gmane.comp.version-control.git

[QGit bug] git user settings not retrieved when launched for Windows explorer

Abdelrazak Younes — 2008-10-07T07:33:28

Dear Marco, When I double click on qgit.exe, the user name and email are not shown in the user settings (for any of the 3 combo values). But if I run qgit from the commandline at the mysysgit bash prompt, the boxes are properly filled. I would like to debug it but, as I reported last week, the MSVC project doesn't work for me. By the way, these two edit boxes are not editable on Windows, is that on purpose? If yes, maybe we could let the user change them is 'Local config' is selected and call the appropriate git function? Thanks, Abdel.

RFH: gitk equivalent of "git log -p --full-diff file.cpp"

Eric Raible — 2008-10-07T02:04:14

Hello All - Hopefully I'm just being dense, but there's a useful tortoisesvn feature that (so far) I haven't found in gitk. Blasphemy! Sniveling windoze luser, etc^2. Anyway what I'm looking for is the gitk equivalent of: git log -p --full-diff file.cpp That is to say: I want to see all commits which touch a given file along with all of the other files changed by that commit. gitk -p --full-diff EntryPoints.cpp doesn't quite do it - it treats the "extra" diffs as part of the commit msg. Anyone got more if a clue than me? Thanks - Eric PS - Aside from aliases, am I missing a simpler spelling of the above git log command?

error pushing stash ?

David Bryson — 2008-10-07T00:34:35

I have a git mirror remote setup on a few of my repositories: [remote "backup"] url = /users/dbryson/backup/janus.git/ fetch = +refs/heads/*:refs/remotes/origin/* receivepack = sudo -u dbryson git-receive-pack mirror = 1 I send my refs to the backup with: $ git push backup Only to find some odd error messages: Counting objects: 133, done. Compressing objects: 100% (109/109), done. Writing objects: 100% (109/109), 31.25 KiB, done. Total 109 (delta 82), reused 0 (delta 0) error: refusing to create funny ref 'refs/stash' remotely To /users/dbryson/backup/janus.git/ 549f8a4..8e93d51 8654 -> 8654 ef6195b..549f8a4 origin/8654 -> origin/8654 + 623e7cb...63d7262 origin/master -> origin/master (forced update) ! [remote rejected] refs/stash -> refs/stash (funny refname) error: failed to push some refs to '/users/dbryson/backup/janus.git/' Should I be concnerned about this or is it normal ? To be honest the fact that the stash isn't pushing doesn't bother me. But maybe it is a symptom of a larger problem ? Dave

merge -s ffonly

Shawn O. Pearce — 2008-10-06T23:56:11

I really don't care about this feature. But Randal's whining on #git made me stop what I was doing and write something that might turn into it. Totally untested code. It might reformat your C:\ drive and install Windows ME. Install as $(git --exec-path)/git-merge-ffonly and call as `git merge -s ffonly`. If you care about this sort of feature, test it, write tests for it, make a formal patch, and send it for review. No, I will not do this for you. As I said, I don't care about this as a feature. --8<-- diff --git a/git-merge-ffonly.sh b/git-merge-ffonly.sh new file mode 100644 index 0000000..24363b5 --- /dev/null +++ b/git-merge-ffonly.sh < at >< at > -0,0 +1,20 < at >< at > +#!/bin/sh +while test $# -gt 0 +do +if test "z$1" = z-- +then +shift +break +else +shift +fi +done + +while test $# -gt 0 +do +if test -n "$(git rev-list $1..HEAD)" +then +exit 2 +fi +shift +done

[PATCH] Replace xmalloc/memset(0) pairs with xcalloc

Brandon Casey — 2008-10-06T23:39:10

Many call sites immediately initialize allocated memory with zero after calling xmalloc. A single call to xcalloc can replace this two-call sequence. Signed-off-by: Brandon Casey nrlssc.navy.mil> --- builtin-merge.c | 3 +-- builtin-pack-objects.c | 4 +--- builtin-unpack-objects.c | 3 +-- merge-tree.c | 3 +-- remote.c | 3 +-- 5 files changed, 5 insertions(+), 11 deletions(-) diff --git a/builtin-merge.c b/builtin-merge.c index 5c65a58..dcf8987 100644 --- a/builtin-merge.c +++ b/builtin-merge.c < at >< at > -123,8 +123,7 < at >< at > static struct strategy *get_strategy(const char *name) exit(1); } -ret = xmalloc(sizeof(struct strategy)); -memset(ret, 0, sizeof(struct strategy)); +ret = xcalloc(1, sizeof(struct strategy)); ret->name = xstrdup(name); return ret; } diff --git a/builtin-pack-objects.c b/builtin-pack-objects.c index 1158e42..59c30d1 100644 --- a/builtin-pack-objects.c +++ b/builtin-pack-objects.c < at >< at > -1369,12 +1369,10 < at >< at > static void find_deltas(struct object_entry **list, unsigned *list_size, int window, int depth, unsigned *processed) { uint32_t i, idx = 0, count = 0; -unsigned int array_size = window * sizeof(struct unpacked); struct unpacked *array; unsigned long mem_usage = 0; -array = xmalloc(array_size); -memset(array, 0, array_size); +array = xcalloc(window, sizeof(struct unpacked)); for (;;) { struct object_entry *entry = *list++; diff --git a/builtin-unpack-objects.c b/builtin-unpack-objects.c index d2796b6..9f4bdd3 100644 --- a/builtin-unpack-objects.c +++ b/builtin-unpack-objects.c < at >< at > -477,8 +477,7 < at >< at > static void unpack_all(void) if (!quiet) progress = start_progress("Unpacking objects", nr_objects); -obj_list = xmalloc(nr_objects * sizeof(*obj_list)); -memset(obj_list, 0, nr_objects * sizeof(*obj_list)); +obj_list = xcalloc(nr_objects, sizeof(*obj_list)); for (i = 0; i < nr_objects; i++) { unpack_one(i); display_progress(progress, i + 1); diff --git a/merge-tree.c b/merge-tree.c index 02fc10f..2d1413e 100644 --- a/merge-tree.c +++ b/merge-tree.c < at >< at > -158,9 +158,8 < at >< at > static int same_entry(struct name_entry *a, struct name_entry *b) static struct merge_list *create_entry(unsigned stage, unsigned mode, const unsigned char *sha1, const char *path) { -struct merge_list *res = xmalloc(sizeof(*res)); +struct merge_list *res = xcalloc(1, sizeof(*res)); -memset(res, 0, sizeof(*res)); res->stage = stage; res->path = path; res->mode = mode; diff --git a/remote.c b/remote.c index c45d96e..a2d7ab1 100644 --- a/remote.c +++ b/remote.c < at >< at > -751,8 +751,7 < at >< at > int remote_find_tracking(struct remote *remote, struct refspec *refspec) struct ref *alloc_ref(unsigned namelen) { -struct ref *ret = xmalloc(sizeof(struct ref) + namelen); -memset(ret, 0, sizeof(struct ref) + namelen); +struct ref *ret = xcalloc(1, sizeof(struct ref) + namelen); return ret; }

