Code

git.git
15 years agoMerge branch 'da/mergetool-lib'
Junio C Hamano [Sat, 23 May 2009 08:41:51 +0000 (01:41 -0700)]
Merge branch 'da/mergetool-lib'

* da/mergetool-lib:
  mergetool--lib: specialize diff options for emerge and ecmerge

15 years agoMerge branch 'cc/bisect' (early part)
Junio C Hamano [Sat, 23 May 2009 08:41:27 +0000 (01:41 -0700)]
Merge branch 'cc/bisect' (early part)

* 'cc/bisect' (early part):
  bisect: make "git bisect" use new "--next-all" bisect-helper function
  bisect: add "check_good_are_ancestors_of_bad" function
  bisect: implement the "check_merge_bases" function
  bisect: automatically sort sha1_array if needed when looking it up
  bisect: make skipped array functions more generic
  bisect: remove too much function nesting
  bisect: use new "struct argv_array" to prepare argv for "setup_revisions"
  bisect: store good revisions in a "sha1_array"
  bisect: implement "rev_argv_push" to fill an argv with revs
  bisect: use "sha1_array" to store skipped revisions
  am: simplify "sq" function by using "git rev-parse --sq-quote"
  bisect: use "git rev-parse --sq-quote" instead of a custom "sq" function
  rev-parse: add --sq-quote to shell quote arguments
  rev-list: remove stringed output flag from "show_bisect_vars"
  bisect--helper: remove "--next-vars" option as it is now useless
  bisect: use "git bisect--helper --next-exit" in "git-bisect.sh"
  bisect--helper: add "--next-exit" to output bisect results
  bisect: move common bisect functionality to "bisect_common"
  rev-list: refactor printing bisect vars
  rev-list: make "estimate_bisect_steps" non static

15 years agoMerge branch 'lt/maint-diff-reduce-lstat'
Junio C Hamano [Sat, 23 May 2009 08:40:33 +0000 (01:40 -0700)]
Merge branch 'lt/maint-diff-reduce-lstat'

* lt/maint-diff-reduce-lstat:
  Teach 'git checkout' to preload the index contents
  Avoid unnecessary 'lstat()' calls in 'get_stat_data()'

15 years agoMerge branch 'jm/format-patch-no-auto-n-when-k-is-given'
Junio C Hamano [Sat, 23 May 2009 08:40:22 +0000 (01:40 -0700)]
Merge branch 'jm/format-patch-no-auto-n-when-k-is-given'

* jm/format-patch-no-auto-n-when-k-is-given:
  format-patch let -k override a config-specified format.numbered

15 years agoMerge branch 'do/maint-merge-recursive-fix'
Junio C Hamano [Sat, 23 May 2009 08:39:50 +0000 (01:39 -0700)]
Merge branch 'do/maint-merge-recursive-fix'

* do/maint-merge-recursive-fix:
  merge-recursive: never leave index unmerged while recursing

15 years agoMerge branch 'rr/forbid-bs-in-ref'
Junio C Hamano [Sat, 23 May 2009 08:39:45 +0000 (01:39 -0700)]
Merge branch 'rr/forbid-bs-in-ref'

* rr/forbid-bs-in-ref:
  Disallow '\' in ref names

15 years agoMerge branch 'hv/sample-update'
Junio C Hamano [Sat, 23 May 2009 08:39:15 +0000 (01:39 -0700)]
Merge branch 'hv/sample-update'

* hv/sample-update:
  Extend sample update hook, disable modifying of existing tags

15 years agoMerge branch 'jk/maint-1.6.0-trace-argv'
Junio C Hamano [Sat, 23 May 2009 08:39:08 +0000 (01:39 -0700)]
Merge branch 'jk/maint-1.6.0-trace-argv'

* jk/maint-1.6.0-trace-argv:
  fix GIT_TRACE segfault with shell-quoted aliases

Conflicts:
alias.c

15 years agoMerge branch 'jk/no-no-no-empty-directory'
Junio C Hamano [Sat, 23 May 2009 08:38:35 +0000 (01:38 -0700)]
Merge branch 'jk/no-no-no-empty-directory'

* jk/no-no-no-empty-directory:
  parseopt: add OPT_NEGBIT

15 years agoMerge branch 'rs/grep-parseopt'
Junio C Hamano [Sat, 23 May 2009 08:38:32 +0000 (01:38 -0700)]
Merge branch 'rs/grep-parseopt'

