Code

git.git
16 years agoDo not rely on the exit status of "unset" for unset variables
H.Merijn Brand [Tue, 4 Dec 2007 22:45:16 +0000 (22:45 +0000)]
Do not rely on the exit status of "unset" for unset variables

POSIX says that exit status "0" means that "unset" successfully unset
the variable.  However, it is kind of ambiguous if an environment
variable which was not set could be successfully unset.

At least the default shell on HP-UX insists on reporting an error in
such a case, so just ignore the exit status of "unset".

[Dscho: extended the patch to git-submodule.sh, as Junio realized that
 this is the only other place where we check the exit status of "unset".]

Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
16 years agogitweb: use Perl built-in utf8 function for UTF-8 decoding.
İsmail Dönmez [Tue, 4 Dec 2007 08:55:41 +0000 (10:55 +0200)]
gitweb: use Perl built-in utf8 function for UTF-8 decoding.

Signed-off-by: İsmail Dönmez <ismail@pardus.org.tr>
Tested-by: Martin Koegler <mkoegler@auto.tuwien.ac.at>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
16 years agot9600: require cvsps 2.1 to perform tests
Jeff King [Tue, 4 Dec 2007 01:41:45 +0000 (20:41 -0500)]
t9600: require cvsps 2.1 to perform tests

git-cvsimport won't run at all with less than cvsps 2.1, because it
lacks the -A flag. But there's no point in preventing people who have an
old cvsps from running the full testsuite.

Tested-by: A Large Angry SCM <gitzilla@gmail.com>
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
16 years agocontrib: Make remotes2config.sh script more robust
Jakub Narebski [Sun, 2 Dec 2007 19:40:43 +0000 (20:40 +0100)]
contrib: Make remotes2config.sh script more robust

The remotes2config.sh script replaced all 'unsafe' characters in repo
name with '.'; include '-' in the 'safe' characters set (the set is
probably even larger).

Script required also space after "URL:", "Push:" and "Pull:" in
remotes file. This for example made the following remote
  URL: git://git.kernel.org/pub/scm/git/git.git
  Pull: refs/heads/master:refs/heads/origin
  Pull:+refs/heads/pu:refs/heads/pu
miss 'pu' branch (forced branch) in config file after conversion.
Allow for any number of whitespace after "URL:", "Push:", "Pull:".

Signed-off-by: Jakub Narebski <jnareb@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
16 years agoUse a strbuf for copying the command line for the reflog.
Kristian Høgsberg [Tue, 4 Dec 2007 07:25:46 +0000 (02:25 -0500)]
Use a strbuf for copying the command line for the reflog.

Signed-off-by: Kristian Høgsberg <krh@redhat.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
16 years agoDocumentation/git.txt: typofix
Junio C Hamano [Tue, 4 Dec 2007 08:47:04 +0000 (00:47 -0800)]
Documentation/git.txt: typofix

Signed-off-by: Junio C Hamano <gitster@pobox.com>
16 years agoDocumentation: rerere is enabled by default these days.
Junio C Hamano [Tue, 4 Dec 2007 08:40:55 +0000 (00:40 -0800)]
Documentation: rerere is enabled by default these days.

Signed-off-by: Junio C Hamano <gitster@pobox.com>
16 years agoFormatting fix for Documentation/git-help.txt
Junio C Hamano [Tue, 4 Dec 2007 08:29:32 +0000 (00:29 -0800)]
Formatting fix for Documentation/git-help.txt

Signed-off-by: Junio C Hamano <gitster@pobox.com>
16 years agoUpdate draft release notes for 1.5.4
Junio C Hamano [Tue, 4 Dec 2007 07:13:02 +0000 (23:13 -0800)]
Update draft release notes for 1.5.4

Signed-off-by: Junio C Hamano <gitster@pobox.com>
16 years agoAllow ':/<oneline-prefix>' syntax to work with save_commit_buffer == 0
Johannes Schindelin [Mon, 3 Dec 2007 18:42:39 +0000 (18:42 +0000)]
Allow ':/<oneline-prefix>' syntax to work with save_commit_buffer == 0

Earlier, ':/<oneline-prefix>' would not work (i.e. die) with commands that
set save_commit_buffer = 0, such as blame, describe, pack-objects, reflog
and bundle.

Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
16 years agoAdd git-fast-export to list of commands.
Junio C Hamano [Tue, 4 Dec 2007 05:56:09 +0000 (21:56 -0800)]
Add git-fast-export to list of commands.

Signed-off-by: Junio C Hamano <gitster@pobox.com>
16 years agofast-export: rename the signed tag mode 'ignore' to 'verbatim'
Johannes Schindelin [Mon, 3 Dec 2007 22:44:39 +0000 (22:44 +0000)]
fast-export: rename the signed tag mode 'ignore' to 'verbatim'

The name 'verbatim' describes much better what this mode does with
signed tags.  While at it, fix the documentation what it actually
does.

Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
16 years agoAdd remote.<name>.proxy
Sam Vilain [Mon, 3 Dec 2007 21:48:54 +0000 (10:48 +1300)]
Add remote.<name>.proxy

As well as allowing a default http.proxy option, allow it to be set
per-remote.

Signed-off-by: Sam Vilain <sam.vilain@catalyst.net.nz>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
16 years agoAllow HTTP proxy to be overridden in config
Sam Vilain [Fri, 23 Nov 2007 00:07:00 +0000 (13:07 +1300)]
Allow HTTP proxy to be overridden in config

The http_proxy / HTTPS_PROXY variables used by curl to control
proxying may not be suitable for git.  Allow the user to override them
in the configuration file.

Signed-off-by: Junio C Hamano <gitster@pobox.com>
16 years agoTrace and quote with argv: get rid of unneeded count argument.
Christian Couder [Mon, 3 Dec 2007 04:51:50 +0000 (05:51 +0100)]
Trace and quote with argv: get rid of unneeded count argument.

Now that str_buf takes care of all the allocations, there is
no more gain to pass an argument count.

So this patch removes the "count" argument from:
- "sq_quote_argv"
- "trace_argv_printf"
and all the callers.

Signed-off-by: Christian Couder <chriscool@tuxfamily.org>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
16 years agoinstall-sh from automake does not like -m without delimiting space
Robert Schiele [Sat, 1 Dec 2007 17:05:40 +0000 (18:05 +0100)]
install-sh from automake does not like -m without delimiting space

The install-sh script as shipped with automake requires a space between
the -m switch and its argument.  Since this is also the regular way of
doing it with other install implementations this change inserts the
missing space in all makefiles.