files missing after converting a cvs repository to git

Adam Mercer — 2008-10-06T22:02:44

Hi One of the prrojects I am involved with is currently looking into migrating from cvs to git, therefore we have been investigating this by setting up a git repository that tracks cvs, however there are some very strange things going on and I was hoping someone could offer some insight into what is going on. I use the following git cvsimport command to import the repository: $ git cvsimport -v -a -i -k -d:pserver:user< at >server:port/path/to/cvs/repo -C /path/to/new/git/repo module this ran successfully with no warnings or errors. However, when I checkout the new git repository that are several files missing from the git checkout that are present in the cvs checkout. Does anyone know why this would happen, or how to find out? This is a major problem because we will be unable to migrate to cvs until this can be solved. Cheers Adam

Mrs Magdarline god

magdarline Kudrat — 2008-10-06T20:16:56

I am Mrs Magdarline Kudrat, from Ivory Coast, Cote D'Ivoire. I was married to Mr. Barry Kudrat who was a contractor with the government of Cote D'Ivoire before he died in the year 2006 after few days in the hospital The doctor said his death was as a result of poison.We were married for so many years without a child. Before his death we were both belive in Almghty. Since his death I decide not to remarry or get a child outside my matrimonial home which is against the will of Almighty.the sum of $(10.000,000,00) Ten Million United State Dollars, ---------------------------------------------------- Kto jest bardziej SEXY?! Tysiące zdjęć i profili najseksowaniejszych Polaków! Dołącz do nich: http://klik.wp.pl/?adr=http%3A%2F%2Fcorto.www.wp.pl%2Fas%2Famisexy.html&sid=512

Mrs Magdarline Kudrat,

magdarline Kudrat — 2008-10-06T20:12:52

Mrs Magdarline Kudrat

magdarline Kudrat — 2008-10-06T20:10:37

I am Mrs Magdarline Kudrat, from Ivory Coast, Cote D'Ivoire. I was married to Mr. Barry Kudrat who was a contractor with the government of Cote D'Ivoire before he died in the year 2006 after few days in the hospital The doctor said his death was as a result of poison.We were married for so many years without a child. Before his death we were both belive in Almghty. Since his death I decide not to remarry or get a child outside my matrimonial home which is against the will of Almighty. When my late husband was alive he ---------------------------------------------------- Sopot Jazz 2008! 9-12 października. Chris Botti z gościnnym występem Anny Marii Jopek, Pfk Sopot z Adamem Makowiczem i Krzysztofem Jabłońskim! Bilety 58/5558451. Zobacz! http://klik.wp.pl/?adr=http%3A%2F%2Fcorto.www.wp.pl%2Fas%2FChrisBotti.html&sid=510

[PATCH] Do not use errno when pread() returns 0

Samuel Tardieu — 2008-10-06T17:28:41

If we use pread() while at the end of the file, it will return 0, which is not an error from the operating system point of view. In this case, errno has not been set and must not be used. Signed-off-by: Samuel Tardieu rfc1149.net> --- index-pack.c | 5 ++++- 1 files changed, 4 insertions(+), 1 deletions(-) diff --git a/index-pack.c b/index-pack.c index 2e4c088..73860bf 100644 --- a/index-pack.c +++ b/index-pack.c < at >< at > -365,8 +365,11 < at >< at > static void *get_data_from_pack(struct object_entry *obj) data = src; do { ssize_t n = pread(pack_fd, data + rdy, len - rdy, from + rdy); -if (n <= 0) +if (n < 0) die("cannot pread pack file: %s", strerror(errno)); +if (!n) +die("premature end of pack file, %lu bytes missing", + len - rdy); rdy += n; } while (rdy < len); data = xmalloc(obj->size);

What's cooking in git/spearce.git (Oct 2008, #01; Mon, 06)

