].

View all issues with New status.

Discussion:

Consider the following code snippet:

#include <vector>
#include <algorithm>

int main() {
  std::vector<int> v1(100, 3);
  std::vector<int> v2(100);
  copy(v1.begin(), v1.end(), v2.begin());
}

It compiles without error on my desktop. Is it required to? I can't find evidence from the standard that it is. In my test std::copy was found by argument-dependent lookup because the implementation I used made std::vector<int>::iterator a user-defined type defined in namespace std. But the standard only requires std::vector<int>::iterator to be an implementation specified random access iterator type. I can't find anything requiring it to be a user-defined type at all (and in fact there are reasonable implementation where it isn't), let alone a user defined type defined in a specific namespace.

Since the defining namespace of container iterators is visible to users, should the standard say anything about what that namespace is?

Proposed resolution:


2286. stringbuf::underflow() underspecified

Section: 27.8.2.4 [stringbuf.virtuals] Status: New Submitter: Sergey Zubkov Opened: 2013-08-29 Last modified: 2016-02-12

Priority: 4

View all other issues in [stringbuf.virtuals].

View all issues with New status.

Discussion:

In 27.8.2.4 [stringbuf.virtuals]/1, basic_stringbuf::underflow() is specified to unconditionally return traits::eof() when a read position is not available.

The semantics of basic_stringbuf require, and existing libraries implement it so that this function makes a read position available if possible to do so, e.g. if some characters were inserted into the stream since the last call to overflow(), resulting in pptr() > egptr(). Compare to the conceptually similar D.5.1.3 [depr.strstreambuf.virtuals]/15.

Proposed resolution:

This wording is relative to N3691.

  1. Change 27.8.2.4 [stringbuf.virtuals] as indicated:

    int_type underflow();
    

    -1- Returns: If the input sequence has a read position available or the function makes a read position available (as described below), returns traits::to_int_type(*gptr()). Otherwise, returns traits::eof(). Any character in the underlying buffer which has been initialized is considered to be part of the input sequence.

    -?- The function can make a read position available only if (mode & ios_base::in) != 0 and if the write next pointer pptr() is not null and is greater than the current read end pointer egptr(). To make a read position available, the function alters the read end pointer egptr() to equal pptr().


2289. constexpr guarantees of defaulted functions still insufficient

Section: 20.4.2 [pairs.pair], 20.5.2.1 [tuple.cnstr], 20.17.5 [time.duration] Status: Open Submitter: Daniel Krügler Opened: 2013-09-09 Last modified: 2016-02-12

Priority: 3

View other active issues in [pairs.pair].

View all other issues in [pairs.pair].

View all issues with Open status.

Discussion:

During the acceptance of N3471 and some similar constexpr papers, specific wording was added to pair, tuple, and other templates that were intended to impose implementation constraints that ensure that the observable constexpr "character" of a defaulted function template is solely determined by the required expressions of the user-provided types when instantiated, for example:

The defaulted move and copy constructor, respectively, of pair shall be a constexpr function if and only if all required element-wise initializations for copy and move, respectively, would satisfy the requirements for a constexpr function.

This wording doesn't require enough, especially since the core language via CWG 1358 does now support constexpr function template instantiations, even if such function cannot appear in a constant expression (as specified in 5.20 [expr.const]) or as a constant initializer of that object (as specified in [basic.start.init]). The wording should be improved and should require valid uses in constant expressions and as constant initializers instead.

[Lenexa 2015-05-05]

STL : notice order of move/copy and copy/move with "respectively".

General word-smithing; ask for updated wording

Are we happy with this with changes we are suggesting?

unanimous

Proposed resolution:

This wording is relative to N3691.

  1. Change 20.4.2 [pairs.pair] p2 as indicated:

    -2- The defaulted move and copy constructor, respectively, of pair shall be a constexpr function if and only if all required element-wise initializations for copy and move, respectively, would satisfy the requirements for a constexpr functionAn invocation of the move or copy constructor of pair shall be a constant expression (5.20 [expr.const]) if all required element-wise initializations would be constant expressions. An invocation of the move or copy constructor of pair shall be a constant initializer for that pair object ( [basic.start.init]) if all required element-wise initializations would be constant initializers for the respective subobjects.

  2. Change 20.5.2.1 [tuple.cnstr] p2 as indicated:

    -2- The defaulted move and copy constructor, respectively, of tuple shall be a constexpr function if and only if all required element-wise initializations for copy and move, respectively, would satisfy the requirements for a constexpr function. The defaulted move and copy constructor of tuple<> shall be constexpr functionsAn invocation of the move or copy constructor of tuple shall be a constant expression (5.20 [expr.const]) if all required element-wise initializations would be constant expressions. An invocation of the move or copy constructor of tuple shall be a constant initializer for that tuple object ( [basic.start.init]) if all required element-wise initializations would be constant initializers for the respective subobjects. An invocation of the move or copy constructor of tuple<> shall be a constant expression, or a constant initializer for that tuple<> object, respectively, if the function argument would be constant expression.

  3. Change 20.17.5 [time.duration] p7 as indicated:

    -7- Remarks: The defaulted copy constructor of duration shall be a constexpr function if and only if the required initialization of the member rep_ for copy and move, respectively, would satisfy the requirements for a constexpr function.An invocation of the copy constructor of duration shall be a constant expression (5.20 [expr.const]) if the required initialization of the member rep_ would be a constant expression. An invocation of the copy constructor of duration shall be a constant initializer for that duration object ( [basic.start.init]) if the required initialization of the member rep_ would be constant initializers for this subobject.


2290. Top-level "SFINAE"-based constraints should get a separate definition in Clause 17

Section: 20.15 [meta] Status: Open Submitter: Daniel Krügler Opened: 2013-09-02 Last modified: 2016-02-12

Priority: 3

View other active issues in [meta].

View all other issues in [meta].

View all issues with Open status.

Discussion:

The current library specification uses at several places wording that is intended to refer to core language template deduction failure at the top-level of expressions (aka "SFINAE"), for example:

The expression declval<T>() = declval<U>() is well-formed when treated as an unevaluated operand (Clause 5). Access checking is performed as if in a context unrelated to T and U. Only the validity of the immediate context of the assignment expression is considered. [Note: The compilation of the expression can result in side effects such as the instantiation of class template specializations and function template specializations, the generation of implicitly-defined functions, and so on. Such side effects are not in the "immediate context" and can result in the program being ill-formed. — end note]

Similar wording can be found in the specification of result_of, is_constructible, and is_convertible, being added to resolve an NB comment by LWG 1390 and 1391 through N3142.

This wording is necessary to limit speculative compilations needed to implement these traits, but it is also lengthy and repetitive.

[2014-05-19, Daniel suggests a descriptive term]

constrictedly well-formed expression:

An expression e depending on a set of types A1, ..., An which is well-formed when treated as an unevaluated operand (Clause 5). Access checking is performed as if in a context unrelated to A1, ..., An. Only the validity of the immediate context of e is considered. [Note: The compilation of the expression can result in side effects such as the instantiation of class template specializations and function template specializations, the generation of implicitly-defined functions, and so on. Such side effects are not in the "immediate context" and can result in the program being ill-formed. — end note]

[2014-05-20, Richard and Jonathan suggest better terms]

Richard suggested "locally well-formed"

Jonathan suggested "contextually well-formed" and then "The expression ... is valid in a contrived argument deduction context"

[2014-06-07, Daniel comments and improves wording]

The 2014-05-19 suggestion did only apply to expressions, but there are two important examples that are not expressions, but instead are involving an object definition (std::is_constructible) and a function definition (std::is_convertible), respectively, instead. Therefore I suggest to rephrase the usage of "expression" into "program construct" in the definition of Jonathan's suggestion of "valid in a contrived argument deduction context".

I would like to point out that given the new definition of "valid in a contrived argument deduction context", there are several other places of the Library specification that could take advantage of this wording to improve the existing specification, such as 20.14.12.2 [func.wrap.func] p2, most functions in 20.10.8.2 [allocator.traits.members], and the **Insertable, EmplaceConstructible, and Erasable definitions in 23.2.1 [container.requirements.general], but given that these are not fully described in terms of the aforementioned wording yet, I would recommend to fix them by a separate issue once the committee has agreed on following the suggestion presented by this issue.

[2015-05-05 Lenexa: Move to Open]

...

MC: I think we like the direction but it isn't quite right: it needs some work

JW: I'm prepared to volunteer to move that further, hopefully with the help of Daniel

Roger Orr: should this be Core wording because it doesn't really have anything to do with libraries - the term could then just be used here

AM: Core has nothing to deal with that, though

HT: it seems there is nothing to imply that allows dropping out with an error - maybe that's a separate issue

MC: I'm not getting what you are getting at: could you write an issue? - any objection to move to Open?

...

Proposed resolution:

This wording is relative to N3936.

  1. Add the following new definition to 17.3 [definitions] as indicated:

    valid in a contrived argument deduction context [defns.valid.contr.context]

    A program construct c depending on a set of types A1, ..., An, and treated as an unevaluated operand (Clause 5) when c is an expression, which is well-formed. Access checking is performed as if in a context unrelated to A1, ..., An. Only the validity of the immediate context (14.8.2 [temp.deduct]) of c is considered. [Note: The compilation of c can result in side effects such as the instantiation of class template specializations and function template specializations, the generation of implicitly-defined functions, and so on. Such side effects are not in the "immediate context" and can result in the program being ill-formed. — end note].

  2. Change Table 49 ("Type property predicates") as indicated:

    Table 49 — Type property predicates
    Template Condition Preconditions
    template <class T, class U>
    struct is_assignable;
    The expression declval<T>() =
    declval<U>()
    is valid in a
    contrived argument deduction context
    ([defns.valid.contr.context]) for types
    T and U.
    well-formed when treated
    as an unevaluated operand
    (Clause 5). Access
    checking is performed as if
    in a context unrelated to T
    and U. Only the validity of
    the immediate context of
    the assignment expression
    is considered. [Note: The
    compilation of the
    expression can result in
    side effects such as the
    instantiation of class
    template specializations
    and function template
    specializations, the
    generation of
    implicitly-defined
    functions, and so on. Such
    side effects are not in the
    "immediate context" and
    can result in the program
    being ill-formed. — end
    note]
    […]
  3. Change 20.15.4.3 [meta.unary.prop] p7 as indicated:

    -7- Given the following function prototype:

    template <class T>
      add_rvalue_reference_t<T> create() noexcept;
    

    the predicate condition for a template specialization is_constructible<T, Args...> shall be satisfied if and only if the following variable definition would be well-formed for some invented variable t would be valid in a contrived argument deduction context ([defns.valid.contr.context]) for types T and Args...:

    T t(create<Args>()...);
    

    [Note: These tokens are never interpreted as a function declaration. — end note] Access checking is performed as if in a context unrelated to T and any of the Args. Only the validity of the immediate context of the variable initialization is considered. [Note: The evaluation of the initialization can result in side effects such as the instantiation of class template specializations and function template specializations, the generation of implicitly-defined functions, and so on. Such side effects are not in the "immediate context" and can result in the program being ill-formed. — end note]

  4. Change Table 57 ("Other transformations") as indicated:

    Table 57 — Other transformations
    Template Condition Comments
    template <class Fn, class... ArgTypes>
    struct result_of<Fn(ArgTypes...)>;
    […] If the expression
    INVOKE(declval<Fn>(),
    declval<ArgTypes>()...)
    is
    valid in a contrived argument deduction
    context ([defns.valid.contr.context]) for types
    Fn and ArgTypes...
    well
    formed when treated as an
    unevaluated operand (Clause 5)
    , the
    member typedef type shall name the
    type
    decltype(INVOKE(declval<Fn>(),
    declval<ArgTypes>()...))
    ;
    otherwise, there shall be no member
    type. Access checking is performed as
    if in a context unrelated to Fn and
    ArgTypes. Only the validity of the
    immediate context of the expression is
    considered. [Note: The compilation of
    the expression can result in side
    effects such as the instantiation of
    class template specializations and
    function template specializations, the
    generation of implicitly-defined
    functions, and so on. Such side effects
    are not in the "immediate context"
    and can result in the program being
    ill-formed. — end note]
  5. Change 20.15.6 [meta.rel] p4 as indicated:

    -4- Given the following function prototype:

    template <class T>
      add_rvalue_reference_t<T> create() noexcept;
    

    the predicate condition for a template specialization is_convertible<From, To> shall be satisfied if and only if the return expression in the following code would be well-formedvalid in a contrived argument deduction context ([defns.valid.contr.context]) for types To and From, including any implicit conversions to the return type of the function:

    To test() {
      return create<From>();
    }
    

    [Note: This requirement gives well defined results for reference types, void types, array types, and function types. — end note] Access checking is performed as if in a context unrelated to To and From. Only the validity of the immediate context of the expression of the return-statement (including conversions to the return type) is considered. [Note: The evaluation of the conversion can result in side effects such as the instantiation of class template specializations and function template specializations, the generation of implicitly-defined functions, and so on. Such side effects are not in the "immediate context" and can result in the program being ill-formed. — end note]


2292. Find a better phrasing for "shall not participate in overload resolution"

Section: 17.5.1.4 [structure.specifications] Status: New Submitter: Jeffrey Yasskin Opened: 2013-09-03 Last modified: 2016-02-12

Priority: 3

View other active issues in [structure.specifications].

View all other issues in [structure.specifications].

View all issues with New status.

Discussion:

The C++14 CD has 25 sections including the phrase "X shall not participate in overload resolution ...". Most of these uses are double negatives, which are hard to interpret. "shall not ... unless" tends to be the easiest to read, since the condition is true when the function is available, but we also have a lot of "if X is not Y, then Z shall not participate", which actually means "You can call Z if X is Y." The current wording is also clumsy and long-winded. We should find a better and more concise phrasing.

As an initial proposal, I'd suggest using "X is enabled if and only if Y" in prose and adding an "Enabled If: ..." element to 17.5.1.4 [structure.specifications].

Daniel:

I suggest to name this new specification element for 17.5.1.4 [structure.specifications] as "Template Constraints:" instead, because the mentioned wording form was intentionally provided starting with LWG 1237 to give implementations more freedom to realize the concrete constraints. Instead of the original std::enable_if-based specifications we can use better forms of "SFINAE" constraints today and it eases the path to possible language-based constraints in the future.

Proposed resolution:


2294. <cstdlib> should declare abs(double)

Section: 26.9 [c.math] Status: Open Submitter: Pete Becker Opened: 2013-09-04 Last modified: 2016-02-12

Priority: 2

View other active issues in [c.math].

View all other issues in [c.math].

View all issues with Open status.

Discussion:

… and abs(float) and abs(long double). And <cmath> should declare abs(int), abs(long), and abs(long long).

As things currently stand, this program is illegal:

#include <cstdlib>

int main() {
  double d = -1.23;
  double dd = std::abs(d);
  return 0;
}

The call is ambiguous because of the various integer overloads, that's because <cstdlib> provides abs(int) but not abs(double).

This lead one commenter on Stackoverflow to state that abs is dangerous, and to recommend using fabs instead.

In general, it makes sense to declare overloaded functions that take user-defined types in the same header as the definition of the user-defined types; it isn't necessary to declare all of the overloads in the same place. But here we're not dealing with any user-defined types; we're dealing with builtin types, which are always defined; all of the overloads should be defined in the same place, to avoid mysterious problems like the one in the code above.

The standard library has six overloads for abs:

int abs(int);  // <cstdlib>
long abs(long); // <cstdlib>
long long abs(long long); // <cstdlib>

float abs(float); // <cmath>
double abs(double); // <cmath>
long double abs(long double); // <cmath>

These should all be declared in both headers.

I have no opinion on <stdlib.h> and <math.h>.

[2013-09 Chicago]

This issue is related to LWG 2192

Move to open

[2014-02-13 Issaquah — Nicolai Josuttis suggest wording]

[2015-03-03, Geoffrey Romer provides improved wording]

See proposed resolution of LWG 2192.

[2015-09-11, Telecon]

Geoff provided combined wording for 2192 after Cologne, Howard to provide updated wording for Kona.

Howard: my notes say I wanted to use is_unsigned instead of 'unsigned integral type'.

Previous resolution from Nicolai [SUPERSEDED]:
  1. Edit 26.9 [c.math] after p7 as indicated:

    -6- In addition to the int versions of certain math functions in <cstdlib>, C++ adds long and long long overloaded versions of these functions, with the same semantics.

    -7- The added signatures are:

    long abs(long);                    // labs()
    long long abs(long long);          // llabs()
    ldiv_t div(long, long);            // ldiv()
    lldiv_t div(long long, long long); // lldiv()
    

    -?- To avoid ambiguities, C++ also adds the following overloads of abs() to <cstdlib>, with the semantics defined in <cmath>:

    float abs(float);
    double abs(double);
    long double abs(long double);
    

    -?- To avoid ambiguities, C++ also adds the following overloads of abs() to <cmath>, with the semantics defined in <cstdlib>:

    int abs(int);
    long abs(long);
    long long abs(long long);
    

Proposed resolution:

See proposed resolution of LWG 2192.


2295. Locale name when the provided Facet is a nullptr

Section: 22.3.1.2 [locale.cons] Status: New Submitter: Juan Soulie Opened: 2013-09-04 Last modified: 2016-02-12

Priority: 3

View all issues with New status.

Discussion:

22.3.1.2 [locale.cons] p14 ends with:

"[…] If f is null, the resulting object is a copy of other."

but the next line p15 says:

"Remarks: The resulting locale has no name."

But both can't be true when other has a name and f is null.

I've tried it on two implementations (MSVC,GCC) and they are inconsistent with each other on this.

Daniel Krügler:

As currently written, the Remarks element applies unconditionally for all cases and thus should "win". The question arises whether the introduction of this element by LWG 424 had actually intended to change the previous Note to a Remarks element. In either case the wording should be improved to clarify this special case.

Proposed resolution:


2303. Explicit instantiation of std::vector<UserType> broken?

Section: 18.6.2.3 [new.delete.placement] Status: New Submitter: Daniel Krügler Opened: 2013-09-18 Last modified: 2016-02-12

Priority: 3

View all other issues in [new.delete.placement].

View all issues with New status.

Discussion:

The library gives explicit permission in 17.6.4.2.1 [namespace.std] p2 that user code may explicitly instantiate a library template provided that the instantiations depend on at least one user-defined type:

A program may explicitly instantiate a template defined in the standard library only if the declaration depends on the name of a user-defined type and the instantiation meets the standard library requirements for the original template.

But it seems that the C++11 library is not specified in a way that guarantees such an instantiation to be well-formed if the minimum requirements of the library is not satisfied.

For example, in general, the first template parameter of std::vector is not required to be DefaultConstructible in general, but due to the split of the single C++03 member function with default argument

void resize(size_type sz, T c = T());

into

void resize(size_type sz);
void resize(size_type sz, const T& c);

the effect is now that for a type ND that is not DefaultConstructible, such as

struct NP { 
  NP(int); 
};

the explicit instantiation of std::vector<ND> is no longer well-formed, because the attempt to instantiate the single-argument overload of resize cannot not succeed, because this function imposes the DefaultInsertable requirements and given the default allocator this effectively requires DefaultConstructible.

But DefaultConstructible is not the only point, what about CopyConstructible versus MoveConstructible alone? It turns out that currently the second resize overload would fail during an explicit instantiation for a type like

struct MO { 
  MO() = default; 
  MO(MO&&) = default; 
};

because it imposes CopyInsertable requirements that end up being equivalent to the CopyConstructible requirements for the default allocator.

Technically a library can solve these issues: For special member functions by defining them in some base class, for others by transforming them effectively into a function template due to the great feature of default template arguments for function templates (At the very moment the validity of the latter approach depends on a resolution of core language issue CWG 1635, though). E.g. the here mentioned resize functions of std::vector could be prevented from instantiation by defining them like this with an implementation:

template<class = void>
void resize(size_type sz) { […] }
template<class = void>
void resize(size_type sz, const T& c) { […] }

In this case, these functions could also be defined in a base class, but the latter approach won't work in all cases.

Basically such an implementation is required to constrain all member functions that are not covered by the general requirements imposed on the actual library template parameters. I tested three different C++11 library implementations and but none could instantiate for example std::list, std::vector, or std::deque with value types that are not DefaultConstructible or only MoveConstructible.

This issue is raised to clarify the current situation in regard to the actual requirements imposed on user-provided types that are used to explicitly instantiate Library-provided templates. For example, the current Container requirements impose very little requirements on the actual value type and it is unclear to which extend library implementations have to respect that.

The minimum solution of this issue should be to at least realize that there is no fundamental requirement on DefaultConstructible for value types of library containers, because we have since C++03 the general statement of 17.6.3.1 [utility.arg.requirements] ("In general, a default constructor is not required."). It is unclear whether CopyConstructible should be required for an explicit instantiation request, but given the careful introduction of move operations in the library it would seem astonishing that a MoveConstructible type wouldn't suffice for value types of the container types.

In any case I can envision at least two approaches to solve this issue:

  1. As indicated in LWG 2292, those function could get an explicit "Template Constraints:" element, albeit this promises more than needed to solve this issue.

  2. The library could introduce a completely new element form, such as "Instantiation Constraints:" that would handle this situation for explicit instantiation situations. This would allow for simpler techniques to solve the issue when explicit instantiation is required compared to the first bullet, because it would not (necessarily) guarantee SFINAE-friendly expression-wellformedness, such as inspecting the expression std::declval<std::vector<ND>&>.resize(0) in an unevaluated context.

It should be noted that the 2013-08-27 comment to LWG 2193 could be resolved by a similar solution as indicated in this issue here.

Proposed resolution:


2307. Should the Standard Library use explicit only when necessary?

Section: 23 [containers] Status: Open Submitter: Zhihao Yuan Opened: 2013-09-26 Last modified: 2016-02-12

Priority: 2

View other active issues in [containers].

View all other issues in [containers].

View all issues with Open status.

Discussion:

LWG 2193 yields explicit for default ctors to allow {}, but not for all cases of uniform initialization. For example:

explicit vector(size_type count, const Allocator& alloc = Allocator());

This prevents {n, alloc()}. Although this use is relatively rare, but the behavior is inconsistent with that of

vector(size_type count, const T& value, const Allocator& alloc = Allocator());

[Urbana 2014-11-07: Move to Open]

Proposed resolution:


2318. basic_string's wording has confusing relics from the copy-on-write era

Section: 21.3.1 [basic.string] Status: New Submitter: Stephan T. Lavavej Opened: 2013-09-21 Last modified: 2016-02-12

Priority: 4

View other active issues in [basic.string].

View all other issues in [basic.string].

View all issues with New status.

Discussion:

21.3.1.4 [string.capacity]/8 specifies basic_string::resize(n, c) with:

Effects: Alters the length of the string designated by *this as follows:

This wording is a relic of the copy-on-write era. In addition to being extremely confusing, it has undesirable implications. Saying "replaces the string designated by *this with a string of length n whose elements are a copy" suggests that the trimming case can reallocate. Reallocation during trimming should be forbidden, like vector.

At least 7 paragraphs are affected: 21.3.1.4 [string.capacity]/8, 21.3.1.6.2 [string::append]/9, 21.3.1.6.3 [string::assign]/3 and /10, 21.3.1.6.4 [string::insert]/11, 21.3.1.6.5 [string::erase]/4, and 21.3.1.6.6 [string::replace]/11 say "replaces the string [designated/controlled] by *this". (21.3.1.6.7 [string::copy]/3 is different — it "replaces the string designated by s".)

Of the affected paragraphs, resize() and erase() are the most important to fix because they should forbid reallocation during trimming.

Proposed resolution:


2321. Moving containers should (usually) be required to preserve iterators

Section: 23.2.1 [container.requirements.general] Status: Open Submitter: Stephan T. Lavavej Opened: 2013-09-21 Last modified: 2016-02-12

Priority: 2

View other active issues in [container.requirements.general].

View all other issues in [container.requirements.general].

View all issues with Open status.

Discussion:

23.2.1 [container.requirements.general]/10 says that unless otherwise specified, "no swap() function invalidates any references, pointers, or iterators referring to the elements of the containers being swapped. [Note: The end() iterator does not refer to any element, so it may be invalidated. — end note]". However, move constructors and move assignment operators aren't given similar invalidation guarantees. The guarantees need several exceptions, so I do not believe that blanket language like /11 "Unless otherwise specified (either explicitly or by defining a function in terms of other functions), invoking a container member function or passing a container as an argument to a library function shall not invalidate iterators to, or change the values of, objects within that container." is applicable.

[2014-02-13 Issaquah]

General agreeement on intent, several wording nits and additional paragraphs to hit.

STL to provide updated wording. Move to Open.

[2015-02, Cologne]

AM: in the proposed wording, I'd like to mention that the iterators now refer to elements of a different container. I think we're saying something like this somewhere. JY: There's some wording like that for swap I think. TK: It's also in list::splice(). DK to JY: 23.2.1p9.

VV: The issue says that STL was going to propose new wording. Has he done that? AM: I believe we're looking at that. GR: The request touches on multiple paragraphs, and this PR has only one new paragraph, so this looks like it's not up-to-date. MC: This was last updated a year ago in Issaquah.

Conclusion: Skip, not up to date.

[2015-06, Telecom]

Still waiting for updated wording

Proposed resolution:

This wording is relative to N3691.

  1. In 23.2.1 [container.requirements.general]/10 change as indicated:

    -10- Unless otherwise specified (see 23.2.4.1, 23.2.5.1, 23.3.3.4, and 23.3.7.5) all container types defined in this Clause meet the following additional requirements:

    • […]

    • no copy constructor or assignment operator of a returned iterator throws an exception.

    • no move constructor (or move assignment operator when allocator_traits<allocator_type>::propagate_on_container_move_assignment::value is true) of a container (except for array) invalidates any references, pointers, or iterators referring to the elements of the source container. [Note: The end() iterator does not refer to any element, so it may be invalidated. — end note]

    • no swap() function throws an exception.

    • no swap() function invalidates any references, pointers, or iterators referring to the elements of the containers being swapped. [Note: The end() iterator does not refer to any element, so it may be invalidated. — end note]


2331. regex_constants::collate's effects are inaccurately summarized

Section: 28.5.1 [re.synopt] Status: Open Submitter: Stephan T. Lavavej Opened: 2013-09-21 Last modified: 2016-02-12

Priority: 3

View other active issues in [re.synopt].

View all other issues in [re.synopt].

View all issues with Open status.

Discussion:

The table in 28.5.1 [re.synopt]/1 says that regex_constants::collate "Specifies that character ranges of the form "[a-b]" shall be locale sensitive.", but 28.13 [re.grammar]/14 says that it affects individual character comparisons too.

[2012-02-12 Issaquah : recategorize as P3]

Marshall Clow: 28.13/14 only applies to ECMAScript

All: we're unsure

Jonathan Wakely: we should ask John Maddock

Move to P3

[2014-5-14, John Maddock response]

The original intention was the original wording: namely that collate only made character ranges locale sensitive. To be frank it's a feature that's probably hardly ever used (though I have no real hard data on that), and is a leftover from early POSIX standards which required locale sensitive collation for character ranges, and then later changed to implementation defined if I remember correctly (basically nobody implemented locale-dependent collation).

So I guess the question is do we gain anything by requiring all character-comparisons to go through the locale when this bit is set? Certainly it adds a great deal to the implementation effort (it's not what Boost.Regex has ever done). I guess the question is are differing code-points that collate identically an important use case? I guess there might be a few Unicode code points that do that, but I don't know how to go about verifying that.

STL:

If this was unintentional, then 28.5.1 [re.synopt]/1's table should be left alone, while 28.13 [re.grammar]/14 should be changed instead.

Jeffrey Yasskin:

This page mentions that [V] in Swedish should match "W" in a perfect world.

However, the most recent version of TR18 retracts both language-specific loose matches and language-specific ranges because "for most full-featured regular expression engines, it is quite difficult to match under code point equivalences that are not 1:1" and "tailored ranges can be quite difficult to implement properly, and can have very unexpected results in practice. For example, languages may also vary whether they consider lowercase below uppercase or the reverse. This can have some surprising results: [a-Z] may not match anything if Z < a in that locale."

ECMAScript doesn't include collation at all.

IMO, +1 to changing 28.13 instead of 28.5.1. It seems like we'd be on fairly solid ground if we wanted to remove regex_constants::collate entirely, in favor of named character classes, but of course that's not for this issue.

Proposed resolution:

This wording is relative to N3691.

  1. In 28.5.1 [re.synopt]/1, Table 138 — "syntax_option_type effects", change as indicated:

    Table 138 — syntax_option_type effects
    Element Effect(s) if set
    collate Specifies that character ranges of the form "[a-b]"comparisons and character range comparisons shall be locale sensitive.

2334. atomic's default constructor requires "uninitialized" state even for types with non-trivial default-constructor

Section: 29.6.5 [atomics.types.operations.req] Status: SG1 Submitter: Daniel Krügler Opened: 2013-10-03 Last modified: 2016-02-12

Priority: Not Prioritized

View all other issues in [atomics.types.operations.req].

View all issues with SG1 status.

Discussion:

According to 29.6.5 [atomics.types.operations.req] p4,

A ::A () noexcept = default;

Effects: leaves the atomic object in an uninitialized state. [Note: These semantics ensure compatibility with C. — end note]

This implementation requirement is OK for POD types, like int, but 29.5 [atomics.types.generic] p1 intentionally allows template arguments of atomic with a non-trivial default constructor ("The type of the template argument T shall be trivially copyable (3.9)"), so this wording can be read in a way that makes the behaviour of the following code undefined:

#include <atomic>
#include <iostream>

struct S {
  S() noexcept : v(42) {}
  int v;
};

int main() {
  std::atomic<S> as; // Default-initialization
  std::cout << as.load().v << std::endl; // ?
}

For a user-defined emulation of atomic the expected outcome would be defined and the program would output "42", but existing implementations differ and the result value is a "random number" for at least one implementation. This seems very surprising to me.

To realize that seemingly existing requirement, an implementation is either required to violate normal language rules internally or to perform specific bit-randomization-techniques after the normal default-initialization that called the default constructor of S.

According to my understanding, the non-normative note in 29.6.5 [atomics.types.operations.req] p4 is intended to refer to types that are valid C-types, but the example type S is not such a type.

To make the mental model of atomic's default constructor more intuitive for user-code, I suggest to clarify the wording to have the effects of default-initialization instead. The current state seems more like an unintended effect of imprecise language used here and has some similarities to wording that was incorrectly used to specify atomic_flag initialization as described by LWG 2159.

[2014-05-17, Daniel comments and provides alternative wording]

The current wording was considered controversial as expressed by reflector discussions. To me, the actual problem is not newly introduced by that wording, but instead is already present in basically all paragraphs specifying semantics of atomic types, since the wording never clearly distinguishes the value of the actual atomic type A and the value of the "underlying", corresponding non-atomic type C. The revised proposed wording attempts to improve the current ambiguity of these two kinds of values.

Previous resolution from Daniel [SUPERSEDED]:

This wording is relative to N3691.

  1. Modify 29.6.5 [atomics.types.operations.req] p4 as indicated: [Editorial note: There is no exposition-only member in atomic, which makes it a bit hard to specify what actually is initialized, but the usage of the term "value" seems consistent with similar wording used to specify the effects of the atomic load functions]

    A ::A () noexcept = default;
    

    -4- Effects: leaves the atomic object in an uninitialized stateThe value of the atomic object is default-initialized (8.6 [dcl.init]). [Note: These semantics ensure compatibility with C. — end note]

[2015-02 Cologne]

Handed over to SG1.

Proposed resolution:

This wording is relative to N3936.

  1. Modify 29.6.5 [atomics.types.operations.req] p2 as indicated: [Editorial note: This is a near-to editorial change not directly affecting this issue, but atomic_address does no longer exist and the pointed to definition is relevant in the context of this issue resolution.]

    -2- In the following operation definitions:

    • an A refers to one of the atomic types.

    • a C refers to its corresponding non-atomic type. The atomic_address atomic type corresponds to the void* non-atomic type.

    • […]

  2. Modify 29.6.5 [atomics.types.operations.req] p4 and the following as indicated: [Editorial note: There is no exposition-only member in atomic, which makes it a bit hard to specify what actually is initialized, but the introductory wording of 29.6.5 [atomics.types.operations.req] p2 b2 defines: "a C refers to its corresponding non-atomic type." which helps to specify the semantics in terms of "the C value referred to by the atomic object"]

    A::A() noexcept = default;
    

    -4- Effects: leaves the atomic object in an uninitialized stateDefault-initializes (8.6 [dcl.init]) the C value referred to by the atomic object. [Note: These semantics ensure compatibility with C. — end note]

    constexpr A::A(C desired) noexcept;
    

    -5- Effects: Direct-iInitializes the C value referred to by the atomic object with the value desired. Initialization is not an atomic operation (1.10). […]

    […]

    void atomic_init(volatile A* object, C desired) noexcept;
    void atomic_init(A* object, C desired) noexcept;
    

    -8- Effects: Non-atomically initializes the C value referred to by *object with value desired. […]

    void atomic_store(volatile A* object, C desired) noexcept;
    […]
    void A::store(C desired, memory_order order = memory_order_seq_cst) noexcept;
    

    -9- […]

    -10- Effects: Atomically replaces the C value pointed to by object or by this with the value of desired. […]

    […]

    C atomic_load(const volatile A* object) noexcept;
    […]
    C A::load(memory_order order = memory_order_seq_cst) const noexcept;
    

    -13- […]

    -14- […]

    -15- Returns: Atomically returns the C value pointed to by object or by this.

    […]

    C atomic_exchange(volatile A* object, C desired) noexcept;
    […]
    C A::exchange(C desired, memory_order order = memory_order_seq_cst) noexcept;
    

    -18- Effects: Atomically replaces the C value pointed to by object or by this with desired. […]

    -19- Returns: Atomically returns the C value pointed to by object or by this immediately before the effects.

    […]

    C atomic_fetch_key(volatile A* object, M operand) noexcept;
    […]
    C A::fetch_key(M operand, memory_order order = memory_order_seq_cst) noexcept;
    

    -28- Effects: Atomically replaces the C value pointed to by object or by this with the result of the computation applied to the C value pointed to by object or by this and the given operand. […]

    -29- Returns: Atomically, returns the C value pointed to by object or by this immediately before the effects.

    […]

  3. Modify 29.7 [atomics.flag] p5 and the following as indicated:

    bool atomic_flag_test_and_set(volatile atomic_flag* object) noexcept;
    […]
    bool atomic_flag::test_and_set(memory_order order = memory_order_seq_cst) noexcept;
    

    -5- Effects: Atomically sets the bool value pointed to by object or by this to true. […]

    -6- Returns: Atomically, returns the bool value of thepointed to by object or by this immediately before the effects.

    void atomic_flag_clear(volatile atomic_flag* object) noexcept;
    […]
    void atomic_flag::clear(memory_order order = memory_order_seq_cst) noexcept;
    

    -7- […]

    -8- Effects: Atomically sets the bool value pointed to by object or by this to false. […]


2335. array<array<int, 3>, 4> should be layout-compatible with int[4][3]

Section: 23.3.7 [array] Status: New Submitter: Jeffrey Yasskin Opened: 2013-10-04 Last modified: 2016-02-12

Priority: 3

View other active issues in [array].

View all other issues in [array].

View all issues with New status.

Discussion:

In order to replace some uses of C arrays with std::array, we need it to be possible to cast from a std::array<> to an equivalent C array. Core wording doesn't appear to be in quite the right state to allow casting, but if we specify that appropriate types are layout-compatible, we can at least write:

union {
  array<array<array<int, 2>, 3>, 4> arr;
  int carr[4][3][2];
};

to view memory as the other type: C++14 CD [class.mem]p18.

