{"thread":{"id":"21839","subject":"[ANNOUNCE] Git 1.6.5.4","startedAt":"2009-12-03T09:21:13Z","lastAt":"2009-12-07T15:18:01Z","messageCount":21,"participants":["Junio C Hamano","Jeff King","Andreas Schwab","Michael J Gruber","Todd Zullinger","Eugene Sajine","Matthieu Moy"],"isPatch":false,"patchVersion":null,"patchTotal":null},"messages":[{"id":"129110","messageId":"7v638o76ra.fsf@alter.siamese.dyndns.org","threadId":"21839","inReplyTo":null,"subject":"[ANNOUNCE] Git 1.6.5.4","fromName":"Junio C Hamano","fromEmail":"gitster@pobox.com","sentAt":"2009-12-03T09:21:13Z","receivedAt":"2009-12-03T09:21:13Z","isPatch":false,"sender":{"key":"gitster@pobox.com","avatar":"https://avatars.githubusercontent.com/u/54884?v=4"},"body":"The latest maintenance release Git 1.6.5.4 is available at the\nusual places:\n\n http://www.kernel.org/pub/software/scm/git/\n\n git-1.6.5.4.tar.{gz,bz2}\t\t\t(source tarball)\n git-htmldocs-1.6.5.4.tar.{gz,bz2}\t\t(preformatted docs)\n git-manpages-1.6.5.4.tar.{gz,bz2}\t\t(preformatted docs)\n\nThe RPM binary packages for a few architectures are found in:\n\n RPMS/$arch/git-*-1.6.5.4-1.fc11.$arch.rpm\t(RPM)\n\nContains some minor fixes that have been accumulated; all of them\nappear in the upcoming 1.6.6 release as well.\n\nThis should fix the problem that man pages formatted on FC11 boxes are\nlittered with \"man.base.url.for.relative.link\" strings reported earlier\ntoday.\n\nGit v1.6.5.4 Release Notes\n==========================\n\nFixes since v1.6.5.3\n--------------------\n\n * \"git help\" (without argument) used to check if you are in a directory\n under git control. There was no breakage in behaviour per-se, but this\n was unnecessary.\n\n * \"git prune-packed\" gave progress output even when its standard error is\n not connected to a terminal; this caused cron jobs that run it to\n produce crufts.\n\n * \"git pack-objects --all-progress\" is an option to ask progress output\n from write-object phase _if_ progress output were to be produced, and\n shouldn't have forced the progress output.\n\n * \"git apply -p --directory=\" did not work well for a\n non-default value of n.\n\n * \"git merge foo HEAD\" was misparsed as an old-style invocation of the\n command and produced a confusing error message. As it does not specify\n any other branch to merge, it shouldn't be mistaken as such. We will\n remove the old style \"git merge HEAD ...\" syntax in\n future versions, but not in this release,\n\n * \"git merge -m ...\" added the standard merge message\n on its own after user-supplied message, which should have overrided the\n standard one.\n\nOther minor documentation updates are included.\n\n----------------------------------------------------------------\n\nChanges since v1.6.5.3 are as follows:\n\nDavid Aguilar (1):\n help: Do not unnecessarily look for a repository\n\nDavid Soria Parra (1):\n Documentation: Document --branch option in git clone synopsis\n\nGreg Price (1):\n Documentation: undocument gc'd function graph_release()\n\nJeff King (1):\n prune-packed: only show progress when stderr is a tty\n\nJunio C Hamano (7):\n builtin-apply.c: pay attention to -p when determining the name\n Do not misidentify \"git merge foo HEAD\" as an old-style invocation\n merge: do not add standard message when message is given with -m option\n Prepare for 1.6.5.4\n Documentation/Makefile: allow man.base.url.for.relative.link to be set from Make\n Unconditionally set man.base.url.for.relative.links\n Git 1.6.5.4\n\nMichael J Gruber (1):\n Documentation: Fix a few i.e./e.g. mix-ups\n\nNicolas Pitre (1):\n pack-objects: split implications of --all-progress from progress activation\n\nStephen Boyd (1):\n instaweb: restart server if already running\n"},{"id":"129111","messageId":"20091203092521.GA3620@coredump.intra.peff.net","threadId":"21839","inReplyTo":"7v638o76ra.fsf@alter.siamese.dyndns.org","subject":"Re: [ANNOUNCE] Git 1.6.5.4","fromName":"Jeff King","fromEmail":"peff@peff.net","sentAt":"2009-12-03T09:25:21Z","receivedAt":"2009-12-03T09:25:21Z","isPatch":false,"sender":{"key":"peff@peff.net","avatar":"https://avatars.githubusercontent.com/u/45925?v=4"},"body":"On Thu, Dec 03, 2009 at 01:21:13AM -0800, Junio C Hamano wrote:\n\n> * \"git pack-objects --all-progress\" is an option to ask progress output\n> from write-object phase _if_ progress output were to be produced, and\n> shouldn't have forced the progress output.