Shawn O. Pearce — 2008-10-06T16:59:43

What's cooking in git/spearce.git (Oct 2008, #01; Mon, 06) -------------------------------------------------- Here are the topics that have been cooking. Commits prefixed with '-' are only in 'pu' while commits prefixed with '+' are in 'next'. The topics list the commits in reverse chronological order. The topics meant to be merged to the maintenance series have "maint-" in their names. Topics aren't sorted by name; I have tried to cluster them together to tell the story of how 'next' and 'pu' have evolved this past week. ---------------------------------------------------------------- [New Topics] * sp/describe-docs (Tue Sep 30 11:55:16 2008 -0700) 1 commit . docs: Improve the description of describe --tags I'm probably going to drop this topic if I can ever get around to formally writing and posting sp/describe-lwtag for review. * sp/describe-lwtag (Tue Sep 30 12:03:41 2008 -0700) 1 commit - [WIP] Change meaning of --tags and --all Modifies "git describe --tags" to allow matching against lightweight tags even when an annotated tag is found in the history. Its not in next because I haven't formally written a commit message and posted the patch for discussion on the list. This is a fairly significant change in behavior. Its most certainly 1.6.1 material, but many argue the change is the _correct_ behavior and the current behavior in 1.6.0 is buggy. * jk/diff-convfilter (Sun Oct 5 17:43:45 2008 -0400) 4 commits - diff: add filter for converting binary to text - diff: introduce diff..binary - diff: unify external diff and funcname parsing code - t4012: use test_cmp instead of cmp A general cleanup on how diff drivers are implemented. Its still missing documentation updates and tests but doesn't break anything current as far as I can tell. It needs more review before it can be slated for 'next'. * pb/rename-rowin32 (Fri Oct 3 12:20:43 2008 +0200) 1 commit - Do not rename read-only files during a push Supposedly fixes pack file renames on Windows, but it makes the test suite fail on Linux. I haven't debugged why yet. Its very likely we need a replacement patch before this can move forward. * gb/formatpatch-autonbr (Thu Oct 2 16:55:39 2008 -0400) 1 commit - format-patch: autonumber by default Fails the test suite; some of the test vectors need to be updated to account for the new default. Someone who cares about this change should follow through on the test suite update, or this may get dropped. * sh/maint-intrebase (Wed Oct 1 01:11:07 2008 -0500) 1 commit + Fix interactive rebase on dropped commits. This merged to next, then was reverted out this morning, due to the introduction of the sh/maint-rebase3. Its dead and will be dropped from the next "What's cooking" release notes. * sh/maint-rebase3 (Sun Oct 5 23:26:52 2008 -0500) 1 commit - rebase--interactive: fix parent rewriting for dropped commits A replacement for sh/maint-intrebase. Its in pu because I have gotten 3 different versions of this patch, two of them posted a full 4 days after I merged the first version into next. I felt burned by the patch author for not keeping up with my tree, so I'm not merging the patch to next. At this point its going to sit in pu until Junio comes back. I think the topic needs a few more days to settle to see if the patch author is going to submit any more revisions. * ns/rebase-noverify (Mon Oct 6 14:14:29 2008 +0900) 1 commit + rebase --no-verify This adds --no-verify to git rebase, to avoid the pre-rebase hook. Lacks documentation but otherwise looks sound. * mv/merge-noff (Fri Oct 3 14:04:47 2008 +0200) 1 commit + builtin-commit: use reduce_heads() only when appropriate Fixes "git merge --no-ff --no-commit". * dp/cywginstat (Tue Sep 30 17:53:47 2008 +0400) 3 commits + cygwin: Use native Win32 API for stat + mingw: move common functionality to win32.h + add have_git_dir() function Performance improvement for Cygwin, bypassing the Cygwin stat function and using one more like MinGW uses. Probably ready for master on the next round. Several Cygwin users have given praise to the series. * js/objc-funchdr (Wed Oct 1 00:46:34 2008 +0100) 1 commit + Teach git diff about Objective-C syntax Adds support for Objective-C function headers. I don't use Objective-C so I can't comment on how good/bad the patterns are for the language. Feedback (and improvements if necessary) from Objective-C users would be appreciated. * dm/svn-branch (Sat Oct 4 19:35:17 2008 -0700) 1 commit + Add git-svn branch to allow branch creation in SVN repositories Adds "git svn branch" to build branches remotely in SVN. Might be ready for master, it has an ACK from Eric. Slated into next just because I wasn't sure if Junio merges such things immediately into master or lets them cook for a bit. * pb/gitweb (Fri Oct 3 07:41:25 2008 -0700) 7 commits + Merge branch 'pb/gitweb-tagcloud' into pb/gitweb + gitweb: Support for simple project search form + gitweb: Make the by_tag filter delve in forks as well + gitweb: Support for tag clouds + gitweb: Add support for extending the action bar with custom links + gitweb: Sort the list of forks on the summary page by age + gitweb: Clean-up sorting of project list I may have been a bad interm maintainer here by shoving together two different gitweb series from Pasky. Overall the combined set looks ready to me. * ph/parseopt (Thu Oct 2 14:59:20 2008 +0200) 3 commits + parse-opt: migrate builtin-merge-file. + parse-opt: migrate git-merge-base. + parse-opt: migrate fmt-merge-msg. * rz/grepz (Wed Oct 1 18:11:15 2008 +0200) 1 commit + git grep: Add "-z/--null" option as in GNU's grep. * mv/merge-refresh (Fri Oct 3 15:02:31 2008 +0200) 1 commit + builtin-merge: refresh the index before calling a strategy * bc/xdiffnl (Wed Oct 1 14:28:26 2008 -0500) 1 commit + xdiff-interface.c: strip newline (and cr) from line before pattern matching * ae/preservemerge (Mon Sep 29 22:28:57 2008 +0200) 1 commit + rebase: Support preserving merges in non-interactive mode These last five all look ready for master, but should cook longer than just a few days. ---------------------------------------------------------------- [Old New Topics] * mw/sendemail (Sun Sep 28 07:51:21 2008 +0300) 10 commits + bash completion: Add --[no-]validate to "git send-email" + send-email: signedoffcc -> signedoffbycc, but handle both + Docs: send-email: Create logical groupings for man text + Docs: send-email: Create logical groupings for --help text + Docs: send-email: Remove unnecessary config variable description + Docs: send-email: --chain_reply_to -> --[no-]chain-reply-to + send-email: change --no-validate to boolean --[no-]validate + Docs: send-email: Man page option ordering + Docs: send-email usage text much sexier + Docs: send-email's usage text and man page mention same options This series looks pretty good to me. It probably can merge to master on the next "What's cooking" cycle. ---------------------------------------------------------------- [Dropped Topics] * sg/merge-options (Sun Apr 