I believe it's sufficient to add "array<T, N> shall be layout-compatible (3.9 [basic.types]) with T[N]." to 23.3.7.1 [array.overview], but we might also need some extension to 9.2 [class.mem] to address the possibility of layout-compatibility between struct and array types.

I checked that libc++ on MacOS already implements this, although it would be good for someone else to double-check; I haven't checked any other standard libraries.

Proposed resolution:


2337. shared_ptr operator*() should not be noexcept

Section: 20.11.2.2.5 [util.smartptr.shared.obs] Status: Tentatively NAD Submitter: Stephan T. Lavavej Opened: 2013-10-05 Last modified: 2016-02-12

Priority: 2

View all other issues in [util.smartptr.shared.obs].

View all issues with Tentatively NAD status.

Discussion:

20.11.1.2.4 [unique.ptr.single.observers]/3: "pointer operator->() const noexcept; Requires: get() != nullptr."

20.11.2.2.5 [util.smartptr.shared.obs]/2: "T& operator*() const noexcept; Requires: get() != 0."

20.11.2.2.5 [util.smartptr.shared.obs]/5: "T* operator->() const noexcept; Requires: get() != 0."

Narrow-contract functions should not be noexcept.

[2014-02-15 Issaquah]

Issue is contentious, raise to P2.

[2015-02 Cologne]

AM: This ship has sailed. JM: What's the issue? AM: operator-> has narrow contract and should never have had noexcept. DK: Not quite. We explicitly called out that for shared_ptr this is fine. You said so in your "narrow contract" paper. GR: This would be a fairly major regression in the design of {unique,shared}_ptr over raw pointers; raw pointer dereferencing is noexcept. It's not a performance regression but a usability regression. AM: Do we expect users to query noexpect on dereference expressions? Room: Yes. VV: We don't just expect it, we have seen it. JM: Yes, users may be querying something like noexcept(x->y) and expect to be checking y, but silently end up checking x->.

Close as NAD, with explanation from GR.

Previous resolution [SUPERSEDED]:

This wording is relative to N3691.

  1. In 20.11.1.2 [unique.ptr.single]/1, class template unique_ptr synopsis for single objects, change as indicated:

    pointer operator->() const noexcept;
    
  2. In 20.11.1.2.4 [unique.ptr.single.observers] change as indicated:

    pointer operator->() const noexcept;
    

    -3- Requires: get() != nullptr.

    -4- Returns: get().

    -?- Throws: Nothing.

    -5- Note: use typically requires that T be a complete type.

  3. In 20.11.2.2 [util.smartptr.shared]/1, class template shared_ptr synopsis, change as indicated:

    T& operator*() const noexcept;
    T* operator->() const noexcept;
    
  4. In 20.11.2.2.5 [util.smartptr.shared.obs] change as indicated:

    T& operator*() const noexcept;
    

    -2- Requires: get() != 0.

    -3- Returns: *get().

    -?- Throws: Nothing.

    -4- Remarks: When T is void, it is unspecified whether this member function is declared. If it is declared, it is unspecified what its return type is, except that the declaration (although not necessarily the definition) of the function shall be well formed.

    T* operator->() const noexcept;
    

    -5- Requires: get() != 0.

    -6- Returns: get().

    -?- Throws: Nothing.

[2015-03-03, Geoffrey provides rationale]

Rationale:

It is by design that these members are noexcept, and changing that now would be a substantial regression in functionality. These classes were designed to substitute for plain pointers as transparently as possible, so since those operations are effectively noexcept on plain pointers, they should be noexcept on unique_ptr and shared_ptr as well. This matters in practice because we expect these members to be used fairly often inside the noexcept operator, and such code could be broken by this change. These design considerations override our general policy against noexcept for narrow-contract functions.

It is notable that N3279, which proposed this policy, did not propose striking noexcept from these operations. It's not clear if the omission of operator* and operator-> was an oversight, or an intentional reflection of the above considerations. N3279 was based on N3248 by the same authors, which states that:

"Most applications of noexcept for unique_ptr and shared_ptr are on functions with wide contracts. However, there are preconditions on the atomic access functions, so these should lose the specification."

Proposed resolution:


2338. §[re.traits]/7 expects of locale facets something not guaranteed by [locale.facet]/4

Section: 28.7 [re.traits], 22.3.1.1.2 [locale.facet] Status: Open Submitter: Sergey Zubkov Opened: 2013-10-15 Last modified: 2016-02-12

Priority: 3

View all other issues in [re.traits].

View all issues with Open status.

Discussion:

28.7 [re.traits]/7, begins with "if typeid(use_facet<collate<charT> >) == typeid(collate_byname<charT>)", which appears to be pseudocode with the intention to convey that the collate facet has not been replaced by the user. Cf. the wording in N1429 "there is no portable way to implement transform_primary in terms of std::locale, since even if the sort key format returned by std::collate_byname<>::transform is known and can be converted into a primary sort key, the user can still install their own custom std::collate implementation into the locale object used, and that can use any sort key format they see fit.".

Taken literally, 28.7 [re.traits]/7 appears to imply that named locales are required to hold their collate facets with dynamic type std::collate_byname<charT>, which is in fact true in some implementations (e.g libc++), but not others (e.g. libstdc++). This does not follow from the description of _byname in 22.3.1.1.2 [locale.facet]/4, which is only required to provide equivalent semantics, to the named locale's facet, not to actually be one.

[2015-05-06 Lenexa: Move to Open]

MC, RP: Consequence of failing to follow the rule is UB.

MC: Tightening of requirements.

RP: It should be this way, we just didn't impose it before.

MC: Second change is a bug fix, original code didn't work.

TK: Doesn't seem to make things worse.

Bring up in larger group tomorrow.

JW arrives.

JW: libstdc++ violates this due to two std::string ABIs.

JW: This prevents installing a type derived from Facet_byname, constrains the implementor from using a smarter derived class version.

JW: Can't look at facet id to detect replacement, because replacements have the same id.

RP: Can you give it multiple ids through multiple inheritance?

JW: No, the facet mechanism wouldn't like that.

JW: We should also ask Martin Sebor, he's implemented this stuff recently.

MC: Sounds like this resolution doesn't work, need a better solution.

JW: Write in words "if the facet has not been replaced by the user", the implementation knows how to detect that, but not like this.

RP: User RE traits need to detect this too.

JW: =(

Move to Open, JW will invite Martin Sebor to join LWG for discussion.

Later ...

JW: This is not needed for user specializations after all.

MC: Agree, [re.traits]/7 only applies to the stdlib traits.

NM: Effects: doesn't make sense.

JW, NM, Martin Sebor to come up with new wording.

Proposed resolution:

This wording is relative to N3691.

  1. Modify 22.3.1.1.2 [locale.facet]/4 as indicated:

    For some standard facets a standard "..._byname" class, derived from it, implements the virtual function semantics equivalent toprovided by that facet of the locale constructed by locale(const char*) with the same name. Each such facet provides a constructor that takes a const char* argument, which names the locale, and a refs argument, which is passed to the base class constructor. Each such facet also provides a constructor that takes a string argument str and a refs argument, which has the same effect as calling the first constructor with the two arguments str.c_str() and refs. If there is no "..._byname" version of a facet, the base class implements named locale semantics itself by reference to other facets. For any locale loc constructed by locale(const char*) and facet Facet that has a corresponding standard Facet_byname class, typeid(use_facet<Facet>(loc)) == typeid(Facet_byname).

  2. Modify 28.7 [re.traits]/7 as indicated:

    template <class ForwardIterator>
      string_type transform_primary(ForwardIterator first, ForwardIterator last) const;
    

    -7- Effects: if typeid(use_facet<collate<charT> >(getloc())) == typeid(collate_byname<charT>) and the form of the sort key returned by collate_byname<charT>::transform(first, last) is known and can be converted into a primary sort key then returns that key, otherwise returns an empty string.


2342. User conversion to wchar_t const* or to wchar_t not invoked for operator<<

Section: 27.7.3.1 [ostream] Status: New Submitter: Alf P. Steinbach Opened: 2013-10-29 Last modified: 2016-02-12

Priority: 4

View all other issues in [ostream].

View all issues with New status.

Discussion:

For wide streams argument types wchar_t const* and wchar_t are supported only as template parameters. User defined conversions are not considered for template parameter matching. Hence inappropriate overloads of operator<< are selected when an implicit conversion is required for the argument, which is inconsistent with the behavior for char const* and char, is unexpected, and is a useless result.

Demonstration:

#include <iostream>

struct Byte_string
{ 
  operator char const*() const { return "Hurray, it works!"; } 
};

struct Wide_string
{ 
  operator wchar_t const*() const { return L"Hurray, it works!"; } 
};

struct Byte_ch
{ 
  operator char() const { return 'X'; } 
};

struct Wide_ch
{ 
  operator wchar_t() const { return L'X'; } 
};

auto main() -> int
{
  using namespace std;
  wcout << "'X' as char value   : " << Byte_ch() << endl;
  wcout << "'X' as wchar_t value: " << Wide_ch() << endl;
  wcout << "Byte string pointer : " << Byte_string() << endl;
  wcout << "Wide string pointer : " << Wide_string() << endl;
}

Example output:

'X' as char value   : X
'X' as wchar_t value: 88
Byte string pointer : Hurray, it works!
Wide string pointer : 000803C8

Proposed resolution:

This wording is relative to N3797.

  1. Modify 27.7.3.1 [ostream], class template basic_ostream synopsis, as indicated:

    namespace std {
    […]
    
    // 27.7.3.6.4 character inserters
    template<class charT, class traits>
      basic_ostream<charT,traits>& operator<<(basic_ostream<charT,traits>&,
                                              charT);
    template<class charT, class traits>
      basic_ostream<charT,traits>& operator<<(basic_ostream<charT,traits>&,
                                              char);
    template<class traits>
      basic_ostream<char,traits>& operator<<(basic_ostream<char,traits>&,
                                             char);
    template<class traits>
      basic_ostream<wchar_t,traits>& operator<<(basic_ostream<wchar_t,traits>&,
                                                wchar_t);
    […]
    
    template<class charT, class traits>
      basic_ostream<charT,traits>& operator<<(basic_ostream<charT,traits>&,
                                              const charT*);
    template<class charT, class traits>
      basic_ostream<charT,traits>& operator<<(basic_ostream<charT,traits>&,
                                              const char*);
    template<class traits>
      basic_ostream<char,traits>& operator<<(basic_ostream<char,traits>&,
                                             const char*);
    template<class traits>
      basic_ostream<wchar_t,traits>& operator<<(basic_ostream<wchar_t,traits>&,
                                                const wchar_t*);
    […]
    }
    
    
  2. Modify 27.7.3.6.4 [ostream.inserters.character] as indicated: [Drafting note: The replacement of os by out in p1 and the insertion of "out." in p4 just fix two obvious typos — end drafting note]

    template<class charT, class traits>
      basic_ostream<charT,traits>& operator<<(basic_ostream<charT,traits>& out,
                                              charT c);
    template<class charT, class traits>
      basic_ostream<charT,traits>& operator<<(basic_ostream<charT,traits>& out,
                                              char c);
    // specialization
    template<class traits>
      basic_ostream<char,traits>& operator<<(basic_ostream<char,traits>& out,
                                             char c);
    template<class traits>
      basic_ostream<wchar_t,traits>& operator<<(basic_ostream<wchar_t,traits>& out,
                                                wchar_t c);
    
    // signed and unsigned
    template<class traits>
      basic_ostream<char,traits>& operator<<(basic_ostream<char,traits>& out,
                                              signed char c);
    template<class traits>
      basic_ostream<char,traits>& operator<<(basic_ostream<char,traits>& out,
                                              unsigned char c);
    

    -1- Effects: Behaves as a formatted output function (27.7.3.6.1 [ostream.formatted.reqmts]) of out. Constructs a character sequence seq. If c has type char and the character type of the stream is not char, then seq consists of out.widen(c); otherwise seq consists of c. Determines padding for seq as described in 27.7.3.6.1 [ostream.formatted.reqmts]. Inserts seq into out. Calls osout.width(0).

    -2- Returns: out.

    template<class charT, class traits>
      basic_ostream<charT,traits>& operator<<(basic_ostream<charT,traits>& out,
                                              const charT* s);
    template<class charT, class traits>
      basic_ostream<charT,traits>& operator<<(basic_ostream<charT,traits>& out,
                                              const char* s);
    template<class traits>
      basic_ostream<char,traits>& operator<<(basic_ostream<char,traits>& out,
                                             const char* s);
    template<class traits>
      basic_ostream<wchar_t,traits>& operator<<(basic_ostream<wchar_t,traits>& out,
                                                const wchar_t* s);
    											
    template<class traits>
      basic_ostream<char,traits>& operator<<(basic_ostream<char,traits>& out,
                                             const signed char* s);
    template<class traits>
      basic_ostream<char,traits>& operator<<(basic_ostream<char,traits>& out,
                                             const unsigned char* s);
    

    -3- Requires: s shall not be a null pointer.

    -4- Effects: Behaves like a formatted inserter (as described in 27.7.3.6.1 [ostream.formatted.reqmts]) of out. Creates a character sequence seq of n characters starting at s, each widened using out.widen() (27.5.5.3), where n is the number that would be computed as if by:

    • traits::length(s) for the following overloads:

      • where the first argument is of type basic_ostream<charT, traits>& and the second is of type const charT*,

      • and also for the overload where the first argument is of type basic_ostream<char, traits>& and the second is of type const char*,

      • where the first argument is of type basic_ostream<wchar_t, traits>& and the second is of type const wchar_t*,

    • std::char_traits<char>::length(s) for the overload where the first argument is of type basic_ostream<charT, traits>& and the second is of type const char*,

    • traits::length(reinterpret_cast<const char*>(s)) for the other two overloads.

    Determines padding for seq as described in 27.7.3.6.1 [ostream.formatted.reqmts]. Inserts seq into out. Calls out.width(0).

    -5- Returns: out.


2343. Is the value of the ECMA-262 RegExp object's multiline property really false?

Section: 28.13 [re.grammar] Status: New Submitter: Nayuta Taga Opened: 2013-10-30 Last modified: 2016-02-12

Priority: 2

View other active issues in [re.grammar].

View all other issues in [re.grammar].

View all issues with New status.

Discussion:

In the following "Multiline" is the value of the ECMA-262 RegExp object's multiline property.

In ECMA-262, there are some definitions that relate to Multiline:

So, the C++11 standard says that Multiline is false. As it is false, ^ matches only the beginning of the string, and $ matches only the end of the string.

However, two flags are defined in 28.5.2 [re.matchflag] Table 139:

match_not_bol: the character ^ in the regular expression shall not match [first,first).

match_not_eol: the character "$" in the regular expression shall not match [last,last).

As Multiline is false, the match_not_bol and the match_not_eol are meaningless because they only make ^ and $ match none.

In my opinion, Multiline should be true.

FYI, Multiline of the existing implementations are as follows:

Multiline=false:

Multiline=true:

[2015-05-22, Daniel comments]

This issue interacts with LWG 2503.

Proposed resolution:


2348. charT('1') is not the wide equivalent of '1'

Section: 20.9 [template.bitset], 27.7.6 [quoted.manip] Status: Open Submitter: Zhihao Yuan Opened: 2013-12-02 Last modified: 2016-02-12

Priority: 3

View other active issues in [template.bitset].

View all other issues in [template.bitset].

View all issues with Open status.

Discussion:

Example: char16_t('1') != u'1' is possible.

The numeric value of char16_t is defined to be Unicode code point, which is same to the ASCII value and UTF-8 for 7-bit chars. However, char is not guaranteed to have an encoding which is compatible with ASCII. For example, '1' in EBCDIC is 241.

I found three places in the standard casting narrow char literals: bitset::bitset, bitset::to_string and quoted.

PJ confirmed this issue and says he has a solution used in their <filesystem> implementation, and he may want to propose it to the standard.

The solution in my mind, for now, is to make those default arguments magical, where the "magic" can be implemented with a C11 _Generic selection (works in clang):

#define _G(T, literal) _Generic(T{}, \
      char: literal, \
      wchar_t: L ## literal, \
      char16_t: u ## literal, \
      char32_t: U ## literal)

  _G(char16_t, '1') == u'1'

[Lenexa 2015-05-05: Move to Open]

Ask for complete PR (need quoted, to string, et al.)

Will then take it up again

Expectation is that this is correct way to fix this

Proposed resolution:

This wording is relative to N3797.

[Drafting note: This is a sample wording fixing only one case; I'm just too lazy to copy-paste it before we discussed whether the solution is worth and sufficient (for example, should the other `charT`s like `unsigned char` just don't compile without supplying those arguments? I hope so). — end drafting note]
  1. Modify 20.9 [template.bitset] p1, class template bitset synopsis, as indicated:

    namespace std {
      template <size_t N> class bitset {
      public:
        […]
        template<class charT, class traits, class Allocator>
          explicit bitset(
            const basic_string<charT,traits,Allocator>& str,
            typename basic_string<charT,traits,Allocator>::size_type pos = 0,
            typename basic_string<charT,traits,Allocator>::size_type n =
              basic_string<charT,traits,Allocator>::npos,
              charT zero = charT('0')see below, charT one = charT('1')see below);
         […]
      };
      […]
    }
    
  2. Modify 20.9.1 [bitset.cons] as indicated:

    template<class charT, class traits, class Allocator>
    explicit 
    bitset(const basic_string<charT, traits, Allocator>& str,
           typename basic_string<charT, traits, Allocator>::size_type pos = 0,
           typename basic_string<charT, traits, Allocator>::size_type n =
             basic_string<charT, traits, Allocator>::npos,
             charT zero = charT('0')see below, charT one = charT('1')see below);
    

    -?- The default values of zero and one compare equal to the character literals 0 and 1 of type charT, respectively.

    -3- Requires:: pos <= str.size().

    […]


2349. Clarify input/output function rethrow behavior

Section: 27.7.2.2.1 [istream.formatted.reqmts] Status: Open Submitter: Zhihao Yuan Opened: 2013-12-06 Last modified: 2016-02-12

Priority: 3

View all other issues in [istream.formatted.reqmts].

View all issues with Open status.

Discussion:

The formatted input function requirement says in 27.7.2.2.1 [istream.formatted.reqmts]:

"If an exception is thrown during input then ios::badbit is turned on in *this's error state. If (exceptions()&badbit) != 0 then the exception is rethrown."

while some formatted function may throw an exception from basic_ios::clear, for example in 20.9.4 [bitset.operators] p6:

"If no characters are stored in str, calls is.setstate(ios_base::failbit) (which may throw ios_base::failure)"

So should this exception be considered as "an exception [...] thrown during input"? And here is an implementation divergence (or you can read the following as "a bug libc++ only has" :)

cin.exceptions(ios_base::failbit);
bitset<N> b;
try {
  cin >> b;  // type 'a' and return
} catch (...)
{}

Now cin.rdstate() is just failbit in libstdc++ (and Dinkumware, by PJ), but failbit & badbit libc++. Similar difference found in other places, like eofbit & badbid after std::getline.

PJ and Matt both agree that the intention (of badbit + rethrow) is "to signify an exception arising in user code, not the iostreams package".

In addition, I found the following words in unformatted input function's requirements (27.7.2.3 [istream.unformatted]):

If an exception is thrown during input then ios::badbit is turned on in *this's error state. (Exceptions thrown from basic_ios<>::clear() are not caught or rethrown.) If (exceptions()&badbit) != 0 then the exception is rethrown.

The content within the parenthesis is added by LWG defect 61, and does fix the ambiguity. However, it only fixed the 1 of 4 requirements, and it lost some context (the word "rethrown" is not seen before this sentence within this section).

[Lenexa 2015-05-07: Marshall to research and report]

Proposed resolution:

This wording is relative to N3797.

[Drafting note: The editor is kindly asked to introduce additional spaces at the following marked occurrences of operator&end drafting note]
  1. Modify 27.7.2.2.1 [istream.formatted.reqmts] p1 as indicated:

    -1- Each formatted input function begins execution by constructing an object of class sentry with the noskipws (second) argument false. If the sentry object returns true, when converted to a value of type bool, the function endeavors to obtain the requested input. If an exception, other than the ones thrown from clear(), if any, is thrown during input then ios::badbit is turned on[Footnote 314] in *this's error state. If (exceptions() & badbit) != 0 then the exception is rethrown. In any case, the formatted input function destroys the sentry object. If no exception has been thrown, it returns *this.

  2. Modify 27.7.3.6.1 [ostream.formatted.reqmts] p1 as indicated:

    -1- Each formatted output function begins execution by constructing an object of class sentry. If this object returns true when converted to a value of type bool, the function endeavors to generate the requested output. If the generation fails, then the formatted output function does setstate(ios_base::failbit), which might throw an exception. If an exception, other than the ones thrown from clear(), if any, is thrown during output, then ios::badbit is turned on[Footnote 327] in *this's error state. If (exceptions() & badbit) != 0 then the exception is rethrown. Whether or not an exception is thrown, the sentry object is destroyed before leaving the formatted output function. If no exception is thrown, the result of the formatted output function is *this.

  3. Modify 27.7.3.7 [ostream.unformatted] p1 as indicated:

    -1- Each unformatted output function begins execution by constructing an object of class sentry. If this object returns true, while converting to a value of type bool, the function endeavors to generate the requested output. If an exception, other than the ones thrown from clear(), if any, is thrown during output, then ios::badbit is turned on[Footnote 330] in *this's error state. If (exceptions() & badbit) != 0 then the exception is rethrown. In any case, the unformatted output function ends by destroying the sentry object, then, if no exception was thrown, returning the value specified for the unformatted output function.

  4. Modify 27.7.2.3 [istream.unformatted] p1 as indicated:

    -1- Each unformatted input function begins execution by constructing an object of class sentry with the default argument noskipws (second) argument true. If the sentry object returns true, when converted to a value of type bool, the function endeavors to obtain the requested input. Otherwise, if the sentry constructor exits by throwing an exception or if the sentry object returns false, when converted to a value of type bool, the function returns without attempting to obtain any input. In either case the number of extracted characters is set to 0; unformatted input functions taking a character array of non-zero size as an argument shall also store a null character (using charT()) in the first location of the array. If an exception, other than the ones thrown from clear(), if any, is thrown during input then ios::badbit is turned on[Footnote 317] in *this's error state. (Exceptions thrown from basic_ios<>::clear() are not caught or rethrown.) If (exceptions() & badbit) != 0 then the exception is rethrown. It also counts the number of characters extracted. If no exception has been thrown it ends by storing the count in a member object and returning the value specified. In any event the sentry object is destroyed before leaving the unformatted input function.


2352. Is a default-constructed std::seed_seq intended to produce a predictable .generate()?

Section: 26.6.7.1 [rand.util.seedseq] Status: New Submitter: Thomas Plum Opened: 2013-12-02 Last modified: 2016-02-12

Priority: 2

View all other issues in [rand.util.seedseq].

View all issues with New status.

Discussion:

With respect to class seed_seq 26.6.7.1 [rand.util.seedseq], is a default-constructed std::seed_seq intended to produce a predictable .generate() sequence?

Implementations differ.

Proposed resolution:


2358. Apparently-bogus definition of is_empty type trait

Section: 20.15.4.3 [meta.unary.prop] Status: New Submitter: Richard Smith Opened: 2014-02-01 Last modified: 2016-02-12

Priority: 3

View other active issues in [meta.unary.prop].

View all other issues in [meta.unary.prop].

View all issues with New status.

Discussion:

The 'Condition' for std::is_empty is listed as:

"T is a class type, but not a union type, with no non-static data members other than bit-fields of length 0, no virtual member functions, no virtual base classes, and no base class B for which is_empty<B>::value is false."

This is incorrect: there is no such thing as a non-static data member that is a bit-field of length 0, since bit-fields of length 0 must be unnamed, and unnamed bit-fields are not members (see 9.2.4 [class.bit] p2).

It also means that classes such as:

struct S {
 int : 3;
};

are empty (because they have no non-static data members). There's implementation divergence on the value of is_empty<S>::value.

I'm not sure what the purpose of is_empty is (or how it could be useful), but if it's desirable for the above type to not be treated as empty, something like this could work:

"T is a class type, but not a union type, with no non-static data members other than, no unnamed bit-fields of non-zero length 0, no virtual member functions, no virtual base classes, and no base class B for which is_empty<B>::value is false."

and if the above type should be treated as empty, then this might be appropriate:

"T is a class type, but not a union type, with no (named) non-static data members other than bit-fields of length 0, no virtual member functions, no virtual base classes, and no base class B for which is_empty<B>::value is false."

Proposed resolution:


2362. unique, associative emplace() should not move/copy the mapped_type constructor arguments when no insertion happens

Section: 23.2.4 [associative.reqmts], 23.2.5 [unord.req] Status: New Submitter: Jeffrey Yasskin Opened: 2014-02-15 Last modified: 2015-09-25

Priority: 3

View other active issues in [associative.reqmts].

View all other issues in [associative.reqmts].

View all issues with New status.

Discussion:

a_uniq.emplace(args) is specified as:

Effects: Inserts a value_type object t constructed with
std::forward<Args>(args)... if and only if there is no element in the
container with key equivalent to the key of t. The bool component of
the returned pair is true if and only if the insertion takes place,
and the iterator component of the pair points to the element with key
equivalent to the key of t.

However, we occasionally find code of the form:

std::unique_ptr<Foo> p(new Foo);
auto res = m.emplace("foo", std::move(p));

where we'd like to avoid destroying the Foo if the insertion doesn't take place (if the container already had an element with the specified key).

N3873 includes a partial solution to this in the form of a new emplace_stable member function, but LEWG's discussion strongly agreed that we'd rather have emplace() Just Work:

Should map::emplace() be guaranteed not to move/copy its arguments if the insertion doesn't happen?

SF: 8 F: 3 N: 0 A: 0 SA: 0

This poll was marred by the fact that we didn't notice or call out that emplace() must construct the key before doing the lookup, and it must not then move the key after it determines whether an insert is going to happen, and the mapped_type instance must live next to the key.

The very similar issue 2006 was previously marked NAD, with N3178 as discussion. However, given LEWG's interest in the alternate behavior, we should reopen the question in this issue.

We will need a paper that describes how to implement this before we can make more progress.

Proposed resolution:


2363. Defect in 30.4.1.4.1 [thread.sharedtimedmutex.class]

Section: 30.4.1.5.1 [thread.sharedtimedmutex.class] Status: Open Submitter: Richard Smith Opened: 2014-02-16 Last modified: 2016-03-06

Priority: 2

View all issues with Open status.

Discussion:

30.4.1.5.1 [thread.sharedtimedmutex.class] paragraph 2:

The class shared_timed_mutex shall satisfy all of the SharedTimedMutex requirements (30.4.1.4). It shall be a standard layout class (Clause 9).

There's no SharedTimedMutex requirements; this name doesn't appear anywhere else in the standard. (Prior to N3891, this was SharedMutex, which was equally undefined.)

I assume this concept should be defined somewhere?

Also, n3891 changes 30.4.1.5 [thread.sharedtimedmutex.requirements] from defining "shared mutex type" to defining "shared timed mutex type", but its paragraph 2 still talks about "shared mutex type". Is that OK? I think you could argue that it's clear enough what it means, but presumably it should use the term that paragraph 1 defined.

30.4.2.3 [thread.lock.shared] paragraph 1 talks about the "shared mutex requirements", which again is a term that isn't defined, and presumably means "the requirements on a shared timed mutex type" or similar (maybe if SharedMutex or SharedTimedMutex were defined it could be reused here).

[2014-05-22, Daniel comments]

As for SharedTimedMutex, there exists a similar problem in regard to TimedMutex referred to in 30.4.1.3.1 [thread.timedmutex.class] p2 and in 30.4.1.3.2 [thread.timedmutex.recursive] p2, but nowhere defined.

Another problem is, that according to 30.4.1.2.1 [thread.mutex.class] p3, "The class mutex shall satisfy all the Mutex requirements (30.4.1 [thread.mutex.requirements]).", but there are no concrete Mutex requirements, 30.4.1 [thread.mutex.requirements] — titled as "Mutex requirements" — describes mutex types, timed mutex types, and shared timed mutex types.

[2014-06-08, Daniel comments and provides wording]

The presented wording adds to the existing mutex types, timed mutex types, and shared timed mutex types terms a new set of corresponding MutexType, TimedMutexType, and SharedTimedMutexType requirements.

The reason for the change of requirement names is two-fold: First, the new name better matches the intention to have a concrete name for the requirements imposed on the corresponding mutex types (This kind of requirement deviate from the more general Lockable requirements, which are not restricted to a explicitly enumerated set of library types). Second, using **MutexType over **Mutex provides the additional advantage that it reduces the chances of confusing named requirements from template parameters named Mutex (such as for unique_lock or shared_lock).

Nonetheless the here presented wording has one unfortunate side-effect: Once applied it would have the effect that types used to instantiate std::shared_lock cannot be user-defined shared mutex types due to 30.4.2.3 [thread.lock.shared]. The reason is based on the currently lack of an existing SharedLockable requirement set, which would complete the existing BasicLockable and Lockable requirements (which are "real" requirements). This restriction is not actually a problem introduced by the provided resolution but instead one that existed before but becomes more obvious now.

[2015-02 Cologne]

Handed over to SG1.

[2015-05 Lenexa, SG1 response]

Thanks to Daniel, and please put it in SG1-OK status. Perhaps open another issue for the remaining problem Daniel points out?

[2015-10 pre-Kona]

SG1 hands this over to LWG for wording review

[2015-10-21 Kona, Daniel comments and adjusts wording to to untimed shared mutex types]

The new wording reflects the addition of the new shared mutex types. The approach used for shared_lock is similar to the one used for unique_lock: The template argument Mutex has a reduced requirement set that is not sufficient for all operations. Only those members that require stronger requirements of SharedTimedMutexType specify that additionally in the Requires element of the corresponding prototype specifications.

The proposed wording could be more general if we would introduce more fundamental requirements set for SharedLockable and SharedTimedLockable types which could be satisfied by user-provided types as well, because the SharedMutexType and SharedTimedMutexType requirements are essentially restricted to an enumerated set of types provided by the Standard Library. But this extension seemed too large for this issue and can be easily fixed later without any harm.

Previous resolution [SUPERSEDED]:

This wording is relative to N3936.

  1. Change 30.4.1.2 [thread.mutex.requirements.mutex] as indicated:

    -1- The mutex types are the standard library types std::mutex, std::recursive_mutex, std::timed_mutex, std::recursive_timed_mutex, and std::shared_timed_mutex. They shall meet the MutexType requirements set out in this section. In this description, m denotes an object of a mutex type.

  2. Change 30.4.1.2.1 [thread.mutex.class] as indicated:

    -3- The class mutex shall satisfy all the MutexType requirements (30.4.1.2 [thread.mutex.requirements.mutex]30.4.1 [thread.mutex.requirements]). It shall be a standard-layout class (Clause 9).

  3. Change 30.4.1.2.2 [thread.mutex.recursive] as indicated:

    -2- The class recursive_mutex shall satisfy all the MutexMutexType requirements (30.4.1.2 [thread.mutex.requirements.mutex]30.4.1 [thread.mutex.requirements]). It shall be a standard-layout class (Clause 9).

  4. Change 30.4.1.3 [thread.timedmutex.requirements] as indicated:

    -1- The timed mutex types are the standard library types std::timed_mutex, std::recursive_timed_mutex, and std::shared_timed_mutex. They shall meet the TimedMutexType requirements set out below. In this description, m denotes an object of a mutex type, rel_time denotes an object of an instantiation of duration (20.12.5), and abs_time denotes an object of an instantiation of time_point (20.12.6).

  5. Change 30.4.1.3.1 [thread.timedmutex.class] as indicated:

    -2- The class timed_mutex shall satisfy all of the TimedMutexType requirements (30.4.1.3 [thread.timedmutex.requirements]). It shall be a standard-layout class (Clause 9).

  6. Change 30.4.1.3.2 [thread.timedmutex.recursive] as indicated:

    -2- The class recursive_timed_mutex shall satisfy all of the TimedMutexType requirements (30.4.1.3 [thread.timedmutex.requirements]). It shall be a standard-layout class (Clause 9).

  7. Change 30.4.1.5 [thread.sharedtimedmutex.requirements] as indicated: [Drafting note: The reference to the timed mutex types requirements has been moved after introducing the new requirement set to ensure that SharedTimedMutexType refine TimedMutexType.]

    -1- The standard library type std::shared_timed_mutex is a shared timed mutex type. Shared timed mutex types shall meet the SharedTimedMutexType requirements of timed mutex types (30.4.1.3 [thread.timedmutex.requirements]), and additionally shall meet the requirements set out below. In this description, m denotes an object of a mutex type, rel_type denotes an object of an instantiation of duration (20.12.5), and abs_time denotes an object of an instantiation of time_point (20.12.6).

    -?- The shared timed mutex types shall meet the TimedMutexType requirements (30.4.1.3 [thread.timedmutex.requirements]).

  8. Change 30.4.1.5.1 [thread.sharedtimedmutex.class] as indicated:

    -2- The class shared_timed_mutex shall satisfy all of the SharedTimedMutexType requirements (30.4.1.5 [thread.sharedtimedmutex.requirements]). It shall be a standard-layout class (Clause 9).

  9. Change 30.4.2.3 [thread.lock.shared] as indicated: [Drafting note: Once N3995 has been applied, the following reference should be changed to the new SharedMutexType requirements ([thread.sharedmutex.requirements]) or even better to some new SharedLockable requirements (to be defined) — end drafting note]

    -1- […] The supplied Mutex type shall meet the shared mutexSharedTimedMutexType requirements (30.4.1.5 [thread.sharedtimedmutex.requirements]).

    -2- [Note: shared_lock<Mutex> meets the TimedLockable requirements (30.2.5.4). — end note]

[2016-02 Jacksonville]

Marshall to review wording.

Proposed resolution:

This wording is relative to N4527.

  1. Change 30.4.1.2 [thread.mutex.requirements.mutex] as indicated:

    -1- The mutex types are the standard library types std::mutex, std::recursive_mutex, std::timed_mutex, std::recursive_timed_mutex, std::shared_mutex, and std::shared_timed_mutex. They shall meet the MutexType requirements set out in this section. In this description, m denotes an object of a mutex type.

    -2- The mutex types shall meet the Lockable requirements (30.2.5.3 [thread.req.lockable.req]).

  2. Change 30.4.1.2.1 [thread.mutex.class] as indicated:

    -3- The class mutex shall satisfy all the MutexType requirements (30.4.1.2 [thread.mutex.requirements.mutex]30.4.1 [thread.mutex.requirements]). It shall be a standard-layout class (Clause 9).

  3. Change 30.4.1.2.2 [thread.mutex.recursive] as indicated:

    -2- The class recursive_mutex shall satisfy all the MutexMutexType requirements (30.4.1.2 [thread.mutex.requirements.mutex]30.4.1 [thread.mutex.requirements]). It shall be a standard-layout class (Clause 9).

  4. Change 30.4.1.3 [thread.timedmutex.requirements] as indicated:

    -1- The timed mutex types are the standard library types std::timed_mutex, std::recursive_timed_mutex, and std::shared_timed_mutex. They shall meet the TimedMutexType requirements set out below. In this description, m denotes an object of a mutex type, rel_time denotes an object of an instantiation of duration (20.12.5), and abs_time denotes an object of an instantiation of time_point (20.12.6).

    -2- The timed mutex types shall meet the TimedLockable requirements (30.2.5.4 [thread.req.lockable.timed]).