\n\nShouldn't this actually be --all-progress-implied? Nico's patch\nintentionally kept --all-progress with the same behavior.\n\n-Peff\n"},{"id":"129116","messageId":"m2hbs85koj.fsf@igel.home","threadId":"21839","inReplyTo":"7v638o76ra.fsf@alter.siamese.dyndns.org","subject":"Re: [ANNOUNCE] Git 1.6.5.4","fromName":"Andreas Schwab","fromEmail":"schwab@linux-m68k.org","sentAt":"2009-12-03T12:03:24Z","receivedAt":"2009-12-03T12:03:24Z","isPatch":false,"sender":{"key":"schwab@linux-m68k.org","avatar":"https://avatars.githubusercontent.com/u/2175493?v=4"},"body":"Junio C Hamano writes:\n\n> Unconditionally set man.base.url.for.relative.links\n\nrm -f git-add.1 && \\\n xmlto -m manpage-normal.xsl --stringparam man.base.url.for.relative.links= man git-add.xml\nxmlto: unrecognized option '--stringparam'\nmake[1]: *** [git-add.1] Error 1\n\nAndreas.\n\n-- \nAndreas Schwab, schwab@linux-m68k.org\nGPG Key fingerprint = 58CA 54C7 6D53 942B 1756 01D3 44D5 214B 8276 4ED5\n\"And now for something completely different.\"\n"},{"id":"129118","messageId":"4B17ABE3.6060003@drmicha.warpmail.net","threadId":"21839","inReplyTo":"m2hbs85koj.fsf@igel.home","subject":"Re: [ANNOUNCE] Git 1.6.5.4","fromName":"Michael J Gruber","fromEmail":"git@drmicha.warpmail.net","sentAt":"2009-12-03T12:15:31Z","receivedAt":"2009-12-03T12:15:31Z","isPatch":false,"sender":{"key":"git@grubix.eu","avatar":"https://avatars.githubusercontent.com/u/233215?v=4"},"body":"Andreas Schwab venit, vidit, dixit 03.12.2009 13:03:\n> Junio C Hamano writes:\n> \n>> Unconditionally set man.base.url.for.relative.links\n> \n> rm -f git-add.1 && \\\n> xmlto -m manpage-normal.xsl --stringparam man.base.url.for.relative.links= man git-add.xml\n> xmlto: unrecognized option '--stringparam'\n> make[1]: *** [git-add.1] Error 1\n> \n> Andreas.\n> \n\nand\n\nuname -a\nxmlto --version\n\nsays?\n\nMichael\n"},{"id":"129119","messageId":"m2d42w5fqq.fsf@igel.home","threadId":"21839","inReplyTo":"4B17ABE3.6060003@drmicha.warpmail.net","subject":"Re: [ANNOUNCE] Git 1.6.5.4","fromName":"Andreas Schwab","fromEmail":"schwab@linux-m68k.org","sentAt":"2009-12-03T13:50:05Z","receivedAt":"2009-12-03T13:50:05Z","isPatch":false,"sender":{"key":"schwab@linux-m68k.org","avatar":"https://avatars.githubusercontent.com/u/2175493?v=4"},"body":"Michael J Gruber writes:\n\n> Andreas Schwab venit, vidit, dixit 03.12.2009 13:03:\n>> Junio C Hamano writes:\n>> \n>>> Unconditionally set man.base.url.for.relative.links\n>> \n>> rm -f git-add.1 && \\\n>> xmlto -m manpage-normal.xsl --stringparam man.base.url.for.relative.links= man git-add.xml\n>> xmlto: unrecognized option '--stringparam'\n>> make[1]: *** [git-add.1] Error 1\n>> \n>> Andreas.\n>> \n>\n> and\n>\n> uname -a\n> xmlto --version\n>\n> says?\n\nxmlto version 0.0.18\n\nAndreas.\n\n-- \nAndreas Schwab, schwab@linux-m68k.org\nGPG Key fingerprint = 58CA 54C7 6D53 942B 1756 01D3 44D5 214B 8276 4ED5\n\"And now for something completely different.\"\n"},{"id":"129120","messageId":"4B17D078.6080000@drmicha.warpmail.net","threadId":"21839","inReplyTo":"m2d42w5fqq.fsf@igel.home","subject":"Re: [ANNOUNCE] Git 1.6.5.4","fromName":"Michael J Gruber","fromEmail":"git@drmicha.warpmail.net","sentAt":"2009-12-03T14:51:36Z","receivedAt":"2009-12-03T14:51:36Z","isPatch":false,"sender":{"key":"git@grubix.eu","avatar":"https://avatars.githubusercontent.com/u/233215?v=4"},"body":"Andreas Schwab venit, vidit, dixit 03.12.2009 14:50:\n> Michael J Gruber writes:\n> \n>> Andreas Schwab venit, vidit, dixit 03.12.2009 13:03:\n>>> Junio C Hamano writes:\n>>>\n>>>> Unconditionally set man.base.url.for.relative.links\n>>>\n>>> rm -f git-add.1 && \\\n>>> xmlto -m manpage-normal.xsl --stringparam man.base.url.for.relative.links= man git-add.xml\n>>> xmlto: unrecognized option '--stringparam'\n>>> make[1]: *** [git-add.1] Error 1\n>>>\n>>> Andreas.\n>>>\n>>\n>> and\n>>\n>> uname -a\n>> xmlto --version\n>>\n>> says?\n> \n> xmlto version 0.0.18\n> \n> Andreas.\n> \n\nNow that predates Git quite a bit (2004-01-21)...\nI think we can require at least 0.0.20, which is in Debian Lenny and\nFedora 9, for example. I think that should have it. (I'm not sure, they\ndon't use a proper vcs ;) ).