6 03:23:47 2008 +0200) 1 commit + merge: remove deprecated summary and diffstat options and config variables This has been reverted from next and is being dropped. Its too soon to remove a feature that was just added in 1.5.6 timeframe. Perhaps in 1.7.0 this can be revisited. ---------------------------------------------------------------- [Stuck Topics] * nd/narrow (Wed Oct 1 11:04:09 2008 +0700) 9 commits - grep: skip files outside sparse checkout area - checkout_entry(): CE_NO_CHECKOUT on checked out entries. - Prevent diff machinery from examining worktree outside sparse checkout - ls-files: Add tests for --sparse and friends - update-index: add --checkout/--no-checkout to update CE_NO_CHECKOUT bit - update-index: refactor mark_valid() in preparation for new options - ls-files: add options to support sparse checkout - Introduce CE_NO_CHECKOUT bit - Extend index to save more flags Recently updated with a repost. I've been too swamped at day-job to actively review the series. There has been some discussion on list about it so its not stalled, but it needs more eyeballs. * gb/gitweb-pathinfo (Thu Oct 2 02:10:34 2008 +0200) 6 commits - gitweb: generate parent..current URLs - gitweb: parse parent..current syntax from pathinfo - gitweb: use_pathinfo filenames start with / - gitweb: generate project/action/hash URLs - gitweb: refactor input parameters parse/validation - gitweb: parse project/action/hash_base:filename PATH_INFO Still going through revisions on the git ML. Its great work and is heading in the right direction. But the discussion and patch generation rate is too high to merge the topic into next just yet. So its waiting for a re-roll of the series once discussion settles down and there is a final patch series available. ---------------------------------------------------------------- [Stalled -- Needs Action to Proceed (or to be dropped)] * pb/submodule (Fri Sep 12 23:09:19 2008 +0200) 1 commit - t7400: Add short "git submodule add" testsuite Still waiting for a reroll. Should be dropped next week if we don't get one during this coming week. * bd/blame (Thu Aug 21 18:22:01 2008 -0500) 5 commits - Use xdiff caching to improve git blame performance - Allow xdiff machinery to cache hash results for a file - Always initialize xpparam_t to 0 - Bypass textual patch generation and parsing in git blame - Allow alternate "low-level" emit function from xdl_diff (jc: Réne had good comments on how the callback should be structured.) * kb/am-directory (Fri Aug 29 15:27:50 2008 -0700) 1 commit - git-am: Pass the --directory option through to git-apply (jc: I think this is still buggy and drops the option when am stops with conflicts.) All three of these are stalled. I'm not going to drop them just yet, but Junio's comments still hold. If there's no action on these next week we may seem them drop off. ---------------------------------------------------------------- [Will be merged to 'master/maint' soon] * jc/add-ita (Thu Aug 21 01:44:53 2008 -0700) 1 commit + git-add --intent-to-add (-N) Teaches "git add" to record only the intent to add a path later. (jc: I rerolled this without the fake empty blob object.) * mg/verboseprune (Mon Sep 29 18:49:52 2008 +0200) 1 commit + make prune report removed objects on -v This is a pretty trivial new feature. * pb/commit-where (Fri Oct 3 22:13:49 2008 -0400) 4 commits + tutorial: update output of git commit + reformat informational commit message + git commit: Reformat output somewhat + builtin-commit.c: show on which branch a commit was added I think this topic is settled for now. * jk/maint-soliconv (Fri Oct 3 02:39:36 2008 -0400) 1 commit + Makefile: do not set NEEDS_LIBICONV for Solaris 8 * sg/maint-intrebase-msghook (Fri Oct 3 11:33:20 2008 +0200) 2 commits + rebase -i: remove leftover debugging + rebase -i: proper prepare-commit-msg hook argument when squashing * bc/maint-stashref (Thu Oct 2 18:52:11 2008 -0500) 1 commit + git-stash.sh: fix flawed fix of invalid ref handling (commit da65e7c1) The latter three are ready for 'maint'. I felt burned by a bad commit into made directly into maint earlier in the week so these wound up getting queued into 'next' first, even though they are certainly maint material. ---------------------------------------------------------------- [Actively Cooking] * tr/workflow-doc (Sat Sep 13 18:11:01 2008 +0200) 2 commits + Documentation: Refer to git-rebase(1) to warn against rewriting + Documentation: new upstream rebase recovery section in git-rebase (jc: My impression from the last round of discusson on the third patch in this series (not queued here) was that as long as we do not present it as "One True Workflow", the description was a good starting point, possibly others to add other recommended flows later.) I haven't looked at this series yet. Based on Junio's remarks above I'm looking for more input on this series before it goes anywhere. * lt/time-reject-fractional-seconds (Sat Aug 16 21:25:40 2008 -0700) 1 commit + date/time: do not get confused by fractional seconds I need to look at this in more detail. I suspect we can merge this to master soon, but its been kicking around since Aug 16th. I need to look at it to see if there's any obvious reason why Junio has no notes on this branch and why it hasn't merged yet. ---------------------------------------------------------------- [On Hold] * jc/post-simplify (Fri Aug 15 01:34:51 2008 -0700) 2 commits - revision --simplify-merges: incremental simplification - revision --simplify-merges: prepare for incremental simplification (jc: I started making this incremental but the progress is not so great.) * jc/stripspace (Sun Mar 9 00:30:35 2008 -0800) 6 commits - git-am --forge: add Signed-off-by: line for the author - git-am: clean-up Signed-off-by: lines - stripspace: add --log-clean option to clean up signed-off-by: lines - stripspace: use parse_options() - Add "git am -s" test - git-am: refactor code to add signed-off-by line for the committer (jc: The one at second from the tip needs reworking.) * jc/send-pack-tell-me-more (Thu Mar 20 00:44:11 2008 -0700) 1 commit - "git push": tellme-more protocol extension This is a great idea, but IIRC its implementation has a deadlock during communication between the peers. Someone needs to pick up this topic and resolve the deadlock before it can continue. * jc/blame (Wed Jun 4 22:58:40 2008 -0700) 2 commits - blame: show "previous" information in --porcelain/--incremental format - git-blame: refactor code to emit "porcelain format" output * jc/merge-whitespace (Sun Feb 24 23:29:36 2008 -0800) 1 commit . WIP: start teaching the --whitespace=fix to merge machinery This has a merge conflict with `next` and isn't even in pu right now. * jk/renamelimit (Sat May 3 13:58:42 2008 -0700) 1 commit - diff: enable "too large a rename" warning when -M/-C is explicitly asked for (jc: This would be the right thing to do for command line use, but gitk will be hit due to tcl/tk's limitation, so I am holding this back for now.)

