{"thread":{"id":"14306","subject":"[ANNOUNCE] GIT 1.5.6.2","startedAt":"2008-07-06T05:34:15Z","lastAt":"2008-07-14T03:13:05Z","messageCount":9,"participants":["Junio C Hamano","Jakub Narebski","Jan Engelhardt","Mikael Magnusson","Michael J Gruber","Jeff King"],"isPatch":false,"patchVersion":null,"patchTotal":null},"messages":[{"id":"82337","messageId":"7vy74fo9t4.fsf@gitster.siamese.dyndns.org","threadId":"14306","inReplyTo":null,"subject":"[ANNOUNCE] GIT 1.5.6.2","fromName":"Junio C Hamano","fromEmail":"gitster@pobox.com","sentAt":"2008-07-06T05:34:15Z","receivedAt":"2008-07-06T05:34:15Z","isPatch":false,"sender":{"key":"gitster@pobox.com","avatar":"https://avatars.githubusercontent.com/u/54884?v=4"},"body":"The latest maintenance release GIT 1.5.6.2 are available at the usual\nplaces:\n\n http://www.kernel.org/pub/software/scm/git/\n\n git-1.5.6.2.tar.{gz,bz2}\t\t\t(source tarball)\n git-htmldocs-1.5.6.2.tar.{gz,bz2}\t\t(preformatted docs)\n git-manpages-1.5.6.2.tar.{gz,bz2}\t\t(preformatted docs)\n\nThe RPM binary packages for a few architectures are also provided as\ncourtesy. As the machine I cut them at kernel.org has been updated to FC9\nrecently, their filenames now have .fc9. in them.\n\n RPMS/$arch/git-*-1.5.6.2-1.fc9.$arch.rpm\t(RPM)\n\nThere still are a handful more fixes queued for 'maint' so we might see\nanother maintenance release, 1.5.6.3, not too far in the future.\n\n----------------------------------------------------------------\n\nChanges since v1.5.6.1 are as follows:\n\nAvery Pennarun (1):\n git-svn: avoid filling up the disk with temp files.\n\nBjörn Steinbrink (1):\n git cat-file: Fix memory leak in batch mode\n\nEric Wong (1):\n git-svn: don't sanitize remote names in config\n\nJeff King (3):\n fetch: report local storage errors in status table\n doc/rev-parse: clarify reflog vs --until for specifying revisions\n fetch: give a hint to the user when local refs fail to update\n\nJochen Voss (1):\n avoid off-by-one error in run_upload_archive\n\nJoey Hess (1):\n fix git config example syntax\n\nJohannes Schindelin (2):\n clone: respect url.insteadOf setting in global configs\n clone: respect the settings in $HOME/.gitconfig and /etc/gitconfig\n\nJunio C Hamano (9):\n Allow \"git-reset path\" when unambiguous\n diff --check: do not discard error status upon seeing a good line\n git-shell: accept \"git foo\" form\n GIT 1.5.4.6\n GIT 1.5.5.5\n Start draft release notes for 1.5.6.2\n Work around gcc warnings from curl headers\n Fix executable bits in t/ scripts\n GIT 1.5.6.2\n\nShawn O. Pearce (1):\n Fix describe --tags --long so it does not segfault\n\nThomas Rast (1):\n Fix 'git show' on signed tag of signed tag of commit\n"},{"id":"82341","messageId":"m38wwffn3b.fsf@localhost.localdomain","threadId":"14306","inReplyTo":"7vy74fo9t4.fsf@gitster.siamese.dyndns.org","subject":"Re: [ANNOUNCE] GIT 1.5.6.2","fromName":"Jakub Narebski","fromEmail":"jnareb@gmail.com","sentAt":"2008-07-06T08:09:08Z","receivedAt":"2008-07-06T08:09:08Z","isPatch":false,"sender":{"key":"jnareb@gmail.com","avatar":"https://avatars.githubusercontent.com/u/2706?v=4"},"body":"Junio C Hamano writes:\n\n> The RPM binary packages for a few architectures are also provided as\n> courtesy. As the machine I cut them at kernel.org has been updated to FC9\n> recently, their filenames now have .fc9. in them.\n> \n> RPMS/$arch/git-*-1.5.6.2-1.fc9.