\n\nMichael\n"},{"id":"129121","messageId":"20091203150323.GI23717@inocybe.localdomain","threadId":"21839","inReplyTo":"4B17D078.6080000@drmicha.warpmail.net","subject":"Re: [ANNOUNCE] Git 1.6.5.4","fromName":"Todd Zullinger","fromEmail":"tmz@pobox.com","sentAt":"2009-12-03T15:03:23Z","receivedAt":"2009-12-03T15:03:23Z","isPatch":false,"sender":{"key":"tmz@pobox.com","avatar":"https://avatars.githubusercontent.com/u/806319?v=4"},"body":"Michael J Gruber wrote:\n>>> Andreas Schwab venit, vidit, dixit 03.12.2009 13:03:\n>> xmlto version 0.0.18\n[...]\n> Now that predates Git quite a bit (2004-01-21)...\n> I think we can require at least 0.0.20, which is in Debian Lenny and\n> Fedora 9, for example. I think that should have it. (I'm not sure, they\n> don't use a proper vcs ;) ).\n\nxmlto-0.0.18 is what's in RHEL/CentOS 5. It would be nice to be able\nto build git docs there, which works pretty well in general (I\nbelieve).\n\nI built 1.6.6.rc1 on CentOS 5.4 last night with Junio's first patch\nfor this, which only called --stringparam if MAN_BASE_URL was set.\n\n-- \nTodd OpenPGP -> KeyID: 0xBEAF0CE3 | URL: www.pobox.com/~tmz/pgp\n~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~\nA common mistake people make when trying to design something\ncompletely foolproof is to underestimate the ingenuity of complete\nfools.\n -- Douglas Adams\n\n"},{"id":"129127","messageId":"7viqco54xh.fsf@alter.siamese.dyndns.org","threadId":"21839","inReplyTo":"20091203150323.GI23717@inocybe.localdomain","subject":"Re: [ANNOUNCE] Git 1.6.5.4","fromName":"Junio C Hamano","fromEmail":"gitster@pobox.com","sentAt":"2009-12-03T17:43:38Z","receivedAt":"2009-12-03T17:43:38Z","isPatch":false,"sender":{"key":"gitster@pobox.com","avatar":"https://avatars.githubusercontent.com/u/54884?v=4"},"body":"Todd Zullinger writes:\n\n> Michael J Gruber wrote:\n>>>> Andreas Schwab venit, vidit, dixit 03.12.2009 13:03:\n>>> xmlto version 0.0.18\n> [...]\n>> Now that predates Git quite a bit (2004-01-21)...\n>> I think we can require at least 0.0.20, which is in Debian Lenny and\n>> Fedora 9, for example. I think that should have it. (I'm not sure, they\n>> don't use a proper vcs ;) ).\n>\n> xmlto-0.0.18 is what's in RHEL/CentOS 5. It would be nice to be able\n> to build git docs there, which works pretty well in general (I\n> believe).\n>\n> I built 1.6.6.rc1 on CentOS 5.4 last night with Junio's first patch\n> for this, which only called --stringparam if MAN_BASE_URL was set.\n\nI did the second one after seeing that both my Debian box and the k.org\nmachine that manpages tarballs are made (FC11 IIRC) had that option, and\nmy impression has been that it usually is safe to say \"even Debian has it,\nand there wouldn't be many things older than that distro\", but that is\napparently not true.\n\nEither we require 0.0.20 or we revert the tip one on this topic. I think\nthe latter is a safe thing to do.\n\nThanks for a quick report and analysis.\n"},{"id":"129141","messageId":"20091203202738.GP23717@inocybe.localdomain","threadId":"21839","inReplyTo":"7viqco54xh.fsf@alter.siamese.dyndns.org","subject":"Re: [ANNOUNCE] Git 1.6.5.4","fromName":"Todd Zullinger","fromEmail":"tmz@pobox.com","sentAt":"2009-12-03T20:27:38Z","receivedAt":"2009-12-03T20:27:38Z","isPatch":false,"sender":{"key":"tmz@pobox.com","avatar":"https://avatars.githubusercontent.com/u/806319?v=4"},"body":"Junio C Hamano wrote:\n> I did the second one after seeing that both my Debian box and the\n> k.org machine that manpages tarballs are made (FC11 IIRC) had that\n> option, and my impression has been that it usually is safe to say\n> \"even Debian has it, and there wouldn't be many things older than\n> that distro\", but that is apparently not true.\n\nHeh, that surely used to be a very good rule of thumb. Perhaps these\ndays a rule of whichever is older Debian stable or current RHEL/CentOS\nwould suffice?\n\n> Either we require 0.0.20 or we revert the tip one on this topic. I\n> think the latter is a safe thing to do.\n\nThat sounds good to me. I'd like to get the EPEL builds for\nRHEL/CentOS updated sometime soon, as they're currently still on\n1.5.5.6 and that lacks too many of the great improvements in newer git\nreleases. Not having to patch for building the docs is one less thing\nto worry about.\n\n-- \nTodd OpenPGP -> KeyID: 0xBEAF0CE3 | URL: www.pobox.com/~tmz/pgp\n~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~\nRowe's Rule: The odds are five to six that the light at the end of the\ntunnel is the headlight of an oncoming train.