What's in git/spearce.git (Oct 2008, #01; Mon, 06)

Shawn O. Pearce — 2008-10-06T16:59:08

What's in git/spearce.git (Oct 2008, #01; Mon, 06) maint e261cf9 (Update release notes for 1.6.0.3) master 276328f (Merge branch 'maint') ------------------------------------------------------------------------ Junio should be returning from vacation this week. I will keep updating my git/spearce.git tree until he gets back and asks for my current topic heads. Once Junio gets back I think we need to focus on getting 1.6.0.3 out the door. We have 77 commits queued up since 1.6.0.2 and it has been almost 4 weeks since our last maint release. So please, test, test, test maint. -------- * The 'maint' branch has these fixes since the last announcement. Alexander Gavrilov (1): builtin-blame: Fix blame -C -C with submodules. David Soria Parra (1): Solaris: Use OLD_ICONV to avoid compile warnings Dmitry Potapov (2): make prefix_path() never return NULL do not segfault if make_cache_entry failed Giuseppe Bilotta (1): gitweb: remove PATH_INFO from $my_url and $my_uri Jakub Narebski (2): gitweb: Fix two 'uninitialized value' warnings in git_tree() gitweb: Add path_info tests to t/t9500-gitweb-standalone-no-errors.sh Linus Torvalds (1): fix bogus "diff --git" header from "diff --no-index" Luc Heinrich (1): git-svn: call 'fatal' correctly in set-tree Michael Prokop (1): Replace svn.foo.org with svn.example.com in git-svn docs (RFC 2606) Nanako Shiraishi (2): docs: describe pre-rebase hook Teach rebase -i to honor pre-rebase hook SZEDER Gábor (3): t0024: add executable permission Documentation: remove '\' in front of short options bash: remove fetch, push, pull dashed form leftovers Shawn O. Pearce (1): Update release notes for 1.6.0.3 Tuncer Ayaz (1): Fix fetch/clone --quiet when stdout is connected martin f. krafft (1): Improve git-log documentation wrt file filters * The 'master' branch has these since the last announcement in addition to the above. Brandon Casey (1): diff.c: remove duplicate bibtex pattern introduced by merge 92bb9785 David Aguilar (1): tests: add a testcase for "git submodule sync" David Bryson (1): Use "git_config_string" to simplify "builtin-gc.c" code where "prune_expire" is set Dmitry Potapov (1): do not segfault if make_cache_entry failed Jeff King (1): tests: grep portability fixes Jonathan del Strother (1): Add OS X support to the pre-auto-gc example hook Nanako Shiraishi (5): dir.c: make dir_add_name() and dir_add_ignored() static run-command.c: remove run_command_v_opt_cd() config.c: make git_parse_long() static commit.c: make read_graft_file() static archive.c: make archiver static Nicolas Pitre (1): fix openssl headers conflicting with custom SHA1 implementations Petr Baudis (3): gitweb: Quote non-displayable characters in hex, not octal config.c: Tolerate UTF8 BOM at the beginning of config file gitweb: Identify all summary metadata table rows Ralf Wildenhues (1): Correct typos in RelNotes-1.6.1 SZEDER Gábor (1): remove vim syntax highlighting in favor of upstream

Git filter branch - removing empty commits

Peter Waller — 2008-10-06T14:57:22

Dear List, I have removed lots of files from my history with a procedure something like this: 1) Clone repository 2) git filter-branch --index-filter 'git rm --quiet -r Archive; git ls-files -z '\''*.png'\'' '\''*.eps'\'' | xargs -0r git rm --quiet; true' 3) git reflog expire --expire=0 --all 4) git prune; git gc The problem is that this leaves many empty commits. The helpful people over at #git advised I do: git filter-branch --commit-filter 'if [ z$1 = z`git rev-parse $3^{tree} 2> /dev/null` ]; then skip_commit "$< at >"; else git commit-tree "$< at >"; fi' But this gets to about commit 70/1300 and grinds to a halt. I modified /usr/libexec/git-core/git-filter-branch to say #!/bin/sh -x, and the result was over 80mb in a few seconds. I have pasted the last few lines here: http://rafb.net/p/ABWvCy44.html It is like it is following some exponential behaviour and not getting anywhere. Any ideas what is wrong? I have tried a couple of scripts I have found around that should do the same thing, for example: skip_commit() { shift while [[ -n $1 ]] ; do shift map "$1" shift done } our_tree="$1" our_parent_tree=$(map $3) if [[ ${our_tree} == $(git rev-parse $(map $3)) ]]; then git commit-tree "$< at >" else skip_commit "$< at >" fi This goes much faster, but when it reaches the final commit, it says this: Ref 'refs/heads/master' was deleted fatal: Not a valid object name HEAD And not much useful seems to have happened. Any help appreciated. Regards, - Peter

Repairing fatal: ref HEAD is not a symbolic ref (git checkout of svn remote)