  5. Change 30.4.1.3.1 [thread.timedmutex.class] as indicated:

    -2- The class timed_mutex shall satisfy all of the TimedMutexType requirements (30.4.1.3 [thread.timedmutex.requirements]). It shall be a standard-layout class (Clause 9).

  6. Change 30.4.1.3.2 [thread.timedmutex.recursive] as indicated:

    -2- The class recursive_timed_mutex shall satisfy all of the TimedMutexType requirements (30.4.1.3 [thread.timedmutex.requirements]). It shall be a standard-layout class (Clause 9).

  7. Change 30.4.1.4 [thread.sharedmutex.requirements] as indicated: [Drafting note: The reference to the mutex types requirements has been moved after introducing the new requirement set to ensure that SharedMutexType refines MutexType.]

    -1- The standard library types std::shared_mutex and std::shared_timed_mutex are shared mutex types. Shared mutex types shall meet the SharedMutexType requirements of mutex types (30.4.1.2 [thread.mutex.requirements.mutex]), and additionally shall meet the requirements set out below. In this description, m denotes an object of a shared mutex type.

    -?- The shared mutex types shall meet the MutexType requirements (30.4.1.2 [thread.mutex.requirements.mutex]).

  8. Change 30.4.1.4.1 [thread.sharedmutex.class] as indicated:

    -2- The class shared_mutex shall satisfy all of the SharedMutexType requirements for shared mutexes (30.4.1.4 [thread.sharedmutex.requirements]). It shall be a standard-layout class (Clause 9).

  9. Change 30.4.1.5 [thread.sharedtimedmutex.requirements] as indicated: [Drafting note: The reference to the timed mutex types requirements has been moved after introducing the new requirement set to ensure that SharedTimedMutexType refines TimedMutexType and SharedMutexType.]

    -1- The standard library type std::shared_timed_mutex is a shared timed mutex type. Shared timed mutex types shall meet the SharedTimedMutexType requirements of timed mutex types (30.4.1.3 [thread.timedmutex.requirements]), shared mutex types (30.4.1.4 [thread.sharedmutex.requirements]), and additionally shall meet the requirements set out below. In this description, m denotes an object of a shared timed mutex type, rel_type denotes an object of an instantiation of duration (20.12.5), and abs_time denotes an object of an instantiation of time_point (20.12.6).

    -?- The shared timed mutex types shall meet the TimedMutexType requirements (30.4.1.3 [thread.timedmutex.requirements]) and the SharedMutexType requirements (30.4.1.4 [thread.sharedmutex.requirements]).

  10. Change 30.4.1.5.1 [thread.sharedtimedmutex.class] as indicated:

    -2- The class shared_timed_mutex shall satisfy all of the SharedTimedMutexType requirements for shared timed mutexes (30.4.1.5 [thread.sharedtimedmutex.requirements]). It shall be a standard-layout class (Clause 9).

  11. Change 30.4.2.3 [thread.lock.shared] as indicated:

    -1- […] The supplied Mutex type shall meet the shared mutexSharedMutexType requirements (30.4.1.5 [thread.sharedtimedmutex.requirements]30.4.1.4 [thread.sharedmutex.requirements]).

    -2- [Note: shared_lock<Mutex> meets the TimedLockable requirements (30.2.5.4). — end note]

  12. Change 30.4.2.3.1 [thread.lock.shared.cons] as indicated:

    template <class Clock, class Duration>
      shared_lock(mutex_type& m,
                  const chrono::time_point<Clock, Duration>& abs_time);
    

    -14- Requires: The supplied Mutex type shall meet the SharedTimedMutexType requirements (30.4.1.5 [thread.sharedtimedmutex.requirements]). The calling thread does not own the mutex for any ownership mode.

    -15- Effects: Constructs an object of type shared_lock and calls m.try_lock_shared_until(abs_time).

    […]

    template <class Rep, class Period>
      shared_lock(mutex_type& m,
                  const chrono::duration<Rep, Period>& rel_time);
    

    -17- Requires: The supplied Mutex type shall meet the SharedTimedMutexType requirements (30.4.1.5 [thread.sharedtimedmutex.requirements]). The calling thread does not own the mutex for any ownership mode.

    -18- Effects: Constructs an object of type shared_lock and calls m.try_lock_shared_for(rel_time).

    […]

  13. Change 30.4.2.3.2 [thread.lock.shared.locking] as indicated:

    template <class Clock, class Duration>
      bool
      try_lock_until(const chrono::time_point<Clock, Duration>& abs_time);
    

    -?- Requires: The supplied Mutex type shall meet the SharedTimedMutexType requirements (30.4.1.5 [thread.sharedtimedmutex.requirements]).

    -8- Effects: pm->try_lock_shared_until(abs_time).

    […]

    template <class Rep, class Period>
      bool try_lock_for(const chrono::duration<Rep, Period>& rel_time);
    

    -?- Requires: The supplied Mutex type shall meet the SharedTimedMutexType requirements (30.4.1.5 [thread.sharedtimedmutex.requirements]).

    -12- Effects: pm->try_lock_shared_for(rel_time).

    […]


2366. istreambuf_iterator end-of-stream equality

Section: 24.6.3 [istreambuf.iterator] Status: New Submitter: Hyman Rosen Opened: 2014-02-19 Last modified: 2015-04-08

Priority: 3

View all other issues in [istreambuf.iterator].

View all issues with New status.

Discussion:

Given the following code,

#include <sstream>

std::stringbuf buf;
std::istreambuf_iterator<char> begin(&buf);
std::istreambuf_iterator<char> end;

it is not clear from the wording of the Standard whether begin.equal(end) must be true. In at least one implementation it is not (CC: Sun C++ 5.10 SunOS_sparc Patch 128228-25 2013/02/20) and in at least one implementation it is (gcc version 4.3.2 x86_64-unknown-linux-gnu).

24.6.3 [istreambuf.iterator] says that end is an end-of-stream iterator since it was default constructed. It also says that an iterator becomes equal to an end-of-stream iterator when end of stream is reached by sgetc() having returned eof(). 24.6.3.5 [istreambuf.iterator::equal] says that equal() returns true iff both iterators are end of stream or not end of stream. But there seems to be no requirement that equal check for end-of-stream by calling sgetc().

Jiahan Zi at BloombergLP discovered this issue through his code failing to work correctly. Dietmar Kühl has opined in a private communication that the iterators should compare equal.

Proposed resolution:


2368. Replacing global operator new

Section: 18.6.2 [new.delete] Status: Open Submitter: Stephen Clamage Opened: 2014-02-20 Last modified: 2016-03-08

Priority: 2

View all other issues in [new.delete].

View all issues with Open status.

Discussion:

Section 18.6.2 [new.delete] and subsections shows:

void* operator new(std::size_t size);
void* operator new[](std::size_t size);

That is, without exception-specifications. (Recall that C++03 specified these functions with throw(std::bad_alloc).)

Section 17.6.5.12 [res.on.exception.handling] the end of paragraph 4 says:

Any other functions defined in the C++ standard library that do not have an exception-specification may throw implementation-defined exceptions unless otherwise specified. An implementation may strengthen this implicit exception-specification by adding an explicit one.

For example, an implementation could provide C++03-compatible declarations of operator new.

Programmers are allowed to replace these operator new functions. But how can you write the definition of these functions when the exception specification can vary among implementations? For example, the declarations

void* operator new(std::size_t size) throw(std::bad_alloc);
void* operator new(std::size_t size);

are not compatible.

From what I have been able to determine, gcc has a hack for the special case of operator new to ignore the differences in (at least) the two cases I show above. But can users expect all compilers to quietly ignore the incompatibility?

The blanket permission to add any explicit exception specification could cause a problem for any user-overridable function. Different implementations could provide incompatible specifications, making portable code impossible to write.

[2016-03, Jacksonville]

STL: Core changes to remove dynamic exception specs would make this moot
Room: This is on track to be resolved by P0003, or may be moot.

Proposed resolution:


2370. Operations involving type-erased allocators should not be noexcept in std::function

Section: 20.14.12.2 [func.wrap.func] Status: Open Submitter: Pablo Halpern Opened: 2014-02-27 Last modified: 2016-01-04

Priority: 3

View other active issues in [func.wrap.func].

View all other issues in [func.wrap.func].

View all issues with Open status.

Discussion:

The following constructors in 20.14.12.2 [func.wrap.func] are declared noexcept, even though it is not possible for an implementation to guarantee that they will not throw:

template <class A> function(allocator_arg_t, const A&) noexcept;
template <class A> function(allocator_arg_t, const A&, nullptr_t) noexcept;

In addition, the following functions are guaranteed not to throw if the target is a function pointer or a reference_wrapper:

template <class A> function(allocator_arg_t, const A& a, const function& f);
template <class F, class A> function(allocator_arg_t, const A& a, F f);

In all of the above cases, the function object might need to allocate memory (an operation that can throw) in order to hold a copy of the type-erased allocator itself. The first two constructors produce an empty function object, but the allocator is still needed in case the object is later assigned to. In this case, we note that the propagation of allocators on assignment is underspecified for std::function. There are three possibilities:

  1. The allocator is never copied on copy-assignment, moved on move-assignment, or swapped on swap.

  2. The allocator is always copied on copy-assignment, moved on move-assignment, and swapped on swap.

  3. Whether or not the allocator is copied, moved, or swapped is determined at run-time based on the propagate_on_container_copy_assignment and propagate_on_container_move_assignment traits of the allocators at construction of the source function, the target function, or both.

Although the third option seems to be the most consistent with existing wording in the containers section of the standard, it is problematic in a number of respects. To begin with, the propagation behavior is determined at run time based on a pair of type-erased allocators, instead of at compile time. Such run-time logic is not consistent with the rest of the standard and is hard to reason about. Additionally, there are two allocator types involved, rather than one. Any set of rules that attempts to rationally interpret the propagation traits of both allocators is likely to be arcane at best, and subtly wrong for some set of codes at worst.

The second option is a non-starter. Historically, and in the vast majority of existing code, an allocator does not change after an object is constructed. The second option, if adopted, would undermine the programmer's ability to construct, e.g., an array of function objects, all using the same allocator.

The first option is (in Pablo's opinion) the simplest and best. It is consistent with historical use of allocators, is easy to understand, and requires minimal wording. It is also consistent with the wording in N3916, which formalizes type-erased allocators.

For cross-referencing purposes: The resolution of this issue should be harmonized with any resolution to LWG 2062, which questions the noexcept specification on the following member functions of std::function:

template <class F> function& operator=(reference_wrapper<F>) noexcept;
void swap(function&) noexcept;

[2015-05, Lenexa]

MC: change to P3 and status to open.

STL: note that noexcept is an issue and large chunks of allocator should be destroyed.

[2015-12-16, Daniel comments]

See 2564 for a corresponding issue addressing library fundamentals v2.

Proposed resolution:

This wording is relative to N3936.

  1. Change 20.14.12.2 [func.wrap.func], class template function synopsis, as indicated:

    template <class A> function(allocator_arg_t, const A&) noexcept;
    template <class A> function(allocator_arg_t, const A&, nullptr_t) noexcept;
    
  2. Change 20.14.12.2.1 [func.wrap.func.con] as indicated:

    -1- When any function constructor that takes a first argument of type allocator_arg_t is invoked, the second argument shall have a type that conforms to the requirements for Allocator (Table 17.6.3.5). A copy of the allocator argument is used to allocate memory, if necessary, for the internal data structures of the constructed function object. For the remaining constructors, an instance of allocator<T>, for some suitable type T, is used to allocate memory, if necessary, for the internal data structures of the constructed function object.

    function() noexcept;
    template <class A> function(allocator_arg_t, const A&) noexcept;
    

    -2- Postconditions: !*this.

    function(nullptr_t) noexcept;
    template <class A> function(allocator_arg_t, const A&, nullptr_t) noexcept;
    

    -3- Postconditions: !*this.

    function(const function& f);
    template <class A> function(allocator_arg_t, const A& a, const function& f);
    

    -4- Postconditions: !*this if !f; otherwise, *this targets a copy of f.target().

    -5- Throws: shall not throw exceptions if f's target is a callable object passed via reference_wrapper or a function pointer. Otherwise, may throw bad_alloc or any exception thrown by the copy constructor of the stored callable object. [Note: Implementations are encouraged to avoid the use of dynamically allocated memory for small callable objects, for example, where f's target is an object holding only a pointer or reference to an object and a member function pointer. — end note]

    template <class A> function(allocator_arg_t, const A& a, const function& f);
    

    -?- Postconditions: !*this if !f; otherwise, *this targets a copy of f.target().

    function(function&& f);
    template <class A> function(allocator_arg_t, const A& a, function&& f);
    

    -6- Effects: If !f, *this has no target; otherwise, move-constructs the target of f into the target of *this, leaving f in a valid state with an unspecified value. If an allocator is not specified, the constructed function will use the same allocator as f.

    template<class F> function(F f);
    template <class F, class A> function(allocator_arg_t, const A& a, F f);
    

    -7- Requires: F shall be CopyConstructible.

    -8- Remarks: These constructors shall not participate in overload resolution unless f is Callable (20.9.11.2) for argument types ArgTypes... and return type R.

    -9- Postconditions: !*this if any of the following hold:

    • f is a null function pointer value.

    • f is a null member pointer value.

    • F is an instance of the function class template, and !f

    -10- Otherwise, *this targets a copy of f initialized with std::move(f). [Note: Implementations are encouraged to avoid the use of dynamically allocated memory for small callable objects, for example, where f's target is an object holding only a pointer or reference to an object and a member function pointer. — end note]

    -11- Throws: shall not throw exceptions when an allocator is not specified and f is a function pointer or a reference_wrapper<T> for some T. Otherwise, may throw bad_alloc or any exception thrown by F's copy or move constructor or by A's allocate function.


2375. Is [iterator.requirements.general]/9 too broadly applied?

Section: 24.2.1 [iterator.requirements.general] Status: New Submitter: Marshall Clow Opened: 2014-03-25 Last modified: 2015-04-08

Priority: 3

View all issues with New status.

Discussion:

24.2.1 [iterator.requirements.general] p9 says:

Destruction of an iterator may invalidate pointers and references previously obtained from that iterator.

But the resolution of LWG issue 2360 specifically advocates returning *--temp; where temp is a local variable.

And 24.2.5 [forward.iterators] p6 says:

If a and b are both dereferenceable, then a == b if and only if *a and *b are bound to the same object.

which disallows "stashing" iterators (i.e, iterators that refer to data inside themselves).

So, I suspect that the restriction in p9 should only apply to input iterators, and can probably be moved into 24.2.3 [input.iterators] instead of 24.2.1 [iterator.requirements.general].

[2014-05-22, Daniel comments]

Given that forward iterators (and beyond) are refinements of input iterator, moving this constraint to input iterators won't help much because it would still hold for all refined forms.

Proposed resolution:


2381. Inconsistency in parsing floating point numbers

Section: 22.4.2.1.2 [facet.num.get.virtuals] Status: New Submitter: Marshall Clow Opened: 2014-04-30 Last modified: 2016-04-15

Priority: 2

View all other issues in [facet.num.get.virtuals].

View all issues with New status.

Discussion:

In 22.4.2.1.2 [facet.num.get.virtuals] we have:

Stage 3: The sequence of chars accumulated in stage 2 (the field) is converted to a numeric value by the rules of one of the functions declared in the header <cstdlib>:

This implies that for many cases, this routine should return true:

bool is_same(const char* p) 
{
  std::string str{p};
  double val1 = std::strtod(str.c_str(), nullptr);
  std::stringstream ss(str);
  double val2;
  ss >> val2;
  return std::isinf(val1) == std::isinf(val2) &&                 // either they're both infinity
         std::isnan(val1) == std::isnan(val2) &&                 // or they're both NaN
         (std::isinf(val1) || std::isnan(val1) || val1 == val2); // or they're equal
}

and this is indeed true, for many strings:

assert(is_same("0"));
assert(is_same("1.0"));
assert(is_same("-1.0"));
assert(is_same("100.123"));
assert(is_same("1234.456e89"));

but not for others

assert(is_same("0xABp-4")); // hex float
assert(is_same("inf"));
assert(is_same("+inf"));
assert(is_same("-inf"));
assert(is_same("nan"));
assert(is_same("+nan"));
assert(is_same("-nan"));

assert(is_same("infinity"));
assert(is_same("+infinity"));
assert(is_same("-infinity"));

These are all strings that are correctly parsed by std::strtod, but not by the stream extraction operators. They contain characters that are deemed invalid in stage 2 of parsing.

If we're going to say that we're converting by the rules of strtold, then we should accept all the things that strtold accepts.

[2016-04, Issues Telecon]

People are much more interested in round-tripping hex floats than handling inf and nan. Priority changed to P2.

Marshall says he'll try to write some wording, noting that this is a very closely specified part of the standard, and has remained unchanged for a long time. Also, there will need to be a sample implementation.

Proposed resolution:


2383. Overflow cannot be ill-formed for chrono::duration integer literals

Section: 20.17.5.8 [time.duration.literals] Status: Open Submitter: Jonathan Wakely Opened: 2014-05-16 Last modified: 2015-04-08

Priority: 3

View all issues with Open status.

Discussion:

20.17.5.8 [time.duration.literals] p3 says:

If any of these suffixes are applied to an integer literal and the resulting chrono::duration value cannot be represented in the result type because of overflow, the program is ill-formed.

Ill-formed requires a diagnostic at compile-time, but there is no way to detect the overflow from unsigned long long to the signed duration<>::rep type.

Overflow could be detected if the duration integer literals were literal operator templates, otherwise overflow can either be undefined or a run-time error, not ill-formed.

[Urbana 2014-11-07: Move to Open]

Proposed resolution:


2391. basic_string is missing non-const data()

Section: 21.3.1 [basic.string] Status: Tentatively Resolved Submitter: Michael Bradshaw Opened: 2014-05-27 Last modified: 2016-05-22

Priority: 3

View other active issues in [basic.string].

View all other issues in [basic.string].

View all issues with Tentatively Resolved status.

Discussion:

Regarding 21.3.1 [basic.string], std::basic_string<charT>::data() returns a const charT* 21.3.1.7.1 [string.accessors]. While this method is convenient, it doesn't quite match std::array<T>::data() 23.3.7.5 [array.data] or std::vector<T>::data() 23.3.11.4 [vector.data], both of which provide two versions (that return T* or const T*). An additional data() method can be added to std::basic_string that returns a charT* so it can be used in similar situations that std::array and std::vector can be used. Without a non-const data() method, std::basic_string has to be treated specially in code that is otherwise oblivious to the container type being used.

Adding a charT* return type to data() would be equivalent to doing &str[0] or &str.front().

Small discussion on the issue can be found here and in the std-discussion thread (which didn't get too much attention).

This requires a small change to std::basic_string's definition in 21.3.1 [basic.string] to add the method to std::basic_string, and another small change in 21.3.1.7.1 [string.accessors] to define the new method.

[2015-02 Cologne]

Back to LEWG.

[2016-05-22]

Marshall says: this issue has been resolved by P0272R1.

Proposed resolution:

This wording is relative to N3936.

  1. Change class template basic_string synopsis, 21.3.1 [basic.string], as indicated:

    namespace std {
      template<class charT, class traits = char_traits<charT>,
      class Allocator = allocator<charT> >
      class basic_string {
      public:
        […]
        // 21.4.7, string operations:
        const charT* c_str() const noexcept;
        const charT* data() const noexcept;
        charT* data() noexcept;
        allocator_type get_allocator() const noexcept;
        […]
      };
    }
    
  2. Add the following sequence of paragraphs following 21.3.1.7.1 [string.accessors] p3, as indicated:

    charT* data() noexcept;
    

    -?- Returns: A pointer p such that p + i == &operator[](i) for each i in [0,size()].

    -?- Complexity: Constant time.

    -?- Requires: The program shall not alter the value stored at p + size().


2392. "character type" is used but not defined

Section: 17.3.16 [defns.ntcts], 22.3.1.1.1 [locale.category], 27.2.2 [iostreams.limits.pos], 27.7.3.6.1 [ostream.formatted.reqmts], 27.7.3.6.4 [ostream.inserters.character] Status: New Submitter: Jeffrey Yasskin Opened: 2014-06-01 Last modified: 2015-09-25

Priority: 3

View all issues with New status.

Discussion:

The term "character type" is used in 17.3.16 [defns.ntcts], 22.3.1.1.1 [locale.category], 27.2.2 [iostreams.limits.pos], 27.7.3.6.1 [ostream.formatted.reqmts], and 27.7.3.6.4 [ostream.inserters.character], but the core language only defines "narrow character types" (3.9.1 [basic.fundamental]).

"wide-character type" is used in 22.5 [locale.stdcvt], but the core language only defines a "wide-character set" and "wide-character literal".

Proposed resolution:


2394. locale::name specification unclear — what is implementation-defined?

Section: 22.3.1.3 [locale.members] Status: New Submitter: Richard Smith Opened: 2014-06-09 Last modified: 2015-04-08

Priority: 3

View all other issues in [locale.members].

View all issues with New status.

Discussion:

22.3.1.3 [locale.members] p5 says:

Returns: The name of *this, if it has one; otherwise, the string "*". If *this has a name, then locale(name().c_str()) is equivalent to *this. Details of the contents of the resulting string are otherwise implementation-defined.

So… what is implementation-defined here, exactly? The first sentence completely defines the behavior of this function in all cases.

Also, the second sentence says (effectively) that all locales with the same name are equivalent: given L1 and L2 that have the same name N, they are both equivalent to locale(N), and since there is no definition of "equivalent" specific to locale, I assume it's the normal transitive equivalence property, which would imply that L1 is equivalent to L2. I'm not sure why this central fact is in the description of locale::name, nor why it's written in this roundabout way.

Proposed resolution:


2398. type_info's destructor shouldn't be required to be virtual

Section: 18.7.2 [type.info] Status: Open Submitter: Stephan T. Lavavej Opened: 2014-06-14 Last modified: 2016-06-08

Priority: 2

View all other issues in [type.info].

View all issues with Open status.

Discussion:

type_info's destructor is depicted as being virtual, which is nearly unobservable to users (since they can't construct or copy this class, they can't usefully derive from it). However, it's technically observable (via is_polymorphic and has_virtual_destructor). It also imposes real costs on implementations, requiring them to store one vptr per type_info object, when RTTI space consumption is a significant concern.

Making this implementation-defined wouldn't affect users (who can observe this only if they're specifically looking for it) and wouldn't affect implementations who need virtual here, but it would allow other implementations to drop virtual and improve their RTTI space consumption.

Richard Smith:

It's observable in a few other ways.

std::map<void*, something> m;
m[dynamic_cast<void*>(&typeid(blah))] = stuff;

... is broken by this change, because you can't dynamic_cast a non-polymorphic class type to void*.

type_info& f();
typeid(f());

... evaluates f() at runtime without this change, and might not do so with this change.

These are probably rare things, but I can imagine at least some forms of the latter being used in SFINAE tricks.

[Lenexa 2015-05-05: Move to Open]

Marshall to poll LEWG for their opinion

[June 2016]

On the reflector, STL wrote:
> We'll prototype this change and report back with data in the future.

Proposed resolution:

This wording is relative to N3936.

  1. Change 18.7.2 [type.info] as indicated:

    namespace std {
      class type_info {
      public:
        virtualsee below ~type_info();
        […]
      };
    }
    

    -1- The class type_info describes type information generated by the implementation. Objects of this class effectively store a pointer to a name for the type, and an encoded value suitable for comparing two types for equality or collating order. The names, encoding rule, and collating sequence for types are all unspecified and may differ between programs. Whether ~type_info() is virtual is implementation-defined.


2412. promise::set_value() and promise::get_future() should not race

Section: 30.6.5 [futures.promise], 30.6.9.1 [futures.task.members] Status: SG1 Submitter: Jonathan Wakely Opened: 2014-06-23 Last modified: 2015-09-25

Priority: Not Prioritized

View other active issues in [futures.promise].

View all other issues in [futures.promise].

View all issues with SG1 status.

Discussion:

The following code has a data race according to the standard:

std::promise<void> p;
std::thread t{ []{
  p.get_future().wait();
}};
p.set_value();
t.join();

The problem is that both promise::set_value() and promise::get_future() are non-const member functions which modify the same object, and we only have wording saying that the set_value() and wait() calls (i.e. calls setting and reading the shared state) are synchronized.

The calls don't actually access the same memory locations, so the standard should allow it. My suggestion is to state that calling get_future() does not conflict with calling the various functions that make the shared state ready, but clarify with a note that this does not imply any synchronization or "happens before", only being free from data races.

[2015-02 Cologne]

Handed over to SG1.

Proposed resolution:

This wording is relative to N3936.

  1. Change 30.6.5 [futures.promise] around p12 as indicated:

    future<R> get_future();
    

    -12- Returns: A future<R> object with the same shared state as *this.

    -?- Synchronization: Calls to this function do not conflict (1.10 [intro.multithread]) with calls to set_value, set_exception, set_value_at_thread_exit, or set_exception_at_thread_exit. [Note: Such calls need not be synchronized, but implementations must ensure they do not introduce data races. — end note]

    -13- Throws: future_error if *this has no shared state or if get_future has already been called on a promise with the same shared state as *this.

    -14- Error conditions: […]

  2. Change 30.6.9.1 [futures.task.members] around p13 as indicated:

    future<R> get_future();
    

    -13- Returns: A future<R> object that shares the same shared state as *this.

    -?- Synchronization: Calls to this function do not conflict (1.10 [intro.multithread]) with calls to operator() or make_ready_at_thread_exit. [Note: Such calls need not be synchronized, but implementations must ensure they do not introduce data races. — end note]

    -14- Throws: a future_error object if an error occurs.

    -15- Error conditions: […]


2413. assert macro is overconstrained

Section: 19.3 [assertions] Status: New Submitter: David Krauss Opened: 2014-06-25 Last modified: 2015-04-08

Priority: 4

View all other issues in [assertions].

View all issues with New status.

Discussion:

When NDEBUG is defined, assert must expand exactly to the token sequence ((void)0), with no whitespace (C99 §7.2/1 and also C11 §7.2/1). This is a lost opportunity to pass the condition along to the optimizer.

The user may observe the token sequence using the stringize operator or discriminate it by making a matching #define directive. There is little chance of practical code doing such things. It's reasonable to allow any expansion that is a void expression with no side effects or semantic requirements, for example, an extension keyword or an attribute-specifier finagled into the context.

Conforming optimizations would still be limited to treating the condition as hint, not a requirement. Nonconformance on this point is quite reasonable though, given user preferences. Anyway, it shouldn't depend on preprocessor quirks.

As for current practice, Darwin OS <assert.h> provides a GCC-style compiler hint __builtin_expect but only in debug mode. Shouldn't release mode preserve hints?

Daniel:

The corresponding resolution should take care not to conflict with the intention behind LWG 2234.

Proposed resolution:


2414. Member function reentrancy should be implementation-defined

Section: 17.6.5.8 [reentrancy] Status: Open Submitter: Stephan T. Lavavej Opened: 2014-07-01 Last modified: 2015-09-27

Priority: 3

View all other issues in [reentrancy].

View all issues with Open status.

Discussion:

N3936 17.6.5.8 [reentrancy]/1 talks about "functions", but that doesn't address the scenario of calling different member functions of a single object. Member functions often have to violate and then re-establish invariants. For example, vectors often have "holes" during insertion, and element constructors/destructors/etc. shouldn't be allowed to observe the vector while it's in this invariant-violating state. The [reentrancy] Standardese should be extended to cover member functions, so that implementers can either say that member function reentrancy is universally prohibited, or selectively allowed for very specific scenarios.

(For clarity, this issue has been split off from LWG 2382.)

[2014-11-03 Urbana]

AJM confirmed with SG1 that they had no special concerns with this issue, and LWG should retain ownership.

AM: this is too overly broad as it also covers calling the exact same member function on a different object
STL: so you insert into a map, and copying the value triggers another insertion into a different map of the same type
GR: reentrancy seems to imply the single-threaded case, but needs to consider the multi-threaded case

Needs more wording.

Move to Open

[2015-07 Telecom Urbana]

Marshall to ping STL for updated wording.

Proposed resolution:

This wording is relative to N3936.

  1. Change 17.6.5.8 [reentrancy] p1 as indicated:

    -1- Except where explicitly specified in this standard, it is implementation-defined which functions (including different member functions called on a single object) in the Standard C++ library may be recursively reentered.


2417. [fund.ts.v2] std::experimental::optional::operator< and LessThanComparable requirement

Section: 99 [fund.ts.v2::optional.relops], 99 [fund.ts.v2::optional.comp_with_t] Status: LEWG Submitter: Daniel Krügler Opened: 2014-06-20 Last modified: 2015-10-27

Priority: Not Prioritized

View all issues with LEWG status.

Discussion:

Addresses: fund.ts.v2

Currently, std::experimental::optional::operator== imposes the EqualityComparable requirement which provides two guarantees: It ensures that operator!= can rely on the equivalence-relation property and more importantly, that the BooleanTestable requirements suggested by issue 2114 are automatically implied.

std::experimental::optional::operator< doesn't provide a LessThanComparable requirement, but there was quite an historic set of changes involved with that family of types: As of N3527 this operator was defined in terms of operator< of the contained type T and imposed the LessThanComparable requirement. In the final acceptance step of optional by the committee, the definition was expressed in terms of std::less and the LessThanComparable requirement had been removed.

The inconsistency between operator== and operator< should be removed. One possible course of action would be to add the LessThanComparable to std::experimental::optional::operator<. The EqualityComparable requirement of operator== could also be removed, but in that case both operators would at least need to require the BooleanTestable requirements (see 2114) for the result type of T's operator== and operator<.

Arguably, corresponding operators for pair and tuple do not impose LessThanComparable (nor EqualityComparable), albeit the definition of the "derived" relation functions depend on properties ensured by LessThanComparable. According to the SGI definition, the intention was to imposed both EqualityComparable and LessThanComparable. If this is not intended, the standard should clarify this position.

[2015-02 Cologne]

VV, DK, JY discuss why and when LessThanComparable was removed. AM: Move to LEWG. Please tell LWG when you look at it.

Proposed resolution:


2419. Clang's libc++ extension to std::tuple

Section: 20.5.2.1 [tuple.cnstr] Status: LEWG Submitter: Akim Demaille Opened: 2014-07-11 Last modified: 2015-04-08

Priority: Not Prioritized

View other active issues in [tuple.cnstr].

View all other issues in [tuple.cnstr].

View all issues with LEWG status.

Discussion:

The issue has been submitted after exchanges with the clang++ team as a consequence of two PR I sent:

Issue 20174

Issue 20175

The short version is shown in the program below:

#include <iostream>
#include <tuple>

struct base
{
  void out(const std::tuple<char, char>& w) const
  {
    std::cerr << "Tuple: " << std::get<0>(w) << std::get<1>(w) << '\n';
  }
};

struct decorator
{
  base b_;

  template <typename... Args>
  auto
  out(Args&&... args)
    -> decltype(b_.out(args...))
  {
    return b_.out(args...);
  }

  void out(const char& w)
  {
    std::cerr << "char: " << w << '\n';
  }
};

int main()
{
  decorator d{base{}};
  char l = 'a';
  d.out(l);
}

This is a stripped down version of a real world case where I wrap objects in decorators. These decorators contributes some functions, and forward all the rest of the API to the wrapped object using perfect forwarding. There can be overloaded names.

Here the inner object provides an

out(const std::tuple<char, char>&) -> void

function, and the wrappers, in addition to perfect forwarding, provides

out(const char&) -> void

The main function then call out(l) where l is a char lvalue.

With (GCC's) libstdc++ I get the expected result: the char overload is run. With (clang++'s) libc++ it is the tuple version which is run.

$ g++-mp-4.9 -std=c++11 bar.cc && ./a.out
char: a
$ clang++-mp-3.5 -std=c++11 bar.cc -Wall && ./a.out
Tuple: a

It turns out that this is the result of an extension of std::tuple in libc++ where they accept constructors with fewer values that tuple elements.

The purpose of this issue is to ask the standard to forbid that this extension be allowed to participate in overload resolution.

[2014-10-05, Daniel comments]

This issue is closely related to LWG 2312.

[2014-11 Urbana]

Moved to LEWG.

Extensions to tuple's design are initially a question for LEWG.

Proposed resolution:


2421. Non-specification of handling zero size in std::align [ptr.align]

Section: 20.10.5 [ptr.align] Status: New Submitter: Melissa Mears Opened: 2014-08-06 Last modified: 2015-04-08

Priority: 3

View all other issues in [ptr.align].

View all issues with New status.

Discussion:

The specification of std::align does not appear to specify what happens when the value of the size parameter is 0. (The question of what happens when alignment is 0 is mentioned in another Defect Report, 2377; it would change the behavior to be undefined rather than potentially implementation-defined.)

The case of size being 0 is interesting because the result is ambiguous. Consider the following code's output:

#include <cstdio>
#include <memory>

int main()
{
  alignas(8) char buffer[8];
  void *ptr = &buffer[1];
  std::size_t space = sizeof(buffer) - sizeof(char[1]);

  void *result = std::align(8, 0, ptr, space);

  std::printf("%d %td\n", !!result, result ? (static_cast<char*>(result) - buffer) : std::ptrdiff_t(-1));
}

There are four straightforward answers as to what the behavior of std::align with size 0 should be:

  1. The behavior is undefined because the size is invalid.

  2. The behavior is implementation-defined. This seems to be the status quo, with current implementations using #3.

  3. Act the same as size == 1, except that if size == 1 would fail but would be defined and succeed if space were exactly 1 larger, the result is a pointer to the byte past the end of the ptr buffer. That is, the "aligned" version of a 0-byte object can be one past the end of an allocation. Such pointers are, of course, valid when not dereferenced (and a "0-byte object" shouldn't be), but whether that is desired is not specified in the Standard's definition of std::align, it appears. The output of the code sample is "1 8" in this case.

  4. Act the same as size == 1; this means that returning "one past the end" is not a possible result. In this case, the code sample's output is "0 -1".

The two compilers I could get working with std::align, Visual Studio 2013 and Clang 3.4, implement #3. (Change %td to %Id on Visual Studio 2013 and earlier. 2014 and later will have %td.)

Proposed resolution:


2423. Missing specification slice_array, gslice_array, mask_array, indirect_array copy constructor

Section: 26.7.5 [template.slice.array], 26.7.7 [template.gslice.array], 26.7.8 [template.mask.array], 26.7.9 [template.indirect.array] Status: New Submitter: Akira Takahashi Opened: 2014-08-12 Last modified: 2015-04-08

Priority: 4

View all other issues in [template.slice.array].

View all issues with New status.

Discussion:

I found a missing specification of the copy constructor of the following class templates:

Proposed resolution:

  1. Before 26.7.5.2 [slice.arr.assign] insert a new sub-clause as indicated:

    -?- slice_array constructors [slice.arr.cons]

    slice_array(const slice_array&);
    

    -?- Effects: The constructed slice refers to the same valarray<T> object to which the argument slice refers.

  2. Before 26.7.7.2 [gslice.array.assign] insert a new sub-clause as indicated:

    -?- gslice_array constructors [gslice.array.cons]

    gslice_array(const gslice_array&);
    

    -?- Effects: The constructed slice refers to the same valarray<T> object to which the argument slice refers.

  3. Before 26.7.8.2 [mask.array.assign] insert a new sub-clause as indicated:

    -?- mask_array constructors [mask.array.cons]

    mask_array(const mask_array&);
    

    -?- Effects: The constructed slice refers to the same valarray<T> object to which the argument slice refers.

  4. Before 26.7.9.2 [indirect.array.assign] insert a new sub-clause as indicated:

    -?- indirect_array constructors [indirect.array.cons]

    indirect_array(const indirect_array&);
    

    -?- Effects: The constructed slice refers to the same valarray<T> object to which the argument slice refers.