$arch.rpm\t(RPM)\n\nThis unfortunately includes also SRPMS (.src.rpm), even though the\nsource RPM from git repository is not distribution specific. I think\nthat src.rpm should drop '.fc9.' from name\n\n-- \nJakub Narebski\nPoland\nShadeHawk on #git\n"},{"id":"82347","messageId":"7vprprgy1f.fsf@gitster.siamese.dyndns.org","threadId":"14306","inReplyTo":"m38wwffn3b.fsf@localhost.localdomain","subject":"Re: [ANNOUNCE] GIT 1.5.6.2","fromName":"Junio C Hamano","fromEmail":"gitster@pobox.com","sentAt":"2008-07-06T09:30:20Z","receivedAt":"2008-07-06T09:30:20Z","isPatch":false,"sender":{"key":"gitster@pobox.com","avatar":"https://avatars.githubusercontent.com/u/54884?v=4"},"body":"Jakub Narebski writes:\n\n> Junio C Hamano writes:\n>\n>> The RPM binary packages for a few architectures are also provided as\n>> courtesy. As the machine I cut them at kernel.org has been updated to FC9\n>> recently, their filenames now have .fc9. in them.\n>> \n>> RPMS/$arch/git-*-1.5.6.2-1.fc9.$arch.rpm\t(RPM)\n>\n> This unfortunately includes also SRPMS (.src.rpm), even though the\n> source RPM from git repository is not distribution specific. I think\n> that src.rpm should drop '.fc9.' from name\n\nIt is whatever rpmbuild command does. Not my problem.\n\nHaving said that, I vaguely recall some people from non Fedora nor RH\ndistros based on RPM were talking about how their .spec file would be\ndifferent from the one for FC. Presumably the resulting SRPM would be\nalso distro specific, wouldn't it?\n"},{"id":"82361","messageId":"alpine.LNX.1.10.0807061444250.28765@fbirervta.pbzchgretzou.qr","threadId":"14306","inReplyTo":"7vy74fo9t4.fsf@gitster.siamese.dyndns.org","subject":"Re: [ANNOUNCE] GIT 1.5.6.2","fromName":"Jan Engelhardt","fromEmail":"jengelh@medozas.de","sentAt":"2008-07-06T12:44:54Z","receivedAt":"2008-07-06T12:44:54Z","isPatch":false,"sender":{"key":"jengelh@medozas.de","avatar":null},"body":"\nOn Sunday 2008-07-06 07:34, Junio C Hamano wrote:\n>Junio C Hamano (9):\n> Allow \"git-reset path\" when unambiguous\n> diff --check: do not discard error status upon seeing a good line\n> git-shell: accept \"git foo\" form\n> GIT 1.5.4.6\n> GIT 1.5.5.5\n> Start draft release notes for 1.5.6.2\n> Work around gcc warnings from curl headers\n> Fix executable bits in t/ scripts\n> GIT 1.5.6.2\n\nThree git versions?\n"},{"id":"82373","messageId":"237967ef0807060754t4bf0ab8el7362e69aac4699b5@mail.gmail.com","threadId":"14306","inReplyTo":"alpine.LNX.1.10.0807061444250.28765@fbirervta.pbzchgretzou.qr","subject":"Re: [ANNOUNCE] GIT 1.5.6.2","fromName":"Mikael Magnusson","fromEmail":"mikachu@gmail.com","sentAt":"2008-07-06T14:54:00Z","receivedAt":"2008-07-06T14:54:00Z","isPatch":false,"sender":{"key":"mikachu@gmail.com","avatar":null},"body":"2008/7/6 Jan Engelhardt :\n>\n> On Sunday 2008-07-06 07:34, Junio C Hamano wrote:\n>>Junio C Hamano (9):\n>> Allow \"git-reset path\" when unambiguous\n>> diff --check: do not discard error status upon seeing a good line\n>> git-shell: accept \"git foo\" form\n>> GIT 1.5.4.6\n>> GIT 1.5.5.5\n>> Start draft release notes for 1.5.6.2\n>> Work around gcc warnings from curl headers\n>> Fix executable bits in t/ scripts\n>> GIT 1.5.6.2\n>\n> Three git versions?\n\nI believe this is related to moving to the dash-less form of receive-pack et al.