\n -- Paul Dickson\n\n"},{"id":"129144","messageId":"7vfx7r4we7.fsf@alter.siamese.dyndns.org","threadId":"21839","inReplyTo":"20091203202738.GP23717@inocybe.localdomain","subject":"Re: [ANNOUNCE] Git 1.6.5.4","fromName":"Junio C Hamano","fromEmail":"gitster@pobox.com","sentAt":"2009-12-03T20:48:00Z","receivedAt":"2009-12-03T20:48:00Z","isPatch":false,"sender":{"key":"gitster@pobox.com","avatar":"https://avatars.githubusercontent.com/u/54884?v=4"},"body":"Todd Zullinger writes:\n\n>> Either we require 0.0.20 or we revert the tip one on this topic. I\n>> think the latter is a safe thing to do.\n>\n> That sounds good to me. I'd like to get the EPEL builds for\n> RHEL/CentOS updated sometime soon, as they're currently still on\n> 1.5.5.6 and that lacks too many of the great improvements in newer git\n> releases. Not having to patch for building the docs is one less thing\n> to worry about.\n\nThis is what I plan to use.\n\n-- >8 ---\nFrom: Junio C Hamano \nDate: Thu, 3 Dec 2009 11:12:32 -0800\nSubject: [PATCH] Documentation: xmlto 0.0.18 does not know --stringparam\n\nNewer DocBook stylesheets want man.base.url.for.relative.links\nparameter set when formatting manpages with external references\nto turn them into full URLs, and leave a helpful \"you should\nset this parameter\" message in the output. Earlier we added\nthe MAN_BASE_URL make variable to specify the value for it.\n\nWhen MAN_BASE_URL is not given, it ought to be safe to set the\nparameter to empty; it would result in an empty leading path for\nolder stylesheets that ignore the parameter, and newer ones\nwould produce the same \"relative URL\" without the message.\n\nUnfortunately, older xmlto (at least version 0.0.18 released in\nearly 2004 that comes with RHEL/CentOS 5) does not understand\nthe --stringparam command line option, so we cannot add the\nparameter definition unconditionally to the command line. Work\nit around by passing the parameter only when set.\n\nIf you do not have a suitable URL prefix, you can pass a quoted empty\nstring to it, like so:\n\n $ make MAN_BASE_URL='\"\"'\n\nSigned-off-by: Junio C Hamano \n---\n Documentation/Makefile | 7 +++++++\n 1 files changed, 7 insertions(+), 0 deletions(-)\n\ndiff --git a/Documentation/Makefile b/Documentation/Makefile\nindex d4c05ca..1c9dfce 100644\n--- a/Documentation/Makefile\n+++ b/Documentation/Makefile\n@@ -108,7 +108,14 @@ endif\n # use MAN_BASE_URL=http://www.kernel.org/pub/software/scm/git/docs/\n # but distros may want to set it to /usr/share/doc/git-core/docs/ or\n # something like that.\n+#\n+# As older stylesheets simply ignore this parameter, it ought to be\n+# safe to set it to empty string when the base URL is not specified,\n+# but unfortunately we cannot do so unconditionally because at least\n+# xmlto 0.0.18 is reported to lack --stringparam option.\n+ifdef MAN_BASE_URL\n XMLTO_EXTRA += --stringparam man.base.url.for.relative.links=$(MAN_BASE_URL)\n+endif\n \n # If your target system uses GNU groff, it may try to render\n # apostrophes as a \"pretty\" apostrophe using unicode. This breaks\n-- \n1.6.6.rc1.5.ge21a85\n"},{"id":"129146","messageId":"20091203220020.GS23717@inocybe.localdomain","threadId":"21839","inReplyTo":"7vfx7r4we7.fsf@alter.siamese.dyndns.org","subject":"Re: [ANNOUNCE] Git 1.6.5.4","fromName":"Todd Zullinger","fromEmail":"tmz@pobox.com","sentAt":"2009-12-03T22:00:20Z","receivedAt":"2009-12-03T22:00:20Z","isPatch":false,"sender":{"key":"tmz@pobox.com","avatar":"https://avatars.githubusercontent.com/u/806319?v=4"},"body":"Junio C Hamano wrote:\n> This is what I plan to use.\n>\n> -- >8 ---\n> From: Junio C Hamano \n> Date: Thu, 3 Dec 2009 11:12:32 -0800\n> Subject: [PATCH] Documentation: xmlto 0.0.18 does not know --stringparam\n>\n> Newer DocBook stylesheets want man.base.url.for.relative.links\n> parameter set when formatting manpages with external references\n> to turn them into full URLs, and leave a helpful \"you should\n> set this parameter\" message in the output. Earlier we added\n> the MAN_BASE_URL make variable to specify the value for it.\n>\n> When MAN_BASE_URL is not given, it ought to be safe to set the\n> parameter to empty; it would result in an empty leading path for\n> older stylesheets that ignore the parameter, and newer ones\n> would produce the same \"relative URL\" without the message.\n>\n> Unfortunately, older xmlto (at least version 0.0.18 released in\n> early 2004 that comes with RHEL/CentOS 5) does not understand\n> the --stringparam command line option, so we cannot add the\n> parameter definition unconditionally to the command line. Work\n> it around by passing the parameter only when set.