Austin Group Reviewers Note teleconference minutes Austin-79 Page 1 of 1 Submitted by Andrew Josey, The Open Group. March 20th, 2001 (corrections applied and actions updated 29/3/2001) Attendees ---------- Don Cragun PASC OR Nick Stoughton WG15 OR Andrew Josey Chair Mark Brown The Open Group OR Bouazza Bachar Compaq Andrew Gollan Sun Joanna Farley Sun The agenda was circulated in mail seq 3017. These minutes record the dispositions for each item (they are not intended as a word-by-word recording of the teleconference). XBD 4112 Notes to Reviewers CLOSED 1003.2 #196 is now complete Action: remove p130 lines 4112-4118 (includes para following the revnote) 4126 Notes to Reviewers There seems little point having a reviewers note saying that an interpretation against .2b should be filed. Better to file one and then we can remove this. Action on the person who wanted the change in XBDd3 ERN 90 Jeff Copeland Action: remove p130 lines 4126-4131 Response: Agreed we can close this item and remove the revnote. ACTION: AJ to notify Jeff Copeland that he can file an interpretation against 2b for this item if he wishes. CLOSED. 9581 Notes to Reviewers (page 276) In we have a Revnote on float_t and double_t that "a floating type" should be changed to "a real-floating type" and that a DR has been filed against C99. It appears to me at least that this change is needed to make sense of 9591-9604, and that since ISO C 6.2.5 defines float, double and long as "real floating types" that we should be able to make this change without the DR. Response: The group agreed with the summary. Action: Make the change as noted and remove the reviewers note 11563 Notes to Reviewers page 330 This is about reviewers checking whether the note about how applications can check for the presence of optional types by checking for a definition of the corresponding limit macro in limits.h. Response: Agreed that this works as described in the note and we should remove the reviewers notes. Action: remove the revnote 11647 Notes to Reviewers page 332 pending ISO C defect report ACTION: AJ to look at TC1, if in TC1 do it otherwise remove it as we will not be able to incorporate it in time 11660 Notes to Reviewers page 332 This put forward an alternate wording for "The argument in any instance of these macros shall be a decimal, octal, or hexadecimal constant with a value that does not exceed the limits for the corresponding type." proposes changing "decimal, octal, or hexadecimal constant" in that to "integer constant" and adding: "The constant shall have one of the following syntactic forms: - Sequence of decimal digits of which the first one is not '0' - Sequence of octal digits of which the first one is '0' - Sequence of hexadecimal digits prefixed with "0x" or "0X" Response: This precludes the use of forms such as a decimal string followed by L, UL etc and we do not think this is appropriate, so do nothing and remove the reviewers note. Action: remove the reviewers note 11671 Notes to Reviewers Pending ISO C defect report ACTION: AJ to look at TC1, if in TC1 do it otherwise remove it as we will not be able to incorporate it in time XSH 1568 Notes to Reviewers page 505 -- set of functions that are reentrant or noninterruptable and async signal safe Response: Leave this open ACTION: AJ to ask Frank if he could check it. 4233 Notes to Reviewers page 581 PASC 1003.1 #118 has confirmed that this is referring to aio_fsync() Is there a need for a leadin? Response: Closed, no change Action: Delete the reviewers note 10793 Notes to Reviewers CLOSED page 801, overcome by XSHd5 ERN 194 delete lines 10793-10797 10864 Notes to Reviewers page 803 fchown on an AF_UNIX socket? Response: The proposal is to For 10862,10863 , pipe or socket [XSR] or an fattach()-ed STREAM [XSR] and remove the reviewers note Action: None pending comments 17495 Notes to Reviewers CLOSED page 1006 , OBE see XSHd5 ERN 582 Action: delete lines 17495-17498 24243 Notes to Reviewers CLOSED page 1227, closed an example was supplied see XSHd5 ERN 742 Action: delete lines 24243-24245 37534 Notes to Reviewers CLOSED page 1709 PASC interp 1003.2 #197 rebutted the proposed change. Action: Delete AGR D5 page 1709 lines 37534-37539. 44872 Notes to Reviewers page 1957 strtod() Is EINVAL an allowable extension to ISO C . A DR was reported to JRB. Not sure if there is any new status Response: Recommended removal of reviewers note Action: remove reviewers note Action AJ to ping John Benito ACTION AJ to review line 44860 as a CX candidate ? 