Minutes of the 9 Sep 2010 Teleconference Austin-496 Page 1 of 1 Submitted by Andrew Josey, The Open Group. September 13 , 2010 Attendees Andrew Josey, The Open Group Don Cragun, PASC OR Nick Stoughton, USENIX, ISO/IEC OR Eric Blake, Red Hat Mark Brown, IBM, TOG OR Geoff Clare, The Open Group Apologies Ulrich Drepper, Red Hat * Old Business We reviewed the Actions carried forward from the last meeting : (The first three items below related to identification of TC1 candidate items) Identify the Aardvark from issue 6 that was not included in the revision (and not logged in SD5/Change History) nor an interpretation. This is likely to be XSHbug2.txt after ERN 234, XBDbug2.txt after ERN 105, XCUbug2.txt after ERN 174. Action: Andrew to identify which ERNs need to be considered. OPEN Identify Issue 6 Mantis Bugs not addressed in the revision. Action: Don will identify candidate items CLOSED Identify Issue 7 interpretations with non controversial changes. Open -- no volunteer assigned yet. Action: Andrew to start a new batch of interpretations for 30 day review. OPEN * New Business We commenced looking at Mantis bugs already closed that were not marked for TC1 to discuss whether they should be included. Bug 0000286: pthread_cond_timedwait() and cond vars http://austingroupbugs.net/view.php?id=286 We believe that this change makes the standard match historic practice. Therefore, this change is suitable for a TC and need not wait for the next revision of the standard. Bug 0000265: more -e contradiction http://austingroupbugs.net/view.php?id=265 We believe that this change makes the standard match historic practice. Therefore, this change is suitable for a TC and need not wait for the next revision of the standard. Bug 0000047: Missing specifiers for date utility http://austingroupbugs.net/view.php?id=47 The resolution of this item was changed from Accepted to Future Enhancement. We agreed that we need to mark all Future Enhancement items with the tag issue8. Geoff agreed to take the action which was closed during the meeting. Bug 0000190: Requiere environment variables and exit status information to man pages http://austingroupbugs.net/view.php?id=190 The resolution of this item was changed from Accepted as Marked to Future Enhancement. Bug 0000191: getopt and leading + in optstring http://austingroupbugs.net/view.php?id=191 The resolution of this item was changed from Accepted as Marked to Future Enhancement. * We picked up on regular Bug processing Bug 0000163: Problems in rationale of trap shell builtin Accepted as Marked This item had been reopened as it was now realized that not all the proposed changes were necessary. The changes were edited and the bug closed as Accepted as Marked. This is a TC1-2008 candidate. Proposed changes... On line 74932 and line 74934 change '$HOME/logout' to '"$HOME"/logout' At line 74956 change The command: trap '$cmd' 0 causes the contents of the shell variable cmd to be executed as a command when the shell exits. Using double-quotes instead of single-quotes might have unexpected behavior, since in theory the value of cmd might be a decimal integer which would be treated as a condition, not an action; or cmd might begin with '-'. Also, using double-quotes will cause the value of cmd to be expanded twice, once when trap is executed, and once when the condition arises. to The command: trap 'eval " $cmd"' 0 causes the contents of the shell variable cmd to be executed as a command when the shell exits. Using trap '$cmd' 0 does not work correctly if cmd contains any special characters such as quoting or redirections. Using trap " $cmd" 0 also works (the leading space character protects against unlikely cases where cmd is a decimal integer or begins with '-'), but it expands the cmd variable when the trap command is executed, not when the exit action is executed. At page 2343 line 74116 section eval, change the APPLICATION USAGE section from None to Since eval is not required to recognize the "--" end of options delimiter, in cases where the argument(s) to eval might begin with '-' it is recommended that the first argument is prefixed by a string that will not alter the commands to be executed, such as a space character: eval " $commands" or eval " $(some_command)" At page 2344 line 74126 section eval, change the RATIONALE section from None to This standard allows, but does not require, eval to recognize "--". Although this means applications cannot use "--" to protect against options supported as an extension (or errors reported for unsupported options), the nature of the eval utility is such that other means can be used to provide this protection (see APPLICATION USAGE above). Bug 000305 Allow RE handling to reject suspicious uses Accepted as Marked. http://austingroupbugs.net/view.php?id=305 This needs an interpretation. Interpretation response The standard states the required behaviour for regular expressions, and conforming implementations must conform to this. However, concerns have been raised about this which are being referred to the sponsor. Rationale: See Mantis Issue 305. Notes to the Editor (not part of this interpretation): At line 5995 (XBD 9.3.5), add a new paragraph: 8. If a bracket expression contains at least three list elements, where the first and last list element are the same single-character element of , , or , then it is unspecified whether the bracket expression will be treated as a collating symbol, equivalence class, or character class, respectively; treated as a matching list expression; or rejected as an error. At line 56541 (XSH regcomp), change: The following constants are defined as error return values: to: The following constants are defined as the minimum set of error return values, although other errors listed as implementation extensions in are possible: At line 116821 (XRAT A.9.3.5), add a new paragraph: The standard specifies three possible behaviors for regular expressions such as "[:alpha:]". One behavior is the traditional implementation, which behaves like "[:ahlp]". Another, for alignment with the tr utility, is to treat it like "[[:alpha:]]". And finally, the standard allows rejecting the regular expression as invalid, as a means of alerting a user to the non-portable aspect of that regular expression. The set of regular expressions with this undefined behavior is limited solely to the expressions where the outer '[' and ']' of the bracket expression can be confused with the missing bracket pair '[' and ']' necessary to form a collating symbol, equivalence class, or character class; thus "[_:alpha:]" or "[::]" do not trigger the unspecified behavior. Bug 0000304: islower() example has undefined behaviour when char is signed Accepted This is a TC1 candidate Bug 0000303: setlocale() LC_MESSAGES description refers to commands and utilities Accepted This is a TC1 candidate. Next Steps ---------- The next call will be on September 16th at 16:00 UK time/08:00 Pacific and will continue processing defect reports. http://austingroupbugs.net See the calendar for the list of dialup numbers. An IRC channel will be available for the meeting irc://irc.freestandards.org #austin ICAL: http://www.google.com/calendar/ical/nvctqtstkuni3fab9k3jqtrt4g@group.calendar.google.com/public/basic XML: http://www.google.com/calendar/feeds/nvctqtstkuni3fab9k3jqtrt4g@group.calendar.google.com/public/basic