* rs/grep-parseopt:
  grep: make callback functions static
  grep: use parseopt
  grep: remove global variable builtin_grep
  parseopt: add PARSE_OPT_NODASH
  parseopt: add OPT_NUMBER_CALLBACK
  parseopt: add OPT_NEGBIT

15 years agoMerge branch 'fl/git-pm'
Junio C Hamano [Sat, 23 May 2009 08:38:28 +0000 (01:38 -0700)]
Merge branch 'fl/git-pm'

* fl/git-pm:
  Git.pm: Always set Repository to absolute path if autodetecting
  Git.pm: Set GIT_WORK_TREE if we set GIT_DIR

15 years agoMerge branch 'mt/submodule-reference'
Junio C Hamano [Sat, 23 May 2009 08:38:24 +0000 (01:38 -0700)]
Merge branch 'mt/submodule-reference'

* mt/submodule-reference:
  Add --reference option to git submodule.

15 years agogitweb: Sanitize title attribute in format_subject_html
Jakub Narebski [Fri, 22 May 2009 15:35:46 +0000 (17:35 +0200)]
gitweb: Sanitize title attribute in format_subject_html

Replace control characters with question mark '?' (like in
chop_and_esc_str).

A little background: some web browsers turn on strict (and
unforgiving) XML validating mode for XHTML documents served using
application/xhtml+xml content type.  This means among others that
control characters are forbidden to appear in gitweb output.

CGI.pm does by default slight escaping (using simple_escape subroutine
from CGI::Util) of all _attribute_ values (depending on the value of
autoEscape, by default on).  This escaping, at least in CGI.pm version
3.10 (most current version at CPAN is 3.43), is minimal: only '"',
'&', '<' and '>' are escaped using named HTML entity references
(&quot;, &amp;, &lt; and &gt; respectively).  But simple_escape does
not do escaping of control characters such as ^X which are invalid in
XHTML (in strict mode).

If by some accident commit message do contain some control character
in first 50 characters (more or less) of first line of commit message,
and this line is longer than 50 characters (so gitweb shortens it for
display), then gitweb would put this control character in title
attribute (and CGI.pm would not remove them).  The tag _contents_ is
safe because it is escaped using esc_html() explicitly, and it
replaces control characters by their printable representation.

While at it: chop_and_escape_str doesn't need capturing group.

Noticed-by: Paul Gortmaker <paul.gortmaker@windriver.com>
Signed-off-by: Jakub Narebski <jnareb@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
15 years agoTerminate argv with NULL before calling setup_revisions()
Nguyễn Thái Ngọc Duy [Thu, 21 May 2009 09:32:44 +0000 (19:32 +1000)]
Terminate argv with NULL before calling setup_revisions()

It is convention that argv should be terminated with NULL, even if
argc is used to specify the size of argv. setup_revisions() requires
this and may segfault otherwise.

This patch makes sure that all argv (that I can find) is NULL terminated.

Signed-off-by: Nguyễn Thái Ngọc Duy <pclouds@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
15 years agodoc/git-rebase.txt: remove mention of multiple strategies
Nguyễn Thái Ngọc Duy [Thu, 21 May 2009 09:47:07 +0000 (19:47 +1000)]
doc/git-rebase.txt: remove mention of multiple strategies

git-rebase.sh does not seem to support this.

Signed-off-by: Nguyễn Thái Ngọc Duy <pclouds@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
15 years agoMerge git://git.bogomips.org/git-svn
Junio C Hamano [Thu, 21 May 2009 14:28:07 +0000 (07:28 -0700)]
Merge git://git.bogomips.org/git-svn

* git://git.bogomips.org/git-svn:
  git-svn: add --authors-prog option
  git-svn: Set svn.authorsfile if it is passed to git svn clone
  git-svn: Correctly report max revision when following deleted paths
  git-svn: Fix for svn paths removed > log-window-size revisions ago
  git-svn testsuite: use standard configuration for Subversion tools

15 years agogit-send-email: Handle quotes when parsing .mailrc files
Eric W. Biederman [Thu, 21 May 2009 02:45:53 +0000 (19:45 -0700)]
git-send-email: Handle quotes when parsing .mailrc files

It is legal and not uncommon to use quotes in a .mailrc file so
you can include a persons fullname as well as their email alias.
Handle this by using quotewords instead of split when parsing
.mailrc files.