Jeff Kowalczyk — 2008-10-06T14:00:58

To test a particular upstream revision, on a git svn remote checkout (i.e. branches/1.2, 76c7af2...), I checked out the equivalent of HEAD^^ (git checkout 4a3d99c0c9). Now back at the branch tip, I get fatal: ref HEAD is not a symbolic ref on git svn rebase: (4a3d99c...) $ git checkout 76c7af2 HEAD is now at 76c7af2... Minor changes to CONTRIBUTORS (76c7af2...) $ git svn rebase fatal: ref HEAD is not a symbolic ref Current branch HEAD is up to date. Segmentation fault How can I manually correct ref HEAD? Thanks. (The segmentation fault is a separate issue)

how to restrict commit for a repo

Dilip M — 2008-10-06T12:09:39

Hi all, I am a new user to GIT. Read few docs in web ..I want to know few things.. If I have a repository, how to prevent push from other repo's into mine master's? I want to prevent the commits from all developers and allow only few ppl to commit to masters.. Thanks for sharing the info on this...

[QGIT PATCH] Rework the commit confirmation box a bit

Abdelrazak Younes — 2008-10-06T09:11:53

The problem was that the dialog was too big for my whenever too many files were changed. Now, the list of changed files is only shown whenever they are less than 20; otherwise it is shown in the detailed text accessible though the 'Show Detail' button. --- src/commitimpl.cpp | 25 ++++++++++++++++++++----- 1 files changed, 20 insertions(+), 5 deletions(-) diff --git a/src/commitimpl.cpp b/src/commitimpl.cpp index 1540947..def5209 100644 --- a/src/commitimpl.cpp +++ b/src/commitimpl.cpp < at >< at > -236,17 +236,32 < at >< at > bool CommitImpl::checkConfirm(SCRef msg, SCRef patchName, SCList selFiles, bool (git->isStGITStack() ? "refresh top patch with" : "amend last commit with") : (git->isStGITStack() ? "create a new patch with" : "commit"); - QString text("Do you want to " + whatToDo + " the following file(s)?\n\n" + - selFiles.join("\n") + "\n\nwith the message:\n\n"); + QString text("Do you want to " + whatToDo); + + bool const fullList = selFiles.size() < 20; + if (fullList) { + text.append(" the following file(s)?\n\n" + selFiles.join("\n") + + "\n\nwith the message:\n\n"); + } else { + text.append(" those " + QString::number(selFiles.size()) + + " files the with the message:\n\n"); + } + text.append(msg); if (git->isStGITStack()) text.append("\n\nAnd patch name: " + patchName); QTextCodec::setCodecForCStrings(tc); - int but = QMessageBox::question(this, "Commit changes - QGit", - text, "&Yes", "&No", QString(), 0, 1); - return (but != 1); + QMessageBox msgBox(this); + msgBox.setWindowTitle("Commit changes - QGit"); + msgBox.setText(text); + if (!fullList) + msgBox.setDetailedText(selFiles.join("\n")); + msgBox.setStandardButtons(QMessageBox::Yes | QMessageBox::No); + msgBox.setDefaultButton(QMessageBox::Yes); + + return msgBox.exec() != QMessageBox::No; } void CommitImpl::pushButtonSettings_clicked() {

(unknown)

Robin Rosenberg — 2008-10-05T23:36:36

This series decorates the graphical and text (jgit log) history listings with tags. For the text command it is optional. Reviewers may want to pay special attention to the changes in the Ref class.