\n\nIs it worth sidestepping the xmlto part entirely? If we set this\ndirectly in a .xsl file, it will work on older systems without any\neffort. Then we can default MAN_BASE_URL to something and let distro\npackagers override it.\n\nI tested with this in Documentation/manpage-base.xsl on a CentOS 5 box\nand it builds fine, leaving no cruft in the man pages regarding the\nman.base.url...\n\n\n/path/to/git/docs\n\nOf course, the relative links looked just like they did in older\ndocbook releases:\n\n 1. Everyday Git\n everyday.html\n\nIs it worth the effort to have @@MAN_BASE_URL@@ in\nDocumentation/manpage-base.xsl or similar and replace it at build\ntime?\n\n-- \nTodd OpenPGP -> KeyID: 0xBEAF0CE3 | URL: www.pobox.com/~tmz/pgp\n~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~\nJust because everything is different doesn't mean anything has\nchanged.\n -- Irene Peter\n\n"},{"id":"129147","messageId":"7vbpif4rn2.fsf@alter.siamese.dyndns.org","threadId":"21839","inReplyTo":"20091203220020.GS23717@inocybe.localdomain","subject":"Re: [ANNOUNCE] Git 1.6.5.4","fromName":"Junio C Hamano","fromEmail":"gitster@pobox.com","sentAt":"2009-12-03T22:30:41Z","receivedAt":"2009-12-03T22:30:41Z","isPatch":false,"sender":{"key":"gitster@pobox.com","avatar":"https://avatars.githubusercontent.com/u/54884?v=4"},"body":"Todd Zullinger writes:\n\n> I tested with this in Documentation/manpage-base.xsl on a CentOS 5 box\n> and it builds fine, leaving no cruft in the man pages regarding the\n> man.base.url...\n>\n> \n> >/path/to/git/docs\n>\n> Of course, the relative links looked just like they did in older\n> docbook releases:\n>\n> 1. Everyday Git\n> everyday.html\n>\n> Is it worth the effort to have @@MAN_BASE_URL@@ in\n> Documentation/manpage-base.xsl or similar and replace it at build\n> time?\n\nI think it depends on the likelihood that a distro has xmlto so old that\nit does not understand --stringparam yet it uses stylesheet so new that\nsetting the parameter makes a positive difference (either it gives the\nfull URL or at least squelches the \"You should define the parameter\"\nnoise) in the output.\n\nI am guessing that the answer would be that is a very unlikely combination\nand not worth worrying about it, but I've been wrong before in this exact\narea ;-)\n"},{"id":"129156","messageId":"76c5b8580912031949k7f4193f9q94f9a2040b877571@mail.gmail.com","threadId":"21839","inReplyTo":"7vbpif4rn2.fsf@alter.siamese.dyndns.org","subject":"Re: [ANNOUNCE] Git 1.6.5.4","fromName":"Eugene Sajine","fromEmail":"euguess@gmail.com","sentAt":"2009-12-04T03:49:54Z","receivedAt":"2009-12-04T03:49:54Z","isPatch":false,"sender":{"key":"euguess@gmail.com","avatar":null},"body":"On Thu, Dec 3, 2009 at 5:30 PM, Junio C Hamano wrote:\n> Todd Zullinger writes:\n>\n>> I tested with this in Documentation/manpage-base.xsl on a CentOS 5 box\n>> and it builds fine, leaving no cruft in the man pages regarding the\n>> man.base.url...\n>>\n>> \n>> >        >/path/to/git/docs\n>>\n>> Of course, the relative links looked just like they did in older\n>> docbook releases:\n>>\n>>        1. Everyday Git\n>>           everyday.html\n>>\n>> Is it worth the effort to have @@MAN_BASE_URL@@ in\n>> Documentation/manpage-base.xsl or similar and replace it at build\n>> time?\n>\n> I think it depends on the likelihood that a distro has xmlto so old that\n> it does not understand --stringparam yet it uses stylesheet so new that\n> setting the parameter makes a positive difference (either it gives the\n> full URL or at least squelches the \"You should define the parameter\"\n> noise) in the output.\n>\n> I am guessing that the answer would be that is a very unlikely combination\n> and not worth worrying about it, but I've been wrong before in this exact\n> area ;-)\n\n\nHello,\n\nWe have RH with xmlto 0.0.18. I was getting ready to update our\ninstallation to 1.6.5.4, but as i understand the documentation will\nnot be fully available untill this issue is resolved. Could you,\nplease, advise if this is going to be in next 1.6.5.5?