Signed-off-by: Robert Schiele <rschiele@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
16 years agoDocumentation: add a new man page for "git-help"
Christian Couder [Tue, 4 Dec 2007 05:44:29 +0000 (06:44 +0100)]
Documentation: add a new man page for "git-help"

Signed-off-by: Christian Couder <chriscool@tuxfamily.org>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
16 years agogit-commit documentation: fix unfinished sentence.
Junio C Hamano [Mon, 3 Dec 2007 21:35:26 +0000 (13:35 -0800)]
git-commit documentation: fix unfinished sentence.

Signed-off-by: Junio C Hamano <gitster@pobox.com>
16 years agogit-commit --allow-empty
Junio C Hamano [Mon, 3 Dec 2007 08:03:10 +0000 (00:03 -0800)]
git-commit --allow-empty

It does not usually make sense to record a commit that has the exact
same tree as its sole parent commit and that is why git-commit prevents
you from making such a mistake, but when data from foreign scm is
involved, it is a different story.  We are equipped to represent such an
(perhaps insane, perhaps by mistake, or perhaps done on purpose) empty
change, and it is better to represent it bypassing the safety valve for
native use.

This is primarily for use by foreign scm interface scripts.

Signed-off-by: Junio C Hamano <gitster@pobox.com>
16 years agogit-commit: Allow to amend a merge commit that does not change the tree
Johannes Sixt [Mon, 3 Dec 2007 07:24:50 +0000 (08:24 +0100)]
git-commit: Allow to amend a merge commit that does not change the tree

Normally, it should not be allowed to generate an empty commit. A merge
commit generated with git 'merge -s ours' does not change the tree (along
the first parent), but merges are not "empty" even if they do not change
the tree. Hence, commit 8588452ceb7 allowed to amend a merge commit that
does not change the tree, but 4fb5fd5d301 disallowed it again in an
attempt to avoid that an existing commit is amended such that it becomes
empty. With this change, a commit can be edited (create a new one or amend
an existing one) either if there are changes or if there are at least two
parents.

Signed-off-by: Johannes Sixt <johannes.sixt@telecom.at>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
16 years agoMerge branch 'js/prune-expire'
Junio C Hamano [Mon, 3 Dec 2007 07:03:38 +0000 (23:03 -0800)]
Merge branch 'js/prune-expire'

* js/prune-expire:
  Add "--expire <time>" option to 'git prune'

16 years agoMerge branch 'js/export-with-assignment'
Junio C Hamano [Mon, 3 Dec 2007 07:01:25 +0000 (23:01 -0800)]
Merge branch 'js/export-with-assignment'

* js/export-with-assignment:
  Replace instances of export VAR=VAL with VAR=VAL; export VAR

16 years agoMerge branch 'js/fast-export'
Junio C Hamano [Mon, 3 Dec 2007 07:01:15 +0000 (23:01 -0800)]
Merge branch 'js/fast-export'

* js/fast-export:
  Add 'git fast-export', the sister of 'git fast-import'

16 years agoMerge branch 'js/rebase-i-rerere'
Junio C Hamano [Mon, 3 Dec 2007 07:01:02 +0000 (23:01 -0800)]
Merge branch 'js/rebase-i-rerere'

* js/rebase-i-rerere:
  rebase -i: give rerere a chance

16 years agoMerge branch 'js/pull-rebase'
Junio C Hamano [Mon, 3 Dec 2007 07:00:43 +0000 (23:00 -0800)]
Merge branch 'js/pull-rebase'

* js/pull-rebase:
  Teach 'git pull' about --rebase

16 years agoMerge branch 'wc/rebase-insn'
Junio C Hamano [Mon, 3 Dec 2007 07:00:28 +0000 (23:00 -0800)]
Merge branch 'wc/rebase-insn'

* wc/rebase-insn:
  Mention that git-rm can be an appropriate resolution as well as git-add.
  revert/cherry-pick: Allow overriding the help text by the calling Porcelain

16 years agoMerge branch 'jc/typebreak'
Junio C Hamano [Mon, 3 Dec 2007 06:59:28 +0000 (22:59 -0800)]
Merge branch 'jc/typebreak'

* jc/typebreak:
  Enable rewrite as well as rename detection in git-status
  rename: Break filepairs with different types.

16 years agoEnable rewrite as well as rename detection in git-status
Jeff King [Mon, 3 Dec 2007 06:58:37 +0000 (22:58 -0800)]
Enable rewrite as well as rename detection in git-status

Signed-off-by: Junio C Hamano <gitster@pobox.com>
16 years agoMerge branch 'jc/color'
Junio C Hamano [Mon, 3 Dec 2007 06:49:58 +0000 (22:49 -0800)]
Merge branch 'jc/color'

* jc/color:
  git-config --get-color: get configured color
  "color.diff = true" is not "always" anymore.

16 years agoAdd 'git fast-export', the sister of 'git fast-import'
Johannes Schindelin [Sun, 2 Dec 2007 14:14:13 +0000 (14:14 +0000)]
Add 'git fast-export', the sister of 'git fast-import'

This program dumps (parts of) a git repository in the format that
fast-import understands.

For clarity's sake, it does not use the 'inline' method of specifying
blobs in the commits, but builds the blobs before building the commits.

Since signed tags' signatures will not necessarily be valid (think
transformations after the export, or excluding revisions, changing
the history), there are 4 modes to handle them: abort (default),
ignore, warn and strip.  The latter just turns the tags into
unsigned ones.

Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
16 years agoMerge branch 'dc/gitweb'
Junio C Hamano [Mon, 3 Dec 2007 03:07:48 +0000 (19:07 -0800)]
Merge branch 'dc/gitweb'

* dc/gitweb:
  gitweb: the commitdiff is very commonly used, it's needed on search page, too

16 years agogitweb: Update and improve gitweb/README file
Jakub Narebski [Sun, 2 Dec 2007 23:22:02 +0000 (00:22 +0100)]
gitweb: Update and improve gitweb/README file

Update list of build configuration variables, add references
to gitweb/INSTALL, add description of runtime and per-repository
runtime configuration.

Signed-off-by: Jakub Narebski <jnareb@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
16 years agogit-stash: Display help message if git-stash is run with wrong sub-commands
Kevin Leung [Mon, 3 Dec 2007 02:34:05 +0000 (10:34 +0800)]
git-stash: Display help message if git-stash is run with wrong sub-commands

The current git-stash behaviour is very error prone to typos. For example,
if you typed "git-stash llist", git-stash would think that you wanted to
save to a stash named "llist", but in fact, you meant "git-stash list".

