Minutes of the 29th November 2018 Teleconference Austin-895 Page 1 of 1 Submitted by Andrew Josey, The Open Group. 4th December 2018 Attendees: Nick Stoughton, USENIX, ISO/IEC JTC 1/SC 22 OR Joerg Schilling, FOKUS Fraunhofer Eric Blake, Red Hat Don Cragun, IEEE PASC OR Geoff Clare, The Open Group Andrew Josey, The Open Group Mike Crowe, (BrightSign) Martin Rehak, Oracle, The Open Group OR Mark Ziegast, SHware Systems (Etherpad only) * General news None * Outstanding actions (Please note that this section has been flushed to shorten the minutes - to locate the previous set of outstanding actions, look to the minutes from 9 March 2018 and earlier) Bug 1077: Recommend support for wide-character regcomp and regexec and/or specify multi-byte behavior OPEN http://austingroupbugs.net/bug_view_page.php?bug_id=1077 Andrew has completed the action to ping his Apple contact and is awaiting a reply. Bug 1122: POSIX should include gettext() and friends OPEN http://austingroupbugs.net/view.php?id=1122 Left open as an action is still in progress to flesh out a complete proposal. * Current Business Bug 1134: Add getentropy interface OPEN http://austingroupbugs.net/view.php?id=1134 Action: Andrew to add boilerplate response regarding a proposal for new interface, and also to ask The Open Group OR if they would sponsor this addition. Both Actions now Completed. (update 19 Oct 2018) There has been no reaction from the Base Working Group and the period for comments has ended now. Bug 1138: Add strsignal(), sig2str() and str2sig() to the standard. OPEN http://austingroupbugs.net/view.php?id=1138 Martin has completed his action. Andrew took an action to chase the Base Working group members for sponsorship of the two interfaces. Update: The action on Andrew is still open. Bug 1216: Adding clockid parameter to functions that accept absolute struct timespec timeouts OPEN http://austingroupbugs.net/view.php?id=1216 There is strong sentiment in favor of this proposal, but it needs more work fleshing out the editing instructions and details. This may help get the proposed interfaces into glibc, therefore providing a reference implementation. Once that happens, it should be easier to find a sponsor. Things to consider, based on points made during the discussion: Look at how openat() is defined in terms of open() and copy that style in writing a new page for each new function. In particular, be sure to include things like ERRORS where you call out all of the same errors as the older function, plus the additional errors of EINVAL for unknown clock, ENOTSUP for clock that can't be used, ... APPLICATION_USAGE should mention that some clocks (like CLOCK_CPUTIME) may not be suitable (precisely because the clock doesn't increment while the interface is blocked) Also, directions to the editor on page/line numbers against the pdf for where to insert other references (such as the new signature for pthread_cond_clockwait() in , in addition to the new page for the documentation of pthread_cond_clockwait() itself) If some functions in the set are difficult to implement, it's okay to document the signature in FUTURE DIRECTIONS, while focusing on condvar as the main one to standardize now Pointing to the glibc mailing list threads with patch proposals matching the POSIX wording proposal will help show that there is existing practice (even if glibc hasn't incorporated the patches yet, the fact that the patches are in the public will count towards an existing practice claim) - the URLs do not have to be part of the proposed normative text additions, but having the cross-link trail in the bug page will help Bug 1145: compound-list does not required a terminating semicolon even in compound commands, which is contradictory with implementations. http://austingroupbugs.net/view.php?id=1145 Closed, Rejected: The command: until a do b; done is not accepted by the grammar because in the grammar do_group requires a "do" keyword to be present, and this command does not have one. (The word "do" in this command is an ordinary word to be passed as an argument to "a"; it is not recognised as a keyword.) This appears to be the same misinterpretation of the shell grammar rules as was made in 0001046. Update: this bug was discussed in the Nov 29, 2018 teleconference and is being rejected for the reason given above. Bug 1146: "total" line in ls -l output Accepted as marked. http://austingroupbugs.net/view.php?id=1146 An interpretation required. This is tagged with the TC3-2008 tag Interpretation response: The standard states that ls lists the space requirements for all files in a directory, and conforming implementations must conform to this. However, concerns have been raised about this which are being referred to the sponsor. Rationale: This is not (and never has been) existing practice. Notes to the Editor (not part of this interpretation): On page 2928 line 96855 section ls, change: ... each list of files within the directory shall be preceded by a status line indicating the number of file system blocks occupied by files in the directory ... to: ... each list of files within a directory shall be preceded by a status line indicating the number of file system blocks occupied by the listed files for that directory ... Add to Application Usage on p2930 after line 96908: The total number provided when using ls -l does not necessarily correspond to the space that would be reclaimed if all the listed files were removed, because of hard links (and symbolic links if -L is present). The space for each listed file is counted in the total regardless of any relationship between the files. Bug 1147: ls -l -F/-p and symlinks OPEN http://austingroupbugs.net/view.php?id=1147 Discussion started on this item and will continue on the next call. Next Steps ---------- The next call is on December 6th 2018 (Thursday). Calls are anchored on US time. (8am Pacific) This call will be for the regular 90 minutes. http://austingroupbugs.net An etherpad is usually up for the meeting, with a URL using the date format as below: https://posix.rhansen.org/p/201x-mm-dd username=posix password=2115756#