Signed-off-by: Eric W. Biederman <ebiederm@xmission.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
15 years agogit-svn: add --authors-prog option
Mark Lodato [Fri, 15 May 2009 01:27:15 +0000 (21:27 -0400)]
git-svn: add --authors-prog option

Add a new option, --authors-prog, to git-svn that allows a more flexible
alternative (or supplement) to --authors-file.  This allows more
advanced username operations than the authors file will allow.  For
example, one may look up Subversion users via LDAP, or may generate the
name and email address from the Subversion username.

Notes:

* If both --authors-name and --authors-prog are given, the former is
  tried first, falling back to the later.

* The program is called once per unique SVN username, and the result is
  cached.

* The command-line argument must be the path to a program, not a generic
  shell command line.  The absolute path to this program is taken at
  startup since the git-svn script changes directory during operation.

* The option is not enabled for `git svn log'.

[ew: fixed case where neither --authors-(name|prog) were defined]
Signed-off-by: Mark Lodato <lodatom@gmail.com>
Acked-by: Eric Wong <normalperson@yhbt.net>
15 years agogit-svn: Set svn.authorsfile if it is passed to git svn clone
Alex Vandiver [Wed, 6 May 2009 20:19:45 +0000 (16:19 -0400)]
git-svn: Set svn.authorsfile if it is passed to git svn clone

Signed-off-by: Alex Vandiver <alexmv@mit.edu>
Acked-by: Eric Wong <normalperson@yhbt.net>
15 years agogit-svn: Correctly report max revision when following deleted paths
Alex Vandiver [Wed, 6 May 2009 20:18:53 +0000 (16:18 -0400)]
git-svn: Correctly report max revision when following deleted paths

Report the maximum found revision in the range, instead of the minimum
changed revision.

Signed-off-by: Alex Vandiver <alexmv@mit.edu>
Acked-by: Eric Wong <normalperson@yhbt.net>
15 years agogit-svn: Fix for svn paths removed > log-window-size revisions ago
Alex Vandiver [Wed, 6 May 2009 20:18:52 +0000 (16:18 -0400)]
git-svn: Fix for svn paths removed > log-window-size revisions ago

Instead of trying to find the end of the commit history only in the
last window, track if we have seen commits yet, and use that to judge
if we need to backtrack and look for a tail.  Otherwise, conversion
can silently lose up to 100 revisions of a branch if it was deleted
>100 revisions ago.

Signed-off-by: Alex Vandiver <alexmv@mit.edu>
Acked-by: Eric Wong <normalperson@yhbt.net>
15 years agogit-svn testsuite: use standard configuration for Subversion tools
Eygene Ryabinkin [Fri, 8 May 2009 08:06:16 +0000 (12:06 +0400)]
git-svn testsuite: use standard configuration for Subversion tools

I have tweaked configuration in my ~/.subversion directory, namely I am
running auto-properties and automatically adding '$Id$' expansion to
every file.  This choke the last test named 'proplist' from
t9101-git-svn-props.sh, because one more property, svn:keywords is
automatically added.

I had just wrapped svn invocation with the svn_cmd that specifies empty
directory via --config-dir argument.  Since the latter is the global
option, it should be recognized by all svn subcommands, so no
regressions will be introduced.

Now svn_cmd is used everywhere, not just in the failed test module: this
should guard us from the future clashes with user-defined configuration
tweaks.

Signed-off-by: Eygene Ryabinkin <rea-git@codelabs.ru>
Acked-by: Eric Wong <normalperson@yhbt.net>
15 years agoMerge branch 'maint'
Junio C Hamano [Thu, 21 May 2009 01:59:09 +0000 (18:59 -0700)]
Merge branch 'maint'

* maint:
  grep: fix word-regexp colouring
  completion: use git rev-parse to detect bare repos
  Cope better with a _lot_ of packs
  for-each-ref: fix segfault in copy_email

15 years agogrep: fix word-regexp colouring
René Scharfe [Wed, 20 May 2009 21:31:53 +0000 (23:31 +0200)]
grep: fix word-regexp colouring

As noticed by Dmitry Gryazin: When a pattern is found but it doesn't
start and end at word boundaries, bol is forwarded to after the match and
the pattern is searched again.  When a pattern is finally found between
word boundaries, the match offsets are off by the number of characters
that have been skipped.

This patch corrects the offsets to be relative to the value of bol as
passed to match_one_pattern() by its caller.