Signed-off-by: Kevin Leung <kevinlsk@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
16 years agogit-am: catch missing author date early.
Junio C Hamano [Sun, 2 Dec 2007 22:38:40 +0000 (14:38 -0800)]
git-am: catch missing author date early.

Even though commit-tree would default to the current time if the incoming
e-mail message somehow did not record the timestamp, it is safer to catch
the breakage sooner.

Signed-off-by: Junio C Hamano <gitster@pobox.com>
16 years agoMerge branch 'maint'
Junio C Hamano [Sun, 2 Dec 2007 19:00:45 +0000 (11:00 -0800)]
Merge branch 'maint'

* maint:
  t9600: test cvsimport from CVS working tree

16 years agot9600: test cvsimport from CVS working tree
Jeff King [Sun, 2 Dec 2007 17:22:19 +0000 (12:22 -0500)]
t9600: test cvsimport from CVS working tree

This test passes with v1.5.3.7, but not with v1.5.3.6.

Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
16 years agogitweb: the commitdiff is very commonly used, it's needed on search page, too
Denis Cheng [Mon, 26 Nov 2007 12:42:06 +0000 (20:42 +0800)]
gitweb: the commitdiff is very commonly used, it's needed on search page, too

Acked-by: Jakub Narebski <jnareb@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
16 years agorename: Break filepairs with different types.
Junio C Hamano [Sat, 1 Dec 2007 06:22:38 +0000 (22:22 -0800)]
rename: Break filepairs with different types.

When we consider if a path has been totally rewritten, we did not
touch changes from symlinks to files or vice versa.  But a change
that modifies even the type of a blob surely should count as a
complete rewrite.

While we are at it, modernise diffcore-break to be aware of gitlinks (we
do not want to touch them).

Signed-off-by: Junio C Hamano <gitster@pobox.com>
16 years agoUpdate draft release notes for 1.5.4
Junio C Hamano [Sun, 2 Dec 2007 07:45:31 +0000 (23:45 -0800)]
Update draft release notes for 1.5.4

Signed-off-by: Junio C Hamano <gitster@pobox.com>
16 years agoConsolidate command list to one.
Junio C Hamano [Sun, 2 Dec 2007 07:39:19 +0000 (23:39 -0800)]
Consolidate command list to one.

The categorized list of commands in git(7) and the list of common
commands in "git help" output were maintained separately, which was
insane.  This consolidates them to a single command-list.txt file.

Signed-off-by: Junio C Hamano <gitster@pobox.com>
16 years agoResurrect peek-remote
Junio C Hamano [Sun, 2 Dec 2007 07:38:35 +0000 (23:38 -0800)]
Resurrect peek-remote

8951d7c1f1ae38f34617b6c2490bf65e73e371f7 (Build in ls-remote) made
peek-remote as a synonym to ls-remote by enhancing the latter, but
at the same time actually _removed_ it, before we officially gave
removal notice.  This was bad.

Resurrect it for v1.5.4.

Signed-off-by: Junio C Hamano <gitster@pobox.com>
16 years agoMerge branch 'tt/help'
Junio C Hamano [Sun, 2 Dec 2007 04:05:49 +0000 (20:05 -0800)]
Merge branch 'tt/help'

* tt/help:
  Remove hint to use "git help -a"
  Make the list of common commands more exclusive

16 years agoMerge branch 'jc/move-gitk'
Junio C Hamano [Sat, 1 Dec 2007 21:58:57 +0000 (13:58 -0800)]
Merge branch 'jc/move-gitk'

* jc/move-gitk:
  Move gitk to its own subdirectory

16 years agoMerge branch 'jc/branch-contains'
Junio C Hamano [Sat, 1 Dec 2007 21:58:54 +0000 (13:58 -0800)]
Merge branch 'jc/branch-contains'

* jc/branch-contains:
  git-branch --contains: doc and test
  git-branch --contains=commit
  parse-options: Allow to hide options from the default usage.

16 years agoMerge branch 'cr/tag-options'
Junio C Hamano [Sat, 1 Dec 2007 21:58:46 +0000 (13:58 -0800)]
Merge branch 'cr/tag-options'

* cr/tag-options:
  git-tag: test that -s implies an annotated tag
  "git-tag -s" should create a signed annotated tag
  builtin-tag: accept and process multiple -m just like git-commit
  Make builtin-tag.c use parse_options.

16 years agoUpdate draft release notes for 1.5.4
Junio C Hamano [Sat, 1 Dec 2007 21:13:11 +0000 (13:13 -0800)]
Update draft release notes for 1.5.4

Signed-off-by: Junio C Hamano <gitster@pobox.com>
16 years agoMerge 1.5.3.7 in
Junio C Hamano [Sat, 1 Dec 2007 20:48:19 +0000 (12:48 -0800)]
Merge 1.5.3.7 in

Signed-off-by: Junio C Hamano <gitster@pobox.com>
16 years agoGIT 1.5.3.7 v1.5.3.7
Junio C Hamano [Sat, 1 Dec 2007 19:20:00 +0000 (11:20 -0800)]
GIT 1.5.3.7

Signed-off-by: Junio C Hamano <gitster@pobox.com>
16 years agoFix typo in t4008 test title
Junio C Hamano [Sat, 1 Dec 2007 19:06:41 +0000 (11:06 -0800)]
Fix typo in t4008 test title

Signed-off-by: Junio C Hamano <gitster@pobox.com>
16 years agogit-svn: Don't create a "master" branch every time rebase is run
Steven Grimm [Thu, 29 Nov 2007 19:54:39 +0000 (11:54 -0800)]
git-svn: Don't create a "master" branch every time rebase is run

If you run "git-svn rebase" while sitting on a topic branch, there is
no need to create a "master" branch if one didn't exist already. The
branch was created implicitly by the automatic checkout after fetching,
which in the case of rebase isn't actually necessary anyway.

Signed-off-by: Steven Grimm <koreth@midwinter.com>
Acked-by: Eric Wong <normalperson@yhbt.net>
16 years agogit-svn: add a show-externals command.
Vineet Kumar [Mon, 19 Nov 2007 22:56:15 +0000 (14:56 -0800)]
git-svn: add a show-externals command.

show-externals can be used by scripts to provide svn:externals-like
functionality.  For example, a script can list all of the externals and then
use check out the listed URLs at the appropriate paths, similar to what the svn
client does.  Said script (or perhaps git-svn itself, in the future) could
simply invoke svn export on the paths, or it could go one further, using
git-svn clone and even git-submodule together to better integrate externals
checkouts.