[ANNOUNCE] cgit 0.8

Lars Hjemli — 2008-10-05T20:10:17

cgit-0.8, another webinterface for git, is now available. clone: git://hjemli.net/pub/git/cgit browse: http://hjemli.net/git/cgit Some of the new features since 0.7.2: * atom feeds for branch and/or path (http://hjemli.net/git/cgit/atom) * builtin clone/fetch (git clone http://hjemli.net/git/cgit) * commit messages with 'encoding' attribute gets converted to utf-8 * commit view includes full diff (http://hjemli.net/git/cgit/commit/) * limited number of entries in cache (no need for manual cleanup) * new plain blob view (http://hjemli.net/git/cgit/plain/cgitrc.5.txt) * new plain patch view (http://hjemli.net/git/cgit/patch/) * searchable, paginated project list * support for PATH_INFO (no need for url rewriting) Big thanks to everyone who submitted patches and feedback! --- Sortlog since v0.7.2 Eric Wong (1): use Host: header to generate cgit_hosturl Harley Laue (3): Print an error if filename is not found in html_include. Added root-desc to default configuration. Add LDFLAGS to makefile. Hiroki Hattori (1): Fix segfault Jonathan Bastien-Filiatrault (4): Add commit->msg_encoding, allocate msg dynamicly. Set msg_encoding according to the header. Add iconv_msg function. Convert subject and message with iconv_msg. Lars Hjemli (142): Use utf8::reencode_string from git Show lines changed as -n/+m in shortlogs Makefile: link with libiconv if NEEDS_LIBICONV is defined Minor css tweaks Create initial testsuite Fix html error detected by test-suite Set commit date on snapshot contents Generate valid html for "downloads" menu header Add support for "robots" meta-tag Gimp my logo Use tables for page layout Add cell-spacing in html Add support for automatic and custom clone urls Add plain patch view Handle missing default branch and error out on invalid branch names Check for NULL-subject in patch view Use GIT-1.5.4.rc4 Use GIT-1.5.4.rc5 Use GIT-1.5.4 Use GIT-1.5.4.1 Introduce struct cgit_context Add all config variables into struct cgit_context Move cgit_repo into cgit_context Improve initialization of git directory Add regression-test for description-less repos Brown paper bag: don't use `grep -v` Make output from `make test` more readable t0010-validate.sh: return on cgit errors tests/setup.sh: cleanup test script output and logging Use GIT-1.5.4.4 Fix segfault in patch view for root commit Introduce html.h Use GIT-1.5.5.rc1 Add struct cgit_page to cgit_context Remove obsolete cacheitem parameter to ui-functions Add command dispatcher Refactor snapshot support Add separate header-files for each page/view Add ui-shared.h Move non-generic functions from shared.c to cgit.c Initial Makefile cleanup Makefile: autobuild dependency rules Move cgit_version from shared.c to cgit.c Makefile: copy the QUIET constructs from the Makefile in git.git Remove global and obsolete cgit_cmd Add cache.h Move function for configfile parsing into configfile.[ch] Use GIT-1.5.5-rc2 Move cgit_parse_query() from parsing.c to html.c as http_parse_querystring() Reset ctx.repo to NULL when the config parser is finished Don't specify mimetype in ui-blob.c Use GIT-1.5.5 Replace sidebar/logo Include diff in commit view Add fixed link to index page from repo header Reintroduce the branch switcher Minor fixup in tree-view css More layout fixes Implement minimal freetext search in the repolist Remove 'patch' link from tab, add to commit view Make repository search case insensitive Fix css font-family cgit.css: set form margins Fix more css ugliness Make branches, tags and log play better together in the summary view Make a few more columns in repolist and log view clickable Fix search in repo index even if caching is enabled Cleanup page header Fix commitdiff annoyance Integrate diffstat with diff ui-diff: remove test on object type Redesign the caching layer Add page 'ls_cache' Prepare for 'about repo' page Move included header-file out of repolist table Re-enable 'index-info' and add support for 'root-desc' in cgitrc Make it possible for a single cmd to work both with and without a repo Prepare for 'about site' page / add 'root-readme' option to cgitrc Add 'about site' and 'about repo' pages Add cgit_index_link() function with support for offset Add a pager on the repolist Add footer with page creation time and cgit version on all pages Add link to index page from repo header, remove page name ui-commit: handle root commits Include commit-id in link from diff-stat Use GIT-1.5.5.1 cache.c: read(2) returns -1 on error, not 0 cache.c: make all io-related functions return errno on error cache.c: use xread()/xwrite() from libgit cache.c: do not ignore errors from print_slot() cache.c: fix error checking in print_slot() ui-tree.c: avoid peeking at GITLINK objects Use GIT-1.5.6 Add support for including a footer on all pages Add a favicon option to cgitrc Use GIT-1.6.0-rc0 Adjust to new calling convention for read_tree_recursive() tests/Makefile: not everyone has `.` in $PATH Makefile: fix git dependency rules Makefile: do not touch the git objects with `make clean` Makefile: remove the `distclean` and `emptycache` targets ui-summary: show clone urls ui-shared: show repo owner along with description Be prepared for empty repositories Modify default value for a few cgitrc options Makefile: another take on git dependency rules Add atom-support Add support for cloning over http Implement plain view ui-tree: link to plain view instead of blob view Makefile: Git dependency, take 3 Use GIT-1.6.0.rc1 Supply status description to html_status() cache.c: use %zd for off_t argument Use GIT-1.6.0 ui-plain: handle subdirectories Use GIT-1.6.0.1 Update Makefile to use GIT-1.6.0.1 parsing.c: be prepared for unexpected content in commit/tag objects ui-tag: show the taggers email Teach cgit how to use PATH_INFO Add support for --scan-tree= option to cgit ui-repolist: enable filtering of repos by path ui-diff: fix links from diffstat ui-diff: make diffstat header a link to the full diff html.c: add html_url_arg ui-shared.c: use html_url_arg() ui-snapshot.c: specify archiver_args.baselen html.c: add html_url_path ui-shared.c: use html_url_path() in repolink() ui-shared.c: add cgit_summary_link() ui-repolist + ui-shared: Use cgit_summary_link() Replace cgitrc with cgitrc.5.txt ui-shared: do not print repo name on the "summary" tab ui-shared: reword the standard page footer Use GIT-1.6.0.2 ui-shared: use html_url_path() to get properly escaped url in form action ui-shared: generate proper links in cgit_object_link() ui-shared: add cgit_tag_link() ui-refs: use cgit_tag_link() ui-summary: use html_url_path() CGIT 0.8 Michael Krelin (2): allow blob extract blobs by head/path combination allow specification of directly linked blobs mimetypes Stefan Naewe (2): Added `local-time` option to cgitrc cgitrc: explain new local-time option