\n\nThanks,\nEugene\n"},{"id":"129158","messageId":"7vocmfxnxh.fsf@alter.siamese.dyndns.org","threadId":"21839","inReplyTo":"76c5b8580912031949k7f4193f9q94f9a2040b877571@mail.gmail.com","subject":"Re: [ANNOUNCE] Git 1.6.5.4","fromName":"Junio C Hamano","fromEmail":"gitster@pobox.com","sentAt":"2009-12-04T06:18:02Z","receivedAt":"2009-12-04T06:18:02Z","isPatch":false,"sender":{"key":"gitster@pobox.com","avatar":"https://avatars.githubusercontent.com/u/54884?v=4"},"body":"Eugene Sajine writes:\n\n> We have RH with xmlto 0.0.18. I was getting ready to update our\n> installation to 1.6.5.4, but as i understand the documentation will\n> not be fully available untill this issue is resolved. Could you,\n> please, advise if this is going to be in next 1.6.5.5?\n\nI've applied the patch in the thread you are responding to already on\n'maint' so it will appear in both 1.6.5.5 and 1.6.6. In the meantime, if\nyou want to run 1.6.5.4 or preferably 1.6.6-rc1, you can locally revert\n8dd35c7 (Unconditionally set man.base.url.for.relative.links, 2009-12-03).\n"},{"id":"129159","messageId":"vpq638ntf7k.fsf@bauges.imag.fr","threadId":"21839","inReplyTo":"7vocmfxnxh.fsf@alter.siamese.dyndns.org","subject":"Re: [ANNOUNCE] Git 1.6.5.4","fromName":"Matthieu Moy","fromEmail":"matthieu.moy@grenoble-inp.fr","sentAt":"2009-12-04T06:39:59Z","receivedAt":"2009-12-04T06:39:59Z","isPatch":false,"sender":{"key":"matthieu.moy@grenoble-inp.fr","avatar":"https://gravatar.com/avatar/72c8a2705971a25dfaff23cece15130d405685845d911aedd5667ace277f3fc5?d=mp&s=160"},"body":"Junio C Hamano writes:\n\n> Eugene Sajine writes:\n>\n>> We have RH with xmlto 0.0.18. I was getting ready to update our\n>> installation to 1.6.5.4, but as i understand the documentation will\n>> not be fully available untill this issue is resolved. Could you,\n>> please, advise if this is going to be in next 1.6.5.5?\n>\n> I've applied the patch in the thread you are responding to already on\n> 'maint' so it will appear in both 1.6.5.5 and 1.6.6. In the meantime, if\n> you want to run 1.6.5.4 or preferably 1.6.6-rc1, you can locally revert\n> 8dd35c7 (Unconditionally set man.base.url.for.relative.links,\n> 2009-12-03).\n\nAlso, one can download the man pages from git.git :\n\n# in a clone of git://git.kernel.org/pub/scm/git/git.git\ngit archive --format=tar origin/man | tar -x -C /usr/share/man/ -vf -\n\n(or leave with 1.6.5.5 and the docs of 1.6.5.4 for some time)\n\n-- \nMatthieu Moy\nhttp://www-verimag.imag.fr/~moy/\n"},{"id":"129169","messageId":"m2r5rb9hes.fsf@igel.home","threadId":"21839","inReplyTo":"7vbpif4rn2.fsf@alter.siamese.dyndns.org","subject":"Re: [ANNOUNCE] Git 1.6.5.4","fromName":"Andreas Schwab","fromEmail":"schwab@linux-m68k.org","sentAt":"2009-12-04T10:12:43Z","receivedAt":"2009-12-04T10:12:43Z","isPatch":false,"sender":{"key":"schwab@linux-m68k.org","avatar":"https://avatars.githubusercontent.com/u/2175493?v=4"},"body":"Junio C Hamano writes:\n\n> I think it depends on the likelihood that a distro has xmlto so old that\n> it does not understand --stringparam yet it uses stylesheet so new that\n> setting the parameter makes a positive difference (either it gives the\n> full URL or at least squelches the \"You should define the parameter\"\n> noise) in the output.\n\nopenSUSE 11.2, for example. Its xmlto has a non-standard --xsltopts\noption that passes its argument down to xsltproc.\n\nAndreas.\n\n-- \nAndreas Schwab, schwab@linux-m68k.org\nGPG Key fingerprint = 58CA 54C7 6D53 942B 1756 01D3 44D5 214B 8276 4ED5\n\"And now for something completely different.\"\n"},{"id":"129215","messageId":"7vzl5ysm11.fsf@alter.siamese.dyndns.org","threadId":"21839","inReplyTo":"m2r5rb9hes.fsf@igel.home","subject":"Re: [ANNOUNCE] Git 1.6.5.4","fromName":"Junio C Hamano","fromEmail":"gitster@pobox.com","sentAt":"2009-12-04T17:10:18Z","receivedAt":"2009-12-04T17:10:18Z","isPatch":false,"sender":{"key":"gitster@pobox.com","avatar":"https://avatars.githubusercontent.com/u/54884?v=4"},"body":"Andreas Schwab writes:\n\n> Junio C Hamano writes:\n>\n>> I think it depends on the likelihood that a distro has xmlto so old that\n>> it does not understand --stringparam yet it uses stylesheet so new that\n>> setting the parameter makes a positive difference (either it gives the\n>> full URL or at least squelches the \"You should define the parameter\"\n>> noise) in the output.\n>\n> openSUSE 11.2, for example. Its xmlto has a non-standard --xsltopts\n> option that passes its argument down to xsltproc.\n\nOk, as I said that I've been wrong before in this area ;-)\n\nI don't think I will have much time for git today, and it would be\nappreciated if somebody can work on this and send a tested patch that\napplies cleanly on top of 'maint' to implement the @@MAN_BASE_URL@@\nreplacement from manpage-base.xsl.in to manpage-base.xsl as Todd suggested\nearlier.