Signed-off-by: Rene Scharfe <rene.scharfe@lsrfire.ath.cx>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
15 years agocompletion: use git rev-parse to detect bare repos
Giuseppe Bilotta [Mon, 18 May 2009 16:24:30 +0000 (18:24 +0200)]
completion: use git rev-parse to detect bare repos

Its check is more robust than a config check for core.bare

Trivially-Acked-by: Shawn O. Pearce <spearce@spearce.org>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
15 years agoCope better with a _lot_ of packs
Johannes Schindelin [Fri, 15 May 2009 18:52:47 +0000 (20:52 +0200)]
Cope better with a _lot_ of packs

You might end up with a situation where you have tons of pack files, e.g.
when using hg2git.  In this situation, all kinds of operations may
end up with a "too many files open" error.  Let's recover gracefully from
that.

Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de>
Looks-right-to-me-by: Shawn O. Pearce <spearce@spearce.org>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
15 years agogrep: make callback functions static
René Scharfe [Wed, 20 May 2009 22:05:22 +0000 (00:05 +0200)]
grep: make callback functions static

Suggested by Stephen Boyd: make the callback functions used for option
parsing static.

Signed-off-by: Rene Scharfe <rene.scharfe@lsrfire.ath.cx>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
15 years agofor-each-ref: fix segfault in copy_email
Jeff King [Mon, 18 May 2009 17:58:11 +0000 (13:58 -0400)]
for-each-ref: fix segfault in copy_email

You can trigger a segfault in git.git by doing:

  git for-each-ref --format='%(taggeremail)' refs/tags/v0.99

The v0.99 tag is special in that it contains no "tagger"
header.

The bug is obvious in copy_email, which carefully checks to
make sure the result of a strchr is non-NULL, but only after
already having used it to perform other work. The fix is to
move the check up.

Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
15 years agoExtend sample pre-commit hook to check for non ascii filenames
Heiko Voigt [Tue, 19 May 2009 20:01:54 +0000 (22:01 +0200)]
Extend sample pre-commit hook to check for non ascii filenames

At the moment non-ascii encodings of filenames are not portably
converted between different filesystems by git. This will most likely
change in the future but to allow repositories to be portable among
different file/operating systems this check is enabled by default.

Signed-off-by: Heiko Voigt <hvoigt@hvoigt.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
15 years agoUnify signedness in hashing calls
Dan McGee [Tue, 19 May 2009 04:34:02 +0000 (23:34 -0500)]
Unify signedness in hashing calls

Our hash_obj and hashtable_index calls and functions were doing a lot of
funny things with signedness. Unify all of it to 'unsigned int'.

Signed-off-by: Dan McGee <dpmcgee@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
15 years agoMerge branch 'np/push-delta'
Junio C Hamano [Mon, 18 May 2009 16:01:16 +0000 (09:01 -0700)]
Merge branch 'np/push-delta'

* np/push-delta:
  allow OFS_DELTA objects during a push

15 years agoMerge branch 'ar/merge-one-file-diag'
Junio C Hamano [Mon, 18 May 2009 16:01:11 +0000 (09:01 -0700)]
Merge branch 'ar/merge-one-file-diag'

* ar/merge-one-file-diag:
  Clarify kind of conflict in merge-one-file helper

15 years agoMerge branch 'ar/unlink-err'
Junio C Hamano [Mon, 18 May 2009 16:01:06 +0000 (09:01 -0700)]
Merge branch 'ar/unlink-err'

* ar/unlink-err:
  print unlink(2) errno in copy_or_link_directory
  replace direct calls to unlink(2) with unlink_or_warn
  Introduce an unlink(2) wrapper which gives warning if unlink failed

15 years agoMerge branch 'jk/maint-add-empty'
Junio C Hamano [Mon, 18 May 2009 16:01:01 +0000 (09:01 -0700)]
Merge branch 'jk/maint-add-empty'

* jk/maint-add-empty:
  add: don't complain when adding empty project root

15 years agoMerge branch 'js/add-edit'
Junio C Hamano [Mon, 18 May 2009 16:00:06 +0000 (09:00 -0700)]
Merge branch 'js/add-edit'

* js/add-edit:
  t3702: fix reliance on SHELL_PATH being '/bin/sh'
  git-add: introduce --edit (to edit the diff vs. the index)