2424. 29.5 should state that atomic types are not trivially copyable

Section: 29.5 [atomics.types.generic] Status: Open Submitter: Jens Maurer Opened: 2014-08-14 Last modified: 2016-03-06

Priority: 2

View other active issues in [atomics.types.generic].

View all other issues in [atomics.types.generic].

View all issues with Open status.

Discussion:

Otherwise, one could use memcpy to save and restore the value according to 3.9p2.

It seems the core language rules in 9 [class]p6 with 12.8 [class.copy]p12 (trivial copy constructor) etc. and 8.4.2 [dcl.fct.def.default]p5 (user-provided) say that the atomic types are trivially copyable, which is bad. We shouldn't rely on future core changes in that area and simply say in the library section 29.5 [atomics.types.generic] that these very special types are not trivially copyable.

[2014-11 Urbana]

Lawrence:Definition of "trivially copyable" has been changing.

Doesn't hurt to add proposed change, even if the sentence is redundant

Move to Review.

[2015-02 Cologne]

GR has a minor problem with the style of the wording. VV has major issues with implementability.

[2015-03-22, Jens Maurer responses to Cologne discussion]

A library implementation could provide a partial specialization for is_trivially_copyable<atomic<T>>, to ensure that any such type query would return false.

Assuming such a specialization would be provided, how could a conforming program observe that per language rules an atomic specialization would actually be trivially copyable if there is no way to call the (deleted) copy constructor or copy assignment operator?

The sole effect of the suggested addition of the constraining sentence is that it would make a user program non-conforming that attempts to invoke memcpy (and the like) on atomic types, since that would invoke undefined behaviour.

[2015-05 Lenexa, SG1 response]

SG1 is fine with P/R (and agrees it's needed), but LWG may want to check the details; it's not entirely an SG1 issue.

[2015-05-05 Lenexa]

Marshall: This was discussed on the telecon. Alisdair was going to write something to Mike and send it to Core.

Hwrd: Core says that deleted copies are trivially copyable, which makes no sense to Library people.

STL: There doesn't appear to be a Core issue about it.

[2015-09-11 Telecom]

Howard: currently std::is_trivially_copyable<std::atomic> is true, so this resolution would contradict reality

Jonathan: changing that is good, we don't want it to be trivially copyable, otherwise users can memcpy them, which we really don't want

Howard: is it reasonable to memcpy something that isn't trivially copy constructible or trivially assignable?

Jonathan: no, it's not, but Core says you can, so this resolution is needed to stop people memcpying atomic

Howard: we should fix the core rule

Marshall: there is a separate issue of whether trivially-copyable makes sense, but this resolution is a net good purely because it stops memcpy of atomics

Howard: so should implementations specialize is_trivially_copyable the trait to meet this?

Jonathan: or add an empty, user-defined destructor.

Howard: should the spec specify that then?

Over-specification.

Howard: without that I fear implementation divergence.

Ville and Jonathan to investigate potential implementation options.

Ville: request a note on the issue saying we need review other types such as condition_variable to see if they are also unintentionally trivially-copyable. N4460 mentions some such types.

[2016-03 Jacksonville]

We think there is something coming from Core to resolve that, and that this will be NAD.
Until then, defer.

Proposed resolution:

  1. Change 29.5 [atomics.types.generic]p3 as indicated:

    Specializations and instantiations of the atomic template shall have a deleted copy constructor, a deleted copy assignment operator, and a constexpr value constructor. They are not trivially copyable types (3.9 [basic.types]).


2430. Heterogeneous container lookup should be enabled using meta-function instead of nested type

Section: 23.2.4 [associative.reqmts] Status: LEWG Submitter: Tomasz Kamiński Opened: 2014-07-14 Last modified: 2015-04-08

Priority: Not Prioritized

View other active issues in [associative.reqmts].

View all other issues in [associative.reqmts].

View all issues with LEWG status.

Discussion:

Currently the heterogeneous lookup in associative container are enabled by presence of is_transparent nested type in the comparator type (23.2.4 [associative.reqmts]). This complicates the definition of call wrapper types that want to define is_transparent if they wrap a callable type that defines is_transparent, and requires the target to be a complete type in cases where an incomplete type would otherwise be ok.

Another problem is that users cannot add the is_transparent member to a third-party comparison type that they do not control, even if they know it supports heterogeneous comparisons.

If the associative containers used a trait instead of checking for an is_transparent member type then it would avoid the requirement for complete types, and would allow customization of the trait without modifying the comparator type. This would also be consistent with the traits is_placeholder and is_bind_expression.

For backward compatibility with the existing design, the default implementation of the is_transparent trait could depend on the presence of the is_transparent nested type.

[2014-11 Urbana]

Move to LEWG

Request for a new metafunction should first be responded to by LEWG.

Proposed resolution:


2431. Missing regular expression traits requirements

Section: 28.3 [re.req] Status: New Submitter: Jonathan Wakely Opened: 2014-09-30 Last modified: 2015-04-08

Priority: 3

View all issues with New status.

Discussion:

The requirements on the traits class in 28.3 [re.req] do not say whether a regular expression traits class is required to be DefaultConstructible, CopyConstructible, CopyAssignable etc.

The std::regex_traits class appears to be all of the above, but can basic_regex assume that for user-defined traits classes?

Should the following statements all leave u in equivalent states?

X u{v};
X u; u = v;
X u; u.imbue(v.getloc();

Whether they are equivalent has implications for basic_regex copy construction and assignment.

Proposed resolution:


2432. initializer_list assignability

Section: 18.9 [support.initlist] Status: EWG Submitter: David Krauss Opened: 2014-09-30 Last modified: 2015-05-22

Priority: 2

View other active issues in [support.initlist].

View all other issues in [support.initlist].

View all issues with EWG status.

Discussion:

std::initializer_list::operator= 18.9 [support.initlist] is horribly broken and it needs deprecation:

std::initializer_list<foo> a = {{1}, {2}, {3}};
a = {{4}, {5}, {6}};
// New sequence is already destroyed.

Assignability of initializer_list isn't explicitly specified, but most implementations supply a default assignment operator. I'm not sure what 17.5 [description] says, but it probably doesn't matter.

[Lenexa 2015-05-05: Send to EWG as discussed in Telecon]

Proposed resolution:

  1. Edit 18.9 [support.initlist] p1, class template initializer_list synopsis, as indicated:

    namespace std {
      template<class E> class initializer_list {
      public:
        […]
        constexpr initializer_list() noexcept;
      
        initializer_list(const initializer_list&) = default;
        initializer_list(initializer_list&&) = default;
        initializer_list& operator=(const initializer_list&) = delete;
        initializer_list& operator=(initializer_list&&) = delete;
        
        constexpr size_t size() const noexcept;
        […]
      };
      […]
    }
    

2443. std::array member functions should be constexpr

Section: 23.3.7 [array] Status: LEWG Submitter: Peter Sommerlad Opened: 2014-10-06 Last modified: 2015-04-08

Priority: Not Prioritized

View other active issues in [array].

View all other issues in [array].

View all issues with LEWG status.

Discussion:

When experimenting with C++14 relaxed constexpr functions I made the observation that I couldn't use std::array to create a table of data at compile time directly using loops in a function. However, a simple substitute I could use instead:

template <typename T, size_t n>
struct ar {
  T a[n];
  constexpr ar() : a{{}}{}
  constexpr auto data() const { return &a[0];}
  constexpr T const & operator[](size_t i) const { return a[i]; }
  constexpr T & operator[](size_t i) { return a[i]; }
};

template <size_t n>
using arr = ar<size_t, n>; // std::array<size_t, n>;

template <size_t n>
constexpr auto make_tab(){
  arr<n> result;
  for(size_t i=0; i < n; ++i)
    result[i] = (i+1)*(i+1); // cannot define operator[] for mutable array...
  return result;
}

template <size_t n>
constexpr auto squares=make_tab< n>();

int main() {
  int dummy[squares<5>[3]];
}

Therefore, I suggest that all member functions of std::array should be made constexpr to make the type usable in constexpr functions.

Wording should be straight forward, may be with the exception of fill, which would require fill_n to be constexpr as well.

[2014-11 Urbana]

Move to LEWG

The extent to which constexpr becomes a part of the Library design is a policy matter best handled initially by LEWG.

Proposed resolution:


2444. Inconsistent complexity for std::sort_heap

Section: 25.5.6.4 [sort.heap] Status: Open Submitter: François Dumont Opened: 2014-10-07 Last modified: 2015-05-22

Priority: 3

View all issues with Open status.

Discussion:

While creating complexity tests for the GNU libstdc++ implementation I stumbled across a surprising requirement for the std::sort_heap algorithm.

In 25.5.6.4 [sort.heap] p3 the Standard states:

Complexity: At most N log(N) comparisons (where N == last - first).

As stated on the libstdc++ mailing list by Marc Glisse sort_heap can be implemented by N calls to pop_heap. As max number of comparisons of pop_heap is 2 * log(N) then sort_heap max limit should be 2 * log(1) + 2 * log(2) + .... + 2 * log(N) that is to say 2 * log(N!). In terms of log(N) we can also consider that this limit is also cap by 2 * N * log(N) which is surely what the Standard wanted to set as a limit.

This is why I would like to propose to replace paragraph 3 by:

Complexity: At most 2N log(N) comparisons (where N == last - first).

[2015-02 Cologne]

Marshall will research the maths and report back in Lenexa.

[2015-05-06 Lenexa]

STL: I dislike exact complexity requirements, they prevent one or two extra checks in debug mode. Would it be better to say O(N log(N)) not at most?

Proposed resolution:

This wording is relative to N3936.

  1. In 25.5.6.4 [sort.heap] p3 the Standard states:

    template<class RandomAccessIterator>
      void sort_heap(RandomAccessIterator first, RandomAccessIterator last);
    template<class RandomAccessIterator, class Compare>
      void sort_heap(RandomAccessIterator first, RandomAccessIterator last,
                     Compare comp);
    

    […]

    -3- Complexity: At most 2N log(N) comparisons (where N == last - first).


2445. "Stronger" memory ordering

Section: 20.11.2.6 [util.smartptr.shared.atomic], 29.6.5 [atomics.types.operations.req] Status: SG1 Submitter: JF Bastien Opened: 2014-10-08 Last modified: 2015-04-08

Priority: Not Prioritized

View all other issues in [util.smartptr.shared.atomic].

View all issues with SG1 status.

Discussion:

The definitions of compare and exchange in 20.11.2.6 [util.smartptr.shared.atomic] p32 and 29.6.5 [atomics.types.operations.req] p20 state:

Requires: The failure argument shall not be memory_order_release nor memory_order_acq_rel. The failure argument shall be no stronger than the success argument.

The term "stronger" isn't defined by the standard.

It is hinted at by 29.6.5 [atomics.types.operations.req] p21:

When only one memory_order argument is supplied, the value of success is order, and the value of failure is order except that a value of memory_order_acq_rel shall be replaced by the value memory_order_acquire and a value of memory_order_release shall be replaced by the value memory_order_relaxed.

Should the standard define a partial ordering for memory orders, where consume and acquire are incomparable with release?

[2014-11 Urbana]

Move to SG1.

Proposed resolution:


2446. Unspecialized std::tuple_size should be defined

Section: 20.5.1 [tuple.general] Status: LEWG Submitter: Nevin Liber Opened: 2014-10-10 Last modified: 2015-04-08

Priority: Not Prioritized

View all issues with LEWG status.

Discussion:

In 20.5.1 [tuple.general] paragraph 2, the unspecialized std::tuple_size is undefined. It would be a lot more useful with SFINAE if it were defined as an empty struct; that way, it can be used with enable_if for determining whether or not it is valid to use tuple_size, tuple_element and get on the corresponding data structure.

[2014-11 Urbana]

Moved to LEWG 42.

This request goes beyond simply making an API respond well to SFINAE, but coupling that with an implication for other tuple APIs. The proper place for such design discussions is LEWG.

Proposed resolution:

This wording is relative to N3936.

  1. Change 20.5.1 [tuple.general] p2, header <tuple> synopsis, as indicated

    […]
    // 20.4.2.5, tuple helper classes:
    template <class T> class tuple_size; // undefined
    […]
    
  2. Change 20.5.2.6 [tuple.helper] as indicated

    […]
    template <class T> struct tuple_size { };
    […]
    

2449. vector::insert invalidates end()?

Section: 23.3.11.5 [vector.modifiers] Status: New Submitter: Marc Glisse Opened: 2014-10-21 Last modified: 2015-04-08

Priority: 3

View other active issues in [vector.modifiers].

View all other issues in [vector.modifiers].

View all issues with New status.

Discussion:

this issue is based on the discussion here.

23.3.11.5 [vector.modifiers] says about vector::insert: "If no reallocation happens, all the iterators and references before the insertion point remain valid." This doesn't seem to guarantee anything about the iterator at the point of insertion.

The question comes from people asking if the following is valid, assuming a sufficient call to reserve() was done first:

v.insert(v.end(), v.begin(), v.end());

It could fail for an implementation using a sentinel for the end of the vector, but I don't know of any (it would be quite inconvenient). And for any implementation using a simple position as iterator (pointer (possibly in a wrapper), or base+offset), this is needlessly restrictive. The fact that this alternative:

v.insert(v.end(), &v[0], &v[0]+v.size())

is arguably valid (again assuming a large enough reserve()) makes it a bit confusing that the first version isn't (23.2.3 [sequence.reqmts] has a precondition that iterator arguments to insert() do not point into the sequence, but vector::insert is more refined and seems to give enough guarantees that it cannot fail).

Then we might as well say that vector iterators act as positions, and that after a reallocation-free operation an iterator points to the same position, whatever may be there now…

Proposed resolution:


2452. is_constructible, etc. and default arguments

Section: 20.15 [meta] Status: Core Submitter: Hubert Tong Opened: 2014-11-04 Last modified: 2015-10-20

Priority: 3

View other active issues in [meta].

View all other issues in [meta].

Discussion:

The BaseCharacteristic for is_constructible is defined in terms of the well-formedness of a declaration for an invented variable. The well-formedness of the described declaration itself may change for the same set of arguments because of the introduction of default arguments.

In the following program, there appears to be conflicting definitions of a specialization of std::is_constructible; however, it seems that this situation is caused without a user violation of the library requirements or the ODR. There is a similar issue with is_convertible, result_of and others.

a.cc:

#include <type_traits>
struct A { A(int, int); };
const std::false_type& x1 = std::is_constructible<A, int>();

int main() { }

b.cc:

#include <type_traits>
struct A { A(int, int); };

inline A::A(int, int = 0) { }

const std::true_type& x2 = std::is_constructible<A, int>();

Presumably this program should invoke undefined behaviour, but the Library specification doesn't say that.

[2015-02 Cologne]

Core wording should say "this kind of thing is ill-formed, no diagnostic required"

Proposed resolution:


2453. §[iterator.range] and now [iterator.container] aren't available via <initializer_list>

Section: 18.9 [support.initlist], 24.7 [iterator.range], 24.8 [iterator.container] Status: New Submitter: Richard Smith Opened: 2014-11-11 Last modified: 2016-02-12

Priority: 3

View other active issues in [support.initlist].

View all other issues in [support.initlist].

View all issues with New status.

Discussion:

These sections define helper functions, some of which apply to initializer_list<T>. And they're available if you include one of a long list of header files, many of which include <initializer_list>. But they are not available if you include <initializer_list>. This seems very odd.

#include <initializer_list>
auto x = {1, 2, 3};
const int *p = data(x); // error, undeclared
#include <vector>
const int *q = data(x); // ok

Proposed resolution:


2457. std::begin() and std::end() do not support multi-dimensional arrays correctly

Section: 24.7 [iterator.range] Status: New Submitter: Janez Žemva Opened: 2014-11-16 Last modified: 2015-04-08

Priority: 3

View all other issues in [iterator.range].

View all issues with New status.

Discussion:

The following code:

#include <algorithm>
#include <iterator>
#include <iostream>
#include <cassert>

int main() 
{
  int a[2][3][4] = { { { 1,  2,  3,  4}, { 5,  6,  7,  8}, { 9, 10, 11, 12} },
                     { {13, 14, 15, 16}, {17, 18, 19, 20}, {21, 22, 23, 24} } };
  int b[2][3][4];

  assert(std::distance(std::begin(a), std::end(a)) == 2 * 3 * 4);
  std::copy(std::begin(a), std::end(a), std::begin(b));
  std::copy(std::begin(b), std::end(b), std::ostream_iterator<int>(std::cout, ","));
}

does not compile.

A possible way to remedy this would be to add the following overloads of begin, end, rbegin, and rend to 24.7 [iterator.range], relying on recursive evaluation:

namespace std {

  template <typename T, size_t M, size_t N>
  constexpr remove_all_extents_t<T>*
  begin(T (&array)[M][N])
  {
    return begin(*array);
  }
  
  template <typename T, size_t M, size_t N>
  constexpr remove_all_extents_t<T>*
  end(T (&array)[M][N])
  {
    return end(array[M - 1]);
  }

  template <typename T, size_t M, size_t N>
  reverse_iterator<remove_all_extents_t<T>*>
  rbegin(T (&array)[M][N])
  {
    return decltype(rbegin(array))(end(array[M - 1]));
  }
  
  template <typename T, size_t M, size_t N>
  reverse_iterator<remove_all_extents_t<T>*>
  rend(T (&array)[M][N])
  {
    return decltype(rend(array))(begin(*array));
  }

}

Proposed resolution:


2460. LWG issue 2408 and value categories

Section: 20.15.7.6 [meta.trans.other], 24.4.1 [iterator.traits] Status: New Submitter: Richard Smith Opened: 2014-11-19 Last modified: 2015-04-08

Priority: 2

View other active issues in [meta.trans.other].

View all other issues in [meta.trans.other].

View all issues with New status.

Discussion:

LWG issue 2408 changes the meat of the specification of common_type to compute:

[..] the type, if any, of an unevaluated conditional expression (5.16) whose first operand is an arbitrary value of type bool, whose second operand is an xvalue of type T1, and whose third operand is an xvalue of type T2.

This has an effect on the specification that I think was unintended. It used to be the case that common_type<T&, U&&> would consider the type of a conditional between an lvalue of type T and an xvalue of type U. It's now either invalid (because there is no such thing as an xvalue of reference type) or considers the type of a conditional between an xvalue of type T and an xvalue of type U, depending on how you choose to read it.

Put another way, this has the effect of changing the usual definition from:

typedef decay_t<decltype(true ? declval<T>() : declval<U>())> type;

to:

typedef decay_t<decltype(true ? declval<remove_reference_t<T>>() : declval<remove_reference_t<U>>())> type;

It also makes common_type underspecified in the case where one of the operands is of type void; in that case, the resulting type depends on whether the expression is a throw-expression, which is not specified (but used to be).

Also on the subject of this wording: the changes to 24.4.1 [iterator.traits] say that iterator_traits<T> "shall have no members" in some cases. That's wrong. It's a class type; it always has at least a copy constructor, a copy assignment operator, and a destructor. Plus this removes the usual library liberty to add additional members with names that don't collide with normal usage (for instance, if a later version of the standard adds members, they can't be present here as a conforming extension). Perhaps this should instead require that the class doesn't have members with any of those five names? That's what 2408 does for common_type's type member.

Proposed resolution:


2461. Interaction between allocators and container exception safety guarantees

Section: 17.6.3.5 [allocator.requirements], 23.3.11.3 [vector.capacity], 23.3.11.5 [vector.modifiers] Status: New Submitter: dyp Opened: 2014-12-06 Last modified: 2015-06-12

Priority: 3

View other active issues in [allocator.requirements].

View all other issues in [allocator.requirements].

View all issues with New status.

Discussion:

When resizing a vector, the accessibility and exception specification of the value type's constructors determines whether the elements are copied or moved to the new buffer. However, the copy/move is performed via the allocator's construct member function, which is assumed, but not required, to call the copy/move constructor and propagate only exceptions from the value type's copy/move constructor. The issue might also affect other classes.

The current wording in N4296 relevant here is from Table 28 — "Allocator requirements" in 17.6.3.5 [allocator.requirements]:

Table 28 — Allocator requirements
Expression Return type Assertion/note
pre-/post-condition
Default
a.construct(c, args) (not used) Effect: Constructs an object of type C at c ::new ((void*)c) C(forward<Args>(args)...)

and from 17.6.3.5 [allocator.requirements] p9:

An allocator may constrain the types on which it can be instantiated and the arguments for which its construct member may be called. If a type cannot be used with a particular allocator, the allocator class or the call to construct may fail to instantiate.

I conclude the following from the wording:

  1. The allocator is not required to call the copy constructor if the arguments (args) is a single (potentially const) lvalue of the value type. Similarly for a non-const rvalue + move constructor. See also 23.2.1 [container.requirements.general] p15 which seems to try to require this, but is not sufficient: That paragraph specifies the semantics of the allocator's operations, but not which constructors of the value type are used, if any.

  2. The allocator may throw exceptions in addition to the exceptions propagated by the constructors of the value type; it can also propagate exceptions from constructors other than a copy/move constructor.

This leads to an issue with the wording of the exception safety guarantees for vector modifiers in 23.3.11.5 [vector.modifiers] p1:

[…]

void push_back(const T& x);
void push_back(T&& x);

Remarks: Causes reallocation if the new size is greater than the old capacity. If no reallocation happens, all the iterators and references before the insertion point remain valid. If an exception is thrown other than by the copy constructor, move constructor, assignment operator, or move assignment operator of T or by any InputIterator operation there are no effects. If an exception is thrown while inserting a single element at the end and T is CopyInsertable or is_nothrow_move_constructible<T>::value is true, there are no effects. Otherwise, if an exception is thrown by the move constructor of a non-CopyInsertable T, the effects are unspecified.

The wording leads to the following problem: Copy and move assignment are invoked directly from vector. For intermediary objects (see 2164), vector also directly invokes the copy and move constructor of the value type. However, construction of the actual element within the buffer is invoked via the allocator abstraction. As discussed above, the allocator currently is not required to call a copy/move constructor. If is_nothrow_move_constructible<T>::value is true for some value type T, but the allocator uses modifying operations for MoveInsertion that do throw, the implementation is required to ensure that "there are no effects", even if the source buffer has been modified.

Similarly, the vector capacity functions specify exception safety guarantees referring to the move constructor of the value type. For example, vector::resize in 23.3.11.3 [vector.capacity] p14:

Remarks: If an exception is thrown other than by the move constructor of a non-CopyInsertable T there are no effects.

The wording leads to the same issue as described above.

Code example:

template<class T>
class allocator;

class pot_reg_type // a type which creates
                   // potentially registered instances
{
private:
  friend class allocator<pot_reg_type>;
  struct register_t {};

  static std::set<pot_reg_type*>& get_registry()
  {
    static std::set<pot_reg_type*> registry;
    return registry;
  }
  void enregister() noexcept(false)
  {
    get_registry().insert(this);
  }
  void deregister()
  {
    get_registry().erase(this);
  }

public:
  pot_reg_type(void               ) noexcept(true) {}
  pot_reg_type(pot_reg_type const&) noexcept(true) {}
  pot_reg_type(pot_reg_type&&     ) noexcept(true) {}

private:
  pot_reg_type(register_t                     ) noexcept(false)
  { enregister(); }
  pot_reg_type(register_t, pot_reg_type const&) noexcept(false)
  { enregister(); }
  pot_reg_type(register_t, pot_reg_type&&     ) noexcept(false)
  { enregister(); }
};

template<class T>
class allocator
{
public:
  using value_type = T;

  value_type* allocate(std::size_t p)
  { return (value_type*) ::operator new(p); }

  void deallocate(value_type* p, std::size_t)
  { ::operator delete(p); }

  void construct(pot_reg_type* pos)
  {
    new((void*)pos) pot_reg_type((pot_reg_type::register_t()));
  }
  void construct(pot_reg_type* pos, pot_reg_type const& source)
  {
    new((void*)pos) pot_reg_type(pot_reg_type::register_t(), source);
  }

  template<class... Args>
  void construct(T* p, Args&&... args)
  {
    new((void*)p) T(std::forward<Args>(args)...);
  }
}; 

The construct member function template is only required for rebinding, which can be required e.g. to store additional debug information in the allocated memory (e.g. VS2013).

Even though the value type has an accessible and noexcept(true) move constructor, this allocator won't call that constructor for rvalue arguments. In any case, it does not call a constructor for which vector has formulated its requirements. An exception thrown by a constructor called by this allocator is not covered by the specification in 23.3.11.5 [vector.modifiers] and therefore is guaranteed not to have any effect on the vector object when resizing.

For an example how this might invalidate the exception safety guarantees, see this post on the std-discussion mailing list.

Another problem arises for value types whose constructors are private, but may be called by the allocator e.g. via friendship. Those value types are not MoveConstructible (is_move_constructible is false), yet they can be MoveInsertable. It is not possible for vector to create intermediary objects (see 2164) of such a type by directly using the move constructor. Current implementations of the single-element forms of vector::insert and vector::emplace do create intermediary objects by directly calling one of the value type's constructors, probably to allow inserting objects from references that alias other elements of the container. As far as I can see, Table 100 — "Sequence container requirements" in 23.2.3 [sequence.reqmts] does not require that the creation of such intermediare objects can be performed by containers using the value type's constructor directly. It is unclear to me if the allocator's construct function could be used to create those intermediary objects, given that they have not been allocated by the allocator.

Two possible solutions:

  1. Add the following requirement to the allocator_traits::construct function: If the parameter pack args consists of a single parameter of the type value_type&&, the function may only propagate exceptions if is_nothrow_move_constructible<value_type>::value is false.

    Requiring alloctor_traits::construct to call a true copy/move constructor of the value type breaks std::scoped_allocator_adapter, as pointed out by Casey Carter in a post on the std-discussion mailing list.

  2. Change vector's criterion whether to move or copy when resizing:

    Instead of testing the value type's constructors via is_move_constructible, check the value of noexcept( allocator_traits<Allocator>::construct(alloc, ptr, rval) ) where alloc is an lvalue of type Allocator, ptr is an expression of type allocator_traits<Allocator>::pointer and rval is a non-const rvalue of type value_type.

A short discussion of the two solutions:

Solution 1 allows keeping is_nothrow_move_constructible<value_type> as the criterion for vector to decide between copying and moving when resizing. It restricts what can be done inside the construct member function of allocators, and requires implementers of allocators to pay attention to the value types used. One could conceive allocators checking the following with a static_assert: If the value type is_nothrow_move_constructible, then the constructor actually called for MoveInsertion within the construct member function is also declared as noexcept.

Solution 2 requires changing both the implementation of the default allocator (add a conditional noexcept) and vector (replace is_move_constructible with an allocator-targeted check). It does not impose additional restrictions on the allocator (other than 23.2.1 [container.requirements.general] p15), and works nicely even if the move constructor of a MoveInsertable type is private or deleted (the allocator might be a friend of the value type).

In both cases, an addition might be required to provide the basic exception safety guarantee. A short discussion on this topic can be found in the std-discussion mailing list. Essentially, if allocator_traits<Allocator>::construct throws an exception, the object may or may not have been constructed. Two solutions are mentioned in that discussion:

  1. allocator_traits<Allocator>::construct needs to tell its caller whether or not the construction was successful, in case of an exception.

  2. If allocator_traits<Allocator>::construct propagates an exception, it shall either not have constructed an object at the specified location, or that object shall have been destroyed (or it shall ensure otherwise that no resources are leaked).

[2015-05-23, Tomasz Kamiński comments]

Solution 1 discussed in this issue also breaks support for the polymorphic_allocator proposed in the part of the Library Fundamentals TS v1, in addition to already mentioned std::scoped_allocator_adapter. Furthermore there is unknown impact on the other user-defined state-full allocators code written in the C++11.

In addition the library resolution proposed in the LWG issues 2089 and N4462, will break the relation between the std::allocator_trait::construct method and copy/move constructor even for the standard std::allocator. As example please consider following class:

struct NonCopyable
{
  NonCopyable() = default;
  NonCopyable(NonCopyable const&) = delete;
  NonCopyable(NonCopyable&&) = delete;
};

struct InitListConstructor : NonCopyable
{
  InitListConstructor() = default;
  InitListConstructor(std::initializer_list<int>);
  operator int() const;
};

For the above declarations following expression are ill-formed:

InitListConstructor copy(std::declval<InitListConstructor const&>());
InitListConstructor move(std::declval<InitListConstructor&&>());

So the class is not CopyConstructible nor MoveConstructible. However the following are well formed:

InitListConstructor copy{std::declval<InitListConstructor const&>()};
InitListConstructor move{std::declval<InitListConstructor&&>()};

And will be used by std::allocator<InitListConstructor>::construct in case of move-insertion and copy-insertion, after appliance of the resolution proposed in mentioned papers:

The gist of the proposed library fix is simple:

As consequence the requirement proposed in the Solution 1:

If the parameter pack args consists of a single parameter of the type value_type&&, the function may only propagate exceptions if is_nothrow_move_constructible<value_type>::value is false.

Will no longer hold for the std::allocator.

Proposed resolution:


2465. SFINAE-friendly common_type is nearly impossible to specialize correctly and regresses key functionality

Section: 20.15.7.6 [meta.trans.other] Status: New Submitter: Eric Niebler Opened: 2015-01-12 Last modified: 2015-04-08

Priority: 2

View other active issues in [meta.trans.other].

View all other issues in [meta.trans.other].

View all issues with New status.

Discussion:

I think there's a defect regarding common_type and its specializations. Unless I've missed it, there is nothing preventing folks from instantiating common_type with cv-qualified types or reference types. In fact, the wording in N3797 explicitly mentions cv void, so presumably at least cv qualifications are allowed.

Users are given license to specialize common_type when at least of of the types is user-defined. (A separate issue is the meaning of user-defined. In core, I believe this is any class/struct/union/enum, but in lib, I think it means any type not defined in std, right?) There is at least one place in the standard that specializes common_type (time.traits.specializations) on time_point and duration. But the specializations are only for non-cv-qualified and non-reference specializations of time_point and duration.

If the user uses, say, common_type<duration<X,Y> const, duration<A,B> const>, they're not going to get the behavior they expect.

Suggest we clarify the requirements of common_type's template parameters. Also, perhaps we can add blanket wording that common_type<A [cv][&], B [cv][&]> is required to be equivalent to common_type<A,B> (if that is in fact the way we intent this to work).

Also, the change to make common_type SFINAE-friendly regressed key functionality, as noted by Agustín K-ballo Bergé in c++std-lib-37178. Since decay_t is not applied until the very end of the type computation, user specializations are very likely to to be found.

Agustín says:

Consider the following snippet:

struct X {};
struct Y { explicit Y(X){} };

namespace std {
  template<> struct common_type<X, Y> { typedef Y type; };
  template<> struct common_type<Y, X> { typedef Y type; };
}

static_assert(is_same<common_type_t<X, Y>, Y>()); // (A)
static_assert(is_same<common_type_t<X, Y, Y>, Y>()); // (B)
static_assert(is_same<common_type_t<X, X, Y>, Y>()); // (C)

Under the original wording, all three assertion holds. Under the current wording,

The discussion following c++std-lib-35636 seemed to cohere around the idea that the primary common_type specialization should have the effect of stripping top-level ref and cv qualifiers by applying std::decay_t to its arguments and, if any of them change as a result of that transformation, re-dispatching to common_type on those transformed arguments, thereby picking up any user-defined specializations. This change to common_type would make the specializations in time.traits.specializations sufficient.

Suggested wording:

I'm afraid I don't know enough to suggest wording. But for exposition, the following is my best shot at implementing the suggested resolution. I believe it also fixes the regression noted by Agustín K-ballo Bergé in c++std-lib-37178.

namespace detail
{
    template<typename T, typename U>
    using default_common_t =
        decltype(true? std::declval<T>() : std::declval<U>());

    template<typename T, typename U, typename Enable = void>
    struct common_type_if
    {};

    template<typename T, typename U>
    struct common_type_if<T, U,
      void_t<default_common_t<T, U>>>
    {
      using type = decay_t<default_common_t<T, U>>;
    };

    template<typename T, typename U,
       typename TT = decay_t<T>, typename UU = decay_t<U>>
    struct common_type2
      : common_type<TT, UU> // Recurse to catch user specializations
    {};

    template<typename T, typename U>
    struct common_type2<T, U, T, U>
      : common_type_if<T, U>
    {};

    template<typename Meta, typename Enable = void>
    struct has_type
      : std::false_type
    {};

    template<typename Meta>
    struct has_type<Meta, void_t<typename Meta::type>>
      : std::true_type
    {};

    template<typename Meta, typename...Ts>
    struct common_type_recurse
      : common_type<typename Meta::type, Ts...>
    {};

    template<typename Meta, typename...Ts>
    struct common_type_recurse_if
      : std::conditional<
          has_type<Meta>::value,
          common_type_recurse<Meta, Ts...>,
          empty
        >::type
    {};
}

template<typename ...Ts>
struct common_type
{};

template<typename T>
struct common_type<T>
{
  using type = std::decay_t<T>;
};

template<typename T, typename U>
struct common_type<T, U>
  : detail::common_type2<T, U>
{};

template<typename T, typename U, typename... Vs>
struct common_type<T, U, Vs...>
  : detail::common_type_recurse_if<common_type<T, U>, Vs...>
{};

Proposed resolution:


2468. Self-move-assignment of library types

Section: 17.6.4.9 [res.on.arguments], 17.6.3.1 [utility.arg.requirements], 17.6.5.15 [lib.types.movedfrom], 23.2.1 [container.requirements.general] Status: Open Submitter: Matt Austern Opened: 2015-01-22 Last modified: 2015-11-04

Priority: 2

View all other issues in [res.on.arguments].

View all issues with Open status.

Discussion:

Suppose we write

vector<string> v{"a", "b", "c", "d"};
v = move(v);

What should be the state of v be? The standard doesn't say anything specific about self-move-assignment. There's relevant text in several parts of the standard, and it's not clear how to reconcile them.

17.6.4.9 [res.on.arguments] writes that, for all functions in the standard library, unless explicitly stated otherwise, "If a function argument binds to an rvalue reference parameter, the implementation may assume that this parameter is a unique reference to this argument." The MoveAssignable requirements table in 17.6.3.1 [utility.arg.requirements] writes that, given t = rv, t's state is equivalent to rv's from before the assignment and rv's state is unspecified (but valid). For containers specifically, the requirements table in 23.2.1 [container.requirements.general] says that, given a = rv, a becomes equal to what rv was before the assignment (and doesn't say anything about rv's state post-assignment).

Taking each of these pieces in isolation, without reference to the other two:

It's not clear from the text how to put these pieces together, because it's not clear which one takes precedence. Maybe 17.6.4.9 [res.on.arguments] wins (it imposes an implicit precondition that isn't mentioned in the MoveAssignable requirements, so v = move(v) is undefined), or maybe 23.2.1 [container.requirements.general] wins (it explicitly gives additional guarantees for Container::operator= beyond what's guaranteed for library functions in general, so v = move(v) is a no-op), or maybe something else.

On the existing implementations that I checked, for what it's worth, v = move(v) appeared to clear the vector; it didn't leave the vector unchanged and it didn't cause a crash.

Proposed wording:

Informally: change the MoveAssignable and Container requirements tables (and any other requirements tables that mention move assignment, if any) to make it explicit that x = move(x) is defined behavior and it leaves x in a valid but unspecified state. That's probably not what the standard says today, but it's probably what we intended and it's consistent with what we've told users and with what implementations actually do.