[PATCH] git init: --bare/--shared overrides system/global config

Deskin Miller — 2008-10-05T19:44:12

From b6144562983703079a1eba8cdac3506c18d751a3 Mon Sep 17 00:00:00 2001 From: Deskin Miller umich.edu> Date: Sat, 4 Oct 2008 20:07:44 -0400 If core.bare or core.sharedRepository are set in /etc/gitconfig or ~/.gitconfig, then 'git init' will read the values when constructing a new config file; reading them, however, will override the values specified on the command line. In the case of --bare, this ends up causing a segfault, without the repository being properly initialised; in the case of --shared, the permissions are set according to the existing config settings, not what was specified on the command line. This fix saves any specified values for --bare and --shared prior to reading existing config settings, and restores them after reading but before writing the new config file. Also includes a testcase which has a specified global config file override, demonstrating the former failure scenario. Signed-off-by: Deskin Miller umich.edu> --- I'm not a great fan of the method I took to save and restore the values specified to init, but it works. Also, I think the testcase is nice (but I'm biased, seeing how I wrote it): in general I'd argue for more testcases which deal with issues caused by the user's complete installation. I based this off of maint, because I think it should be applied there, but it applies cleanly to master if you feel that's better. builtin-init-db.c | 8 ++++++++ t/t0001-init.sh | 17 +++++++++++++++++ 2 files changed, 25 insertions(+), 0 deletions(-) diff --git a/builtin-init-db.c b/builtin-init-db.c index 8140c12..38e282c 100644 --- a/builtin-init-db.c +++ b/builtin-init-db.c < at >< at > -17,6 +17,9 < at >< at > #define TEST_FILEMODE 1 #endif +static int init_is_bare_repository = 0; +static int init_shared_repository = PERM_UMASK; + static void safe_create_dir(const char *dir, int share) { if (mkdir(dir, 0777) < 0) { < at >< at > -191,6 +194,8 < at >< at > static int create_default_files(const char *template_path) copy_templates(template_path); git_config(git_default_config, NULL); +is_bare_repository_cfg = init_is_bare_repository; +shared_repository = init_shared_repository; /* * We would have created the above under user's umask -- under < at >< at > -277,6 +282,9 < at >< at > int init_db(const char *template_dir, unsigned int flags) safe_create_dir(get_git_dir(), 0); +init_is_bare_repository = is_bare_repository(); +init_shared_repository = shared_repository; + /* Check to see if the repository version is right. * Note that a newly created repository does not have * config file, so this will not fail. What we are catching diff --git a/t/t0001-init.sh b/t/t0001-init.sh index 620da5b..6a6bca0 100755 --- a/t/t0001-init.sh +++ b/t/t0001-init.sh < at >< at > -167,4 +167,21 < at >< at > test_expect_success 'init with --template (blank)' ' ! test -f template-blank/.git/info/exclude ' +test_expect_success 'init --bare/--shared overrides system/global config' ' +( +HOME="`pwd`" && +export HOME && +test_config="$HOME"/.gitconfig && +unset GIT_CONFIG_NOGLOBAL && +git config -f "$test_config" core.bare false && +git config -f "$test_config" core.sharedRepository 0640 && +mkdir init-bare-shared-override && +cd init-bare-shared-override && +git init --bare --shared=0666 +) && +check_config init-bare-shared-override true unset && +test 0666 = \ +`git config -f init-bare-shared-override/config core.sharedRepository` +' + test_done

[PATCH] fix bogus "diff --git" header from "diff --no-index"

Jeff King — 2008-10-05T19:35:15

From: Linus Torvalds linux-foundation.org> When "git diff --no-index" is given an absolute pathname, it would generate a diff header with the absolute path prepended by the prefix, like: diff --git a/dev/null b/foo Not only is this nonsensical, and not only does it violate the description of diffs given in git-diff(1), but it would produce broken binary diffs. Unlike text diffs, the binary diffs don't contain the filenames anywhere else, and so "git apply" relies on this header to figure out the filename. This patch just refuses to use an invalid name for anything visible in the diff. Now, this fixes the "git diff --no-index --binary a /dev/null" kind of case (and we'll end up using "a" as the basename), but some other insane cases are impossible to handle. If you do git diff --no-index --binary a /bin/echo you'll still get a patch like diff --git a/a b/bin/echo old mode 100644 new mode 100755 index ... and "git apply" will refuse to apply it for a couple of reasons, and the diff is simply bogus. And that, btw, is no longer a bug, I think. It's impossible to know whethe the user meant for the patch to be a rename or not. And as such, refusing to apply it because you don't know what name you should use is probably _exactly_ the right thing to do! Original problem reported by Imre Deak. Test script and problem description by Jeff King. Signed-off-by: Jeff King peff.net> --- diff.c | 4 ++++ t/t4012-diff-binary.sh | 21 +++++++++++++++++++++ 2 files changed, 25 insertions(+), 0 deletions(-) diff --git a/diff.c b/diff.c index 4e4e439..02e948c 100644 --- a/diff.c +++ b/diff.c < at >< at > -1509,6 +1509,10 < at >< at > static void builtin_diff(const char *name_a, b_prefix = o->b_prefix; } +/* Never use a non-valid filename anywhere if at all possible */ +name_a = DIFF_FILE_VALID(one) ? name_a : name_b; +name_b = DIFF_FILE_VALID(two) ? name_b : name_a; + a_one = quote_two(a_prefix, name_a + (*name_a == '/')); b_two = quote_two(b_prefix, name_b + (*name_b == '/')); lbl[0] = DIFF_FILE_VALID(one) ? a_one : "/dev/null"; diff --git a/t/t4012-diff-binary.sh b/t/t4012-diff-binary.sh index b8ec6e9..421f4bb 100755 --- a/t/t4012-diff-binary.sh +++ b/t/t4012-diff-binary.sh < at >< at > -77,4 +77,25 < at >< at > test_expect_success 'apply binary patch' \ tree1=`git write-tree` && test "$tree1" = "$tree0"' +q_to_nul() { +perl -pe 'y/Q/\000/' +} + +nul_to_q() { +perl -pe 'y/\000/Q/' +} + +test_expect_success 'diff --no-index with binary creation' ' +echo Q | q_to_nul >binary && +(:# hide error code from diff, which just indicates differences + git diff --binary --no-index /dev/null binary >current || + true +) && +rm binary && +git apply --binary expected && +nul_to_q actual && +test_cmp expected actual +' + test_done

git svn: Bad URL passed to RA layer: Unrecognized URL scheme

Jeff Kowalczyk — 2008-10-05T17:24:04

I made an ill-advised upgrade to from subversion-1.4.6 to subversion-1.5.2. While I had subversion-1.5.2 installed, git svn rebase ended with a 'segmentation fault' error. Now back at subversion-1.4.6, git repositories with svn remotes updated during the aforentioned upgrade now give an error: (master) $ git svn fetch Bad URL passed to RA layer: Unrecognized URL scheme for 'http://code.djangoproject.com/svn' at /usr/libexec/git-core/git-svn line 1400 (master) $ git svn rebase Bad URL passed to RA layer: Unrecognized URL scheme for 'http://code.djangoproject.com/svn' at /usr/libexec/git-core/git-svn line 1909 Others which were not updated recently work fine for git svn operations. How can I repair the affected git repositories? Thanks.