\n\n-- \nMikael Magnusson\n"},{"id":"82441","messageId":"g4t342$9m8$1@ger.gmane.org","threadId":"14306","inReplyTo":"7vy74fo9t4.fsf@gitster.siamese.dyndns.org","subject":"Test breakage [Re: [ANNOUNCE] GIT 1.5.6.2]","fromName":"Michael J Gruber","fromEmail":"michaeljgruber+gmane@fastmail.fm","sentAt":"2008-07-07T12:44:49Z","receivedAt":"2008-07-07T12:44:49Z","isPatch":false,"sender":{"key":"git@grubix.eu","avatar":"https://avatars.githubusercontent.com/u/233215?v=4"},"body":"Junio C Hamano venit, vidit, dixit 06.07.2008 07:34:\n> The latest maintenance release GIT 1.5.6.2 are available at the usual\n> places:\n...\n\nOn Kubuntu 7.10 (not my choice, this is at work...), \"make test\" gives:\n\n*** t9600-cvsimport.sh ***\n* ok 1: setup cvsroot\n* ok 2: setup a cvs module\n* ok 3: import a trivial module\n* ok 4: pack refs\n* ok 5: update cvs module\n* FAIL 6: update git module\n\n\n cd module-git &&\n git cvsimport -a -z 0 module &&\n git merge origin &&\n cd .. &&\n test_cmp module-cvs/o_fortuna module-git/o_fortuna\n\n\n* ok 7: update cvs module\n* ok 8: cvsimport.module config works\n* ok 9: import from a CVS working tree\n* failed 1 among 9 test(s)\n\nOn the second (and subsequent) run, \"make test\" succeeds. I tried again \nwith a fresh copy from the tarball, same effect: failure on 1, success \non 2 and following.\n\nThis is with LANG=de_DE.UTF-8, whereas with LANG=C the tests succeed the \nfirst time already. Are tests supposed to be done in C locale only? Do \nthey reuse data from a previous test run? I didn't notice this with 1.5.6.1.\n\nMichael\n"},{"id":"82552","messageId":"20080708051622.GC2037@sigill.intra.peff.net","threadId":"14306","inReplyTo":"g4t342$9m8$1@ger.gmane.org","subject":"Re: Test breakage [Re: [ANNOUNCE] GIT 1.5.6.2]","fromName":"Jeff King","fromEmail":"peff@peff.net","sentAt":"2008-07-08T05:16:22Z","receivedAt":"2008-07-08T05:16:22Z","isPatch":false,"sender":{"key":"peff@peff.net","avatar":"https://avatars.githubusercontent.com/u/45925?v=4"},"body":"On Mon, Jul 07, 2008 at 02:44:49PM +0200, Michael J Gruber wrote:\n\n> On Kubuntu 7.10 (not my choice, this is at work...), \"make test\" gives:\n>\n> *** t9600-cvsimport.sh ***\n> [...]\n> * FAIL 6: update git module\n> cd module-git &&\n> git cvsimport -a -z 0 module &&\n> git merge origin &&\n> cd .. &&\n> test_cmp module-cvs/o_fortuna module-git/o_fortuna\n>\n> On the second (and subsequent) run, \"make test\" succeeds. I tried again \n> with a fresh copy from the tarball, same effect: failure on 1, success on \n> 2 and following.\n\nI tried reproducing on my Debian box with various weaks of $LANG, but I\ncouldn't. Can you please try running \"./t9600-cvsimport.sh -v -i\" and\nsend us the output of a failing instance. It's not clear which step in\nfailing test is causing the problem, or if cvsps is giving off any\nuseful errors.\n\n> This is with LANG=de_DE.UTF-8, whereas with LANG=C the tests succeed the \n> first time already. Are tests supposed to be done in C locale only? Do \n> they reuse data from a previous test run? I didn't notice this with \n> 1.5.6.1.\n\nThey should be perfectly repeatable, but I have run into problems in the\npast with cruft in my ~/.cvsps directory. However, we should be setting\n$HOME properly in t9600 to avoid this, so I'm not sure what the culprit\nis.