47807 Notes to Reviewers page 2056 ttyname() ERANGE should be E2BIG Whilst it may be inconsistent it is from the Base standard suggest delete reviewers note and if submitter feels strongly they should file a PASC interp. Response Do nothing, as per summary above Action: remove the reviewers note 49207 Notes to Reviewers page 2106, waitid() omission case WHOHANG and no children to be waited for -> Base WG? Response: Agreed to pass it to the Base WG Action: AJ->defer to BWG CLOSED 51164 Notes to Reviewers page 2173, write() pwrite and pread need to be limited to seekable devices Response: we believe this is covered page 2171 51084-51085 cover the case of pwrite to a pipe or FIFO page 1681 36663 covers pread Agreed to close, no action. delete reviewers note. Action: delete reviewers note XCU 194 Notes to Reviewers CLOSED page 2218, overcome by XCUd5 ERN 9 Action: delete lines 194-196 931 Notes to Reviewers page 2235 asked folks to check STDOUT/STDIN sections of every utility for completeness and correctness Response: Leave as a global reviewers note for D6 but remove from here. Action: AJ add to global reviewers notes for D6 Action: delete the reviewers note here 5117 Notes to Reviewers CLOSED page 2351, ar utility, obe by PASC 1003.2 interp # 198 Action: delete lines 5117-5121 8320 Notes to Reviewers page 2428, _POSIX_V6 macros , is there a better name? Response: We agreed that we could not come up with a better name, no change, delete the reviewers note 8354 Notes to Reviewers p 2429 ditto Response: We agreed that we could not come up with a better name, no change, delete the reviewers note 8414 Notes to Reviewers p 2430 ditto Response: We agreed that we could not come up with a better name, no change, delete the reviewers note 10307 Notes to Reviewers CLOSED p 2481 , closed by PASC interp 1003.2 #194 delete 10307-10310 (other changes conveyed to editor in interp) 22039 Notes to Reviewers p 2788, m4 utility, Response: Proposal follows- Replace 22036-22052 The ifelse macro takes 3n+0 or 3n+1 arguments (for n>0). If the first two arguments compare as equal strings the defining text shall be the third argument. If the first two arguments do not compare as equal strings and one argument remains the default text shall be the fourth argument. If the first two arguments do not compare as equal strings and three or more arguments remain the first three arguments shall be discarded and processing shall restart with the remaining arguments. If two arguments remain the results are unspecified. ACTION: AJ to pass the proposal to the Base WG. CLOSED 25896 Notes to Reviewers p 2888, od , long long, do we need to invent a new character notation for it ? Propose NO. Response Agreed that no change was necessary. it is possible to specify it using a length modifier Action: delete the reviewers note 27331 Notes to Reviewers CLOSED p 2923, pax, closed by PASC interp 1003.2 #195 27349 Notes to Reviewers CLOSED p 2923, pax, closed by PASC interp 1003.2 #195 36025 Notes to Reviewers p 3150, umask, default output for umask needs to be specified looks like invention Response: The base standard is clear and we believe no change is required Action: delete revnote 38567 Notes to Reviewers p 3221, vi description of the R command I agree with the statement here that the description of the R command looks wrong. ACTION: AJ to file an interp. Once the interp is filed I will include the interp number. CLOSED Other Business ---------------- The ORs agreed to resolve issue 7 on the issues list The ORs took a vote to confirm that _POSIX_V6 is a suitable name for the macro and that the issue be closed (Issue 7 in the issues list). All in favor. The ORs took a vote on item 34 on the consent list (as modified below) 34. AGREED that all implementations shall support one or more environments where the widths of the following types are no greater than the width of type long: blksize_t, cc_t, mode_t, nfds_t, pid_t, ptrdiff_t, size_t, speed_t, ssize_t, suseconds_t, tcflag_t, useconds_t, wchar_t, wint_t Action: AJ. update SD/2. CLOSED After discussion with the ORs as a consequence it was agreed to action email 3011 for XBD ERN 322 Next call =========== The reviewers notes teleconference on March 27th has been cancelled. ----- Andrew Josey The Open Group Austin Group Chair Apex Plaza,Forbury Road,