15 years agoMerge branch 'mh/diff-stat-color'
Junio C Hamano [Mon, 18 May 2009 15:59:54 +0000 (08:59 -0700)]
Merge branch 'mh/diff-stat-color'

* mh/diff-stat-color:
  diff: do not color --stat output like patch context

15 years agoMerge branch 'mh/show-branch-color'
Junio C Hamano [Mon, 18 May 2009 15:59:48 +0000 (08:59 -0700)]
Merge branch 'mh/show-branch-color'

* mh/show-branch-color:
  bash completion: show-branch color support
  show-branch: color the commit status signs

Conflicts:
contrib/completion/git-completion.bash

15 years agoMerge branch 'ac/graph-horizontal-line'
Junio C Hamano [Mon, 18 May 2009 15:59:30 +0000 (08:59 -0700)]
Merge branch 'ac/graph-horizontal-line'

* ac/graph-horizontal-line:
  graph API: Use horizontal lines for more compact graphs

15 years agoMerge branch 'ae/anon-fetch-info'
Junio C Hamano [Mon, 18 May 2009 15:59:27 +0000 (08:59 -0700)]
Merge branch 'ae/anon-fetch-info'

* ae/anon-fetch-info:
  fetch: Strip usernames from url's before storing them

15 years agoMerge branch 'maint'
Junio C Hamano [Mon, 18 May 2009 15:59:20 +0000 (08:59 -0700)]
Merge branch 'maint'

* maint:
  show-branch: Fix die message in parse_reflog_param()

15 years agoshow-branch: Fix die message in parse_reflog_param()
Stephen Boyd [Sun, 17 May 2009 10:47:02 +0000 (03:47 -0700)]
show-branch: Fix die message in parse_reflog_param()

Commit 76a44c5 (show-branch --reflog: show the reflog message at the
top, 2007-01-19) introduced parse_reflog_param(). The die() call was
incorrectly passed arg + 9, when it should have been passed arg.

Signed-off-by: Stephen Boyd <bebarino@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
15 years agoMerge branch 'maint'
Junio C Hamano [Sun, 17 May 2009 06:08:49 +0000 (23:08 -0700)]
Merge branch 'maint'

* maint:
  completion: add missing options to show-branch and show
  dir.c: clean up handling of 'path' parameter in read_directory_recursive()
  Fix type-punning issues

15 years agocompletion: add missing options to show-branch and show
Stephen Boyd [Sun, 17 May 2009 03:42:43 +0000 (20:42 -0700)]
completion: add missing options to show-branch and show

Add --oneline and --abbrev-commit to show and --sparse to show-branch.

Signed-off-by: Stephen Boyd <bebarino@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
15 years agodir.c: clean up handling of 'path' parameter in read_directory_recursive()
Linus Torvalds [Thu, 14 May 2009 20:05:03 +0000 (13:05 -0700)]
dir.c: clean up handling of 'path' parameter in read_directory_recursive()

Right now we pass two different pathnames ('path' and 'base') down to
read_directory_recursive(), and the only real reason for that is that we
want to allow an empty 'base' parameter, but when we do so, we need the
pathname to "opendir()" to be "." rather than the empty string.

And rather than handle that confusion in the caller, we can just fix
read_directory_recursive() to handle the case of an empty path itself,
by just passing opendir() a "." ourselves if the path is empty.

This would allow us to then drop one of the pathnames entirely from the
calling convention, but rather than do that, we'll start separating them
out as a "filesystem pathname" (the one we use for filesystem accesses)
and a "git internal base name" (which is the name that we use for git
internally).

That will eventually allow us to do things like handle different
encodings (eg the filesystem pathnames might be Latin1, while git itself
would use UTF-8 for filename information).

Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
15 years agoFix type-punning issues
Dan McGee [Tue, 12 May 2009 01:17:38 +0000 (20:17 -0500)]
Fix type-punning issues

In these two places we are casting part of our unsigned char sha1 array into
an unsigned int, which violates GCCs strict-aliasing rules (and probably
other compilers).