\n\n-Peff\n"},{"id":"82570","messageId":"g4vd41$psa$1@ger.gmane.org","threadId":"14306","inReplyTo":"20080708051622.GC2037@sigill.intra.peff.net","subject":"Re: Test breakage [Re: [ANNOUNCE] GIT 1.5.6.2]","fromName":"Michael J Gruber","fromEmail":"michaeljgruber+gmane@fastmail.fm","sentAt":"2008-07-08T09:47:45Z","receivedAt":"2008-07-08T09:47:45Z","isPatch":false,"sender":{"key":"git@grubix.eu","avatar":"https://avatars.githubusercontent.com/u/233215?v=4"},"body":"Jeff King venit, vidit, dixit 08.07.2008 07:16:\n> On Mon, Jul 07, 2008 at 02:44:49PM +0200, Michael J Gruber wrote:\n> \n>> On Kubuntu 7.10 (not my choice, this is at work...), \"make test\" gives:\n>>\n>> *** t9600-cvsimport.sh ***\n>> [...]\n>> * FAIL 6: update git module\n>> cd module-git &&\n>> git cvsimport -a -z 0 module &&\n>> git merge origin &&\n>> cd .. &&\n>> test_cmp module-cvs/o_fortuna module-git/o_fortuna\n>>\n>> On the second (and subsequent) run, \"make test\" succeeds. I tried again \n>> with a fresh copy from the tarball, same effect: failure on 1, success on \n>> 2 and following.\n> \n> I tried reproducing on my Debian box with various weaks of $LANG, but I\n> couldn't. Can you please try running \"./t9600-cvsimport.sh -v -i\" and\n> send us the output of a failing instance. It's not clear which step in\n> failing test is causing the problem, or if cvsps is giving off any\n> useful errors.\n> \n>> This is with LANG=de_DE.UTF-8, whereas with LANG=C the tests succeed the \n>> first time already. Are tests supposed to be done in C locale only? Do \n>> they reuse data from a previous test run? I didn't notice this with \n>> 1.5.6.1.\n> \n> They should be perfectly repeatable, but I have run into problems in the\n> past with cruft in my ~/.cvsps directory. However, we should be setting\n> $HOME properly in t9600 to avoid this, so I'm not sure what the culprit\n> is.\n\nSure enough, I can't reproduce it today. I reproduced that effect \nseveral times yesterday, or else I would not have reported anything. \nTrying the very same steps today the tests succeed on the first run. \nOnly differences between yesterday and today are be:\n\n- I installed git 1.5.6.2 now (had 1.5.6.1 before) under $HOME, with \nPATH, PERL5LIB, LD_LIBRARY_PATH set accordingly; but tests should pick \nup git stuff from the local build dir, right?\n\n- There may have been an automated update to kubuntu meanwhile, but I \ndon't think so; cvsps is under $HOME anyways, and what else would \nmatter? Perl version maybe?\n\nSo it's WORKSFORME now. Sorry for the noise. Next time I'll not only try \nto reproduce (as I did) but wait a day also!\n\nBest\nMichael\n"},{"id":"83209","messageId":"7v1w1xma4e.fsf@gitster.siamese.dyndns.org","threadId":"14306","inReplyTo":"7vy74fo9t4.fsf@gitster.siamese.dyndns.org","subject":"[ANNOUNCE] GIT 1.5.6.3","fromName":"Junio C Hamano","fromEmail":"gitster@pobox.com","sentAt":"2008-07-14T03:13:05Z","receivedAt":"2008-07-14T03:13:05Z","isPatch":false,"sender":{"key":"gitster@pobox.com","avatar":"https://avatars.githubusercontent.com/u/54884?v=4"},"body":"The latest maintenance release GIT 1.5.6.3 is available at the\nusual places:\n\n http://www.kernel.org/pub/software/scm/git/\n\n git-1.5.6.3.tar.{gz,bz2}\t\t\t(source tarball)\n git-htmldocs-1.5.6.3.tar.{gz,bz2}\t\t(preformatted docs)\n git-manpages-1.5.6.3.tar.{gz,bz2}\t\t(preformatted docs)\n\nThe RPM binary packages for a few architectures are also provided\nas courtesy. As the machine I cut them at kernel.org has been updated to FC9\nrecently, their filenames now have \"fc9\" in them.\n\n RPMS/$arch/git-*-1.5.6.3-1.fc9.$arch.rpm\t(RPM)\n\nThe changes since the previous maintenance release is getting smaller.