[2015-10, Kona Saturday afternoon]

JW: So far, the library forbids self-assignment since it assumes that anything bound to an rvalue reference has no aliases. But self-assignment can happen in real code, and it can be implemented. So I want to add an exception to the Standard that this should be allowed and leave the object in a valid-but-unspecified state.

STL: When this is resolved, I want to see a) VBU for library types after self-move, but also b) requirements on user types for self-moves. E.g. should algorithms be required to avoid self-assignments (since a user-defined type might blow up)? HH: In other words, should we require that you can assign from moved-from values.

WEB: What can one generally do with moved-from values?

VV: Call any member function that has no preconditions.

JW: That's certainly the library requirement, and it's also good guidance for user types.

JW: I'm writing wording. I care about this.

Move to Open; Jonathan to provide wording

Proposed resolution:


2471. copy_n's number of InputIterator increments unspecified

Section: 25.4.1 [alg.copy] Status: New Submitter: Jonathan Wakely Opened: 2015-01-28 Last modified: 2015-05-04

Priority: 3

View other active issues in [alg.copy].

View all other issues in [alg.copy].

View all issues with New status.

Discussion:

It's unspecified how many times copy_n increments the InputIterator. uninitialized_copy_n is specified to increment it exactly n times, which means if an istream_iterator is used then the next character after those copied is read from the stream and then discarded, losing data.

I believe all three of Dinkumware, libc++ and libstdc++ implement copy_n with n - 1 increments of the InputIterator, which avoids reading and discarding a character when used with istream_iterator, but is inconsistent with uninitialized_copy_n and causes surprising behaviour with istreambuf_iterator instead, because copy_n(in, 2, copy_n(in, 2, out)) is not equivalent to copy_n(in, 4, out)

Proposed resolution:


2472. Heterogeneous comparisons in the standard library can result in ambiguities

Section: 20.5.2.8 [tuple.rel], 20.10.9.2 [allocator.globals], 20.11.1.5 [unique.ptr.special], 20.11.2.2.7 [util.smartptr.shared.cmp], 20.17.5.6 [time.duration.comparisons], 20.17.6.6 [time.point.comparisons], 20.13.5 [scoped.adaptor.operators], 24.5.1.3.13 [reverse.iter.op==], 24.5.3.3.13 [move.iter.op.comp] Status: New Submitter: Richard Smith Opened: 2015-02-07 Last modified: 2015-05-04

Priority: 3

View other active issues in [tuple.rel].

View all other issues in [tuple.rel].

View all issues with New status.

Discussion:

The standard library specifies a lot of heterogeneous comparison operators. For instance:

template<class... TTypes, class... UTypes>
constexpr bool operator!=(const tuple<TTypes...>&, const tuple<UTypes...>&);

This has an unfortunate consequence:

#include <tuple>
#include <utility>

using namespace std::rel_ops;
std::tuple<int> a(0);
bool b = a != a;

The last line here is ill-formed due to ambiguity: it might be rel_ops::operator!=, and it might be the heterogeneous tuple operator!=. These are not partially ordered, because they have different constraints: rel_ops requires the types to match, whereas the tuple comparison requires both types to be tuples (but not to match). The same thing happens for user code that defines its own unconstrained 'template<typename T> operator!=(const T&, const T&)' rather than using rel_ops.

One straightforward fix would be to add a homogeneous overload for each heterogeneous comparison:

template<class... TTypes>
constexpr bool operator!=(const tuple<TTypes...>&, const tuple<TTypes...>&);

This is then unambiguously chosen over the other options in the preceding case. FWIW, libstdc++ already does this in some cases.

Proposed resolution:


2475. Allow overwriting of std::basic_string terminator with charT() to allow cleaner interoperation with legacy APIs

Section: 21.3.1.5 [string.access] Status: New Submitter: Matt Weber Opened: 2015-02-21 Last modified: 2015-05-04

Priority: 3

View all other issues in [string.access].

View all issues with New status.

Discussion:

It is often desirable to use a std::basic_string object as a buffer when interoperating with libraries that mutate null-terminated arrays of characters. In many cases, these legacy APIs write a null terminator at the specified end of the provided buffer. Providing such a function with an appropriately-sized std::basic_string results in undefined behavior when the charT object at the size() position is overwritten, even if the value remains unchanged.

Absent the ability to allow for this, applications are forced into pessimizations such as: providing appropriately-sized std::vectors of charT for interoperating with the legacy API, and then copying the std::vector to a std::basic_string; providing an oversized std::basic_string object and then calling resize() later.

A trivial example:

#include <string>
#include <vector>

void legacy_function(char *out, size_t count) {
  for (size_t i = 0; i < count; ++i) {
    *out++ = '0' + (i % 10);
  }
  *out = '\0'; // if size() == count, this results in undefined behavior
}

int main() {
  std::string s(10, '\0');
  legacy_function(&s[0], s.size()); // undefined behavior

  std::vector<char> buffer(11);
  legacy_function(&buffer[0], buffer.size() - 1);
  std::string t(&buffer[0], buffer.size() - 1); // potentially expensive copy

  std::string u(11, '\0');
  legacy_function(&u[0], u.size() - 1);
  u.resize(u.size() - 1); // needlessly complicates the program's logic
}

A slight relaxation of the requirement on the returned object from the element access operator would allow for this interaction with no semantic change to existing programs.

Proposed resolution:

This wording is relative to N4296.

  1. Edit 21.3.1.5 [string.access] as indicated:

    const_reference operator[](size_type pos) const;
    reference operator[](size_type pos);
    

    -1- Requires: […]

    -2- Returns: *(begin() + pos) if pos < size(). Otherwise, returns a reference to an object of type charT with value charT(), where modifying the object to any value other than charT() leads to undefined behavior.

    […]


2478. Unclear how wstring_convert uses cvtstate

Section: 22.3.3.2.2 [conversions.string] Status: New Submitter: Jonathan Wakely Opened: 2015-03-04 Last modified: 2015-05-04

Priority: 4

View other active issues in [conversions.string].

View all other issues in [conversions.string].

View all issues with New status.

Discussion:

How do wstring_convert::from_bytes and wstring_convert::to_bytes use the cvtstate member?

Is it passed to the codecvt member functions? Is a copy of it passed to the member functions? "Otherwise it shall be left unchanged" implies a copy is used, but if that's really what's intended there are simpler ways to say so.

Proposed resolution:


2479. Unclear how wbuffer_convert uses cvtstate

Section: 22.3.3.2.3 [conversions.buffer] Status: New Submitter: Jonathan Wakely Opened: 2015-03-04 Last modified: 2015-05-04

Priority: 4

View other active issues in [conversions.buffer].

View all other issues in [conversions.buffer].

View all issues with New status.

Discussion:

How does wbuffer_convert use the cvtstate member?

Is the same conversion state object used for converting both the get and put areas? That means a read which runs out of bytes halfway through a multibyte character will leave some shift state in cvtstate, which would then be used by a following write, even though the shift state of the get area is unrelated to the put area.

Proposed resolution:


2480. Error handling of wbuffer_convert unclear

Section: 22.3.3.2.3 [conversions.buffer] Status: New Submitter: Jonathan Wakely Opened: 2015-03-04 Last modified: 2015-05-04

Priority: 4

View other active issues in [conversions.buffer].

View all other issues in [conversions.buffer].

View all issues with New status.

Discussion:

If a codecvt conversion returns codecvt_base::error should that be treated as EOF? An exception? Should all the successfully converted characters before a conversion error be available to the users of the wbuffer_convert and/or the internal streambuf, or does a conversion error lose information?

Proposed resolution:


2481. wstring_convert should be more precise regarding "byte-error string" etc.

Section: 22.3.3.2.2 [conversions.string] Status: New Submitter: Jonathan Wakely Opened: 2015-03-04 Last modified: 2015-05-04

Priority: 4

View other active issues in [conversions.string].

View all other issues in [conversions.string].

View all issues with New status.

Discussion:

Paragraph 4 of 22.3.3.2.2 [conversions.string] introduces byte_err_string as "a byte string to display on errors". What does display mean? The string is returned on error, it's not displayed anywhere.

Paragraph 14 says "Otherwise, if the object was constructed with a byte-error string, the member function shall return the byte-error string." The term byte-error string is not used anywhere else.

Paragraph 17 talks about storing "default values in byte_err_string". What default value? Is "Hello, world!" allowed? If it means default-construction it should say so. If paragraph 14 says it won't be used what does it matter how it's initialized? The end of the paragraph refers to storing "byte_err in byte_err_string". This should be more clearly related to the wording in paragraph 14.

It might help if the constructor (and destructor) was specified before the other member functions, so it can more formally define the difference between being "constructed with a byte-error string" and not.

All the same issues apply to the wide_err_string member.

Proposed resolution:


2490. <regex> needs lots of noexcept

Section: 28 [re] Status: New Submitter: Stephan T. Lavavej Opened: 2015-03-27 Last modified: 2015-05-04

Priority: 3

View other active issues in [re].

View all other issues in [re].

View all issues with New status.

Discussion:

Only 4 functions are marked noexcept in all of Clause 28. Many more need to be marked — for example, regex_error::code(), basic_regex::swap(), and sub_match::length().

Proposed resolution:


2491. std::less<T*> in constant expression

Section: 20.14.6 [comparisons] Status: New Submitter: Agustín K-ballo Bergé Opened: 2015-04-01 Last modified: 2015-05-04

Priority: 3

View other active issues in [comparisons].

View all other issues in [comparisons].

View all issues with New status.

Discussion:

It is not entirely clear if and when the specializations of std::less (and friends) for pointer types can be used in a constant expression. Consider the following code:

#include <functional>

struct foo {};
foo x, y;
constexpr bool b = std::less<foo*>{}(&x, &y); // [1]

foo z[] = {{}, {}};
constexpr bool ba = std::less<foo*>{}(&z[0], &z[1]); // [2]

Comparing the address of unrelated objects is not a constant expression since the result is unspecified, so it could be expected for [1] to fail and [2] to succeed. However, std::less specialization for pointer types is well-defined and yields a total order, so it could just as well be expected for [1] to succeed. Finally, since the implementation of such specializations is not mandated, [2] could fail as well (This could happen, if an implementation would provide such a specialization and if that would use built-in functions that would not be allowed in constant expressions, for example). In any case, the standard should be clear so as to avoid implementation-defined constexpr-ness.

Proposed resolution:


2493. initializer_list supports incomplete classes

Section: 18.9 [support.initlist] Status: New Submitter: David Krauss Opened: 2015-04-27 Last modified: 2015-05-04

Priority: 4

View other active issues in [support.initlist].

View all other issues in [support.initlist].

View all issues with New status.

Discussion:

The typical use-case of std::initializer_list<T> is for a pass-by-value parameter of T's constructor. However, this contravenes 17.6.4.8 [res.on.functions]/2.5 because initializer_list doesn't specifically allow incomplete types (as do for example std::unique_ptr (20.11.1 [unique.ptr]/5) and std::enable_shared_from_this (20.11.2.5 [util.smartptr.enab]/2)).

A resolution would be to copy-paste the relevant text from such a paragraph.

Proposed resolution:


2496. Certain hard-to-avoid errors not in the immediate context are not allowed to be triggered by the evaluation of type traits

Section: 20.15.4.3 [meta.unary.prop] Status: New Submitter: Hubert Tong Opened: 2015-05-07 Last modified: 2015-08-03

Priority: 3

View other active issues in [meta.unary.prop].

View all other issues in [meta.unary.prop].

View all issues with New status.

Discussion:

I do not believe that the wording in 20.15.4.3 [meta.unary.prop] paragraph 3 allows for the following program to be ill-formed:

#include <type_traits>

template <typename T> struct B : T { };
template <typename T> struct A { A& operator=(const B<T>&); };

std::is_assignable<A<int>, int> q;

In particular, I do not see where the wording allows for the "compilation of the expression" declval<T>() = declval<U>() to occur as a consequence of instantiating std::is_assignable<T, U> (where T and U are, respectively, A<int> and int in the example code).

Instantiating A<int> as a result of requiring it to be a complete type does not trigger the instantiation of B<int>; however, the "compilation of the expression" in question does.

Proposed resolution:


2497. Use of uncaught_exception()

Section: 27.7.3.4 [ostream::sentry] Status: New Submitter: Roger Orr Opened: 2015-05-08 Last modified: 2015-09-27

Priority: 3

View all other issues in [ostream::sentry].

View all issues with New status.

Discussion:

In the current 27.7.3.4 [ostream::sentry], p4 refers to the now deprecated std::uncaught_exception(): D.9 [depr.uncaught].

If ((os.flags() & ios_base::unitbuf) && !uncaught_exception() && os.good()) is true, calls os.rdbuf()->pubsync().

This needs to be changed, for example to use std::uncaught_exceptions() and to capture the value on entry and compare with the saved value on exit.

[2015-06, Telecom]

JW: I already added an 's' here to make it use the new function, but that doesn't resolve Roger's suggestion to capture the value on entry and check it.

Proposed resolution:


2498. operator>>(basic_istream&&, T&&) returns basic_istream&, but should probably return basic_istream&&

Section: 27.7.2.6 [istream.rvalue] Status: New Submitter: Richard Smith Opened: 2015-05-08 Last modified: 2015-09-27

Priority: 3

View all other issues in [istream.rvalue].

View all issues with New status.

Discussion:

Consider:

auto& is = make_istream() >> x; // oops, istream object is already gone

With a basic_istream&& return type, the above would be ill-formed, and generally we'd preserve the value category properly.

[2015-06, Telecom]

JW: think this needs proper consideration, it would make

stream() >> x >> y >> z
go from 3 operator>> calls to 6 operator>> calls, and wouldn't prevent dangling references (change the example to auto&&)

Proposed resolution:


2499. operator>>(basic_istream&, CharT*) makes it hard to avoid buffer overflows

Section: 27.7.2.2.3 [istream::extractors] Status: Open Submitter: Richard Smith Opened: 2015-05-08 Last modified: 2015-11-04

Priority: 2

View all other issues in [istream::extractors].

View all issues with Open status.

Discussion:

We removed gets() (due to an NB comment and C11 — bastion of backwards compatibility — doing the same). Should we remove this too?

Unlike gets(), there are legitimate uses:

char buffer[32];
char text[32] = // ...
ostream_for_buffer(text) >> buffer; // ok, can't overrun buffer

… but the risk from constructs like "std::cin >> buffer" seems to outweigh the benefit.

The issue had been discussed on the library reflector starting around c++std-lib-35541.

[2015-06, Telecom]

VV: Request a paper to deprecate / remove anything

[2015-10, Kona Saturday afternoon]

STL: This overload is evil and should probably die.

VV: I agree with that, even though I don't care.

STL: Say that we either remove it outright following the gets() rationale, or at least deprecate it.

Move to Open; needs a paper.

Proposed resolution:


2501. std::function requires POCMA/POCCA

Section: 20.14.12.2 [func.wrap.func] Status: New Submitter: David Krauss Opened: 2015-05-20 Last modified: 2015-08-03

Priority: 3

View other active issues in [func.wrap.func].

View all other issues in [func.wrap.func].

View all issues with New status.

Discussion:

The idea behind propagate_on_container_move_assignment is that you can keep an allocator attached to a container. But it's not really designed to work with polymorphism, which introduces the condition where the current allocator is non-POCMA and the RHS of assignment, being POCMA, wants to replace it. If function were to respect the literal meaning, any would-be attached allocator is at the mercy of every assignment operation. So, std::function is inherently POCMA, and passing a non-POCMA allocator should be ill-formed.

The other alternative, and the status quo, is to ignore POCMA and assume it is true. This seems just dangerous enough to outlaw. It is, in theory, possible to properly support POCMA as far as I can see, albeit with difficulty and brittle results. It would require function to keep a throwing move constructor, which otherwise can be noexcept.

The same applies to propagate_on_container_copy_assignment. This presents more difficulty because std::allocator does not set this to true. Perhaps it should. For function to respect this would require inspecting the POCCA of the source allocator, slicing the target from the erasure of the source, slicing the allocation from the erasure of the destination, and performing a copy with the destination's allocator with the source's target. This comes out of the blue for the destination allocator, which might not support the new type anyway. Theoretically possible, but brittle and not very practical. Again, current implementations quietly ignore the issue but this isn't very clean.

The following code example is intended to demonstrate the issue here:

#include <functional>
#include <iostream>
#include <vector>

template <typename T>
struct diag_alloc 
{
  std::string name;

  T* allocate(std::size_t n) const 
  {
    std::cout << '+' << name << '\n';
    return static_cast<T*>(::operator new(n * sizeof(T)));
  }
  
  void deallocate(T* p, std::size_t) const 
  {
    std::cout << '-' << name << '\n';
    return ::operator delete(p);
  }

  template <typename U>
  operator diag_alloc<U>() const { return {name}; }

  friend bool operator==(const diag_alloc& a, const diag_alloc& b)
  { return a.name == b.name; }
  
  friend bool operator!=(const diag_alloc& a, const diag_alloc& b)
  { return a.name != b.name; }

  typedef T value_type;
  
  template <typename U>
  struct rebind { typedef diag_alloc<U> other; };
};

int main() {
  std::cout << "VECTOR\n";
  std::vector<int, diag_alloc<int>> foo({1, 2}, {"foo"}); // +foo
  std::vector<int, diag_alloc<int>> bar({3, 4}, {"bar"}); // +bar

  std::cout << "move\n";
  foo = std::move(bar); // no message

  std::cout << "more foo\n";
  foo.reserve(40); // +foo -foo
  std::cout << "more bar\n";
  bar.reserve(40); // +bar -bar

  std::cout << "\nFUNCTION\n";
  int bigdata[100];
  auto bigfun = [bigdata]{};
  typedef decltype(bigfun) ft;
  std::cout << "make fizz\n";
  std::function<void()> fizz(std::allocator_arg, diag_alloc<ft>{"fizz"}, bigfun); // +fizz
  std::cout << "another fizz\n";
  std::function<void()> fizz2;
  fizz2 = fizz; // +fizz as if POCCA
  std::cout << "make buzz\n";
  std::function<void()> buzz(std::allocator_arg, diag_alloc<ft>{"buzz"}, bigfun); // +buzz
  std::cout << "move\n";
  buzz = std::move(fizz); // -buzz as if POCMA

  std::cout << "\nCLEANUP\n";
}

Proposed resolution:


2502. std::function does not use allocator::construct

Section: 20.14.12.2 [func.wrap.func] Status: New Submitter: David Krauss Opened: 2015-05-20 Last modified: 2015-08-03

Priority: 3

View other active issues in [func.wrap.func].

View all other issues in [func.wrap.func].

View all issues with New status.

Discussion:

It is impossible for std::function to construct its target object using the construct method of a type-erased allocator. More confusingly, it is possible when the allocator and the target are created at the same time. The means of target construction should be specified.

Proposed resolution:


2503. multiline option should be added to syntax_option_type

Section: 28.5.1 [re.synopt] Status: New Submitter: Nozomu Katō Opened: 2015-05-22 Last modified: 2015-09-27

Priority: 2

View other active issues in [re.synopt].

View all other issues in [re.synopt].

View all issues with New status.

Discussion:

The specification of ECMAScript defines the Multiline property for its RegExp and the regular expressions ^ and $ behave differently according to the value of this property. Thus, this property should be available also in the ECMAScript compatible engine in std::regex.

[2015-05-22, Daniel comments]

This issue interacts somewhat with LWG 2343.

[Telecom 2015-07]

Set the priority to match LWG 2343.

Proposed resolution:

This wording is relative to N4431.

  1. Change 28.5.1 [re.synopt] as indicated:

    namespace std::regex_constants {
      typedef T1 syntax_option_type;
      constexpr syntax_option_type icase = unspecified ;
      constexpr syntax_option_type nosubs = unspecified ;
      constexpr syntax_option_type optimize = unspecified ;
      constexpr syntax_option_type collate = unspecified ;
      constexpr syntax_option_type ECMAScript = unspecified ;
      constexpr syntax_option_type basic = unspecified ;
      constexpr syntax_option_type extended = unspecified ;
      constexpr syntax_option_type awk = unspecified ;
      constexpr syntax_option_type grep = unspecified ;
      constexpr syntax_option_type egrep = unspecified ;
      constexpr syntax_option_type multiline = unspecified ;
    }
    
  2. Change 28.5.2 [re.matchflag], Table 138 — "syntax_option_type effects" as indicated:

    Table 138 — syntax_option_type effects
    Element Effect(s) if set
    multiline Specifies that ^ shall match the beginning of a line and $ shall match the end of a line, if the ECMAScript engine is selected.

2504. basic_streambuf is not an abstract class

Section: 27.6.3 [streambuf] Status: New Submitter: Jonathan Wakely Opened: 2015-05-28 Last modified: 2015-08-03

Priority: 3

View other active issues in [streambuf].

View all other issues in [streambuf].

View all issues with New status.

Discussion:

27.6.3 [streambuf] p1 says:

The class template basic_streambuf<charT, traits> serves as an abstract base class for deriving various stream buffers whose objects each control two character sequences: […]

The term "abstract base class" is not defined in the standard, but "abstract class" is (10.4 [class.abstract]).

According to the synopsis basic_streambuf has no pure virtual functions so is not an abstract class and none of libstdc++, libc++, or dinkumware implement it as an abstract class. I don't believe the wording was ever intended to require it to be an abstract class, but it could be read that way.

I suggest the wording be changed to "polymorphic base class" or something else that can't be seen to imply a normative requirement to make it an abstract class.

Proposed resolution:


2505. auto_ptr_ref creation requirements underspecified

Section: 99 [auto.ptr.conv] Status: Open Submitter: Hubert Tong Opened: 2015-05-28 Last modified: 2016-05-17

Priority: 4

View all other issues in [auto.ptr.conv].

View all issues with Open status.

Discussion:

In C++14 sub-clause 99 [auto.ptr.conv], there appears to be no requirement that the formation of an auto_ptr_ref<Y> from an auto_ptr<X> is done only when X* can be implicitly converted to Y*.

For example, I expect formation of the auto_ptr_ref<A> from the prvalue of type auto_ptr<B> to be invalid in the case below (but the wording does not seem to be there):

#include <memory>

struct A { };
struct B { } b;

std::auto_ptr<B> apB() { return std::auto_ptr<B>(&b); }
int main() {
  std::auto_ptr<A> apA(apB());
  apA.release();
}

The behaviour of the implementation in question on the case presented above is to compile and execute it successfully (which is what the C++14 wording implies). The returned value from apA.release() is essentially reinterpret_cast<A*>(&b).

There is nothing in the specification of

template <class X>
template <class Y> operator auto_ptr<X>::auto_ptr_ref<Y>() throw();

which implies that X* should be implicitly convertible to Y*.

The implementation in question uses the reinterpret_cast interpretation even when Y is an accessible, unambiguous base class of X; the result thereof is that no offset adjustment is performed.

[2015-07, Telecon]

Marshall to resolve.

[2016-03-16, Alisdair Meredith comments]

This issue is a defect in a component we have actively removed from the standard. I can't think of a clearer example of something that is no longer a defect!

Proposed resolution:

This wording is relative to ISO/IEC 14882:2014(E).

  1. Change 99 [auto.ptr.conv] as indicated:

    template<class Y> operator auto_ptr_ref<Y>() throw();
    

    -?- Requires: X* can be implicitly converted to Y*.

    -3- Returns: An auto_ptr_ref<Y> that holds *this.

    -?- Notes: Because auto_ptr_ref is present for exposition only, the only way to invoke this function is by calling one of the auto_ptr conversions which take an auto_ptr_ref as an argument. Since all such conversions will call release() on *this (in the form of the auto_ptr that the auto_ptr_ref holds a reference to), an implementation of this function may cause instantiation of said release() function without changing the semantics of the program.

    template<class Y> operator auto_ptr<Y>() throw();
    

    -?- Requires: X* can be implicitly converted to Y*.

    -4- Effects: Calls release().

    -5- Returns: An auto_ptr<Y> that holds the pointer returned from release().


2506. Underspecification of atomics

Section: 1.10 [intro.multithread], 29.5 [atomics.types.generic], 18.10 [support.runtime] Status: New Submitter: Geoffrey Romer Opened: 2015-05-29 Last modified: 2015-09-25

Priority: 3

View all other issues in [intro.multithread].

View all issues with New status.

Discussion:

The concurrency libraries specified in clauses 29 and 30 do not adequately specify how they relate to the concurrency model specified in 1.10 [intro.multithread]. In particular:

1.10 [intro.multithread] specifies "atomic objects" as having certain properties. I can only assume that instances of the classes defined in Clause 29 are intended to be "atomic objects" in this sense, but I can't find any wording to specify that, and it's genuinely unclear whether Clause 30 objects are atomic objects. In fact, on a literal reading the C++ Standard doesn't appear to provide any portable way to create an atomic object, or even determine whether an object is an atomic object.

(It's not clear if the term "atomic object" is actually needed, given that atomic objects can have non-atomic operations, and non-atomic objects can have atomic operations. But even if the term itself goes away, there still needs to be some indication that Clause 29 objects have the properties currently attributed to atomic objects).

Similarly, 1.10 [intro.multithread] uses "atomic operation" as a term of art, but the standard never unambiguously identifies any operation as an "atomic operation" (although in one case it unambiguously identifies an operation that is not atomic). It does come close in a few cases, but not close enough:

Proposed resolution:


2507. codecvt_mode should be a bitmask type

Section: 22.5 [locale.stdcvt] Status: New Submitter: Jonathan Wakely Opened: 2015-06-08 Last modified: 2015-08-03

Priority: 3

View all other issues in [locale.stdcvt].

View all issues with New status.

Discussion:

The enumeration type codecvt_mode is effectively a bitmask type (17.5.2.1.3 [bitmask.types]) with three elements, but isn't defined as such.

This harms usability because bitmask types are required to work well with bitwise operators, but codecvt_mode doesn't have overloaded operators, making it very inconvenient to combine values:

std::codecvt_utf16<char32_t, 0x10FFFF,
  static_cast<std::codecvt_mode>(std::little_endian|std::generate_header)>
cvt;

The static_cast harms readability and should not be necessary.

I suggest that either codecvt_mode is specified to be a bitmask type, or as a minimal fix we provide an overloaded operator| that returns the right type.

Proposed resolution:


2508. §[new.delete.dataraces] wording needs to be updated

Section: 18.6.2.4 [new.delete.dataraces] Status: New Submitter: Hans Boehm Opened: 2015-06-09 Last modified: 2016-02-12

Priority: 3

View all other issues in [new.delete.dataraces].

View all issues with New status.

Discussion:

18.6.2.4 [new.delete.dataraces] uses obsolete wording.

It should introduce a "synchronizes with" relationship. "Happens before" is too weak, since that may not composes with sequenced before.

The "shall not introduce a data race" wording is probably not technically correct either. These may race with other (non-allocation/deallocation) concurrent accesses to the object being allocated or deallocated.

Proposed resolution:


2510. Tag types should not be DefaultConstructible

Section: 18.6 [support.dynamic], 20.2 [utility], 20.4.5 [pair.piecewise], 20.10.2 [memory.syn], 20.10.6 [allocator.tag], 30.4 [thread.mutex] Status: Open Submitter: Ville Voutilainen Opened: 2015-06-13 Last modified: 2016-06-28

Priority: 2

View all other issues in [support.dynamic].

View all issues with Open status.

Discussion:

std::experimental::optional, for certain reasons, specifies its nullopt type to not be DefaultConstructible. It doesn't do so for its tag type in_place_t and neither does the standard proper for any of its tag types. That turns out to be very unfortunate, consider the following:

#include <memory>
#include <array>

void f(std::array<int, 1>, int) {} // #1
void f(std::allocator_arg_t, int) {} // #2

int main()
{
  f({}, 666); // #3
}

The call at #3 is ambiguous. What's even worse is that if the overload #1 is removed, the call works just fine. The whole point of a tag type is that it either needs to mentioned in a call or it needs to be a forwarded argument, so being able to construct a tag type like that makes no sense.

Making the types have an explicit default constructor might have helped, but CWG 1518 is going against that idea.

[optional.nullopt]/3 solves this problem for nullopt:

Type nullopt_t shall not have a default constructor. It shall be a literal type. Constant nullopt shall be initialized with an argument of literal type.

[2015-06, Telecom]

Move to Tentatively Ready.

[2015-10, Kona Saturday afternoon]

Move back to Open

JW: The linked Core issue (CWG 1518) gives us a better tool to solve this (explicit default constructors). [The CWG Issue means that an explicit default constructor will no longer match "{}".] JW explains that it's important that tag types cannot be constructed from "{}" (e.g. the allocator tag in the tuple constructors).

WEB: Should we now go back and update our constructors? JW: For tag types, yes.

VV: The guideline is that anything that does not mention the type name explicitly should not invoke an explicit constructor.

Ville will provide wording.

Discussion about pair/tuple's default constructor - should they now be explicit?

[2016-01-31]

Ville provides revised wording.

Previous resolution [SUPERSEDED]:

This wording is relative to N4527.

  1. In 18.6 [support.dynamic]/1, change the header <new> synopsis:

    […]
    struct nothrow_t {}; see below
    extern const nothrow_t nothrow;
    […]
    
  2. Add a new paragraph after 18.6 [support.dynamic]/1 (following the header <new> synopsis):

    -?- Type nothrow_t shall not have a default constructor.

  3. In 20.2 [utility]/2, change the header <utility> synopsis:

    […]
    // 20.3.5, pair piecewise construction
    struct piecewise_construct_t { }; see below
    constexpr piecewise_construct_t piecewise_construct{ unspecified };
    […]
    
  4. Add a new paragraph after 20.2 [utility]/2 (following the header <utility> synopsis):

    -?- Type piecewise_construct_t shall not have a default constructor. It shall be a literal type. Constant piecewise_construct shall be initialized with an argument of literal type.

  5. In 20.4.5 [pair.piecewise], apply the following edits:

    struct piecewise_construct_t { };
    constexpr piecewise_construct_t piecewise_construct{ unspecified };
    
  6. In 20.10.2 [memory.syn]/1, change the header <memory> synopsis:

    […]
    // 20.7.6, allocator argument tag
    struct allocator_arg_t { }; see below
    constexpr allocator_arg_t allocator_arg{ unspecified };
    […]
    
  7. Add a new paragraph after 20.10.2 [memory.syn]/1 (following the header <memory> synopsis):

    -?- Type allocator_arg_t shall not have a default constructor. It shall be a literal type. Constant allocator_arg shall be initialized with an argument of literal type.

  8. In 20.10.6 [allocator.tag], apply the following edits:

    namespace std {
      struct allocator_arg_t { };
      constexpr allocator_arg_t allocator_arg{ unspecified };
    }
    

    Editorial drive-by: piecewise_construct_t is written, in 20.4.5 [pair.piecewise] like

    struct piecewise_construct_t { };
    constexpr piecewise_construct_t piecewise_construct{};
    

    whereas other tag types such as allocator_construct_t are, in e.g. 20.10.6 [allocator.tag], written like

    namespace std {
      struct allocator_arg_t { };
      constexpr allocator_arg_t allocator_arg{};
    }
    

    We should decide whether or not to write out the std namespace in such paragraphs. I would suggest not to write it out.

  9. In 30.4 [thread.mutex]/1, change the header <mutex> synopsis:

    […]
    struct defer_lock_t { }; see below
    struct try_to_lock_t { }; see below
    struct adopt_lock_t { }; see below
    
    constexpr defer_lock_t defer_lock { unspecified  };
    constexpr try_to_lock_t try_to_lock { unspecified  };
    constexpr adopt_lock_t adopt_lock { unspecified  };
    […]
    
  10. Add three new paragraphs after [thread.mutex]/1 (following the header <mutex> synopsis):

    -?- Type defer_lock_t shall not have a default constructor. It shall be a literal type. Constant defer_lock shall be initialized with an argument of literal type.

    -?- Type try_to_lock_t shall not have a default constructor. It shall be a literal type. Constant try_to_lock shall be initialized with an argument of literal type.

    -?- Type adopt_lock_t shall not have a default constructor. It shall be a literal type. Constant adopt_lock shall be initialized with an argument of literal type.

[2016-03 Jacksonville]

AM: should have note about compatibility in Annex C
HH: like this idiom well enough that I've started using it in my own code
AM: why are pair and tuple involved here?
GR: they are the only types which forward explicitness with EXPLICIT
AM: British spelling of behaviour
AM: happy to drop my issue about Annex C

[Oulu, 2016-06]

This is waiting on Core issue 1518

Saturday: Core 1518 was resolved in Oulu

Proposed resolution:

This wording is relative to N4567.

  1. In 18.6 [support.dynamic]/1, change the header <new> synopsis:

    […]
    struct nothrow_t { explicit nothrow_t() = default; };
    extern const nothrow_t nothrow;
    […]
    
  2. In 20.2 [utility]/2, change the header <utility> synopsis:

    […]
    // 20.3.5, pair piecewise construction
    struct piecewise_construct_t { explicit piecewise_construct_t() = default; };
    constexpr piecewise_construct_t piecewise_construct{};
    […]
    
  3. In 20.4.2 [pairs.pair], change the class template pair synopsis:

    […]
    pair(pair&&) = default;
    EXPLICIT constexpr pair();
    EXPLICIT constexpr pair(const T1& x, const T2& y);
    […]
    
  4. Around 20.4.2 [pairs.pair] p3, apply the following edits:

    EXPLICIT constexpr pair();
    

    -3- Effects: Value-initializes first and second.

    -4- Remarks: This constructor shall not participate in overload resolution unless is_default_constructible<first_type>::value is true and is_default_constructible<second_type>::value is true. [Note: This behaviour can be implemented by a constructor template with default template arguments. — end note] The constructor is explicit if and only if either first_type or second_type is not implicitly default-constructible. [Note: This behaviour can be implemented with a trait that checks whether a const first_type& or a const second_type& can be initialized with {}. — end note]

  5. In 20.4.5 [pair.piecewise], apply the following edits:

    struct piecewise_construct_t { explicit piecewise_construct_t() = default; };
    constexpr piecewise_construct_t piecewise_construct{};
    
  6. In 20.5.2 [tuple.tuple], change the class template tuple synopsis:

    […]
    // 20.4.2.1, tuple construction
    EXPLICIT constexpr tuple();
    EXPLICIT constexpr tuple(const Types&...); // only if sizeof...(Types) >= 1
    […]
    
  7. Around 20.5.2.1 [tuple.cnstr] p4, apply the following edits:

    EXPLICIT constexpr tuple();
    

    -4- Effects: Value initializes each element.

    -5- Remarks: This constructor shall not participate in overload resolution unless is_default_constructible<Ti>::value is true for all i. [Note: This behaviour can be implemented by a constructor template with default template arguments. — end note] The constructor is explicit if and only if Ti is not implicitly default-constructible for at least one i. [Note: This behaviour can be implemented with a trait that checks whether a const Ti& can be initialized with {}. — end note]

  8. In 20.10.2 [memory.syn]/1, change the header <memory> synopsis:

    […]
    // 20.7.6, allocator argument tag
    struct allocator_arg_t { explicit allocator_arg_t() = default; };
    constexpr allocator_arg_t allocator_arg{};
    […]
    
  9. In 20.10.6 [allocator.tag], apply the following edits:

    namespace std {
      struct allocator_arg_t { explicit allocator_arg_t() = default; };
      constexpr allocator_arg_t allocator_arg{};
    }
    

    Editorial drive-by: piecewise_construct_t is written, in 20.4.5 [pair.piecewise] like

    struct piecewise_construct_t { };
    constexpr piecewise_construct_t piecewise_construct{};
    

    whereas other tag types such as allocator_construct_t are, in e.g. 20.10.6 [allocator.tag], written like

    namespace std {
      struct allocator_arg_t { };
      constexpr allocator_arg_t allocator_arg{};
    }
    

    We should decide whether or not to write out the std namespace in such paragraphs. I would suggest not to write it out.

