summaryrefslogtreecommitdiff
path: root/ext/POSIX
Commit message (Collapse)AuthorAgeFilesLines
* Add conditional code to initialise RETVAL, to avoid compiler warnings.Nicholas Clark2008-05-311-0/+5
| | | | | (There was never an error, as croak() was called before the return). p4raw-id: //depot/perl@33956
* My recent changes to POSIX.xs forgot that WEXITSTATUS etc may not evenNicholas Clark2008-05-251-0/+24
| | | | | | | be defined. This fix changes the error message from "Your vendor has not defined POSIX macro %s, used" to "POSIX::%s not implemented on this architecture", which I assume is not going to break anything. p4raw-id: //depot/perl@33936
* Replaced the WEXITSTATUS, WIFEXITED, WIFSIGNALED, WIFSTOPPED, WSTOPSIGNicholas Clark2008-05-211-20/+31
| | | | | | | | and WTERMSIG wrappers with one wrapper using the XS "ALIAS" feature. This gets the shared object size back below the size before the removal of int_macro_int. It looks like there are other space savings to be made this way. p4raw-id: //depot/perl@33897
* Eliminate POSIX::int_macro_int, and all the complex AUTOLOAD fandangoNicholas Clark2008-05-212-165/+27
| | | | | | | | that creates closures round it. Instead, wrap WEXITSTATUS, WIFEXITED, WIFSIGNALED, WIFSTOPPED, WSTOPSIG and WTERMSIG directly with XS. The shared library is slightly larger, but dynamic memory usage savings beat this, even within one thread of one process. Simpler code too. p4raw-id: //depot/perl@33896
* Remove POSIX's internal implementation of S_ISBLK, S_ISCHR, S_ISDIR,Nicholas Clark2008-05-152-73/+12
| | | | | | | S_ISFIFO and S_ISREG, and pull them in from Fcntl. Spotted as a result of bug #54186, but there has been a redefined subroutine warning for ages if you elected to import all of POSIX and Fcntl's exports. p4raw-id: //depot/perl@33826
* It transpires that POSIX.xs also duplicated several constants definedNicholas Clark2008-05-152-6/+6
| | | | | | | by Fcntl but only conditionally exported by Fcntl. The most obvious were SEEK_CUR, SEEK_END and SEEK_SET, as reported in bug #54186. So add them to the list of constants that POSIX imports from Fcntl. p4raw-id: //depot/perl@33825
* $VERSION++ for all the non-dual life modules in ext/ thatNicholas Clark2008-03-311-1/+1
| | | | | Porting/cmpVERSION.pl reports differ from the 5.10.0 release. p4raw-id: //depot/perl@33621
* Make ext/POSIX/t/sysconf.t use File::Spec->tmpdir() for pathconf/Nicholas Clark2008-02-291-6/+10
| | | | | | fpathconf testing rather than ->curdir(), as the latter can be on networked storage that fails the syscall (and the tests). p4raw-id: //depot/perl@33401
* mistaken variable in podYitzchak Scott-Thoennes2008-01-291-1/+1
| | | | | | From: "Yitzchak Scott-Thoennes" <sthoenna@efn.org> Message-ID: <40783.64.81.167.122.1201556369.squirrel@webmail.efn.org> p4raw-id: //depot/perl@33101
* From: "Robin Barker" <Robin.Barker@npl.co.uk>Robin Barker2008-01-111-2/+2
| | | | | | | Message-ID: <46A0F33545E63740BC7563DE59CA9C6D0939C3@exchsvr2.npl.ad.local> More consting. p4raw-id: //depot/perl@32947
* ext/ constingRobin Barker2007-12-221-2/+2
| | | | | | | From: "Robin Barker" <Robin.Barker@npl.co.uk> Message-ID: <46A0F33545E63740BC7563DE59CA9C6D09399D@exchsvr2.npl.ad.local> Date: Sat, 22 Dec 2007 00:39:47 -0000 p4raw-id: //depot/perl@32703
* Warning shutdowns, by Robin BarkerRafael Garcia-Suarez2007-11-281-3/+3
| | | p4raw-id: //depot/perl@32539
* Re: fpathconf test failures on QNXMatt Kraai2007-11-281-8/+12
| | | | | Message-ID: <20071127190019.GA6319@ftbfs.org> p4raw-id: //depot/perl@32524