\n"},{"id":"129233","messageId":"20091204193355.GC4629@inocybe.localdomain","threadId":"21839","inReplyTo":"7vzl5ysm11.fsf@alter.siamese.dyndns.org","subject":"Re: [ANNOUNCE] Git 1.6.5.4","fromName":"Todd Zullinger","fromEmail":"tmz@pobox.com","sentAt":"2009-12-04T19:33:55Z","receivedAt":"2009-12-04T19:33:55Z","isPatch":false,"sender":{"key":"tmz@pobox.com","avatar":"https://avatars.githubusercontent.com/u/806319?v=4"},"body":"Junio C Hamano wrote:\n> Andreas Schwab writes:\n>\n>> Junio C Hamano writes:\n>>\n>>> I think it depends on the likelihood that a distro has xmlto so old that\n>>> it does not understand --stringparam yet it uses stylesheet so new that\n>>> setting the parameter makes a positive difference (either it gives the\n>>> full URL or at least squelches the \"You should define the parameter\"\n>>> noise) in the output.\n>>\n>> openSUSE 11.2, for example. Its xmlto has a non-standard --xsltopts\n>> option that passes its argument down to xsltproc.\n>\n> Ok, as I said that I've been wrong before in this area ;-)\n>\n> I don't think I will have much time for git today, and it would be\n> appreciated if somebody can work on this and send a tested patch that\n> applies cleanly on top of 'maint' to implement the @@MAN_BASE_URL@@\n> replacement from manpage-base.xsl.in to manpage-base.xsl as Todd suggested\n> earlier.\n\nSomething like this perhaps? I tested it with 1.6.5.4 on Fedora 10,\n12 and CentOS 5.4 but it could surely use more eyes and testing to\nensure it works as it should and doesn't have unintended negative\neffects on make clean and such.\n\nThis does set MAN_BASE_URL unconditionally, pointing to kernel.org.\nThat way anyone building with recent DocBook and taking no new action\nwill have something useful in the man page links.\n\nI noticed on Fedora 12 that email addresses get added to the NOTES\nsection. For example, git-branch(1) has:\n\nDocumentation by Junio C Hamano and the git-list .\n...\nNOTES\n...\n4. git@vger.kernel.org\n mailto:git@vger.kernel.org\n\nThat must be something new in DocBook, as it doesn't occur in the\nFedora 10 builds. It's a little extraneous, but not harmful I guess.\n\n--- >8 ---\nFrom: Todd Zullinger \nDate: Fri, 4 Dec 2009 12:53:21 -0500\nSubject: [PATCH] Documentation: Avoid use of xmlto --stringparam\n\nThe --stringparam option is not available on older xmlto versions.\nInstead, set man.base.url.for.relative.links via a .xsl file. Older\ndocbook versions will ignore this without causing grief to users of\nolder xmlto versions.\n\nSigned-off-by: Todd Zullinger \n---\n Documentation/.gitignore | 1 +\n Documentation/Makefile | 23 ++++++++++++-----------\n Documentation/manpage-base-url.xsl.in | 10 ++++++++++\n 3 files changed, 23 insertions(+), 11 deletions(-)\n create mode 100644 Documentation/manpage-base-url.xsl.in\n\ndiff --git a/Documentation/.gitignore b/Documentation/.gitignore\nindex d8edd90..1c3a9fe 100644\n--- a/Documentation/.gitignore\n+++ b/Documentation/.gitignore\n@@ -8,3 +8,4 @@ gitman.info\n howto-index.txt\n doc.dep\n cmds-*.txt\n+manpage-base-url.xsl\ndiff --git a/Documentation/Makefile b/Documentation/Makefile\nindex 1c9dfce..1c867fa 100644\n--- a/Documentation/Makefile\n+++ b/Documentation/Makefile\n@@ -104,18 +104,15 @@ XMLTO_EXTRA += -m manpage-suppress-sp.xsl\n endif\n \n # Newer DocBook stylesheet emits warning cruft in the output when\n-# this is not set, and if set it shows an absolute link. We can\n-# use MAN_BASE_URL=http://www.kernel.org/pub/software/scm/git/docs/\n-# but distros may want to set it to /usr/share/doc/git-core/docs/ or\n-# something like that.\n+# this is not set, and if set it shows an absolute link. Older\n+# stylesheets simply ignore this parameter.\n #\n-# As older stylesheets simply ignore this parameter, it ought to be\n-# safe to set it to empty string when the base URL is not specified,\n-# but unfortunately we cannot do so unconditionally because at least\n-# xmlto 0.0.18 is reported to lack --stringparam option.\n-ifdef MAN_BASE_URL\n-XMLTO_EXTRA += --stringparam man.base.url.for.relative.links=$(MAN_BASE_URL)\n+# Distros may want to use MAN_BASE_URL=file:///path/to/git/docs/\n+# or similar.