The implementation is shamelessly copied from show-ignores.  A more general
command to list user-specified properties is probably a better idea.

Signed-off-by: Vineet Kumar <vineet@doorstop.net>
Acked-by: Eric Wong <normalperson@yhbt.net>
16 years agogit-svn: Remove unnecessary Git::SVN::Util package
David D. Kilzer [Thu, 22 Nov 2007 19:18:00 +0000 (11:18 -0800)]
git-svn: Remove unnecessary Git::SVN::Util package

Digest::MD5 is loaded regardless of the package in which it's
declared, so move its 'use' statement and the md5sum() function
into the main package.

Signed-off-by: David D. Kilzer <ddkilzer@kilzer.net>
Acked-by: Eric Wong <normalperson@yhbt.net>
16 years agogit-svn: add support for pulling author from From: and Signed-off-by:
Andy Whitcroft [Thu, 22 Nov 2007 13:44:42 +0000 (13:44 +0000)]
git-svn: add support for pulling author from From: and Signed-off-by:

Add support for pulling the real author of a commit from the From:
and first Signed-off-by: fields of the SVN commit message.

Signed-off-by: Andy Whitcroft <apw@shadowen.org>
Acked-by: Eric Wong <normalperson@yhbt.net>
16 years agogit-svn now reads settings even if called in subdirectory
Gustaf Hendeby [Sat, 24 Nov 2007 13:47:56 +0000 (14:47 +0100)]
git-svn now reads settings even if called in subdirectory

Previously, git-svn first read the .git/config file for settings as if
current working directory was the repository top-directory, and after
that made sure to cd into top-directory.  The result was a silent
failur to read configuration settings.  This patch changes the order
these two things are done.

Signed-off-by: Gustaf Hendeby <hendeby@isy.liu.se>
Acked-by: Eric Wong <normalperson@yhbt.net>
16 years agoMerge branch 'maint'
Junio C Hamano [Sat, 1 Dec 2007 00:21:33 +0000 (16:21 -0800)]
Merge branch 'maint'

* maint:
  Replace the word 'update-cache' by 'update-index' everywhere
  cvsimport: fix usage of cvsimport.module
  t7003-filter-branch: Fix test of a failing --msg-filter.
  cvsimport: miscellaneous packed-ref fixes
  cvsimport: use rev-parse to support packed refs
  Add basic cvsimport tests

16 years agoFix a pathological case in git detecting proper renames
Linus Torvalds [Fri, 30 Nov 2007 00:41:09 +0000 (16:41 -0800)]
Fix a pathological case in git detecting proper renames

On Thu, 29 Nov 2007, Jeff King wrote:
>
> I think it will get worse, because you are simultaneously calculating
> all of the similarity scores bit by bit rather than doing a loop. Though
> perhaps you mean at the end you will end up with a list of src/dst pairs
> sorted by score, and you can loop over that.

Well, after thinking about this a bit, I think there's a solution that may
work well with the current thing too: instead of looping just *once* over
the list of rename pairs, loop twice - and simply refuse to do copies on
the first loop.

This trivial patch does that, and turns Kumar's test-case into a perfect
rename list.

It's not pretty, it's not smart, but it seems to work. There's something
to be said for keeping it simple and stupid.

And it should not be nearly as expensive as it may _look_. Yes, the loop
is "(i = 0; i < num_create * num_src; i++)", but the important part is
that the whole array is sorted by rename score, and we have a

if (mx[i].score < minimum_score)
break;

in it, so uthe loop actually would tend to terminate rather quickly.

Anyway, Kumar, the thing to take away from this is:

 - git really doesn't even *care* about the whole "rename detection"
   internally, and any commits you have done with renames are totally
   independent of the heuristics we then use to *show* the renames.

 - the rename detection really is for just two reasons: (a) keep humans
   happy, and keep the diffs small and (b) help automatic merging across
   renames. So getting renames right is certainly good, but it's more of a
   "politeness" issue than a "correctness" issue, although the merge
   portion of it does matter a lot sometimes.

 - the important thing here is that you can commit your changes and not
   worry about them being somehow "corrupted" by lack of rename detection,
   even if you commit them with a version of git that doesn't do rename
   detection the way you expected it. The rename detection is an
   "after-the-fact" thing, not something that actually gets saved in the
   repository, which is why we can change the heuristics _after_ seeing
   examples, and the examples magically correct themselves!

 - try out the two patches I've posted, and see if they work for you. They
   pass the test-suite, and the output for your example commit looks sane,
   but hey, if you have other test-cases, try them out.

Here's Kumar's pretty diffstat with both my patches:

 Makefile                                         |    6 +++---
 board/{cds => freescale}/common/cadmus.c         |    0
 board/{cds => freescale}/common/cadmus.h         |    0
 board/{cds => freescale}/common/eeprom.c         |    0
 board/{cds => freescale}/common/eeprom.h         |    0
 board/{cds => freescale}/common/ft_board.c       |    0
 board/{cds => freescale}/common/via.c            |    0
 board/{cds => freescale}/common/via.h            |    0
 board/{cds => freescale}/mpc8541cds/Makefile     |    0
 board/{cds => freescale}/mpc8541cds/config.mk    |    0
 board/{cds => freescale}/mpc8541cds/init.S       |    0
 board/{cds => freescale}/mpc8541cds/mpc8541cds.c |    0
 board/{cds => freescale}/mpc8541cds/u-boot.lds   |    4 ++--
 board/{cds => freescale}/mpc8548cds/Makefile     |    0
 board/{cds => freescale}/mpc8548cds/config.mk    |    0
 board/{cds => freescale}/mpc8548cds/init.S       |    0
 board/{cds => freescale}/mpc8548cds/mpc8548cds.c |    0
 board/{cds => freescale}/mpc8548cds/u-boot.lds   |    4 ++--
 board/{cds => freescale}/mpc8555cds/Makefile     |    0
 board/{cds => freescale}/mpc8555cds/config.mk    |    0
 board/{cds => freescale}/mpc8555cds/init.S       |    0
 board/{cds => freescale}/mpc8555cds/mpc8555cds.c |    0
 board/{cds => freescale}/mpc8555cds/u-boot.lds   |    4 ++--
 23 files changed, 9 insertions(+), 9 deletions(-)

and here it is before:

 Makefile                                           |    6 +-
 board/cds/mpc8548cds/Makefile                      |   60 -----
 board/cds/mpc8555cds/Makefile                      |   60 -----
 board/cds/mpc8555cds/init.S                        |  255 --------------------
 board/cds/mpc8555cds/u-boot.lds                    |  150 ------------
 board/{cds => freescale}/common/cadmus.c           |    0
 board/{cds => freescale}/common/cadmus.h           |    0
 board/{cds => freescale}/common/eeprom.c           |    0
 board/{cds => freescale}/common/eeprom.h           |    0
 board/{cds => freescale}/common/ft_board.c         |    0
 board/{cds => freescale}/common/via.c              |    0
 board/{cds => freescale}/common/via.h              |    0
 board/{cds => freescale}/mpc8541cds/Makefile       |    0
 board/{cds => freescale}/mpc8541cds/config.mk      |    0
 board/{cds => freescale}/mpc8541cds/init.S         |    0
 board/{cds => freescale}/mpc8541cds/mpc8541cds.c   |    0
 board/{cds => freescale}/mpc8541cds/u-boot.lds     |    4 +-
 .../mpc8541cds => freescale/mpc8548cds}/Makefile   |    0
 board/{cds => freescale}/mpc8548cds/config.mk      |    0
 board/{cds => freescale}/mpc8548cds/init.S         |    0
 board/{cds => freescale}/mpc8548cds/mpc8548cds.c   |    0
 board/{cds => freescale}/mpc8548cds/u-boot.lds     |    4 +-
 .../mpc8541cds => freescale/mpc8555cds}/Makefile   |    0
 board/{cds => freescale}/mpc8555cds/config.mk      |    0
 .../mpc8541cds => freescale/mpc8555cds}/init.S     |    0
 board/{cds => freescale}/mpc8555cds/mpc8555cds.c   |    0
 .../mpc8541cds => freescale/mpc8555cds}/u-boot.lds |    4 +-
 27 files changed, 9 insertions(+), 534 deletions(-)

so it certainly makes the diffs prettier.

Linus

Signed-off-by: Junio C Hamano <gitster@pobox.com>
16 years agoFix a pathological case in git detecting proper renames
Linus Torvalds [Thu, 29 Nov 2007 21:30:13 +0000 (13:30 -0800)]
Fix a pathological case in git detecting proper renames

Kumar Gala had a case in the u-boot archive with multiple renames of files
with identical contents, and git would turn those into multiple "copy"
operations of one of the sources, and just deleting the other sources.

This patch makes the git exact rename detection prefer to spread out the
renames over the multiple sources, rather than do multiple copies of one
source.

NOTE! The changes are a bit larger than required, because I also renamed
the variables named "one" and "two" to "target" and "source" respectively.
That makes the logic easier to follow, especially as the "one" was
illogically the target and not the soruce, for purely historical reasons
(this piece of code used to traverse over sources and targets in the wrong
order, and when we fixed that, we didn't fix the names back then. So I
fixed them now).

The important part of this change is just the trivial score calculations
for when files have identical contents:

/* Give higher scores to sources that haven't been used already */
score = !source->rename_used;
score += basename_same(source, target);

and when we have multiple choices we'll now pick the choice that gets the
best rename score, rather than only looking at whether the basename
matched.

It's worth noting a few gotchas:

 - this scoring is currently only done for the "exact match" case.

   In particular, in Kumar's example, even after this patch, the inexact
   match case is still done as a copy+delete rather than as two renames:

 delete mode 100644 board/cds/mpc8555cds/u-boot.lds
 copy board/{cds => freescale}/mpc8541cds/u-boot.lds (97%)
 rename board/{cds/mpc8541cds => freescale/mpc8555cds}/u-boot.lds (97%)

   because apparently the "cds/mpc8541cds/u-boot.lds" copy looked
   a bit more similar to both end results. That said, I *suspect* we just
   have the exact same issue there - the similarity analysis just gave
   identical (or at least very _close_ to identical) similarity points,
   and we do not have any logic to prefer multiple renames over a
   copy/delete there.

   That is a separate patch.

 - When you have identical contents and identical basenames, the actual
   entry that is chosen is still picked fairly "at random" for the first
   one (but the subsequent ones will prefer entries that haven't already
   been used).

   It's not actually really random, in that it actually depends on the
   relative alphabetical order of the files (which in turn will have
   impacted the order that the entries got hashed!), so it gives
   consistent results that can be explained. But I wanted to point it out
   as an issue for when anybody actually does cross-renames.

   In Kumar's case the choice is the right one (and for a single normal
   directory rename it should always be, since the relative alphabetical
   sorting of the files will be identical), and we now get:

 rename board/{cds => freescale}/mpc8541cds/init.S (100%)
 rename board/{cds => freescale}/mpc8548cds/init.S (100%)

   which is the "expected" answer. However, it might still be better to
   change the pedantic "exact same basename" on/off choice into a more
   graduated "how similar are the pathnames" scoring situation, in order
   to be more likely to get the exact rename choice that people *expect*
   to see, rather than other alternatives that may *technically* be
   equally good, but are surprising to a human.

It's also unclear whether we should consider "basenames are equal" or
"have already used this as a source" to be more important. This gives them
equal weight, but I suspect we might want to just multiple the "basenames
are equal" weight by two, or something, to prefer equal basenames even if
that causes a copy/delete pair. I dunno.

Anyway, what I'm just saying in a really long-winded manner is that I
think this is right as-is, but it's not the complete solution, and it may
want some further tweaking in the future.

Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
16 years agoAdd "--expire <time>" option to 'git prune'
Johannes Schindelin [Thu, 29 Nov 2007 20:59:55 +0000 (20:59 +0000)]
Add "--expire <time>" option to 'git prune'

Earlier, 'git prune' would prune all loose unreachable objects.
This could be quite dangerous, as the objects could be used in
an ongoing operation.

This patch adds a mode to expire only loose, unreachable objects
which are older than a certain time.  For example, by

git prune --expire 14.days

you can prune only those objects which are loose, unreachable
and older than 14 days (and thus probably outdated).

The implementation uses st.st_mtime rather than st.st_ctime,
because it can be tested better, using 'touch -d <time>' (and
omitting the test when the platform does not support that
command line switch).

Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
16 years agoReplace the word 'update-cache' by 'update-index' everywhere
Johannes Schindelin [Fri, 30 Nov 2007 11:35:23 +0000 (11:35 +0000)]
Replace the word 'update-cache' by 'update-index' everywhere

Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
16 years agocvsimport: fix usage of cvsimport.module
Jeff King [Fri, 30 Nov 2007 22:22:12 +0000 (17:22 -0500)]
cvsimport: fix usage of cvsimport.module

There were two problems:

  1. We only look at the config variable if there is no module
     given on the command line. We checked this by comparing
     @ARGV == 0. However, at the time of the comparison, we
     have not yet parsed the dashed options, meaning that
     "git cvsimport" would read the variable but "git
     cvsimport -a" would not. This is fixed by simply moving
     the check after the call to getopt.

  2. If the config variable did not exist, we were adding an
     empty string to @ARGV. The rest of the script, rather
     than barfing for insufficient input, would then try to
     import the module '', leading to rather confusing error
     messages. Based on patch from Emanuele Giaquinta.

Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
16 years agoreceive-pack: allow deletion of corrupt refs
Johannes Schindelin [Thu, 29 Nov 2007 01:02:53 +0000 (01:02 +0000)]
receive-pack: allow deletion of corrupt refs

Occasionally, in some setups (*cough* forks on repo.or.cz *cough*) some
refs go stale, e.g. when the forkee rebased and lost some objects needed
by the fork.  The quick & dirty way to deal with those refs is to delete
them and push them again.

However, git-push first would first fetch the current commit name for the
ref, would receive a null sha1 since the ref does not point to a valid
object, then tell receive-pack that it should delete the ref with this
commit name.  delete_ref() would be subsequently be called, and check that
resolve_ref() (which does _not_ check for validity of the object) returns
the same commit name.  Which would fail.

The proper fix is to avoid corrupting repositories, but in the meantime
this is a good fix in any case.

Incidentally, some instances of "cd .." in the test cases were fixed, so
that subsequent test cases run in t/trash/ irrespective of the outcome of
the previous test cases.

Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
16 years agoMerge branch 'jk/maint-cvsimport-fix' into maint
Junio C Hamano [Fri, 30 Nov 2007 22:22:54 +0000 (14:22 -0800)]
Merge branch 'jk/maint-cvsimport-fix' into maint

* jk/maint-cvsimport-fix:
  cvsimport: miscellaneous packed-ref fixes
  cvsimport: use rev-parse to support packed refs
  Add basic cvsimport tests

16 years agot7003-filter-branch: Fix test of a failing --msg-filter.
Johannes Sixt [Tue, 27 Nov 2007 12:10:19 +0000 (13:10 +0100)]
t7003-filter-branch: Fix test of a failing --msg-filter.

The test passed for the wrong reason: If the script given to --msg-filter
fails, it is expected that git-filter-branch aborts. But the test forgot
to tell the branch name to rewrite, and so git-filter-branch failed due to
incorrect usage.

Signed-off-by: Johannes Sixt <johannes.sixt@telecom.at>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
16 years agoMention that git-rm can be an appropriate resolution as well as git-add.
David Symonds [Fri, 23 Nov 2007 13:38:50 +0000 (00:38 +1100)]
Mention that git-rm can be an appropriate resolution as well as git-add.

Especially when using git-cherry-pick, removing files that are unmerged can be
a logical action. This patch merely changes the informative text to be less
confusing.

Signed-off-by: David Symonds <dsymonds@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
16 years agorevert/cherry-pick: Allow overriding the help text by the calling Porcelain
Wincent Colaiuta [Wed, 28 Nov 2007 08:06:36 +0000 (00:06 -0800)]
revert/cherry-pick: Allow overriding the help text by the calling Porcelain

A Porcelain command that uses cherry-pick or revert may make a commit
out of resolved index itself, in which case telling the user to commit
the result is not appropriate at all.  This allows GIT_CHERRY_PICK_HELP
environment variable to be set by the calling Porcelain in order to
override the built-in help text.

[jc: this is heavily modified from the original but should be equivalent
 in spirit]

Signed-off-by: Junio C Hamano <gitster@pobox.com>
16 years agoAdd config_int() method to the Git perl module
Jakub Narebski [Fri, 23 Nov 2007 18:04:52 +0000 (19:04 +0100)]
Add config_int() method to the Git perl module

Integer variables can have optional 'k', 'm' or 'g' suffix.
config_int() method will return simple decimal number, taking
care of those suffixes.

Signed-off-by: Jakub Narebski <jnareb@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
16 years agoError out when user doesn't have access permission to the repository
André Goddard Rosa [Thu, 22 Nov 2007 22:22:23 +0000 (20:22 -0200)]
Error out when user doesn't have access permission to the repository

Signed-off-by: André Goddard Rosa <andre.goddard@gmail.com>
16 years agoper-directory-exclude: lazily read .gitignore files
Junio C Hamano [Thu, 29 Nov 2007 10:17:44 +0000 (02:17 -0800)]
per-directory-exclude: lazily read .gitignore files

Operations that walk directories or trees, which potentially need to
consult the .gitignore files, used to always try to open the .gitignore
file every time they entered a new directory, even when they ended up
not needing to call excluded() function to see if a path in the
directory is ignored.  This was done by push/pop exclude_per_directory()
functions that managed the data in a stack.

This changes the directory walking API to remove the need to call these
two functions.  Instead, the directory walk data structure caches the
data used by excluded() function the last time, and lazily reuses it as
much as possible.  Among the data the last check used, the ones from
deeper directories that the path we are checking is outside are
discarded, data from the common leading directories are reused, and then
the directories between the common directory and the directory the path
being checked is in are checked for .gitignore file.  This is very
similar to the way gitattributes are handled.

This API change also fixes "ls-files -c -i", which called excluded()
without setting up the gitignore data via the old push/pop functions.

Signed-off-by: Junio C Hamano <gitster@pobox.com>
16 years agodir.c: minor clean-up
Junio C Hamano [Thu, 29 Nov 2007 09:11:46 +0000 (01:11 -0800)]
dir.c: minor clean-up

Replace handcrafted reallocation with ALLOC_GROW().
Reindent "file_exists()" helper function.

Signed-off-by: Junio C Hamano <gitster@pobox.com>
16 years agorebase -i: give rerere a chance
Johannes Schindelin [Thu, 22 Nov 2007 11:18:10 +0000 (11:18 +0000)]
rebase -i: give rerere a chance

Like non-interactive rebase, interactive mode now calls rerere when
appropriate.

Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
16 years agogit-config --get-color: get configured color
Junio C Hamano [Wed, 28 Nov 2007 06:41:05 +0000 (22:41 -0800)]
git-config --get-color: get configured color

This new option allows scripts to grab color setting from the user
configuration, translated to ANSI color escape sequence.

Signed-off-by: Junio C Hamano <gitster@pobox.com>
16 years ago"color.diff = true" is not "always" anymore.
Junio C Hamano [Mon, 26 Nov 2007 22:30:28 +0000 (14:30 -0800)]
"color.diff = true" is not "always" anymore.

Too many people got burned by setting color.diff and color.status to
true when they really should have set it to "auto".

This makes only "always" to do the unconditional colorization, and
change the meaning of "true" to the same as "auto": colorize only when
we are talking to a terminal.

Signed-off-by: Junio C Hamano <gitster@pobox.com>
16 years agoTeach 'git pull' about --rebase
Johannes Schindelin [Wed, 28 Nov 2007 13:11:07 +0000 (13:11 +0000)]
Teach 'git pull' about --rebase

When calling 'git pull' with the '--rebase' option, it performs a
fetch + rebase instead of a fetch + merge.

This behavior is more desirable than fetch + pull when a topic branch
is ready to be submitted and needs to be update.

fetch + rebase might also be considered a better workflow with shared
repositories in any case, or for contributors to a centrally managed
repository, such as WINE's.

As a convenience, you can set the default behavior for a branch by
defining the config variable branch.<name>.rebase, which is
interpreted as a bool.  This setting can be overridden on the command
line by --rebase and --no-rebase.

Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
16 years agoReplace instances of export VAR=VAL with VAR=VAL; export VAR
Johannes Schindelin [Wed, 28 Nov 2007 15:56:11 +0000 (15:56 +0000)]
Replace instances of export VAR=VAL with VAR=VAL; export VAR

It might be POSIX, but there are shells that do not like the
expression 'export VAR=VAL'.  To be on the safe side, rewrite them
into 'VAR=VAL' and 'export VAR'.

Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
16 years agoMerge branch 'maint'
Junio C Hamano [Thu, 29 Nov 2007 01:06:57 +0000 (17:06 -0800)]
Merge branch 'maint'

* maint:
  scripts: do not get confused with HEAD in work tree
  Improve description of git-branch -d and -D in man page.

16 years agoscripts: do not get confused with HEAD in work tree
Junio C Hamano [Thu, 29 Nov 2007 00:15:04 +0000 (16:15 -0800)]
scripts: do not get confused with HEAD in work tree

When you have a file called HEAD in your work tree, many commands that
our scripts feed "HEAD" to would complain about the rev vs path
ambiguity.  A solution is to form command line more carefully by
appending -- to them, which makes it clear that we mean HEAD rev not
HEAD file.

This patch would apply to maint.

Signed-off-by: Junio C Hamano <gitster@pobox.com>
16 years agobash completion: add diff options
Johannes Schindelin [Fri, 23 Nov 2007 01:11:35 +0000 (01:11 +0000)]
bash completion: add diff options

I use "git diff" (the porcelain) really often, and am almost as often
annoyed that the completions do not know how to complete something simple
as --cached.  Now they do.

Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
16 years agogit checkout's reflog: even when detaching the HEAD, say from where
Johannes Schindelin [Fri, 23 Nov 2007 00:20:35 +0000 (00:20 +0000)]
git checkout's reflog: even when detaching the HEAD, say from where

When checking out another ref, the reflogs already record from which
branch you switched.  Do that also when switching to a detached HEAD.

While at it, record also when coming _from_ a detached HEAD.

Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
16 years agosha1_file.c: Fix size_t related printf format warnings
Steffen Prohaska [Wed, 21 Nov 2007 20:27:19 +0000 (21:27 +0100)]
sha1_file.c: Fix size_t related printf format warnings

The old way of fixing warnings did not succeed on MinGW.  MinGW
does not support C99 printf format strings for size_t [1].  But
gcc on MinGW issues warnings if C99 printf format is not used.
Hence, the old stragegy to avoid warnings fails.

[1] http://www.mingw.org/MinGWiki/index.php/C99

This commits passes arguments of type size_t through a tiny
helper functions that casts to the type expected by the format
string.

Signed-off-by: Steffen Prohaska <prohaska@zib.de>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
16 years agoRevert "t5516: test update of local refs on push"
Jeff King [Wed, 21 Nov 2007 07:19:34 +0000 (02:19 -0500)]
Revert "t5516: test update of local refs on push"

This reverts commit 09fba7a59d38d1cafaf33eadaf1d409c4113b30c.

These tests are superseded by the ones in t5404 (added in
6fa92bf3 and 8736a848), which are more extensive and better
organized.

Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
16 years agoImprove description of git-branch -d and -D in man page.
Jan Hudec [Sat, 17 Nov 2007 19:51:44 +0000 (20:51 +0100)]
Improve description of git-branch -d and -D in man page.

Some users expect that deleting a remote-tracking branch would prevent
fetch from creating it again, so be explcit about that it's not the case.
Also be a little more explicit about what fully merged means.

Signed-off-by: Jan Hudec <bulb@ucw.cz>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
16 years agofilter-branch: fix dirty way to provide the helpers to commit filters
Johannes Schindelin [Wed, 28 Nov 2007 15:50:38 +0000 (15:50 +0000)]
filter-branch: fix dirty way to provide the helpers to commit filters

The helper functions 'map' and 'skip_commit' were provided to commit
filters by sourcing filter-branch itself.  This was done with a certain
environment variable set to indicate that only the functions should be
defined, and the script should return then.

This was really hacky, and it did not work all that well, since the
full path to git-filter-branch was not known at all times.

Avoid that by putting the functions into a variable, and eval'ing
that variable.  The commit filter gets these functions by prepending
the variable to the specified commands.

Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
16 years agocvsimport: miscellaneous packed-ref fixes
Jeff King [Wed, 28 Nov 2007 18:56:28 +0000 (13:56 -0500)]
cvsimport: miscellaneous packed-ref fixes

These were found with a grep for '$git_dir'; they all
replace a direct access of "$git_dir/refs/..." with a call
to git-rev-parse or git-update-ref.

Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
16 years agocvsimport: use rev-parse to support packed refs
Jeff King [Wed, 28 Nov 2007 18:56:11 +0000 (13:56 -0500)]
cvsimport: use rev-parse to support packed refs

Previously, if refs were packed, git-cvsimport would assume
that particular refs did not exist. This could lead to, for
example, overwriting previous 'origin' commits that were
packed.

Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
16 years agoAdd basic cvsimport tests
Jeff King [Wed, 28 Nov 2007 18:55:46 +0000 (13:55 -0500)]
Add basic cvsimport tests

We weren't even testing basic things before, so let's at
least try importing and updating a trivial repository, which
will catch total breakage.

Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
16 years agogit-stash: do not get fooled with "color.diff = true"
Pascal Obry [Mon, 26 Nov 2007 22:04:28 +0000 (23:04 +0100)]
git-stash: do not get fooled with "color.diff = true"

When colors are set to "true" on the repository, the git log output
will contain control characters to set/reset the colors, even when
the output is to a pipe. This makes list_stash() fail as the
downstream sed does not see what it is expecting.

Signed-off-by: Junio C Hamano <gitster@pobox.com>
17 years agoUse is_absolute_path() in diff-lib.c, lockfile.c, setup.c, trace.c
Steffen Prohaska [Sun, 25 Nov 2007 22:29:03 +0000 (23:29 +0100)]
Use is_absolute_path() in diff-lib.c, lockfile.c, setup.c, trace.c

Using the helper function to test for absolute paths makes porting easier.

Signed-off-by: Steffen Prohaska <prohaska@zib.de>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
17 years agoFix typo in draft 1.5.4 release notes
Wincent Colaiuta [Mon, 26 Nov 2007 08:34:48 +0000 (09:34 +0100)]
Fix typo in draft 1.5.4 release notes

Signed-off-by: Wincent Colaiuta <win@wincent.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
17 years agogit-tag: test that -s implies an annotated tag
Jeff King [Mon, 26 Nov 2007 04:50:58 +0000 (23:50 -0500)]
git-tag: test that -s implies an annotated tag

This detects a regression introduced while moving git-tag to a C
builtin.

Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
17 years agoMerge branch 'maint'
Junio C Hamano [Mon, 26 Nov 2007 03:10:01 +0000 (19:10 -0800)]
Merge branch 'maint'

* maint:
  user-manual: recovering from corruption
  user-manual: clarify language about "modifying" old commits
  user-manual: failed push to public repository
  user-manual: define "branch" and "working tree" at start
  git-checkout: describe detached head correctly

17 years agoMerge branch 'maint' of git://linux-nfs.org/~bfields/git into maint
Junio C Hamano [Mon, 26 Nov 2007 03:04:27 +0000 (19:04 -0800)]
Merge branch 'maint' of git://linux-nfs.org/~bfields/git into maint

* 'maint' of git://linux-nfs.org/~bfields/git:
  user-manual: recovering from corruption
  user-manual: clarify language about "modifying" old commits
  user-manual: failed push to public repository
  user-manual: define "branch" and "working tree" at start

17 years agouser-manual: recovering from corruption
J. Bruce Fields [Sun, 4 Mar 2007 03:53:37 +0000 (22:53 -0500)]
user-manual: recovering from corruption

Some instructions on dealing with corruption of the object database.

Most of this text is from an example by Linus, identified by Nicolas
Pitre <nico@cam.org> with a little further editing by me.

Signed-off-by: "J. Bruce Fields" <bfields@citi.umich.edu>
17 years agoPrint the real filename that we failed to open.
André Goddard Rosa [Wed, 21 Nov 2007 20:59:14 +0000 (18:59 -0200)]
Print the real filename that we failed to open.

When we fail to open a temporary file to be renamed to something else,
we reported the final filename, not the temporary file we failed to
open.

Signed-off-by: André Goddard Rosa <andre.goddard@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
17 years agouser-manual: clarify language about "modifying" old commits
J. Bruce Fields [Mon, 26 Nov 2007 00:01:57 +0000 (19:01 -0500)]
user-manual: clarify language about "modifying" old commits

It's important to remember that git doesn't really allowing "editing" or
"modifying" commits, only replacing them by new commits.  Redo some of
the language to make this clearer.

Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
17 years agorevert/cherry-pick: do not mention the original ref
Junio C Hamano [Sun, 25 Nov 2007 23:15:48 +0000 (15:15 -0800)]
revert/cherry-pick: do not mention the original ref

When you cherry-pick or revert a commit, naming it with an annotated
tag, we added a comment, attempting to repeat what we got from the end
user, to the message.

But this was inconsistent.  When we got "cherry-pick branch", we
recorded the object name (40-letter SHA-1) without saying anything like
"original was 'branch'".  There was no need to.  Also recent rewrite to
use parse-options made it impossible to parrot the original command line
without "unparsing".

This removes the code that implements the misguided "we dereferenced the
tag so record that in the commit message" behaviour.

Signed-off-by: Junio C Hamano <gitster@pobox.com>
17 years ago"git-tag -s" should create a signed annotated tag
Junio C Hamano [Sun, 25 Nov 2007 23:21:42 +0000 (15:21 -0800)]
"git-tag -s" should create a signed annotated tag

The earlier patch 396865859918e9c7bf8ce74aae137c57da134610 (Make
builtin-tag.c use parse_options.) broke "git-tag -s".

Signed-off-by: Junio C Hamano <gitster@pobox.com>
17 years agouser-manual: failed push to public repository
J. Bruce Fields [Sun, 25 Nov 2007 22:54:19 +0000 (17:54 -0500)]
user-manual: failed push to public repository

More details on the case of a failed push to a public (non-shared)
repository.

Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
17 years agouser-manual: define "branch" and "working tree" at start
J. Bruce Fields [Sun, 25 Nov 2007 18:53:37 +0000 (13:53 -0500)]
user-manual: define "branch" and "working tree" at start

Some explanation here might help.

Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
17 years agogit-checkout: describe detached head correctly
Junio C Hamano [Sun, 25 Nov 2007 18:34:27 +0000 (10:34 -0800)]
git-checkout: describe detached head correctly

When you have a file called HEAD in the work tree, the code to report
where the HEAD is at when "git checkout $commit^0" is done triggered
unnecessary ambiguity checking.

Explicitly mark the command line with "--" and make it clear that we are
talking about a revision.

Signed-off-by: Junio C Hamano <gitster@pobox.com>
17 years agoUpdate draft release notes for 1.5.4
Junio C Hamano [Sun, 25 Nov 2007 01:42:36 +0000 (17:42 -0800)]
Update draft release notes for 1.5.4

Signed-off-by: Junio C Hamano <gitster@pobox.com>