  10. In 30.4 [thread.mutex]/1, change the header <mutex> synopsis:

    […]
    struct defer_lock_t { explicit defer_lock_t() = default; };
    struct try_to_lock_t { explicit try_to_lock_t() = default; };
    struct adopt_lock_t { explicit adopt_lock_t() = default; };
    
    constexpr defer_lock_t defer_lock { };
    constexpr try_to_lock_t try_to_lock { };
    constexpr adopt_lock_t adopt_lock { };
    […]
    

2511. scoped_allocator_adaptor piecewise construction does not require CopyConstructible

Section: 20.13.4 [allocator.adaptor.members] Status: New Submitter: David Krauss Opened: 2015-06-16 Last modified: 2015-08-03

Priority: 3

View other active issues in [allocator.adaptor.members].

View all other issues in [allocator.adaptor.members].

View all issues with New status.

Discussion:

20.13.4 [allocator.adaptor.members]/10 requires that the argument types in the piecewise-construction tuples all be CopyConstructible. These tuples are typically created by std::forward_as_tuple, such as in ¶13. So they will be a mix of lvalue and rvalue references, the latter of which are not CopyConstructible.

My guess is that CopyConstructible was specified to feed the tuple_cat, before that function could handle rvalues. Since the argument tuple is already moved in ¶11, the requirement is obsolete. It should either be changed to MoveConstructible, or perhaps better, convert the whole tuple to references (i.e. form tuple<Args1&&...>) so nothing needs to be moved. After all, this is a facility for handling non-movable types.

It appears that the resolution of DR 2203, which added std::move to ¶11, simply omitted the change to ¶10.

Proposed resolution:


2512. Y2K bites; what is an "unambiguous year identifier"?

Section: 22.4.5.1.2 [locale.time.get.virtuals] Status: New Submitter: Hubert Tong Opened: 2015-06-19 Last modified: 2015-08-03

Priority: 2

View all other issues in [locale.time.get.virtuals].

View all issues with New status.

Discussion:

I recently encountered a failure related to questionable use of do_get_year. The platform where the code happened to work had an implementation which handled certain three-digit "year identifiers" as the number of years since 1900 (this article describes such an implementation).

22.4.5.1.2 [locale.time.get.virtuals] makes it implementation defined whether two-digit years are accepted, etc., but does not say anything specifically about three-digit years.

The implementation freedom to not report errors in 22.4.5.1 [locale.time.get] paragraph 1 also seems to be too broad.

See also the discussion following c++std-lib-38042.

Proposed resolution:


2513. Missing requirements for basic_string::value_type

Section: 21.1 [strings.general] Status: New Submitter: Jonathan Wakely Opened: 2015-06-26 Last modified: 2016-07-06

Priority: 4

View all other issues in [strings.general].

View all issues with New status.

Discussion:

The allocator-aware container requirements in Table 98 impose no MoveAssignable requirements on the value_type when propagate_on_container_move_assignment is true, because typically the container's storage would be moved by just exchanging some pointers.

However for a basic_string using the small string optimization move assignment may need to assign individual characters into the small string buffer, even when the allocator propagates.

The only requirement on the char-like objects stored in a basic_string are that they are non-array POD types and Destructible, which means that a POD type with a deleted move assignment operator should be usable in a basic_string, despite it being impossible to move assign:

#include <string>

struct odd_pod 
{
  odd_pod() = default;
  odd_pod& operator=(odd_pod&&) = delete;
};

static_assert(std::is_pod<odd_pod>::value, "POD");

int main()
{
  using S = std::basic_string<odd_pod>;
  S s;
  s = S{};       // fails
}

Using libstdc++ basic_string<odd_pod> cannot even be default-constructed because the constructor attempts to assign the null terminator to the first element of the small string buffer.

Similar problems exist with POD types with a deleted default constructor.

I believe that basic_string should require its value_type to be at least DefaultConstructible and MoveAssignable.

[2016-06, Oulu]

This should be resolved by P0178

Note: P0178 was sent back to LEWG in Oulu.

Proposed resolution:


2514. Type traits must not be final

Section: 20.15.1 [meta.rqmts] Status: New Submitter: Jonathan Wakely Opened: 2015-07-03 Last modified: 2015-08-03

Priority: 3

View all issues with New status.

Discussion:

We should make it clear that all standard UnaryTypeTraits, BinaryTypeTraits and TransformationTraits are not final.

Otherwise it is not safe to use them as arguments to a template like this:

template<typename C1, typename C2>
struct conjunction
  : conditional_t<C1::value, C2, C1>
{ };

Proposed resolution:


2518. [fund.ts.v2] Non-member swap for propagate_const should call member swap

Section: 99 [fund.ts.v2::propagate_const.algorithms] Status: New Submitter: Tim Song Opened: 2015-07-08 Last modified: 2016-02-29

Priority: 3

View all issues with New status.

Discussion:

Addresses: fund.ts.v2

For consistency with the rest of the standard library, the non-member swap for propagate_const should call member swap.

[2015-07, Telecom]

Both P3 and NAD were suggested.

[2016-02-20, Ville comments]

Feedback from an implementation:

The implementation of propagate_const in libstdc++ calls propagate_const's member swap from the non-member swap.

Proposed resolution:

This wording is relative to N4529.

  1. Edit [propagate_const.algorithms] as indicated:

    -1- template <class T>
    constexpr void swap(propagate_const<T>& pt1, propagate_const<T>& pt2) noexcept(see below)
    

    -2- The constant-expression in the exception-specification is noexcept(swap(pt1.t_, pt2.t_)pt1.swap(pt2)).

    -3- Effects: swap(pt1.t_, pt2.t_)pt1.swap(pt2).


2519. Iterator operator-= has gratuitous undefined behaviour

Section: 24.2.7 [random.access.iterators] Status: New Submitter: Hubert Tong Opened: 2015-07-15 Last modified: 2015-09-14

Priority: 2

View all other issues in [random.access.iterators].

View all issues with New status.

Discussion:

In subclause 24.2.7 [random.access.iterators], Table 110, the operational semantics for the expression "r -= n" are defined as

return r += -n;

Given a difference_type of a type int with range [-32768, 32767], if the value of n is -32768, then the evaluation of -n causes undefined behaviour (Clause 5 [expr] paragraph 4).

The operational semantics may be changed such that the undefined behaviour is avoided.

Suggested wording:

Replace the operational semantics for "r -= n" with:

{ 
  difference_type m = n;
  if (m >= 0)
    while (m--)
      --r;
  else
    while (m++)
      ++r;
  return r; 
}

Jonathan Wakely:

I'm now convinced we don't want to change the definition of -= and instead we should explicitly state the (currently implicit) precondition that n != numeric_limits<difference_type>::min().

Proposed resolution:

This wording is relative to N4527.

  1. Change Table 110 "Random access iterator requirements (in addition to bidirectional iterator)" as indicated:

    Table 110 — Random access iterator requirements (in addition to bidirectional iterator)
    Expression Return type Operational
    semantics
    Assertion/note
    pre-/post-condition
    r -= n X& return r += -n; pre: the absolute value of n is in the range of representable values of difference_type.

2521. [fund.ts.v2] weak_ptr's converting move constructor should be modified as well for array support

Section: 99 [fund.ts.v2::memory.smartptr.weak.const] Status: New Submitter: Tim Song Opened: 2015-07-25 Last modified: 2016-01-04

Priority: 2

View all issues with New status.

Discussion:

Addresses: fund.ts.v2

[memory.smartptr.weak.const] altered the constraints on weak_ptr's constructor from const weak_ptr<Y>& and const shared_ptr<Y>&. The constraints on the converting move constructor from weak_ptr<Y>&& was not, but should be, similarly modified.

[2015-10-26]

Daniel adjusts wording to lib. fund. v2. As link to the originating proposal: The discussion in this issue refers to wording changes that were requested by N3920.

Proposed resolution:

This wording is relative to N4529.

  1. At the end of [memory.smartptr.weak.const], add:
    [Drafting note: The current paragraph [memory.smartptr.weak.const] p2 is incorrectly declared as Requires element, but it does not describe a requirement, instead it describes a "template constraint" which are elsewhere always specified within a Remarks element because it describes constraints that an implementation (and not the user) has to meet. See LWG 2292 for a suggestion to introduce a separate new specification element for this situation. This has also been fixed in the current working draft. — end drafting note]

    weak_ptr(weak_ptr&& r) noexcept;
    template<class Y> weak_ptr(weak_ptr<Y>&& r) noexcept;
    

    -?- Remark: The second constructor shall not participate in overload resolution unless Y* is compatible with T*.

    -?- Effects: Move-constructs a weak_ptr instance from r.

    -?- Postconditions: *this shall contain the old value of r. r shall be empty. r.use_count() == 0.


2524. generate_canonical can occasionally return 1.0

Section: 26.6.8.4.2 [rand.dist.pois.exp] Status: Open Submitter: Michael Prähofer Opened: 2015-08-20 Last modified: 2015-11-04

Priority: 2

View all issues with Open status.

Discussion:

Original title was: exponential_distribution<float> sometimes returns inf.

The random number distribution class template exponential_distribution<float> may return "inf" as can be seen from the following example program:

// compiled with
// g++ -std=c++11 Error_exp_distr.cpp

#include <iostream>
#include <random>
#include <bitset>

int main(){
  unsigned long long h;
  std::mt19937_64 mt1(1);
  std::mt19937_64 mt2(1);
  mt1.discard(517517);
  mt2.discard(517517);
  std::exponential_distribution<float> dis(1.0);
  h = mt2();
  std::cout << std::bitset<64>(h) << " " << (float) -log(1 - h/pow(2, 64)) << " " 
            << -log(1 - (float) h/pow(2, 64)) << " " << dis(mt1) << std::endl;
  h = mt2();
  std::cout << std::bitset<64>(h) << " " << (float) -log(1 - h/pow(2, 64)) << " " 
            << -log(1 - (float) h/pow(2, 64)) << " " << dis(mt1) << std::endl;
}

output:

0110010110001001010011000111000101001100111110100001110011100001 0.505218 0.505218 0.505218
1111111111111111111111111101010011000110011110011000110101100110 18.4143 inf inf

The reason seems to be that converting a double x in the range [0, 1) to float may result in 1.0f if x is close enough to 1. I see two possibilities to fix that:

  1. use internally double (or long double?) and then convert the result at the very end to float.

  2. take only 24 random bits and convert them to a float x in the range [0, 1) and then return -log(1 - x).

I have not checked if std::exponential_distribution<double> has the same problem: For float on the average 1 out of 224 (~107) draws returns "inf", which is easily confirmed. For double on the average 1 out of 253 (~1016) draws might return "inf", which I have not tested.

Marshall:
I don't think the problem is in std::exponential_distribution; but rather in generate_canonical.

Consider:

std::mt19937_64 mt2(1);
mt2.discard(517517);
std::cout << std::hexfloat << std::generate_canonical<float, std::numeric_limits<float>::digits>(mt2) << std::endl;
std::cout << std::hexfloat << std::generate_canonical<float, std::numeric_limits<float>::digits>(mt2) << std::endl;
std::cout << std::hexfloat << std::generate_canonical<float, std::numeric_limits<float>::digits>(mt2) << std::endl;

which outputs:

0x1.962532p-2
0x1p+0
0x1.20d0cap-3
but generate_canonical is defined to return a result in the range [0, 1).

[2015-10, Kona Saturday afternoon]

Options:

WEB: The one thing we cannot tolerate is any output range other than [0, 1).

WEB: I believe there may be a documented algorithm for the generator, and perhaps it's possible to discover en-route that the algorithm produces the wrong result and fix it.

MC: No. I analyzed this once, and here it is: the algorithm is in [rand.util.canonical], and it's all fine until p5. The expression S/R^k is mathematically less than one, but it may round to one.

GR: Could we change the rounding mode for the computation?

HH: No, because the rounding mode is global, not thread-local.

AM: SG1 wants to get rid of the floating point environment.

STL: The problem is that the standard specifies the implementation, and the implementation doesn't work.

MC: I'm not sure if nudging it down will introduce a subtle bias.

EF: I worry about how the user's choice of floating point environment affects the behaviour.

MS offers to run the topic past colleagues.

MC: Will set the status to open. STL wants to rename the issue. WEB wants to be able to find the issue by its original name still.

Mike Spertus to run the options past his mathematical colleagues, and report back.

Proposed resolution:


2525. [fund.ts.v2] get_memory_resource should be const and noexcept

Section: 99 [fund.ts.v2::func.wrap.func], 99 [fund.ts.v2::futures.promise], 99 [fund.ts.v2::futures.task] Status: New Submitter: Tim Song Opened: 2015-08-04 Last modified: 2015-10-27

Priority: 3

View other active issues in [fund.ts.v2::func.wrap.func].

View all other issues in [fund.ts.v2::func.wrap.func].

View all issues with New status.

Discussion:

Addresses: fund.ts.v2

There doesn't seem to be any reason why this member function cannot be called on a const object, or why it would ever throw. I discussed this with Pablo Halpern, the author of N3916, and he agrees that this appears to have been an oversight.

[2015-10-26]

Daniel adjusts wording to lib. fund. v2.

Proposed resolution:

This wording is relative to N4529.

  1. Edit each of the synposes in 99 [func.wrap.func], 99 [futures.promise], and 99 [futures.task] as indicated:

    pmr::memory_resource* get_memory_resource() const noexcept;
    

2527. [fund.ts.v2] ALLOCATOR_OF for function::operator= has incorrect default

Section: 99 [fund.ts.v2::func.wrap.func.con] Status: New Submitter: Tim Song Opened: 2015-08-04 Last modified: 2015-10-27

Priority: 3

View all other issues in [fund.ts.v2::func.wrap.func.con].

View all issues with New status.

Discussion:

Addresses: fund.ts.v2

According to the table in [memory.type.erased.allocator], if no allocator argument is specified at the time of construction, the memory resource pointer used is the value of experimental::pmr::get_default_resource() at the time of construction.

Yet in 99 [func.wrap.func.con], ALLOCATOR_OF is specified to return allocator<char>() if no allocator was specified at the time of construction, which seems incorrect, especially as the user can change the default memory resource pointer to something other than new_delete_resource().

[2015-10-26]

Daniel adjusts wording to lib. fund. v2.

Proposed resolution:

This wording is relative to N4529.

  1. Edit 99 [func.wrap.func.con]/p2 as indicated:

    -2- In the following descriptions, let ALLOCATOR_OF(f) be the allocator specified in the construction of function f, or allocator<char>()the value of experimental::pmr::get_default_resource() at the time of the construction of f if no allocator was specified.


2528. Order of std::tuple construction unspecified

Section: 20.5.2.1 [tuple.cnstr] Status: New Submitter: Brian Rodriguez Opened: 2015-08-25 Last modified: 2015-09-14

Priority: 3

View other active issues in [tuple.cnstr].

View all other issues in [tuple.cnstr].

View all issues with New status.

Discussion:

The std::tuple order of element construction is unspecified. It is either in the same order of the type list or in reverse.

Consider the following program:

#include <iostream>
#include <tuple>

struct X 
{
  X(int) { std::cout << "X constructor\n"; }
};

struct Y 
{
  Y(int) { std::cout << "Y constructor\n"; }
};

int main()
{
  std::tuple<X, Y> t(1, 2);
}

Here is a link to two sample compilations. The first uses libstdc++ and constructs in reverse order, and the second uses libc++ and constructs in in-order.

A std::tuple mimics both a struct and type-generic container and should thus follow their standards. Construction is fundamentally different from a function call, and it has been historically important for a specific order to be guaranteed; namely: whichever the developer may decide. Mandating construction order will allow developers to reference younger elements later on in the chain as well, much like a struct allows you to do with its members.

There are implementation issues as well. Reversed lists will require unnecessary overhead for braced-initializer-list initialization. Since lists are evaluated from left to right, the initializers must be placed onto the stack to respect the construction order. This issue could be significant for large tuples, deeply nested tuples, or tuples with elements that require many constructor arguments.

I propose that the std::tuple<A, B, ..., Y, Z>'s constructor implementation be standardized, and made to construct in the same order as its type list e.g. A{}, B{}, ..., Y{}, Z{}.

Daniel:

When N3140 became accepted, wording had been added that gives at least an indication of requiring element initialization in the order of the declaration of the template parameters. This argumentation can be based on 20.5.2.1 [tuple.cnstr] p3 (emphasize mine):

-3- In the constructor descriptions that follow, let i be in the range [0,sizeof...(Types)) in order, Ti be the ith type in Types, and Ui be the ith type in a template parameter pack named UTypes, where indexing is zero-based.

But the current wording needs to be improved to make that intention clearer and an issue like this one is necessary to be sure that the committee is agreeing (or disagreeing) with that intention, especially because N3140 didn't really point out the relevance of the element construction order in the discussion, and because not all constructors explicitly refer to the ordered sequence of numbers generated by the variable i (The move constructor does it right, but most other don't do that).

Proposed resolution:


2529. Assigning to enable_shared_from_this::__weak_this twice

Section: 20.11.2.5 [util.smartptr.enab] Status: Tentatively Resolved Submitter: Jonathan Wakely Opened: 2015-08-26 Last modified: 2016-05-17

Priority: 3

View other active issues in [util.smartptr.enab].

View all other issues in [util.smartptr.enab].

View all issues with Tentatively Resolved status.

Discussion:

It is unclear what should happen if a pointer to an object with an enable_shared_from_this base is passed to two different shared_ptr constructors.

#include <memory>

using namespace std;

int main()
{
  struct X : public enable_shared_from_this<X> { };
  auto xraw = new X;
  shared_ptr<X> xp1(xraw);  // #1
  {
    shared_ptr<X> xp2(xraw, [](void*) { });  // #2
  }
  xraw->shared_from_this();  // #3
}

This is similar to LWG 2179, but involves no undefined behaviour due to the no-op deleter, and the question is not whether the second shared_ptr should share ownership with the first, but which shared_ptr shares ownership with the enable_shared_from_this::__weak_this member.

With all three of the major std::shared_ptr implementations the xp2 constructor modifies the __weak_this member so the last line of the program throws bad_weak_ptr, even though all the requirements on the shared_from_this() function are met (20.11.2.5 [util.smartptr.enab])/7:

Requires: enable_shared_from_this<T> shall be an accessible base class of T. *this shall be a subobject of an object t of type T. There shall be at least one shared_ptr instance p that owns &t.

Boost doesn't update __weak_this, leaving it sharing with xp1, so the program doesn't throw. That change was made to boost::enable_shared_from_this because someone reported exactly this issue as a bug, see Boost issue 2584.

On the reflector Peter Dimov explained that there are real-world use cases that rely on the Boost behaviour, and none which rely on the behaviour of the current std::shared_ptr implementations. We should specify the behaviour of enable_shared_from_this more precisely, and resolve this issue one way or another.

[2016-03-16, Alisdair comments]

This issues should be closed as Resolved by paper p0033r1 at Jacksonville.

Proposed resolution:


2530. Clarify observable side effects of releasing a shared state

Section: 30.6.4 [futures.state] Status: New Submitter: Agustín K-ballo Bergé Opened: 2015-09-03 Last modified: 2015-10-20

Priority: 3

View all other issues in [futures.state].

View all issues with New status.

Discussion:

When a shared-state is released, it may be necessary to execute user defined code for the destructor of a stored value or exception. It is unclear whether the execution of said destructor constitutes an observable side effect.

While discussing N4445 in Lenexa, Nat Goodspeed pointed out that 30.6.4 [futures.state]/5.1 does not explicitly mention the destruction of the result, so implementations should be allowed to release (or reuse) a shared state ahead of time under the "as-if" rule.

The standard should clarify whether the execution of destructors is a visible side effect of releasing a shared state.

Proposed resolution:


2531. future::get should explicitly state that the shared state is released

Section: 30.6.6 [futures.unique_future] Status: New Submitter: Agustín K-ballo Bergé Opened: 2015-09-03 Last modified: 2015-10-20

Priority: 3

View other active issues in [futures.unique_future].

View all other issues in [futures.unique_future].

View all issues with New status.

Discussion:

The standard is usually very explicit on when a shared state is released, except for future::get for which it only states valid() == false as a postcondition.

Proposed resolution:

This wording is relative to N4527.

  1. Modify 30.6.6 [futures.unique_future] as indicated:

    R future::get();
    R& future<R&>::get();
    void future<void>::get();
    

    -14- Note: as described above, the template and its two required specializations differ only in the return type and return value of the member function get.

    -15- Effects:

    • wait()s until the shared state is ready, then retrieves the value stored in the shared state.;

    • releases any shared state (30.6.4 [futures.state]).

    […]


2532. Satisfying a promise at thread exit

Section: 30.6.5 [futures.promise] Status: New Submitter: Agustín K-ballo Bergé Opened: 2015-09-03 Last modified: 2015-10-20

Priority: 3

View other active issues in [futures.promise].

View all other issues in [futures.promise].

View all issues with New status.

Discussion:

promise::set_value_at_thread_exit and promise::set_exception_at_thread_exit operate on a shared state at thread exit, without making the thread participate in the ownership of such shared state.

Consider the following snippet:

std::promise<int>{}.set_value_at_thread_exit(42);

Arguably, since the promise abandons its shared state without actually making it ready, a broken_promise error condition should be stored in the shared state. Implementations diverge, they either crash at thread exit by dereferencing an invalid pointer, or keep the shared state around until thread exit.

Proposed resolution:


2533. [concurr.ts] Constrain threads where future::then can run a continuation

Section: 99 [concurr.ts::futures.unique_future] Status: SG1 Submitter: Agustín K-ballo Bergé Opened: 2015-09-03 Last modified: 2015-10-20

Priority: Not Prioritized

View other active issues in [concurr.ts::futures.unique_future].

View all other issues in [concurr.ts::futures.unique_future].

View all issues with SG1 status.

Discussion:

Addresses: concurr.ts

In N4538, the continuation given to future::then can be run "on an unspecified thread of execution". This is too broad, as it allows the continuation to be run on the main thread, a UI thread, or any other thread. In comparison, functions given to async run "as if in a new thread of execution", while the Parallelism TS gives less guarantees by running "in either the invoking thread or in a thread implicitly created by the library to support parallel algorithm execution". The threads on which the continuation given to future::then can run should be similarly constrained.

Proposed resolution:


2534. Constrain rvalue stream operators

Section: 27.7.3.9 [ostream.rvalue], 27.7.2.6 [istream.rvalue] Status: New Submitter: Robert Haberlach Opened: 2015-09-08 Last modified: 2015-09-25

Priority: 3

View other active issues in [ostream.rvalue].

View all other issues in [ostream.rvalue].

View all issues with New status.

Discussion:

The rvalue stream insertion and extraction operators should be constrained to not participate in overload resolution unless the expression they evaluate is well-formed. Programming code that tests the validity of stream insertions (or extractions) using SFINAE can result in false positives, as the present declarations accept virtually any right-hand side argument. Moreover, there is no need for pollution of the candidate set with ill-formed specializations.

Proposed resolution:

This wording is relative to N4527.

  1. Modify 27.7.3.9 [ostream.rvalue] as indicated:

    template <class charT, class traits, class T>
      basic_ostream<charT, traits>&
      operator<<(basic_ostream<charT, traits>&& os, const T& x);
    

    -1- Effects: os << x

    -2- Returns: os

    -?- Remarks: This overload shall not participate in overload resolution unless the expression os << x, considered as an unevaluated operand, is well-formed.

  2. Modify 27.7.2.6 [istream.rvalue] as indicated:

    template <class charT, class traits, class T>
      basic_istream<charT, traits>&
      operator>>(basic_istream<charT, traits>&& is, T& x);
    

    -1- Effects: is >> x

    -2- Returns: is

    -?- Remarks: This overload shall not participate in overload resolution unless the expression is >> x, considered as an unevaluated operand, is well-formed.


2535. Inconsistency between ostream::write and ostream::operator<<

Section: 21.3.2.9 [string.io], 27.7.3.7 [ostream.unformatted] Status: New Submitter: Marshall Clow Opened: 2015-09-10 Last modified: 2015-09-14

Priority: 2

View all other issues in [string.io].

View all issues with New status.

Discussion:

Consider the following program:

#include <iostream>
#include <ostream>
#include <string>

template <class CharT>
class testbuf : public std::basic_streambuf<CharT> 
{
public:
  testbuf() {}

protected:
  virtual std::streamsize xsputn(const CharT *s, std::streamsize n)
  {
    std::cout << "xsputn('" << s << "', " << n << ")\n";
    return n;
  }
};

int main () 
{
  testbuf<char> sb;
  std::ostream os (&sb);
  
  std::string s1{"abc"};
  os.write(s1.data(), s1.size());
  
  os.write(s1.data(), 0);
  
  std::string s2{"def"};
  os << s2;
  
  std::string s3{""};
  os << s3;
  
  os << "";
}

What should it print?

libc++:

xsputn('abc', 3)
xsputn('def', 3)

libstdc++:

xsputn('abc', 3)
xsputn('abc', 0)
xsputn('def', 3)
xsputn('', 0)
xsputn('', 0)

VS:

xsputn('abc', 3)
xsputn('def', 3)
xsputn('', 0)
xsputn('', 0)

21.3.2.9 [string.io]/5 seems to say that an implementation is required to call sputn (which calls xsputn) even if there's nothing to output (in the case of ostream::operator<<(basic_string)).

27.7.3.7 [ostream.unformatted]/5.1 implies that it's OK to not call sputn if there's nothing to output (in the case of ostream::write)

Backstory: A user has a ostream with a subclass of basic_streambuf. it creates an output file on first write. Sometimes, he calls ostream::write(p, 0), and expects this to create the file. This doesn't work in libc++, and then he pointed out the inconsistency between operator<< and write.

For reference to a bug report see here.

There are two obvious possible resolutions:

  1. a) require all output functions to call sputn, even if there are no characters to output. In practice, this reduces to "string-like" things which are empty (string, string_view, char*, etc), and write(ptr, len).

  2. b) remove the requirement that ostream::operator<< call sputn when there are no characters to output.

Proposed resolution:


2536. What should <complex.h> do?

Section: D.4 [depr.c.headers] Status: New Submitter: Richard Smith Opened: 2015-09-10 Last modified: 2015-09-14

Priority: 2

View all other issues in [depr.c.headers].

View all issues with New status.

Discussion:

LWG issue 1134 removed the resolution of LWG 551, leaving an incorrect specification for the behavior of <complex.h>. This header is currently required to make std::complex (and associated functions) visible in the global namespace, but should not be so required.

Proposed resolution:

This wording is relative to N4527.

  1. Add a new paragraph before D.4 [depr.c.headers]/2:

    -?- The header <complex.h> behaves as if it simply includes the header <ccomplex>.

  2. Change in D.4 [depr.c.headers]/2:

    -2- Every other C header, each of which has a name of the form name.h, behaves as if each name placed in the standard library namespace by the corresponding cname header is placed within the global namespace scope. It is unspecified whether these names are first declared or defined within namespace scope (3.3.6) of the namespace std and are then injected into the global namespace scope by explicit using-declarations (7.3.3).


2540. unordered_multimap::insert hint iterator

Section: 23.2.5 [unord.req] Status: New Submitter: Isaac Hier Opened: 2015-09-16 Last modified: 2015-10-20

Priority: 3

View other active issues in [unord.req].

View all other issues in [unord.req].

View all issues with New status.

Discussion:

I have been wondering about the C++ standard requirements regarding the hint iterator for insertion into an unordered_multimap (and I imagine a similar question could be asked of unordered_map, but I have not researched that topic). As far as I can tell, it seems perfectly valid for an implementation to allow only valid dereferencable iterators to be used as the hint argument for this member function. If that is correct, it means that one could not expect the end iterator to be used as a valid hint nor could one use the begin iterator of an empty unordered_multimap as the hint. However, this essentially precludes all uses of inserter on an empty unordered_multimap seeing as the inserter functor requires that a hint iterator be passed to its constructor.

Howard Hinnant:

The intent of the standard is that the iterator produced from container c by c.end() is a valid (but non-dereferenceable) iterator into container c. It is reachable by every other iterator into c.

It appears to me that you and the Bloomberg implementation have fallen victim to a type-o in the Unordered associative container requirements, Table 102. The row containing:

a.insert(q, t);

should read instead:

a.insert(p, t);

The distinction is that p is valid, and q is both valid and dereferenceable. The correction of this type-o would make unordered container insert consistent with unordered emplace_hint, associative insert, and associative emplace_hint.

Proposed resolution:

Change the insert-with-hint row in Table 102 Unordered associative container requirements like so:

a.insert(qp, t);
iterator
Requires: If t is a non-const
...
Average Case
...

2541. [parallel.ts] Headers for ExecutionPolicy algorithm overloads

Section: 99 [parallel.ts::parallel.alg.overloads] Status: New Submitter: Tim Song Opened: 2015-09-26 Last modified: 2015-10-20

Priority: 1

View all issues with New status.

Discussion:

Addresses: parallel.ts

99 [parallel.alg.overloads] provides parallel algorithm overloads for many algorithms in the standard library, but I can't find any normative wording specifying which headers these new overloads live in. Presumably, if the original algorithm is in <meow>, the new overloads should be in <experimental/meow>.

Proposed resolution:


2543. LWG 2148 (hash support for enum types) seems under-specified

Section: 20.14.14 [unord.hash] Status: Open Submitter: Ville Voutilainen Opened: 2015-09-27 Last modified: 2016-06-20

Priority: 2

View other active issues in [unord.hash].

View all other issues in [unord.hash].

View all issues with Open status.

Discussion:

The rationale in issue 2148 says:

This proposed resolution doesn't specify anything else about the primary template, allowing implementations to do whatever they want for non-enums: static_assert nicely, explode horribly at compiletime or runtime, etc.

libc++ seems to implement it by defining the primary template and static_asserting is_enum inside it. However, that brings forth a problem; there are reasonable SFINAE uses broken by it:

#include <type_traits>
#include <functional>

class S{}; // No hash specialization

template<class T>
auto f(int) -> decltype(std::hash<T>(), std::true_type());

template<class T>
auto f(...) -> decltype(std::false_type());

static_assert(!decltype(f<S>(0))::value, "");

MSVC doesn't seem to accept that code either.

There is a way to implement LWG 2148 so that hash for enums is supported without breaking that sort of SFINAE uses:

  1. Derive the main hash template from a library-internal uglified-named base template that takes a type and a bool, pass as argument for the base the result of is_enum.

  2. Partially specialize that base template so that the false-case has a suitable set of private special member function declarations so that it's not an aggregate nor usable in almost any expression.

[2015-10, Kona Saturday afternoon]

EricWF to come back with wording; move to Open

[2016-05-08, Eric Fiselier & Ville provide wording]

[2016-05-25, Tim Song comments]

I see two issues with this P/R:

  1. "for which neither the library nor the user provides an explicit specialization" should probably be "for which neither the library nor the user provides an explicit or partial specialization".

  2. Saying that the specialization "is not DefaultConstructible nor MoveAssignable" is not enough to guarantee that common SFINAE uses will work. Both of those requirements have several parts, and it's not too hard to fail only some of them. For instance, not meeting the assignment postcondition breaks MoveAssignable, but is usually not SFINAE-detectible. And for DefaultConstructible, it's easy to write something in a way that breaks T() but not T{} (due to aggregate initialization in the latter case).

[2016-06-14, Daniel comments]

The problematic part of the P/R is that it describes constraints that would be suitable if they were constraints for user-code, but they are not suitable as requirements imposed on implementations to provide certain guarantees for clients of the Library. The guarantees should be written in terms of testable compile-time expressions, e.g. based on negative results of is_default_constructible<hash<X>>::value, std::is_copy_constructible<hash<X>>::value, and possibly also std::is_destructible<hash<X>>::value. How an implementation realizes these negative results shouldn't be specified, though, but the expressions need to be well-formed and well-defined.

Proposed resolution:

This wording is relative to N4582.

  1. Insert a new paragraph after 20.14.14 [unord.hash]/2

    -2- The template specializations shall meet the requirements of class template hash (20.12.14).

    -?- For any type that is not of integral or enumeration type, or for which neither the library nor the user provides an explicit specialization of the class template hash, the specialization of hash does not meet any of the Hash requirements, and is not DefaultConstructible nor MoveAssignable. [Note: this means that the specialization of hash exists, but any attempts to use it as a Hash will be ill-formed. — end note]


2544. istreambuf_iterator(basic_streambuf<charT, traits>* s) effects unclear when s is 0

Section: 24.6.3.2 [istreambuf.iterator.cons] Status: New Submitter: S. B. Tam Opened: 2015-10-05 Last modified: 2015-10-23

Priority: 3

View all issues with New status.

Discussion:

N4527 24.6.3.2 [istreambuf.iterator.cons] does not mention what the effect of calling istreambuf_iterator(basic_streambuf<charT, traits>* s) is when s is a null pointer. It should be made clear that this case is well-formed and the result is a end-of-stream iterator.

Daniel:

According to 24.6.3 [istreambuf.iterator] p1:

[…] The default constructor istreambuf_iterator() and the constructor istreambuf_iterator(0) both construct an end-of-stream iterator object suitable for use as an end-of-range. […]

This indicates that the described constructor creates an end-of-stream iterator, but this wording is part of the introductory wording and I recommend to make 24.6.3.2 [istreambuf.iterator.cons] clearer, because the existing specification is already flawed, e.g. it never specifies when and how the exposition-only-member sbuf_ is initialized. The proposed wording below attempts to solve these problems as well.

Previous resolution [SUPERSEDED]:

This wording is relative to N4527.

  1. Change 24.6.3.2 [istreambuf.iterator.cons] as indicated:

    [Editorial note: The proposed wording changes also performs some editorial clean-up of the existing mismatches of the declarations in the class template synopsis and the individual member specifications. The below wording intentionally does not say anything about the concrete value of sbuf_ for end-of-stream iterator values, because that was never specified before; in theory, this could be some magic non-null pointer that can be used in constant expressions. But the wording could be drastically simplified by requiring sbuf_ to be a null pointer for an end-of-stream iterator value, since I have not yet seen any implementation where this requirement does not hold. — end editorial note]

    constexpr istreambuf_iterator() noexcept;
    

    -1- Effects: Constructs the end-of-stream iterator.

    istreambuf_iterator(basic_istream<charT,traits>istream_type& s) noexcept;
    istreambuf_iterator(basic_streambuf<charT,traits>* s) noexcept;
    

    -2- Effects: If s.rdbuf() is a null pointer, constructs an end-of-stream iterator; otherwise initializes sbuf_ with s.rdbuf() and constructs an istreambuf_iterator that uses the streambuf_type object *sbuf_Constructs an istreambuf_iterator<> that uses the basic_streambuf<> object *(s.rdbuf()), or *s, respectively. Constructs an end-of-stream iterator if s.rdbuf() is null.

    istreambuf_iterator(streambuf_type* s) noexcept;
    

    -?- Effects: If s is a null pointer, constructs an end-of-stream iterator; otherwise initializes sbuf_ with s and constructs an istreambuf_iterator that uses the streambuf_type object *sbuf_.

    istreambuf_iterator(const proxy& p) noexcept;
    

    -3- Effects: Initializes sbuf_ with p.sbuf_ and constructs an istreambuf_iterator that uses the streambuf_type object *sbuf_Constructs a istreambuf_iterator<> that uses the basic_streambuf<> object pointed to by the proxy object's constructor argument p.

[2015-10-20, Daniel provides alternative wording]

Proposed resolution:

This wording is relative to N4527.