\n+ifndef MAN_BASE_URL\n+MAN_BASE_URL = http://www.kernel.org/pub/software/scm/git/docs/\n endif\n+XMLTO_EXTRA += -m manpage-base-url.xsl\n \n # If your target system uses GNU groff, it may try to render\n # apostrophes as a \"pretty\" apostrophe using unicode. This breaks\n@@ -244,6 +241,7 @@ clean:\n \t$(RM) howto-index.txt howto/*.html doc.dep\n \t$(RM) technical/api-*.html technical/api-index.txt\n \t$(RM) $(cmds_txt) *.made\n+\t$(RM) manpage-base-url.xsl\n \n $(MAN_HTML): %.html : %.txt\n \t$(QUIET_ASCIIDOC)$(RM) $@+ $@ && \\\n@@ -251,7 +249,10 @@ $(MAN_HTML): %.html : %.txt\n \t\t$(ASCIIDOC_EXTRA) -agit_version=$(GIT_VERSION) -o $@+ $< && \\\n \tmv $@+ $@\n \n-%.1 %.5 %.7 : %.xml\n+manpage-base-url.xsl: manpage-base-url.xsl.in\n+\tsed \"s|@@MAN_BASE_URL@@|$(MAN_BASE_URL)|\" $< > $@\n+\n+%.1 %.5 %.7 : %.xml manpage-base-url.xsl\n \t$(QUIET_XMLTO)$(RM) $@ && \\\n \txmlto -m $(MANPAGE_XSL) $(XMLTO_EXTRA) man $<\n \ndiff --git a/Documentation/manpage-base-url.xsl.in b/Documentation/manpage-base-url.xsl.in\nnew file mode 100644\nindex 0000000..e800904\n--- /dev/null\n+++ b/Documentation/manpage-base-url.xsl.in\n@@ -0,0 +1,10 @@\n+\n+\n+\n+\n+@@MAN_BASE_URL@@\n+\n+\n-- \n1.6.6.rc1\n\n-- \nTodd OpenPGP -> KeyID: 0xBEAF0CE3 | URL: www.pobox.com/~tmz/pgp\n~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~\nThe world keeps ending but new people too dumb to know it keep showing\nup as if the fun's just started.\n"},{"id":"129257","messageId":"7vws116fsf.fsf@alter.siamese.dyndns.org","threadId":"21839","inReplyTo":"20091204193355.GC4629@inocybe.localdomain","subject":"Re: [ANNOUNCE] Git 1.6.5.4","fromName":"Junio C Hamano","fromEmail":"gitster@pobox.com","sentAt":"2009-12-05T07:28:16Z","receivedAt":"2009-12-05T07:28:16Z","isPatch":false,"sender":{"key":"gitster@pobox.com","avatar":"https://avatars.githubusercontent.com/u/54884?v=4"},"body":"Todd Zullinger writes:\n\n> This does set MAN_BASE_URL unconditionally, pointing to kernel.org.\n> That way anyone building with recent DocBook and taking no new action\n> will have something useful in the man page links.\n\nI'd rather see it set to an empty string by default which will essentially\nkeep the old output. An alternative would be to default to the location\nthe install-html target installs, i.e. $HOME/share/doc/git-doc pointed by\n$(htmldir) in Documentation/Makefile.\n\nDistro people should be strongly encouraged to repoint the variable to\nsomewhere local that does not require connections to k.org or anywhere\noutside the end-user's installation, but that is a slightly separate and\ndifferent topic.\n"},{"id":"129367","messageId":"7vzl5vbp7y.fsf@alter.siamese.dyndns.org","threadId":"21839","inReplyTo":"20091204193355.GC4629@inocybe.localdomain","subject":"Re: [ANNOUNCE] Git 1.6.5.4","fromName":"Junio C Hamano","fromEmail":"gitster@pobox.com","sentAt":"2009-12-07T00:30:09Z","receivedAt":"2009-12-07T00:30:09Z","isPatch":false,"sender":{"key":"gitster@pobox.com","avatar":"https://avatars.githubusercontent.com/u/54884?v=4"},"body":"Todd Zullinger writes:\n\n> Something like this perhaps? I tested it with 1.6.5.4 on Fedora 10,\n> 12 and CentOS 5.4 but it could surely use more eyes and testing to\n> ensure it works as it should and doesn't have unintended negative\n> effects on make clean and such.\n\nThanks (and thanks Andreas for pointing out a yet another distro that is\ndifferent). I think the patch makes sense.\n\n> This does set MAN_BASE_URL unconditionally, pointing to kernel.org.\n\nI'd change this to point at \"file://$(htmldir)/\", though.\n"},{"id":"129425","messageId":"20091207151800.GI4629@inocybe.localdomain","threadId":"21839","inReplyTo":"7vzl5vbp7y.fsf@alter.siamese.dyndns.org","subject":"Re: [ANNOUNCE] Git 1.6.5.4","fromName":"Todd Zullinger","fromEmail":"tmz@pobox.com","sentAt":"2009-12-07T15:18:01Z","receivedAt":"2009-12-07T15:18:01Z","isPatch":false,"sender":{"key":"tmz@pobox.com","avatar":"https://avatars.githubusercontent.com/u/806319?v=4"},"body":"Junio C Hamano wrote:\n>> This does set MAN_BASE_URL unconditionally, pointing to kernel.org.\n>\n> I'd change this to point at \"file://$(htmldir)/\", though.\n\nThanks. I can delete the mail with an updated patch that I started\nover the weekend and didn't finish, which did that as well. ;)\n\nMany thanks for all your hard work Junio.\n\n-- \nTodd OpenPGP -> KeyID: 0xBEAF0CE3 | URL: www.pobox.com/~tmz/pgp\n~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~\nAim Low, Reach Your Goals, Avoid Disappointment.\n\n"}]}