\nAbout half the diff is in Documentation/ section.\n\n Documentation/RelNotes-1.5.6.3.txt | 52 ++++++++++++++++++++++++++++++++\n Documentation/git-cvsimport.txt | 6 ++++\n Documentation/git-cvsserver.txt | 3 ++\n Documentation/git-mailinfo.txt | 5 ++-\n Documentation/git-rev-parse.txt | 4 +-\n Documentation/git.txt | 3 +-\n Documentation/gitattributes.txt | 33 ++++++++++----------\n Documentation/gitcvs-migration.txt | 5 +++\n Documentation/pretty-formats.txt | 8 ++--\n RelNotes | 2 +-\n builtin-clone.c | 8 +++--\n builtin-commit.c | 11 ++----\n builtin-log.c | 16 ++-------\n builtin-mailinfo.c | 3 +-\n builtin-send-pack.c | 3 +-\n builtin-tag.c | 2 +-\n contrib/completion/git-completion.bash | 30 ++++++++++++++++++\n convert.c | 25 +++++----------\n diff.c | 8 +----\n git-am.sh | 3 +-\n git-submodule.sh | 3 +-\n git-svn.perl | 2 +\n http.c | 36 +++++++--------------\n path.c | 2 +-\n read-cache.c | 16 ++++++++++\n refs.c | 2 +-\n run-command.c | 2 +\n t/t0004-unwritable.sh | 19 ++++++-----\n t/t1301-shared-repo.sh | 23 ++++++++++++++\n t/t5404-tracking-branches.sh | 7 ++++\n t/t5601-clone.sh | 20 +++++++++++-\n transport.c | 4 ++-\n 32 files changed, 253 insertions(+), 113 deletions(-)\n\n----------------------------------------------------------------\n\nGIT v1.5.6.3 Release Notes\n==========================\n\nFixes since v1.5.6.2\n--------------------\n\n* Setting core.sharerepository to traditional \"true\" value was supposed to make\n the repository group writable but should not affect permission for others.\n However, since 1.5.6, it was broken to drop permission for others when umask is\n 022, making the repository unreadable by others.\n\n* Setting GIT_TRACE will report spawning of external process via run_command().\n\n* Using an object with very deep delta chain pinned memory needed for extracting\n intermediate base objects unnecessarily long, leading to excess memory usage.\n\n* Bash completion script did not notice '--' marker on the command\n line and tried the relatively slow \"ref completion\" even when\n completing arguments after one.\n\n* Registering a non-empty blob racily and then truncating the working\n tree file for it confused \"racy-git avoidance\" logic into thinking\n that the path is now unchanged.\n\n* The section that describes attributes related to git-archive were placed\n in a wrong place in the gitattributes(5) manual page.\n\n* \"git am\" was not helpful to the users when it detected that the committer\n information is not set up properly yet.\n\n* \"git clone\" had a leftover debugging fprintf().\n\n* \"git clone -q\" was not quiet enough as it used to and gave object count\n and progress reports.\n\n* \"git clone\" marked downloaded packfile with .keep; this could be a\n good thing if the remote side is well packed but otherwise not,\n especially for a project that is not really big.\n\n* \"git daemon\" used to call syslog() from a signal handler, which\n could raise signals of its own but generally is not reentrant. This\n was fixed by restructuring the code to report syslog() after the handler\n returns.\n\n* When \"git push\" tries to remove a remote ref, and corresponding\n tracking ref is missing, we used to report error (i.e. failure to\n remove something that does not exist).\n\n* \"git mailinfo\" (hence \"git am\") did not handle commit log messages in a\n MIME multipart mail correctly.\n\nContains other various documentation fixes.\n"}]}