* Correct parentheses. Calling Win32::GetOSVersion() on non-win32 wouldNicholas Clark2007-08-161-2/+2
| | | | | be bad, m'kay. p4raw-id: //depot/perl@31725
* Skip the POSIX::strftime() time test with a 60sec parameter on Vista:Steve Hay2007-08-151-3/+5
| | | | | | | | | | | it seems to have the same problem as when built with the VC8 CRT (Fixes perl #44589) Also, change the %D to the equivalent %m/%d/%y because %D isn't supported at all by Microsoft's CRT and just converts to nothing with VC6's CRT and causes another invalid parameter error with VC8 and Vista p4raw-id: //depot/perl@31721
* Fix POSIX::setlocale(): the CRT function returns a pointer to aSteve Hay2007-06-271-3/+12
| | | | | | | | buffer that may be overwritten by subsequent calls to the CRT function, so we must make a safe copy of that buffer for our own use. This fixes lib/locale.t on Win32 with the Borland compiler, but presumably could affect other compilers too. p4raw-id: //depot/perl@31482
* DBL_EPSILON DBL_MIN FLT_EPSILON FLT_MIN are not C constant expressionsNicholas Clark2007-05-291-7/+6
| | | | | on z/OS, so move them to the list of non const constants. p4raw-id: //depot/perl@31308
* Documentation fixes for EGID/GID confusionDavid Leadbeater2007-05-041-1/+1
| | | | | Message-ID: <20070503172713.GA23180@sirius.otherwize.co.uk> p4raw-id: //depot/perl@31137
* Fix spelling nit.Steve Peters2007-04-111-1/+1
| | | p4raw-id: //depot/perl@30912
* Cast needed to get POSIX compiling with g++ on FreeBSD.Nicholas Clark2007-04-101-1/+1
| | | p4raw-id: //depot/perl@30898
* Only use proxy constant subroutines for 5.9.x.Nicholas Clark2007-03-211-1/+1
| | | p4raw-id: //depot/perl@30671
* Make the isdst argument to asctime and mktime default to -1Rafael Garcia-Suarez2007-03-153-6/+6
| | | | | instead of 0, as suggested by Mike Schilli. p4raw-id: //depot/perl@30590
* Fix ext/POSIX/t/sysconf.t failures on Cygwin.Steve Peters2007-01-261-1/+8
| | | p4raw-id: //depot/perl@30014
* do $file; won't propagate errors from die, as do is an implicit eval.Nicholas Clark2007-01-082-2/+2
| | | | | So need to propagate errors with $@. p4raw-id: //depot/perl@29723
* Add a regression test to verify that POSIX::AUTOLOAD works,Rafael Garcia-Suarez2006-12-192-1/+5
| | | | | and silence a potential warning from it. p4raw-id: //depot/perl@29592
* Disable strictures in POSIX AUTOLOADRafael Garcia-Suarez2006-12-191-0/+1
| | | p4raw-id: //depot/perl@29591
* Ensure that POSIX.pm is strict and warnings compliant throughout ifNicholas Clark2006-12-181-3/+3
| | | | | the __END__ is temporarily removed. p4raw-id: //depot/perl@29587
* Given that POSIX already has AutoLoader loaded, move as much asNicholas Clark2006-12-181-57/+61
| | | | | | possible of POSIX::SigRt out to AutoLoadLand, so that it won't be loaded unless someone starts using the realtime signals interface. p4raw-id: //depot/perl@29586
* POSIX::SigAction::new can be moved into the AUTOLOAD section.Nicholas Clark2006-12-181-1/+1
| | | p4raw-id: //depot/perl@29571
* Silence VC++ 8 warnings about "possible loss of data"Steve Hay2006-11-281-1/+1
| | | p4raw-id: //depot/perl@29408
* POSIX::remove() and directoriesPeter Dintelmann2006-11-221-1/+1
| | | | | | From: "Dintelmann, Peter" <Peter.Dintelmann@Dresdner-Bank.com> Message-ID: <3852726AFA94DE4F87E616F13F416A5F0A912F@naimucu1.ffz00k.rootdom.net> p4raw-id: //depot/perl@29353
* VC8 regards 60 seconds as an invalid parameter for strftime()Steve Hay2006-11-221-1/+6
| | | | | so skip the test in that case. p4raw-id: //depot/perl@29352
* Re: Off by one in the trie code?Yves Orton2006-10-191-1/+1