  1. Change 24.6.3.2 [istreambuf.iterator.cons] as indicated:

    [Editorial note: The proposed wording changes also performs some editorial clean-up of the existing mismatches of the declarations in the class template synopsis and the individual member specifications. The below wording is simplified by requiring sbuf_ to be a null pointer for an end-of-stream iterator value, since I have not yet seen any implementation where this requirement does not hold. Even if there were such an implementation, this would still be conforming, because concrete exposition-only member values are not part of public API. — end editorial note]

    For each istreambuf_iterator constructor in this section, an end-of-stream iterator is constructed if and only if the exposition-only member sbuf_ is initialized with a null pointer value.

    constexpr istreambuf_iterator() noexcept;
    

    -1- Effects: Initializes sbuf_ with nullptrConstructs the end-of-stream iterator.

    istreambuf_iterator(basic_istream<charT,traits>istream_type& s) noexcept;
    istreambuf_iterator(basic_streambuf<charT,traits>* s) noexcept;
    

    -2- Effects: Initializes sbuf_ with s.rdbuf()Constructs an istreambuf_iterator<> that uses the basic_streambuf<> object *(s.rdbuf()), or *s, respectively. Constructs an end-of-stream iterator if s.rdbuf() is null.

    istreambuf_iterator(streambuf_type* s) noexcept;
    

    -?- Effects: Initializes sbuf_ with s.

    istreambuf_iterator(const proxy& p) noexcept;
    

    -3- Effects: Initializes sbuf_ with p.sbuf_Constructs a istreambuf_iterator<> that uses the basic_streambuf<> object pointed to by the proxy object's constructor argument p.


2546. Implementability of locale-sensitive UnicodeEscapeSequence matching

Section: 28.13 [re.grammar] Status: New Submitter: Hubert Tong Opened: 2015-10-08 Last modified: 2015-10-20

Priority: 4

View other active issues in [re.grammar].

View all other issues in [re.grammar].

View all issues with New status.

Discussion:

In 28.13 [re.grammar] paragraph 2:

basic_regex member functions shall not call any locale dependent C or C++ API, including the formatted string input functions. Instead they shall call the appropriate traits member function to achieve the required effect.

Yet, the required interface for a regular expression traits class (28.3 [re.req]) does not appear to have any reliable method for determining whether a character as encoded for the locale associated with the traits instance is the same as a character represented by a UnicodeEscapeSequence, e.g., assuming a sane ru_RU.koi8r locale:

#include <stdio.h>
#include <stdlib.h>
#include <regex>

const char data[] = "\xB3";
const char matchCyrillicCaptialLetterYo[] = R"(\u0401)";

int main(void) 
{
  try {
    std::regex myRegex;
    myRegex.imbue(std::locale("ru_RU.koi8r"));

    myRegex.assign(matchCyrillicCaptialLetterYo, std::regex_constants::ECMAScript);
    printf("(%s)\n", std::regex_replace(std::string(data), myRegex, std::string("E")).c_str());

    myRegex.assign("[[:alpha:]]", std::regex_constants::ECMAScript);
    printf("(%s)\n", std::regex_replace(std::string(data), myRegex, std::string("E")).c_str());
  } catch (std::regex_error& e) {
    abort();
  }
  return 0;
}

The implementation I tried prints:

(Ё)
(E)

Which means that the character class matching worked, but not the matching to the UnicodeEscapeSequence.

Proposed resolution:


2547. Container requirements (and other library text) should say "strict total order", not just "total order"

Section: 20.14.6 [comparisons], 23.2.1 [container.requirements.general], 30.3.1.1 [thread.thread.id] Status: New Submitter: Matt Austern Opened: 2015-10-08 Last modified: 2015-10-20

Priority: 3

View other active issues in [comparisons].

View all other issues in [comparisons].

View all issues with New status.

Discussion:

A number of places in the library, including 20.14.6 [comparisons]/14, the Optional container requirements in 23.2.1 [container.requirements.general], and 30.3.1.1 [thread.thread.id]/8, use the phrase "total order". Unfortunately, that phrase is ambiguous. In mathematics, the most common definition is that a relation is a total order if it's total, transitive, and antisymmetric in the sense that x≤y ∧ y≤x ⇒ x=y. What we really want is a strict total order: a relation < is a strict total order if it's total, transitive, and antisymmetric in the sense that exactly one of x<y, y<x, and x=y holds.

The non-normative note in 25.5 [alg.sorting]/4 correctly uses the phrase "strict total ordering" rather than simply "total ordering".

We could address this issue by replacing "total order" with "strict total order" everywhere it appears, since I think there are no cases where we actually want a non-strict total order, or we could add something in Clause 17 saying that we always mean strict total order whenever we say total order.

Proposed resolution:


2548. Missing vfscanf from <cstdio>

Section: 27.11 [c.files] Status: New Submitter: Richard Smith Opened: 2015-10-09 Last modified: 2016-06-28

Priority: 3

View all other issues in [c.files].

View all issues with New status.

Discussion:

C's vfscanf function is not present in C++'s <cstdio>, and presumably should be. It looks like this is the only missing member of C's [v]{f,s,sn}[w]{printf,scanf} family.

Proposed resolution:

This wording is relative to N4527.

Modify Table 133 as follows:

ungetc
vfprintf
vfscanf
vprintf
vscanf


2556. Wide contract for future::share()

Section: 30.6.6 [futures.unique_future] Status: New Submitter: Agustín K-ballo Bergé Opened: 2015-11-05 Last modified: 2016-03-06

Priority: 3

View other active issues in [futures.unique_future].

View all other issues in [futures.unique_future].

View all issues with New status.

Discussion:

future::share() is not noexcept, it has a narrow contact requiring valid() as per the blanket wording in 30.6.6 [futures.unique_future] p3. Its effects, however, are return shared_future<R>(std::move(*this)), which is noexcept as it has a wide contract. If the source future isn't valid then the target shared_future simply won't be valid either. There appears to be no technical reason preventing future::share() from having a wide contract, and thus being noexcept.

Proposed resolution:

This wording is relative to N4567.

  1. Change 30.6.6 [futures.unique_future] as indicated:

    -3- The effect of calling any member function other than the destructor, the move-assignment operator, share, or valid on a future object for which valid() == false is undefined. [Note: Implementations are encouraged to detect this case and throw an object of type future_error with an error condition of future_errc::no_state. — end note]

    namespace std {
      template <class R>
      class future {
      public:
        […]
        shared_future<R> share() noexcept;
        […]
      };
    }
    

    […]

    shared_future<R> share() noexcept;
    

    -12- Returns: shared_future<R>(std::move(*this)).

    -13- Postcondition: valid() == false.

    […]


2561. [fund.ts.v2] Incorrect exception specifications for 'swap' in C++ Extensions for Library Fundamentals

Section: 99 [fund.ts.v2::optional.object.swap], 99 [fund.ts.v2::propagate_const.modifiers] Status: New Submitter: Daniel Krügler Opened: 2015-11-14 Last modified: 2016-03-06

Priority: 3

View all issues with New status.

Discussion:

Addresses: fund.ts.v2

As pointed out in N4511, the Library fundamentals are affected by a similar problem as described in LWG 2456. First, it is caused by optional's member swap (99 [optional.object.swap]):

void swap(optional<T>& rhs) noexcept(see below);

with

The expression inside noexcept is equivalent to:

is_nothrow_move_constructible_v<T> && noexcept(swap(declval<T&>(), declval<T&>()))

Again, the unqualified lookup for swap finds the member swap instead of the result of a normal argument-depending lookup, making this ill-formed.

A second example of such a problem recently entered the arena with the addition of the propagate_const template with another member swap (99 [propagate_const.modifiers]):

constexpr void swap(propagate_const& pt) noexcept(see below);

-2- The constant-expression in the exception-specification is noexcept(swap(t_, pt.t_)).

A working approach is presented in N4511. By adding a new trait to the standard library and referencing this by the library fundamentals (A similar approach had been applied in the file system specification where the quoted manipulator from C++14 had been referred to, albeit the file system specification is generally based on the C++11 standard), optional's member swap exception specification could be rephrased as follows:

The expression inside noexcept is equivalent to:

is_nothrow_move_constructible_v<T> && is_nothrow_swappable_v<T>noexcept(swap(declval<T&>(), declval<T&>()))

and propagate_const's member swap exception specification could be rephrased as follows:

constexpr void swap(propagate_const& pt) noexcept(see below);

-2- The constant-expression in the exception-specification is is_nothrow_swappable_v<T>noexcept(swap(t_, pt.t_)).

[2016-02-20, Ville comments]

Feedback from an implementation:

libstdc++ already applies the proposed resolution for propagate_const, but not for optional.

[2016-02-20, Daniel comments]

A recent paper update has been provided: P0185R0.

[2016-03, Jacksonville]

Add a link to 2456

Proposed resolution:


2562. Consistent total ordering of pointers by comparison functors

Section: 20.14.6 [comparisons] Status: New Submitter: Casey Carter Opened: 2015-11-18 Last modified: 2016-05-22

Priority: 3

View other active issues in [comparisons].

View all other issues in [comparisons].

View all issues with New status.

Discussion:

N4567 20.14.6 [comparisons]/14 specifies that the comparison functors provide a total ordering for pointer types:

For templates greater, less, greater_equal, and less_equal, the specializations for any pointer type yield a total order, even if the built-in operators <, >, <=, >= do not.

It notably does not specify:

All of which are important for sane semantics and provided by common implementations, since the built-in operators provide a total order and the comparison functors yield that same order.

It would be extremely confusing — if not outright insane — for e.g.:

Consistent semantics for the various comparison functors and the built-in operators is so intuitive as to be assumed by most programmers.

Related issues: 2450, 2547.

Previous resolution [SUPERSEDED]:

This wording is relative to N4567.

  1. Alter 20.14.6 [comparisons]/14 to read:

    For templates greater, less, greater_equal, and less_equal, the specializations for any pointer type yield athe same total order, even if the built-in operators <, >, <=, >= do not. The total order shall respect the partial order imposed by the built-in operators.

[2016-05-20, Casey Carter comments and suggests revised wording]

The new proposed wording is attempting to address the issue raised in the 2016-02-04 telecon.

The real issue I'm trying to address here is ensure that "weird" implementations provide the same kind of consistency for pointer orderings as "normal" implementations that use a flat address spaces and have totally ordered <. If a < b is true for int pointers a and b, then less<int*>(a, b), less_equal<int*>(a, b), less<char*>(a, b), less<void*>(a, b), and greater<int*>(b, a) should all hold. I think this wording is sufficient to provide that.

Proposed resolution:

This wording is relative to N4582.

  1. Alter 20.14.6 [comparisons] to read:

    -14- For templates greater, less, greater_equal, and less_equal, the specializations for any pointer type yield athe same total order. That total order is consistent with the partial order imposed by, even if the built-in operators <, >, , and > do not. [Note: When a < b is well-defined for pointers a and b of type P, this implies (a < b) == less<P>(a, b), (a > b) == greater<P>(a, b), and so forth. — end note] For template specializations greater<void>, less<void>, greater_equal<void>, and less_equal<void>, if the call operator calls a built-in operator comparing pointers, the call operator yields a total order.


2563. LWG 2259 relaxes requirements, perhaps unintentionally

Section: 17.6.5.5 [member.functions] Status: New Submitter: Ville Voutilainen Opened: 2015-11-29 Last modified: 2016-05-22

Priority: 2

View other active issues in [member.functions].

View all other issues in [member.functions].

View all issues with New status.

Discussion:

The combination of 17.6.5.5 [member.functions], paragraphs 2 and 3 that LWG 2259 does seems to drop a requirement that any call behaves as if no overloads were added. Paragraph 3 used to say "A call to a member function signature described in the C ++ standard library behaves as if the implementation declares no additional member function signatures." whereas the new wording says "provided that any call to the member function that would select an overload from the set of declarations described in this standard behaves as if that overload were selected."

This can be read as meaning that if there's no default constructor specified, like for instance for std::ostream, an implementation is free to add it. It can also be read as meaning that an implementation is free to add any overloads that wouldn't change the overload resolution result of any call expression that would select a specified overload. That's vastly different from allowing extensions that add new functions rather than new overloads.

Was this relaxation intentional?

[2016-04, Issues Telecon]

Ville provides a motivating example.

#include <iostream>

class my_stream : std::ostream
{
};

int main()
{
    my_stream ms;
}

This example is accepted by libstdc++, msvc rejects it, and clang+libc++ segfault on melpon.org/wandbox o_O. An earlier clang+libc++ just accepts it. I don't think the implementation divergence is caused by the acceptance of the referred-to 2259, but it certainly seems to increasingly bless the implementation divergence.

[2016-05 Issues Telecom]

This is related to issue 2695.

Proposed resolution:


2564. [fund.ts.v2] std::experimental::function constructors taking allocator arguments may throw exceptions

Section: 99 [fund.ts.v2::func.wrap.func] Status: New Submitter: Tim Song Opened: 2015-12-05 Last modified: 2016-02-05

Priority: 3

View other active issues in [fund.ts.v2::func.wrap.func].

View all other issues in [fund.ts.v2::func.wrap.func].

View all issues with New status.

Discussion:

Addresses: fund.ts.v2

[This is essentially LWG 2370, but deals with the fundamentals TS version rather than the one in the standard]

In 99 [func.wrap.func] of library fundamentals TS, the constructors

template<class A> function(allocator_arg_t, const A&) noexcept;
template<class A> function(allocator_arg_t, const A&, nullptr_t) noexcept;

must type-erase and store the provided allocator, since the operator= specification requires using the "allocator specified in the construction of" the std::experimental::function object. This may require a dynamic allocation and so cannot be noexcept. Similarly, the following constructors

template<class A> function(allocator_arg_t, const A&, const function&); 
template<class A> function(allocator_arg_t, const A&, function&&);
template<class F, class A> function(allocator_arg_t, const A&, F);

cannot satisfy the C++14 requirement that they "shall not throw exceptions if [the function object to be stored] is a callable object passed via reference_wrapper or a function pointer" if they need to type-erase and store the allocator.

Proposed resolution:

This wording is relative to N4562.

  1. Edit 99 [func.wrap.func], class template function synopsis, as follows:

    namespace std {
      namespace experimental {
      inline namespace fundamentals_v2 {
    
        […]
    
        template<class R, class... ArgTypes>
        class function<R(ArgTypes...)> {
        public:    
          […]
          template<class A> function(allocator_arg_t, const A&) noexcept;
          template<class A> function(allocator_arg_t, const A&,
            nullptr_t) noexcept;
          […]
        };
    
        […]
    
      } // namespace fundamentals_v2
      } // namespace experimental
    
      […]
    
    } // namespace std
    
  2. Insert the following paragraphs after 99 [func.wrap.func.con]/1:

    [Drafting note: This just reproduces the wording from C++14 with the "shall not throw exceptions for reference_wrapper/function pointer" provision deleted. — end drafting note]

    -1- When a function constructor that takes a first argument of type allocator_arg_t is invoked, the second argument is treated as a type-erased allocator (8.3). If the constructor moves or makes a copy of a function object (C++14 §20.9), including an instance of the experimental::function class template, then that move or copy is performed by using-allocator construction with allocator get_memory_resource().

    template <class A> function(allocator_arg_t, const A& a);
    template <class A> function(allocator_arg_t, const A& a, nullptr_t);
    

    -?- Postconditions: !*this.

    template <class A> function(allocator_arg_t, const A& a, const function& f);
    

    -?- Postconditions: !*this if !f; otherwise, *this targets a copy of f.target().

    -?- Throws: May throw bad_alloc or any exception thrown by the copy constructor of the stored callable object. [Note: Implementations are encouraged to avoid the use of dynamically allocated memory for small callable objects, for example, where f's target is an object holding only a pointer or reference to an object and a member function pointer. — end note]

    template <class A> function(allocator_arg_t, const A& a, function&& f);
    

    -?- Effects: If !f, *this has no target; otherwise, move-constructs the target of f into the target of *this, leaving f in a valid state with an unspecified value.

    template <class F, class A> function(allocator_arg_t, const A& a, F f);
    

    -?- Requires: F shall be CopyConstructible.

    -?- Remarks: This constructor shall not participate in overload resolution unless f is Callable (C++14 §20.9.11.2) for argument types ArgTypes... and return type R.

    -?- Postconditions: !*this if any of the following hold:

    • f is a null function pointer value.

    • f is a null member pointer value.

    • F is an instance of the function class template, and !f.

    -?- Otherwise, *this targets a copy of f initialized with std::move(f). [Note: Implementations are encouraged to avoid the use of dynamically allocated memory for small callable objects, for example, where f's target is an object holding only a pointer or reference to an object and a member function pointer. — end note]

    -?- Throws: May throw bad_alloc or any exception thrown by F's copy or move constructor.

    -2- In the following descriptions, let ALLOCATOR_OF(f) be the allocator specified in the construction of function f, or allocator<char>() if no allocator was specified.

    […]


2567. Specification of logical operator traits uses BaseCharacteristic, which is defined only for UnaryTypeTraits and BinaryTypeTraits

Section: 20.15.8 [meta.logical] Status: New Submitter: Tim Song Opened: 2015-12-10 Last modified: 2016-02-07

Priority: 2

View other active issues in [meta.logical].

View all other issues in [meta.logical].

View all issues with New status.

Discussion:

The specification of conjunction and disjunction uses the term BaseCharacteristic, which is problematic in several ways:

Proposed resolution:


2568. [fund.ts.v2] Specification of logical operator traits uses BaseCharacteristic, which is defined only for UnaryTypeTraits and BinaryTypeTraits

Section: 99 [fund.ts.v2::meta.logical] Status: New Submitter: Tim Song Opened: 2015-12-10 Last modified: 2016-02-07

Priority: 2

View other active issues in [fund.ts.v2::meta.logical].

View all other issues in [fund.ts.v2::meta.logical].

View all issues with New status.

Discussion:

Addresses: fund.ts.v2

The specification of conjunction and disjunction uses the term BaseCharacteristic, which is problematic in several ways:

Proposed resolution:


2569. conjunction and disjunction requirements are too strict

Section: 20.15.8 [meta.logical] Status: New Submitter: Tim Song Opened: 2015-12-11 Last modified: 2016-02-07

Priority: 2

View other active issues in [meta.logical].

View all other issues in [meta.logical].

View all issues with New status.

Discussion:

20.15.8 [meta.logical]/2 and /5 impose the following requirement on the arguments of conjunction and disjunction:

Every template type argument shall be usable as a base class and shall have a static data member value which is convertible to bool, is not hidden, and is unambiguously available in the type.

Since the requirement is unconditional, it applies even to type arguments whose instantiation is not required due to short circuiting. This seems contrary to the design intent, expressed in P0013R1, that it is valid to write conjunction_v<is_class<T>, is_foo<T>> even if instantiating is_foo<T>::value is ill-formed for non-class types.

Proposed resolution:


2570. [fund.ts.v2] conjunction and disjunction requirements are too strict

Section: 99 [fund.ts.v2::meta.logical] Status: New Submitter: Tim Song Opened: 2015-12-11 Last modified: 2016-06-28

Priority: 2

View other active issues in [fund.ts.v2::meta.logical].

View all other issues in [fund.ts.v2::meta.logical].

View all issues with New status.

Discussion:

Addresses: fund.ts.v2

99 [meta.logical]/2 and /5 impose the following requirement on the arguments of conjunction and disjunction:

Every template type argument shall be usable as a base class and shall have a static data member value which is convertible to bool, is not hidden, and is unambiguously available in the type.

Since the requirement is unconditional, it applies even to type arguments whose instantiation is not required due to short circuiting. This seems contrary to the design intent, expressed in P0013R1, that it is valid to write conjunction_v<is_class<T>, is_foo<T>> even if instantiating is_foo<T>::value is ill-formed for non-class types.

[Oulu, 2016-06]

Alisdair has a paper in progress addressing this

Proposed resolution:


2578. Iterator requirements should reference iterator traits

Section: 24.2 [iterator.requirements], 24.4.1 [iterator.traits] Status: New Submitter: Ville Voutilainen Opened: 2016-01-05 Last modified: 2016-04-16

Priority: 3

View other active issues in [iterator.requirements].

View all other issues in [iterator.requirements].

View all issues with New status.

Discussion:

See this reflector discussion for background.

24.2 [iterator.requirements] attempts to establish requirements for iterators, but 24.4.1 [iterator.traits]/1 establishes further requirements that must be met in order to author a portable iterator that works with existing implementations. Failing to meet the requirements of the latter will fail to work in practice. The former requirements should reference the latter, normatively.

Proposed resolution:

After [iterator.requirements.general]/5, insert the following new paragraph:

-?- In addition to these requirements, the nested typedefs specified in ([iterator.traits]) shall be provided for the iterator type. [Note: Either the iterator type must provide the typedefs directly (in which case iterator_traits pick them up automatically), or an iterator_traits specialization must provide them. -end note]


2584. <regex> ECMAScript IdentityEscape is ambiguous

Section: 28.13 [re.grammar] Status: New Submitter: Billy O'Neal III Opened: 2016-01-13 Last modified: 2016-03-06

Priority: 2

View other active issues in [re.grammar].

View all other issues in [re.grammar].

View all issues with New status.

Discussion:

Stephan and I are seeing differences in implementation for how non-special characters should be handled in the IdentityEscape part of the ECMAScript grammar. For example:

#include <stdio.h>
#include <iostream>
#ifdef USE_BOOST
#include <boost/regex.hpp>
using namespace boost;
#else
#include <regex>
#endif
using namespace std;

int main() {
  try {
    const regex r("\\z");
    cout << "Constructed \\z." << endl;
    if (regex_match("z", r))
      cout << "Matches z" << endl;
  } catch (const regex_error& e) {
      cout << e.what() << endl;
  }
}

libstdc++, boost, and browsers I tested with (Microsoft Edge, Google Chrome) all happily interpret \z, which otherwise has no meaning, as an identity character escape for the letter z. libc++ and msvc++ say that this is invalid, and throw regex_error with error_escape.

ECMAScript 3 (which is what C++ currently points to) seems to agree with libc++ and msvc++:

IdentityEscape ::
  SourceCharacter but not IdentifierPart

IdentifierPart ::
  IdentifierStart
  UnicodeCombiningMark
  UnicodeDigit
  UnicodeConnectorPunctuation
  \ UnicodeEscapeSequence

IdentifierStart ::
  UnicodeLetter
  $
  _
  \ UnicodeEscapeSequence

But this doesn't make any sense — it prohibits things like \$ which users absolutely need to be able to escape. So let's look at ECMAScript 6. I believe this says much the same thing, but updates the spec to better handle Unicode by referencing what the Unicode standard says is an identifier character:

IdentityEscape ::
  SyntaxCharacter
  /
  SourceCharacter but not UnicodeIDContinue
  
UnicodeIDContinue ::
  any Unicode code point with the Unicode property "ID_Continue", "Other_ID_Continue", or "Other_ID_Start"

However, ECMAScript 6 has an appendix B defining "additional features for web browsers" which says:

IdentityEscape ::
  SourceCharacter but not c

which appears to agree with what libstdc++, boost, and browsers are doing.

What should be the correct behavior here?

Proposed resolution:

This wording is relative to N4567.

  1. Change 28.13 [re.grammar]/3 as indicated:

    -3- The following productions within the ECMAScript grammar are modified as follows:

    ClassAtom ::
      -
      ClassAtomNoDash
      ClassAtomExClass
      ClassAtomCollatingElement
      ClassAtomEquivalence
      
    IdentityEscape ::
      SourceCharacter but not c
    

2587. "Convertible to bool" requirement in conjunction and disjunction

Section: 20.15.8 [meta.logical] Status: New Submitter: Tim Song Opened: 2016-01-18 Last modified: 2016-02-07

Priority: 3

View other active issues in [meta.logical].

View all other issues in [meta.logical].

View all issues with New status.

Discussion:

The specification of conjunction and disjunction in 20.15.8 [meta.logical] p2 and p5 requires Bi::value to be convertible to bool, but nothing in the specification of the actual behavior of the templates, which instead uses the expressions Bi::value == false and Bi::value != false instead, actually requires this conversion.

If the intention of this requirement is to allow implementations to pass Bi::value directly to std::conditional, like the sample implementation in P0013R1:

template<class B1, class B2>
struct and_<B1, B2> : conditional_t<B1::value, B2, B1> { };

then it's insufficient in at least two ways:

  1. Nothing in the specification requires the result of comparing Bi::value with false to be consistent with the result of the implicit conversion. This is similar to LWG 2114, though I don't think the BooleanTestable requirements in that issue's P/R covers Bi::value == false and Bi::value != false.

  2. More importantly, the above implementation is ill-formed for, e.g., std::conjunction<std::integral_constant<int, 2>, std::integral_constant<int, 4>>, because converting 2 to bool is a narrowing conversion that is not allowed for non-type template arguments (see 5.20 [expr.const]/4). (Note that GCC currently doesn't diagnose this error at all, and Clang doesn't diagnose it inside system headers.) It's not clear whether such constructs are intended to be supported, but if they are not, the current wording doesn't prohibit it.

Proposed resolution:


2588. [fund.ts.v2] "Convertible to bool" requirement in conjunction and disjunction

Section: 99 [fund.ts.v2::meta.logical] Status: New Submitter: Tim Song Opened: 2016-01-18 Last modified: 2016-02-07

Priority: 3

View other active issues in [fund.ts.v2::meta.logical].

View all other issues in [fund.ts.v2::meta.logical].

View all issues with New status.

Discussion:

Addresses: fund.ts.v2

The specification of conjunction and disjunction in 99 [meta.logical] p2 and p5 requires Bi::value to be convertible to bool, but nothing in the specification of the actual behavior of the templates, which instead uses the expressions Bi::value == false and Bi::value != false instead, actually requires this conversion.

If the intention of this requirement is to allow implementations to pass Bi::value directly to std::conditional, like the sample implementation in P0013R1:

template<class B1, class B2>
struct and_<B1, B2> : conditional_t<B1::value, B2, B1> { };

then it's insufficient in at least two ways:

  1. Nothing in the specification requires the result of comparing Bi::value with false to be consistent with the result of the implicit conversion. This is similar to LWG 2114, though I don't think the BooleanTestable requirements in that issue's P/R covers Bi::value == false and Bi::value != false.

  2. More importantly, the above implementation is ill-formed for, e.g., std::conjunction<std::integral_constant<int, 2>, std::integral_constant<int, 4>>, because converting 2 to bool is a narrowing conversion that is not allowed for non-type template arguments (see 5.20 [expr.const]/4). (Note that GCC currently doesn't diagnose this error at all, and Clang doesn't diagnose it inside system headers.) It's not clear whether such constructs are intended to be supported, but if they are not, the current wording doesn't prohibit it.

Proposed resolution:


2589. match_results can't satisfy the requirements of a container

Section: 28.10 [re.results] Status: New Submitter: S. B. Tam Opened: 2016-01-24 Last modified: 2016-05-19

Priority: 3

View other active issues in [re.results].

View all other issues in [re.results].

View all issues with New status.

Discussion:

N4567 28.10 [re.results] p2 mentions

The class template match_results shall satisfy the requirements of an allocator-aware container and of a sequence container, as specifed in 23.2.3, except that only operations defined for const-qualified sequence containers are supported.

However, this is impossible because match_results has a operator== whose semantics differs from the one required in Table 95 — "Container requirements".

Table 95 requires that a == b is an equivalence relation and means equal(a.begin(), a.end(), b.begin(), b.end()). But for match_results, a == b and equal(a.begin(), a.end(), b.begin(), b.end()) can give different results. For example:

#include <iostream>
#include <regex>
#include <string>
#include <algorithm>

int main()
{
  std::regex re("a*");
  std::string target("baaab");
  std::smatch a;

  std::regex_search(target, a, re);

  std::string target2("raaau");
  std::smatch b;

  std::regex_search(target2, b, re);

  std::cout << std::boolalpha;
  std::cout << (a == b) << '\n'; // false
  std::cout << std::equal(a.begin(), a.end(), b.begin(), b.end()) << '\n'; // true
}

[2016-02, Issues Telecon]

Marshall: The submitter is absolutely right, but the proposed resolution is insufficient. We should avoid "shall", for once.
Jonathan: This is NAD, because the container comparison functions say "unless otherwise stated", 23.3.1p14 and table 97.
Ville: wrong, table 95 is relevant for ==.
Jonathan: good point

2016-05: Marshall cleans up the wording around the change

Proposed resolution:

Previous resolution [SUPERSEDED]:

This wording is relative to N4567.

  1. Change 28.10 [re.results] p2 as indicated:

    -2- The class template match_results shall satisfy the requirements of an allocator-aware container and of a sequence container, as specified in 23.2.3, except that only operations defined for const-qualified sequence containers are supported and that the semantics of comparison functions are different from those required for a container.

This wording is relative to N4567.

  1. Change 28.10 [re.results] p2 as indicated:

    -2- The class template match_results shall satisfysatisfies the requirements of an allocator-aware container and of a sequence container, as specified in (23.2.3), except that only operations defined for const-qualified sequence containers are supported and the semantics of comparison functions are different from those required for a container.


2591. std::function's member template target() should not lead to undefined behaviour

Section: 20.14.12.2.5 [func.wrap.func.targ] Status: New Submitter: Daniel Krügler Opened: 2016-01-31 Last modified: 2016-02-07

Priority: 3

View all other issues in [func.wrap.func.targ].

View all issues with New status.

Discussion:

This issue is a spin-off of LWG 2393, it solely focuses on the pre-condition of 20.14.12.2.5 [func.wrap.func.targ] p2:

Requires: T shall be a type that is Callable (20.9.12.2) for parameter types ArgTypes and return type R.

Originally, the author of this issue here had assumed that simply removing the precondition as a side-step of fixing LWG 2393 would be uncontroversial. Discussions on the library reflector indicated that this is not the case, although it seemed that there was agreement on removing the undefined behaviour edge-case.

There exist basically the following positions:

  1. The constraint should be removed completely, the function is considered as having a wide contract.

  2. The pre-condition should be replaced by a Remarks element, that has the effect of making the code ill-formed, if T is a type that is not Lvalue-Callable (20.9.11.2) for parameter types ArgTypes and return type R. Technically this approach is still conforming with a wide contract function, because the definition of this contract form depends on runtime constraints.

Not yet explicitly discussed, but a possible variant of bullet (2) could be:

  1. The pre-condition should be replaced by a Remarks element, that has the effect of SFINAE-constraining this member: "This function shall not participate in overload resolution unless T is a type that is Lvalue-Callable (20.9.11.2) for parameter types ArgTypes and return type R".

The following describes a list of some selected arguments that have been provided for one or the other position using corresponding list items. Unless explicitly denoted, no difference has been accounted for option (3) over option (2).

    1. It reflects existing implementation practice, Visual Studio 2015 SR1, gcc 6 libstdc++, and clang 3.8.0 libc++ do accept the following code:

      #include <functional>
      #include <iostream>
      #include <typeinfo>
      #include "boost/function.hpp"
      
      void foo(int) {}
      
      int main() {
        std::function<void(int)> f(foo);
        std::cout << f.target<void(*)()>() << std::endl;
        boost::function<void(int)> f2(foo);
        std::cout << f2.target<void(*)()>() << std::endl;
      }
      

      and consistently output the implementation-specific result for two null pointer values.

    2. The current Boost documentation does not indicate any precondition for calling the target function, so it is natural that programmers would expect similar specification and behaviour for the corresponding standard component.

    3. There is a consistency argument in regard to the free function template get_deleter

      template<class D, class T> 
      D* get_deleter(const shared_ptr<T>& p) noexcept;
      

      This function also does not impose any pre-conditions on its template argument D.

    1. Programmers have control over the type they're passing to target<T>(). Passing a non-callable type can't possibly retrieve a non-null target, so it seems highly likely to be programmer error. Diagnosing that at compile time seems highly preferable to allowing this to return null, always, at runtime.

    2. If T is a reference type then the return type T* is ill-formed anyway. This implies that one can't blindly call target<T> without knowing what T is.

    3. It has been pointed out that some real world code, boiling down to

      void foo() {}
      
      int main() {
        std::function<void()> f = foo;
        if (f.target<decltype(foo)>()) {
          // fast path
        } else {
          // slow path
        }
      }
      

      had manifested as a performance issue and preparing a patch that made the library static_assert in that case solved this problem (Note that decltype(foo) evaluates to void(), but a proper argument of target() would have been the function pointer type void(*)(), because a function type void() is not any Callable type).

It might be worth adding that if use case (2 c) is indeed an often occurring idiom, it would make sense to consider to provide an explicit conversion to a function pointer (w/o template parameters that could be provided incorrectly), if the std::function object at runtime conditions contains a pointer to a real function, e.g.

R(*)(ArgTypes...) target_func_ptr() const noexcept;

Proposed resolution:

This wording is relative to N4567.

  1. Change 20.14.12.2.5 [func.wrap.func.targ] p2 as indicated:

    template<class T> T* target() noexcept;
    template<class T> const T* target() const noexcept;
    

    -2- Requires: T shall be a type that is Callable (20.14.12.2 [func.wrap.func]) for parameter types ArgTypes and return type R.

    -3- Returns: If target_type() == typeid(T) a pointer to the stored function target; otherwise a null pointer.


2592. Require that chrono::duration_casts from smaller durations to larger durations do not overflow

Section: 20.17.2 [time.syn] Status: New Submitter: Andy Giese Opened: 2016-02-05 Last modified: 2016-05-17

Priority: 4

View all other issues in [time.syn].

View all issues with New status.

Discussion:

Currently 20.17.2 [time.syn] states

// convenience typedefs
typedef duration<signed integer type of at least 64 bits,        nano> nanoseconds;
typedef duration<signed integer type of at least 55 bits,       micro> microseconds;
typedef duration<signed integer type of at least 45 bits,       milli> milliseconds;
typedef duration<signed integer type of at least 35 bits             > seconds;
typedef duration<signed integer type of at least 29 bits, ratio<  60>> minutes;
typedef duration<signed integer type of at least 23 bits, ratio<3600>> hours;

However, a duration_cast<minutes>(seconds::max()) would cause overflow if the underlying signed integers only met the minimums specified.

The standard should specify that implementations guarantee that a duration_cast from any smaller duration in these "convenience typedefs" will not overflow any larger duration. That is, hours should be able to hold the maximum of minutes, which should be able to hold the maximum of seconds and so on.

More formally, if the ratio typedef A and typedef B is 1:Y where Y > 1 (e.g., 1 : 60 in case of minutes : seconds), then #bitsA-1 must be at least ceil(log2(2#bitsB-1)/Y)).

In the case of minutes : seconds, X = 1, Y = 60. Let #bitsseconds = 32. Therefore:

Therefore, a minimum of 27 bits would be needed to store minutes if 32 were used to store seconds.

I propose to change the definitions of the convenience typedefs as follows:

// convenience typedefs
typedef duration<signed integer type of at least 64 bits,        nano> nanoseconds;
typedef duration<signed integer type of at least 55 bits,       micro> microseconds;
typedef duration<signed integer type of at least 46 bits,       milli> milliseconds;
typedef duration<signed integer type of at least 37 bits             > seconds;
typedef duration<signed integer type of at least 32 bits, ratio<  60>> minutes;
typedef duration<signed integer type of at least 27 bits, ratio<3600>> hours;

These bits were chosen to satisfy the above formula. Note that minimums only increased, so larger ranges could be held. A nice outcome of this choice is that minutes does not go above 32 bits.

[2016-04-23, Tim Song comments]

The P/R of LWG 2592 doesn't fix the issue it wants to solve, because the actual underlying type will likely have more bits than the specified minimum.

Consider seconds, which the P/R requires to have at least 37 bits. On a typical system this implies using a 64-bit integer. To ensure that casting from seconds::max() to minutes doesn't overflow in such a system, it is necessary for the latter to have at least 59 bits (which means, in practice, 64 bits too), not just 32 bits. Thus, just changing the minimum number of bits will not be able to provide the desired guarantee that casting from a smaller unit to a larger one never overflow.