| | | | | | | | Message-ID: <9b18b3110610181151i3ca438cdied769ebaa4255079@mail.gmail.com> change test files that do a require "./test.pl"; without a BEGIN block to ensure prototypes are seen, plus fix any breakage this reveals. p4raw-id: //depot/perl@29056
* return value of -1 without errno set is ok in Craig A. Berry2006-09-161-38/+33
| | | | | | ext/POSIX/t/sysconf.t (it just means the feature is not implemented, not defined, or has no limit) p4raw-id: //depot/perl@28851
* Additional changes to get C++ a little closer to a clean compileSteve Peters2006-08-162-2/+2
| | | | | of Perl. p4raw-id: //depot/perl@28729
* Make sysconf tests handle unimplemented success indicationsCraig A. Berry2006-07-141-3/+4
| | | | | for the benefit of Mac OS X and VMS. p4raw-id: //depot/perl@28574
* Skip tests for pathconf() and fpathconf() on HP-UX for Steve Peters2006-07-141-8/+16
| | | | | | _PC_CHOWN_RESTRICTED. The HP-UX manpage suggests not trying to do it, and that the errno will not be set on failure. p4raw-id: //depot/perl@28572
* sysconf.t: still failing in tru64, try harder to skipJarkko Hietaniemi2006-07-111-10/+32
| | | | | Message-Id: <200607110623.k6B6N8Ni488812@kosh.hut.fi> p4raw-id: //depot/perl@28540
* Skip tests of a POSIX constant on Mac OS X because saved IDs are borkedDominic Dunlop2006-07-101-4/+8
| | | | | Message-Id: <F1942C06-CDE3-47C6-AC5A-4358960E0F6E@mac.com> p4raw-id: //depot/perl@28520
* Skip some POSIX tests when the thing they are testing is unimplementedYves Orton2006-07-082-9/+22
| | | | | Message-ID: <9b18b3110607080855g73407c4fx41a578815c2a6da2@mail.gmail.com> p4raw-id: //depot/perl@28508
* POSIX test improvements on True64Jarkko Hietaniemi2006-07-083-15/+94
| | | | | | Subject: [PATCH] the new POSIX tests Message-ID: <44AF7019.3070509@iki.fi> p4raw-id: //depot/perl@28505
* Test scripts for I18N::Langinfo and POSIXSébastien Aperghis-Tramoni2006-07-072-0/+111
| | | | | | | | | | Message-ID: <1152262951.44ae23272ffa0@imp3-g19.free.fr> Only includes changes to: * ext/I18N/Langinfo/t/Langinfo.t * ext/POSIX/t/sysconf.t * ext/POSIX/t/termios.t p4raw-id: //depot/perl@28503
* Add constants needed for recvmsg()/sendmsg() support.Steve Peters2006-07-051-1/+2
| | | p4raw-id: //depot/perl@28486
* Two signed array indicies that Coverity spots should be unsigned.Nicholas Clark2006-05-091-2/+2
| | | p4raw-id: //depot/perl@28142
* Disable SIGRTMAX and SIGRTMIN in POSIX if Configure's probes found thatNicholas Clark2006-05-011-2/+18
| | | | | they aren't viable. (Certain glibcs are "interesting") p4raw-id: //depot/perl@28039
* Need to update test count from change #27440.Steve Peters2006-03-091-1/+1
| | | | | p4raw-link: @27440 on //depot/perl: 516d25e8e8c09c6c60bf2f46703fc4d5add0f5fb p4raw-id: //depot/perl@27441
* croak in POSIX::sigaction() when passed a negative signal insteadSteve Peters2006-03-092-0/+7
| | | | | dumping core. p4raw-id: //depot/perl@27440
* Solaris seems not to enjoy initialising constants from SIG_??? on 64Nicholas Clark2006-01-291-3/+3
| | | | | bit builds. p4raw-id: //depot/perl@26996
* Change all NEWSV() to newSV() in the core and non-dual-lived modules.Steve Hay2006-01-181-1/+1
| | | | | | | | Keep NEWSV() itself for backwards-compatibility outside of the core, but don't advertise it any more. (cf. change #25101). p4raw-link: @25101 on //depot/perl: a02a5408b2f199007c4dcb74559cc79066307ada p4raw-id: //depot/perl@26901
* A few more places that can use hv_fetchs().Gisle Aas2006-01-111-8/+8
| | | | | Ref change 26676. p4raw-id: //depot/perl@26795
View Lorry Controller Status