If such a guarantee is to be provided, it needs to be spelled out directly. Note that the difference here is 9 bits (for the 1000-fold case) and 5 bits (for the 60-fold case), which is less than the size difference between integer types on common systems, so such a requirement would effectively require those convenience typedefs to use the same underlying integer type.

Proposed resolution:

This wording is relative to N4567.

  1. Change 20.17.2 [time.syn], header <chrono> synopsis, as indicated

    […]
    
    // convenience typedefs
    typedef duration<signed integer type of at least 64 bits,        nano> nanoseconds;
    typedef duration<signed integer type of at least 55 bits,       micro> microseconds;
    typedef duration<signed integer type of at least 4645 bits,       milli> milliseconds;
    typedef duration<signed integer type of at least 3735 bits             > seconds;
    typedef duration<signed integer type of at least 3229 bits, ratio<  60>> minutes;
    typedef duration<signed integer type of at least 2723 bits, ratio<3600>> hours;
    
    […]
    

2593. Moved-from state of Allocators

Section: 17.6.3.5 [allocator.requirements] Status: New Submitter: David Krauss Opened: 2016-02-19 Last modified: 2016-06-20

Priority: 4

View other active issues in [allocator.requirements].

View all other issues in [allocator.requirements].

View all issues with New status.

Discussion:

17.6.3.5 [allocator.requirements] suggests that the moved-from state of an allocator may be unequal to its previous state. Such a move constructor would break most container implementations, which move-construct the embedded allocator along with a compressed pair. Even if a moved-from container is empty, it should still subsequently allocate from the same resource pool as it did before.

std::vector<int, pool> a(500, my_pool);
auto b = std::move(a); // b uses my_pool too.
a.resize(500); // should still use my_pool.

[2016-02, Jacksonville]

Marshall will see if this can be resolved editorially.

After discussion, the editors and I decided that this could not be handled editorially. The bit about a moved-from state of an allocator being the same as the original state is a normative change. I submitted a pull request to handle the mismatched variables in the table.

Previous resolution [SUPERSEDED]:

This wording is relative to N4567.

  1. Change 17.6.3.5 [allocator.requirements], Table 28 — "Allocator requirements" as indicated:

    Note there's an editorial error in Table 28 in that line and the surrounding ones. The left column was apparently updated to use u and the right column is still using a/a1/b.

    Table 28 — Allocator requirements
    Expression Return type Assertion/note
    pre-/post-condition
    Default
    X u(move(a));
    X u = move(a);
    Shall not exit via an exception. post: u is equal to a and equal to the prior value of aa1 equals the prior value of a.

[2016-06-20, Oulu, Daniel comments]

According to the current working draft, the situation has changed due to changes performed by the project editor, the revised resolution has been adjusted to N4594.

Proposed resolution:

This wording is relative to N4594.

  1. Change 17.6.3.5 [allocator.requirements], Table 28 — "Allocator requirements" as indicated:

    Table 28 — Allocator requirements
    Expression Return type Assertion/note
    pre-/post-condition
    Default
    X u(std::move(a));
    X u = std::move(a);
    Shall not exit via an exception. post: u is equal to a and equal to the prior value of au is equal to the prior value of a..

2594. Contradicting definition of empty shared_ptr on shared_ptr(nullptr, d)

Section: 20.11.2.2 [util.smartptr.shared] Status: New Submitter: Kazutoshi Satoda Opened: 2016-02-20 Last modified: 2016-06-20

Priority: 3

View other active issues in [util.smartptr.shared].

View all other issues in [util.smartptr.shared].

View all issues with New status.

Discussion:

Latest draft (N4567) 20.11.2.2 [util.smartptr.shared] p1 says:

A shared_ptr object is empty if it does not own a pointer.

Please note that it says "own a pointer". This definition was added as the resolution for LWG defect 813.

20.11.2.2.1 [util.smartptr.shared.const] p8 says about the effect of shared_ptr(nullptr_t p, D d):

Effects: Constructs a shared_ptr object that owns the object p and the deleter d.

Please note that it says "owns the object". This was intentionally changed from "the pointer" as a part of resolution for LWG defect 758, to cover nullptr_t case.

Since shared_ptr(nullptr, d) owns an object of type nullptr_t, but does not own a pointer, it is said as "empty" by a strict reading of the above mentioned definition in 20.11.2.2 [util.smartptr.shared] p1.

These cause a contradiction:

20.11.2.2.1 [util.smartptr.shared.const] p9 sets a postcondition use_count() == 1 on shared_ptr(nullptr, d). But 20.11.2.2.5 [util.smartptr.shared.obs] p7 says that the return value of use_count() is "0 when *this is empty".

Proposed wording changes:

Replace the last 2 words in 20.11.2.2 [util.smartptr.shared] p1 from

[…] empty if it does not own a pointer.

to

[…] empty if it does not own an object.

Note that shared_ptr(nullptr_t) is defined to be empty in synopsis in 20.11.2.2 [util.smartptr.shared].

constexpr shared_ptr(nullptr_t) noexcept : shared_ptr() { }

It could be less confusing if shared_ptr(nullptr, d) could be defined to be empty. But it seems too late to change that (which means changing whether the deleter is called or not, see this Stackoverflow article). Then I'm proposing just fix the contradiction.

Proposed resolution:

This wording is relative to N4594.

  1. Change 20.11.2.2 [util.smartptr.shared] p1 as indicated:

    -1- The shared_ptr class template stores a pointer, usually obtained via new. shared_ptr implements semantics of shared ownership; the last remaining owner of the pointer is responsible for destroying the object, or otherwise releasing the resources associated with the stored pointer. A shared_ptr object is empty if it does not own an objecta pointer.


2595. reverse_iterator::operator[]'s return type revisited

Section: 24.5.1.1 [reverse.iterator], 24.5.1.3.12 [reverse.iter.opindex] Status: New Submitter: Robert Haberlach Opened: 2016-02-28 Last modified: 2016-04-15

Priority: 3

View all other issues in [reverse.iterator].

View all issues with New status.

Discussion:

Issue 386 changed the return type of reverse_iterator::operator[] to unspecified. However, as of N3066, the return type of a random access iterator's operator[] shall be convertible to reference; thus the return type of reverse_iterator::operator[] should be reference (and it is in all common implementations).

Suggested resolution: Adjust 24.5.1.1 [reverse.iterator]'s synopsis and 24.5.1.3.12 [reverse.iter.opindex] to use reference instead of unspecified.

Proposed resolution:

This wording is relative to N4582.

  1. Edit 24.5.1.1 [reverse.iterator], class template synopsis, as indicated:

    namespace std {
      template <class Iterator>
      class reverse_iterator {
      public:
        […]
        typedef typename iterator_traits<Iterator>::reference reference;
        […]
        constexpr referenceunspecified operator[](difference_type n) const;
        […]
      };
    }
    
  2. Change 24.5.1.3.12 [reverse.iter.opindex] before p1 as indicated:

    constexpr referenceunspecified operator[](
      typename reverse_iterator<Iterator>::difference_type n) const;
    

2597. std::log misspecified for complex numbers

Section: 26.5.8 [complex.transcendentals] Status: New Submitter: Thomas Koeppe Opened: 2016-03-01 Last modified: 2016-04-15

Priority: 3

View all other issues in [complex.transcendentals].

View all issues with New status.

Discussion:

The current specification of std::log is inconsistent for complex numbers, specifically, the Returns clause (26.5.8 [complex.transcendentals]). On the one hand, it states that the imaginary part of the return value lies in the closed interval [-i π, +i π]. On the other hand, it says that "the branch cuts are along the negative real axis" and "the imaginary part of log(x) is when x is a negative real number".

The inconsistency lies in the difference between the mathematical concept of a branch cut and the nature of floating point numbers in C++. The corresponding specification in the C standard makes it clearer that if x is a real number, then log(x + 0i) = +π, but log(x - 0i) = -π, i.e. they consider positive and negative zero to represent the two different limits of approaching the branch cut from opposite directions. In other words, the term "negative real number" is misleading, and in fact there are two distinct real numbers, x + 0i and x - 0i, that compare equal but whose logarithms differ by 2 π i.

The resolution should consist of two parts:

  1. Double-check that our usage and definition of "branch cut" is sufficiently unambiguous. The C standard contains a lot more wording around this that we don't have in C++.

  2. Change the Returns clause of log appropriately. For example: "When x is a negative real number, imag(log(x + 0i)) is π, and imag(log(x - 0i)) is ."

Current implementations seem to behave as described in (2). (Try-it-at-home link)

Proposed resolution:


2598. addressof works on temporaries

Section: 20.10.10.1 [specialized.addressof] Status: New Submitter: Brent Friedman Opened: 2016-03-06 Last modified: 2016-04-15

Priority: 3

View all other issues in [specialized.addressof].

View all issues with New status.

Discussion:

LWG issue 970 removed the rvalue reference overload for addressof. This allows const prvalues to bind to a call to addressof, which is dissimilar from the behavior of operator&.

const vector<int> a();

void b()
{
  auto x = addressof(a()); // "ok"
  auto y = addressof<const int>(0); // "ok"
  auto z = &a(); //error: cannot take address of a temporary
}

Proposed resolution:

This wording is relative to N4582.

  1. Change 20.10.2 [memory.syn], header <memory> synopsis, as indicated:

    […]
    // 20.9.12, specialized algorithms:
    template <class T> constexpr T* addressof(T& r) noexcept;
    template <class T> const T* addressof(const T&& elem) = delete;
    […]
    
  2. Change 20.10.10.1 [specialized.addressof] p1 as indicated:

    template <class T> constexpr T* addressof(T& r) noexcept;
    template <class T> const T* addressof(const T&& elem) = delete;
    

    -1- Returns: The actual address of the object or function referenced by r, even in the presence of an overloaded operator&.


2599. Library incomplete type permission phrase is unclear

Section: 20.2.6 [declval], 20.11.1 [unique.ptr], 20.11.1.1.1 [unique.ptr.dltr.general], 20.11.2.2 [util.smartptr.shared], 20.11.2.3 [util.smartptr.weak], 20.11.2.5 [util.smartptr.enab] Status: New Submitter: Zhihao Yuan Opened: 2016-03-08 Last modified: 2016-04-15

Priority: 3

View all issues with New status.

Discussion:

Currently the phrase to grant this permission is:

The template parameter T of LibraryTemplate may be an incomplete type.

Two problems:

  1. The timing is unclear. We always allow specializations like LibraryTemplate<Incomp>* p;

  2. To the users of a template, the correct terminology should be "argument" rather than "parameter".

Suggested resolution:

In an instantiation of LibraryTemplate, an incomplete type may be used as the template argument for the template parameter T.

as shown here.

Or, to copy N4510's wording:

An incomplete type T may be used when instantiating LibraryTemplate.

Proposed resolution:


2600. ios_base must store inaccessible iostate flags

Section: 27.5.3.5 [ios.base.storage] Status: LEWG Submitter: David Krauss Opened: 2016-03-14 Last modified: 2016-04-15

Priority: Not Prioritized

View all other issues in [ios.base.storage].

View all issues with LEWG status.

Discussion:

DR 41, "Ios_base needs clear(), exceptions()" stopped short of providing the interface suggested in its title, but it did require the underlying state to be stored in ios_base. Because rdstate() is also missing, ios_base manipulators relying on iword and pword cannot detect failure. The only safe alternative is to manipulate a derived class, which must be a template.

libc++ already provides the interface as a nonconforming extension. libstdc++ implements the internal state but leaves it frustratingly inaccessible, as specified. Any conforming implementation should be able to provide the interface without ABI problems.

[2016-04, Issues Telecon]

This is really a request for an (feature) API. Passing to LEWG.

Proposed resolution:


2663. Enable efficient retrieval of file size from directory_entry

Section: 27.10.15.14 [fs.op.file_size] Status: New Submitter: Gor Nishanov Opened: 2014-05-22 Last modified: 2016-04-14

Priority: 2

View all issues with New status.

Discussion:

On Windows, the FindFileData WIN32_FIND_DATA structure, which is the underlying data type for directory_entry, contains the file size as one of the fields. Thus efficient enumeration of files and getting their sizes is possible without doing a separate query for the file size.

[17 Jun 2014 Rapperswil LWG will investigate issue at a subsequent meeting.]

[23 Nov 2015 Editorally correct name of data structure mentioned in discussion.]

[Mar 2016 Jacksonville Beman to provide paper about this]

[Apr 2016 Issue updated to address the C++ Working Paper. Previously addressed File System TS]

Proposed resolution:

In 27.10.12 [class.directory_entry] Class directory_entry add the following observer declarations:

      uintmax_t file_size();
      uintmax_t file_size(error_code& ec) noexcept;
    

In directory_entry observers 27.10.12.3 [directory_entry.obs] add the following:

      uintmax_t file_size();
      uintmax_t file_size(error_code& ec) noexcept;
    

Returns: if *this contains a cached file size, return it. Otherwise return file_size(path()) or file_size(path(), ec) respectively.

Throws: As specified in Error reporting (7).


2664. operator/ (and other append) semantics not useful if argument has root

Section: 27.10.8.4.3 [path.append], 27.10.8.6 [path.non-member] Status: New Submitter: Peter Dimov Opened: 2014-05-30 Last modified: 2016-07-06

Priority: 2

View other active issues in [path.append].

View all other issues in [path.append].

View all issues with New status.

Discussion:

In a recent discussion on the Boost developers mailing list, the semantics of operator / and other append operations were questioned:

In brief, currently p1 / p2 is required to concatenate the lexical representation of p1 and p2, inserting a preferred separator as needed.

This means that, for example, "c:\x" / "d:\y" gives "c:\x\d:\y", and that "c:\x" / "\\server\share" gives "c:\x\\server\share". This is rarely, if ever, useful.

An alternative interpretation of p1 / p2 could be that it yields a path that is the approximation of what p2 would mean if interpreted in an environment in which p1 is the starting directory. Under this interpretation, "c:\x" / "d:\y" gives "d:\y", which is more likely to match what was intended.

I am not saying that this second interpretation is the right one, but I do say that we have reasons to suspect that the first one (lexical concatenation using a separator) may not be entirely correct. This leads me to think that the behavior of p1 / p2, when p2 has a root, needs to be left implementation-defined, so that implementations are not required to do the wrong thing, as above.

This change will not affect the ordinary use case in which p2 is a relative, root-less, path.

[17 Jun 2014 Rapperswil LWG will investigate issue at a subsequent meeting.]

[2016-02, Jacksonville]

Beman to provide wording.

[2016-06-13, Beman provides wording and rationale]

Rationale: The purpose of the append operations is to provide a simple concatenation facility for users wishing to extend a path by appending one or more additional elements, and to do so without worrying about the details of when a separator is needed. In that context it makes no sense to provide an argument that has a root-name. The simplest solution is simply to require !p.has_root_name(). The other suggested solutions IMO twist the functions into something harder to reason about yet any advantages for users are purely speculative. The concatenation functions can be used instead for corner cases.

[Apr 2016 Issue updated to address the C++ Working Paper. Previously addressed File System TS]

[2016-07-03, Daniel comments]

The same wording area is touched by LWG 2732.

Proposed resolution:

This wording is relative to N4594.

  1. Change 27.10.8.4.3 [path.append] path appends as indicated:

    path& operator/=(const path& p);

    -?- Requires: !p.has_root_name().

    -2- Effects: Appends path::preferred_separator to pathname unless:

    • an added directory-separator would be redundant, or

    • an added directory-separator would change a relative path to an absolute path [Note: An empty path is relative. — end note], or

    • p.empty() is true, or

    • *p.native().cbegin() is a directory-separator.

    Then appends p.native() to pathname.

    -3- Returns: *this.

    template <class Source>
      path& operator/=(const Source& source);
    template <class Source>
      path& append(const Source& source);
    template <class InputIterator>
      path& append(InputIterator first, InputIterator last);

    -?- Requires: !source.has_root_name() or !*first.has_root_name(), respectively.

    -4- Effects: Appends path::preferred_separator to pathname, converting format and encoding if required (27.10.8.2 [path.cvt]), unless:

    • an added directory-separator would be redundant, or

    • an added directory-separator would change a relative path to an absolute path, or

    • source.empty() is true, or

    • *source.native().cbegin() is a directory-separator.

    Then appends the effective range of source (27.10.8.3 [path.req]) or the range [first, last) to pathname, converting format and encoding if required (27.10.8.2 [path.cvt]).

    -5- Returns: *this.

  2. Change 27.10.8.6 [path.non-member] path non-member functions as indicated:

    path operator/(const path& lhs, const path& rhs);

    -?- Requires: !rhs.has_root_name().

    -13- Returns: path(lhs) /= rhs.


2665. remove_filename() post condition is incorrect

Section: 27.10.8.4.5 [path.modifiers] Status: New Submitter: Eric Fiselier Opened: 2014-06-07 Last modified: 2016-04-15

Priority: 3

View all issues with New status.

Discussion:

remove_filename() specifies !has_filename() as the post condition. This post condition is not correct. For example the path "/foo" has a filename of "foo". If we remove the filename we get "/", and "/" has a filename of "/".

[2014-06-08 Beman supplies an Effects: element.]

[17 Jun 2014 Rapperswil LWG will investigate issue at a subsequent meeting.]

[Apr 2016 Issue updated to address the C++ Working Paper. Previously addressed File System TS]

[2016-04, Issues Telecon]

There was concern that the effects wording is not right. Jonathan provided updated wording.

Previous resolution [SUPERSEDED]:

    path& remove_filename();
  

Postcondition: !has_filename().

Effects: *this = parent_path(), except that if parent_path() == root_path(), clear().

Returns: *this.

[Example:

        std::cout << path("/foo").remove_filename();  // outputs "/"
        std::cout << path("/").remove_filename();     // outputs ""
      

—end example]

Proposed resolution:

  path& remove_filename();

Postcondition: !has_filename().

Effects: If *this == root_path(), then clear(). Otherwise, *this = parent_path().

Returns: *this.

[Example:

      std::cout << path("/foo").remove_filename();  // outputs "/"
      std::cout << path("/").remove_filename();     // outputs ""
    

—end example]


2668. path::operator+= is defined, but not operator+

Section: 27.10.8 [class.path] Status: New Submitter: Jonathan Wakely Opened: 2014-07-03 Last modified: 2016-04-14

Priority: 3

View all other issues in [class.path].

View all issues with New status.

Discussion:

This doesn't seem to be in Boost.Filesystem, so maybe it isn't needed, but since path += path2 works then it seems reasonable to expect path1 + path2 to work as well.

[04 Jul 2014 Beman Dawes comments:]

The 12 overloads required by basic_string operator+ scared me off, and I never came back to the issue.

[22 Nov 2015 Beman supplies proposed resolution wording.]

[Apr 2016 Issue updated to address the C++ Working Paper. Previously addressed File System TS]

Proposed resolution:

NAD, Future.

It is not necessary to provide every possible string operation for class path because it is always possible to convert to a string, perform the operation, and convert back to a path. The most commonly needed string operations are provided for class path as a convenience, but every added function comes at the cost of increased interface complexity. In this case, the cost is judged to outweigh the convenience.

Future changes to the language, such as concepts, and changes to the library, such as basic_string_view, may allow reduction of the complexity of the class path interface. The LWG may wish to reconsider this issue at that time.


2672. Should is_empty use error_code in its specification?

Section: 27.10.15.19 [fs.op.is_empty] Status: New Submitter: Jonathan Wakely Opened: 2014-08-01 Last modified: 2016-05-17

Priority: 3

View all issues with New status.

Discussion:

The descriptions of most functions are explicit about the use of the ec argument, either saying something like "foo(p) or foo(p, ec), respectively" or using the ec argument like foo(p[, ec]), but is_empty does not.

27.10.15.19 [fs.op.is_empty]/2 refers to ec unconditionally, but more importantly 27.10.15.19 [fs.op.is_empty]/3 doesn't pass ec to the directory_iterator constructor or the file_size function.

[ 9 Oct 2014 Beman supplies proposed wording. ]

[Apr 2016 Issue updated to address the C++ Working Paper. Previously addressed File System TS]

Proposed resolution:

  1.     bool is_empty(const path& p);
        bool is_empty(const path& p, error_code& ec) noexcept;
      

    Effects:

    • Determine file_status s, as if by status(p) or status(p, ec), respectively.
    • For the signature with argument ec, return false if an error occurred.
    • Otherwise, if is_directory(s):
      • Create directory_iterator itr, as if by directory_iterator(p) or directory_iterator(p, ec), respectively.
      • For the signature with argument ec, return false if an error occurred.
      • Otherwise, return itr == directory_iterator().
    • Otherwise:
      • Determine uintmax_t sz, as if by file_size(p) or file_size(p, ec), respectively .
      • For the signature with argument ec, return false if an error occurred.
      • Otherwise, return sz == 0.

    Remarks: The temporary objects described in Effects are for exposition only. Implementations are not required to create them.

    Returns: is_directory(s) ? directory_iterator(p) == directory_iterator() : file_size(p) == 0; See Effects.

    The signature with argument ec returns false if an error occurs.

    Throws: As specified in Error reporting (7).


2675. register_callback can fail

Section: 27.5.3.6 [ios.base.callback] Status: New Submitter: David Krauss Opened: 2016-03-14 Last modified: 2016-04-15

Priority: 3

View all issues with New status.

Discussion:

register_callback allocates memory and so it can fail, but the case is unspecified. libc++ sets badbit, which is consistent with iword and pword. libstdc++ throws std::bad_alloc.

Proposed resolution:


2676. Provide filesystem::path overloads for File-based streams

Section: 27.9 [file.streams] Status: New Submitter: Beman Dawes Opened: 2016-03-16 Last modified: 2016-04-15

Priority: 2

View all other issues in [file.streams].

View all issues with New status.

Discussion:

The constructor and open functions for File-based streams in 27.9 [file.streams] currently provide overloads for const char* and const string& arguments that specify the path for the file to be opened. With the addition of the File System TS to the standard library, these constructors and open functions need to be overloaded for const filesystem::path& so that file-based streams can take advantage of class filesystem::path features such as support for strings of character types wchar_t, char16_t, and char32_t.

The const filesystem::path& p overload for these functions is like the existing const string& overload; it simply calls the overload of the same function that takes a C-style string.

For operating systems like POSIX that traffic in char strings for filenames, nothing more is required. For operating systems like Windows that traffic in wchar_t strings for filenames, an additional C-style string overload is required. The overload's character type needs to be specified as std::filesystem::path::value_type to also support possible future operating systems that traffic in char16_t or char32_t characters.

Not recommended:

Given the proposed constructor and open signatures taking const filesystem::path&, it would in theory be possible to remove some of the other signatures. This is not proposed because it would break ABI's, break user code depending on user-defined automatic conversions to the existing argument types, and invalidate existing documentation, books, and tutorials.

Implementation experience:

The Boost Filesystem library has provided header <boost/filesystem/fstream.hpp> implementing the proposed resolution for over a decade.

The Microsoft/Dinkumware implementation of standard library header <fstream> has provided the const wchar_t* overloads for many years.

Proposed resolution:

At the end of 27.9.1 File streams [fstreams] add a paragraph:

In this subclause, member functions taking arguments of const std::filesystem::path::value_type* shall only be provided on systems where std::filesystem::path::value_type ([class.path]) is not char. [Note: These functions enable class path support for systems with a wide native path character type, such as wchar_t. — end note]

Change 27.9.1.1 Class template basic_filebuf [filebuf] as indicated:

basic_filebuf<charT,traits>* open(const char* s,
    ios_base::openmode mode);
basic_filebuf<charT,traits>* open(const std::filesystem::path::value_type* s,
    ios_base::openmode mode);  // wide systems only; see [fstreams] 
basic_filebuf<charT,traits>* open(const string& s,
   ios_base::openmode mode);
basic_filebuf<charT,traits>* open(const filesystem::path& p,
   ios_base::openmode mode);

Change 27.9.1.4 Member functions [filebuf.members] as indicated:

basic_filebuf<charT,traits>* open(const char* s,
   ios_base::openmode mode);
basic_filebuf<charT,traits>* open(const std::filesystem::path::value_type* s,
   ios_base::openmode mode);  // wide systems only; see [fstreams]

To 27.9.1.4 Member functions [filebuf.members] add:

basic_filebuf<charT,traits>* open(const filesystem::path& p,
   ios_base::openmode mode);

Returns: open(p.c_str(), mode);

Change 27.9.1.6 Class template basic_ifstream [ifstream] as indicated:

explicit basic_ifstream(const char* s,
    ios_base::openmode mode = ios_base::in);
explicit basic_ifstream(const std::filesystem::path::value_type* s,
    ios_base::openmode mode = ios_base::in);  // wide systems only; see [fstreams]
explicit basic_ifstream(const string& s,
    ios_base::openmode mode = ios_base::in);
explicit basic_ifstream(const filesystem::path& p,
    ios_base::openmode mode = ios_base::in);
...
void open(const char* s,
    ios_base::openmode mode = ios_base::in);
void open(const std::filesystem::path::value_type* s,
    ios_base::openmode mode = ios_base::in);  // wide systems only; see [fstreams]
void open(const string& s,
    ios_base::openmode mode = ios_base::in);
void open(const filesystem::path& p,
    ios_base::openmode mode = ios_base::in);

Change 27.9.1.7 basic_ifstream constructors [ifstream.cons] as indicated:

explicit basic_ifstream(const char* s,
    ios_base::openmode mode = ios_base::in);
explicit basic_ifstream(const std::filesystem::path::value_type* s,
    ios_base::openmode mode = ios_base::in);  // wide systems only; see [fstreams]

To 27.9.1.7 basic_ifstream constructors [ifstream.cons] add:

explicit basic_ifstream(const filesystem::path& p,
    ios_base::openmode mode = ios_base::in);

Effects: the same as basic_ifstream(p.c_str(), mode).

Change 27.9.1.9 Member functions [ifstream.members] as indicated:

void open(const char* s,
    ios_base::openmode mode = ios_base::in);
void open(const std::filesystem::path::value_type* s,
    ios_base::openmode mode = ios_base::in);  // wide systems only; see [fstreams]

To 27.9.1.9 Member functions [ifstream.members] add:

void open(const filesystem::path& p,
    ios_base::openmode mode = ios_base::in);

Effects: calls open(p.c_str(), mode).

Change 27.9.1.10 Class template basic_ofstream [ofstream] as indicated:

explicit basic_ofstream(const char* s,
    ios_base::openmode mode = ios_base::out);
explicit basic_ofstream(const std::filesystem::path::value_type* s,
    ios_base::openmode mode = ios_base::out);  // wide systems only; see [fstreams]
explicit basic_ofstream(const string& s,
    ios_base::openmode mode = ios_base::out);
explicit basic_ofstream(const filesystem::path& p,
    ios_base::openmode mode = ios_base::out);
...
void open(const char* s,
    ios_base::openmode mode = ios_base::out);
void open(const std::filesystem::path::value_type* s,
    ios_base::openmode mode = ios_base::out);  // wide systems only; see [fstreams]
void open(const string& s,
    ios_base::openmode mode = ios_base::out);
void open(const filesystem::path& p,
    ios_base::openmode mode = ios_base::out);

Change 27.9.1.11 basic_ofstream constructors [ofstream.cons] as indicated:

explicit basic_ofstream(const char* s,
    ios_base::openmode mode = ios_base::out);
explicit basic_ofstream(const std::filesystem::path::value_type* s,
    ios_base::openmode mode = ios_base::out);  // wide systems only; see [fstreams]

To 27.9.1.11 basic_ofstream constructors [ofstream.cons] add:

explicit basic_ofstream(const filesystem::path& p,
    ios_base::openmode mode = ios_base::out);

Effects: the same as basic_ofstream(p.c_str(), mode).

Change 27.9.1.13 Member functions [ofstream.members] as indicated:

void open(const char* s,
    ios_base::openmode mode = ios_base::out);
void open(const std::filesystem::path::value_type* s,
    ios_base::openmode mode = ios_base::out);  // wide systems only; see [fstreams]

To 27.9.1.13 Member functions [ofstream.members] add:

void open(const filesystem::path& p,
    ios_base::openmode mode = ios_base::out);

Effects: calls open(p.c_str(), mode).

Change 27.9.1.14 Class template basic_fstream [fstream] as indicated:

explicit basic_fstream(const char* s,
    ios_base::openmode mode = ios_base::in|ios_base::out);
explicit basic_fstream(const std::filesystem::path::value_type* s,
    ios_base::openmode mode = ios_base::in|ios_base::out);  // wide systems only; see [fstreams]
explicit basic_fstream(const string& s,
    ios_base::openmode mode = ios_base::in|ios_base::out);
explicit basic_fstream(const filesystem::path& p,
    ios_base::openmode mode = ios_base::in|ios_base::out);
...
void open(const char* s,
    ios_base::openmode mode = ios_base::in|ios_base::out);
void open(const std::filesystem::path::value_type* s,
    ios_base::openmode mode = ios_base::in|ios_base::out);  // wide systems only; see [fstreams]
void open(const string& s,
    ios_base::openmode mode = ios_base::in|ios_base::out);
void open(const filesystem::path& p,
    ios_base::openmode mode = ios_base::in|ios_base::out);

Change 27.9.1.15 basic_fstream constructors [fstream.cons] as indicated:

explicit basic_fstream(const char* s,
    ios_base::openmode mode = ios_base::in|ios_base::out);
explicit basic_fstream(const std::filesystem::path::value_type* s,
    ios_base::openmode mode = ios_base::in|ios_base::out);  // wide systems only; see [fstreams]

To 27.9.1.15 basic_fstream constructors [fstream.cons] add:

explicit basic_fstream(const filesystem::path& p,
    ios_base::openmode mode = ios_base::in|ios_base::out);

Effects: the same as basic_fstream(p.c_str(), mode).

Change 27.9.1.17 Member functions [fstream.members] as indicated:

void open(const char* s,
    ios_base::openmode mode = ios_base::in|ios_base::out);
void open(const std::filesystem::path::value_type* s,
    ios_base::openmode mode = ios_base::in|ios_base::out);  // wide systems only; see [fstreams]

To 27.9.1.17 Member functions [fstream.members] add:

void open(const filesystem::path& p,
    ios_base::openmode mode = ios_base::in|ios_base::out);

Effects: calls open(p.c_str(), mode).


2677. directory_entry::status is not allowed to be cached as a quality-of-implementation issue

Section: 27.10.12.3 [directory_entry.obs] Status: New Submitter: Billy O'Neal Opened: 2016-03-03 Last modified: 2016-04-15

Priority: 2

View all issues with New status.

Discussion:

To fix multi-threading problems in directory_entry, caching behavior was removed from the type. This is bad for performance reasons, because the values can no longer be cached from the result of readdir on POSIX platforms, or from FindFirstFile/FindNextFile on Windows.

It appears that the intent was to allow implementers to fill in the values for directory_entry::status inside directory_iterator, but as currently specified:

Returns: status(path()[, ec]).

This is not allowed to be cached, because the target of the path can change. For example, consider the following program:

#include <stdio.h>
#include <stdlib.h>
#include <filesystem>
#include <fstream>

using namespace std;
namespace fs = ::std::filesystem;

void verify(const bool b, const char * const msg) {
    if (!b) {
        printf("fail: %s\n", msg);
        exit(1);
    }
}

void touch_file(const char * const filename) {
    ofstream f(filename);
    f << '\n' << endl;
    verify(f.good(), "File write failed");
}

int main() {
    fs::remove_all("subDir");
    fs::create_directory("subDir");
    fs::create_directory("subDir/child");
    touch_file("subDir/child/child");
    fs::current_path("./subDir");
    fs::directory_iterator dir(".");
    ++dir;
    fs::directory_entry entry = *dir;

    verify(entry == "./child",
      "unexpected subdirectory"); //enumerating "subDir" returned the directory "child"

    fs::file_status status = entry.status();
    verify(status.type() == fs::file_type::directory,
        "subDir/child was not a directory");
    fs::current_path("./child");
    status = entry.status(); // REQUIRED to re-stat() on POSIX,
                             // GetFileAttributes() on Windows
    verify(status.type() == fs::file_type::regular,
        "subDir/child/child was not a regular file");
    return 0;
}

directory_entry should be re-specified to allow implementers to cache the value of status(path) at the time irectory_iterator was incremented to avoid repeated stat() / GetFileAttributes calls. (This may mean additional constructors are necessary for directory_entry as well)

[2016-04, Issues Telecon]

Beman is working on a paper to address this.

Proposed resolution:


2678. std::filesystem enum classes overspecified

Section: 27.10.10.1 [enum.file_type], 27.10.10.2 [enum.copy_options], 27.10.10.4 [enum.directory_options] Status: New Submitter: Richard Smith Opened: 2016-03-19 Last modified: 2016-05-17

Priority: 3

View all issues with New status.

Discussion:

The enum class std::filesystem::file_type specifies a collection of enumerators for describing the type of a file, but also specifies (1) the numeric values of the enumerators, and (2) that an implementation cannot extend the list with additional types known to it (and by extension, the standard cannot extend the list in future versions without breakage).

These both seem like mistakes in the design. I would suggest we remove the specification of numerical values, and maybe also

Similar overspecification exists for std::filesystem::copy_options and std::filesystem::directory_options, where again precise numerical values are specified.

Proposed resolution:

  1. Strike the "Value" column from:

  2. Change 27.10.10.1 [enum.file_type] Table 145 Enum class file_type as indicated:

    Constant Meaning
    none The type of the file has not been determined or an error occurred while trying to determine the type.
    not_found Pseudo-type indicating the file was not found. [ Note: The file not being found is not considered an error while determining the type of a file. —end note ]
    regular Regular file
    directory Directory file
    symlink Symbolic link file
    block Block special file
    character Character special file
    fifo FIFO or pipe file
    socket Socket file
    implementation-
    defined
    Implementations that support file systems having file types in addition to the above file_type types shall supply implementation-defined file_type constants to separately identify each of those additional file types
    unknown The file does exist, but is of an operating system dependent type not covered by any of the other cases or the process does not have permission to query the file type The file exists but the type could not be determined

2679. Inconsistent Use of Effects and Equivalent To

Section: 17.5.1.4 [structure.specifications] Status: New Submitter: Dawn Perchik Opened: 2016-04-14 Last modified: 2016-05-17

Priority: 3

View other active issues in [structure.specifications].

View all other issues in [structure.specifications].

View all issues with New status.

Discussion:

[2016-04, Issues Telecon]

The PR is fine; but bullet #1 in the report really should have a list of places to change.

Jonathan checked throughout the library for bullet #2 and found two problems in [string.access], which have been added to the PR.

Proposed resolution:

  1. Change [structure.specifications]/4 as indicated:

    "[…] if F has no Returns: element, a non-void return from F is specified by the Returns: elementsreturn statements in the code sequence."

  2. Add two return keywords to [string.access]:

    const charT& front() const;
    charT& front();
    

    -7- Requires: !empty().

    -8- Effects: Equivalent to return operator[](0).

    const charT& back() const;
    charT& back();
    

    -9- Requires: !empty().

    -10- Effects: Equivalent to return operator[](size() - 1).


2680. Add "Equivalent to" to filesystem

Section: 27.10 [filesystems] Status: New Submitter: Beman Dawes Opened: 2016-04-15 Last modified: 2016-05-17

Priority: 2

View all issues with New status.

Discussion:

The IS project editors have requested that filesystem Effects elements specified purely as C++ code include the phrase "Equivalent to" if this is the intent. They do not consider this change to be editorial.

Proposed resolution:

27.10.8.4.5 [path