Signed-off-by: Dan McGee <dpmcgee@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
15 years agoMerge branch 'maint'
Junio C Hamano [Sun, 17 May 2009 02:49:42 +0000 (19:49 -0700)]
Merge branch 'maint'

* maint:
  test: checkout shouldn't say that HEAD has moved if it didn't
  completion: enhance "current branch" display
  completion: simplify "current branch" in __git_ps1()
  completion: fix PS1 display during a merge on detached HEAD
  builtin-checkout: Don't tell user that HEAD has moved before it has
  pre-commit.sample: don't print incidental SHA1
  tests: Add tests for missing format-patch long options
  api-parse-options.txt: use 'func' instead of 'funct'
  Turn on USE_ST_TIMESPEC for OpenBSD
  ls-tree manpage: output of ls-tree is compatible with update-index
  ls-tree manpage: use "unless" instead of "when ... is not"

15 years agotest: checkout shouldn't say that HEAD has moved if it didn't
Nanako Shiraishi [Sun, 17 May 2009 02:43:08 +0000 (11:43 +0900)]
test: checkout shouldn't say that HEAD has moved if it didn't

Signed-off-by: しらいしななこ <nanako3@lavabit.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
15 years agocompletion: enhance "current branch" display
Junio C Hamano [Sun, 10 May 2009 08:56:21 +0000 (01:56 -0700)]
completion: enhance "current branch" display

Introduce GIT_PS1_DESCRIBE option you can set to "contains", "branch", or
"describe" to tweak the way how a detached HEAD is described.

The default behaviour is to describe only exact match with some tag
(otherwise use the first 7 hexdigits) as before.

Signed-off-by: Junio C Hamano <gitster@pobox.com>
15 years agocompletion: simplify "current branch" in __git_ps1()
Junio C Hamano [Sun, 10 May 2009 08:53:19 +0000 (01:53 -0700)]
completion: simplify "current branch" in __git_ps1()

As I very often work on a detached HEAD, I found it pretty confusing
when __git_ps1() said 'some-name'.  Did I create a branch with that name
by mistake, or do I happen to be on a commit with that exact tag?

This patch fixes the issue by enclosing non branch names in a pair of
parentheses when used to substitute %s token in __git_ps1() argument.

It also fixes a small bug where the branch part is left empty when
.git/HEAD is unreadable for whatever reason.  The output now says
"(unknown)".

Signed-off-by: Junio C Hamano <gitster@pobox.com>
15 years agocompletion: fix PS1 display during a merge on detached HEAD
Junio C Hamano [Sat, 16 May 2009 18:46:22 +0000 (11:46 -0700)]
completion: fix PS1 display during a merge on detached HEAD

If your merge stops in a conflict while on a detached HEAD, recent
completion code fails to show anything.  This was because various cases
added to support the operation-in-progress markers (e.g. REBASE, MERGING)
forgot that they need to set the variable "b" to something for the result
they computed to be displayed at all.

Probably not many people make trial merges on a detached HEAD (which is
tremendously useful feature of git, by the way), and that may be why this
was not noticed for a long time.

Acked-By: Shawn O. Pearce <spearce@spearce.org>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
15 years agoImprove the naming of guessed target repository for git clone
Alex Riesen [Wed, 13 May 2009 16:32:06 +0000 (18:32 +0200)]
Improve the naming of guessed target repository for git clone

Strip leading and trailing spaces off guessed target directory, and
replace sequences of whitespace and 'control' characters with one
space character.

User still can have any name by specifying it explicitely after url.

Signed-off-by: Alex Riesen <raa.lkml@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
15 years agoUse UTF-8 instead of utf8 for backward compatibility
Alex Riesen [Thu, 14 May 2009 12:55:54 +0000 (14:55 +0200)]
Use UTF-8 instead of utf8 for backward compatibility

An old iconv (GNU libiconv 1.11) does not know about utf8, it does know
UTF-8 though, which is also understood by all newer iconv implementations.

Signed-off-by: Alex Riesen <raa.lkml@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
15 years agobuiltin-checkout: Don't tell user that HEAD has moved before it has
Daniel Cordero [Sat, 16 May 2009 17:54:45 +0000 (10:54 -0700)]
builtin-checkout: Don't tell user that HEAD has moved before it has

Previously, checkout would tell the user this message before moving HEAD,
without regard to whether the upcoming move will result in success.
If the move failed, this causes confusion.

Show the message after the move, unless the move failed.

Signed-off-by: Daniel Cordero <theappleman@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
15 years agopre-commit.sample: don't print incidental SHA1
Jim Meyering [Sat, 16 May 2009 10:21:50 +0000 (12:21 +0200)]
pre-commit.sample: don't print incidental SHA1

Make the sample pre-commit hook script discard
all git-rev-parse output, not just stderr.
Otherwise, it would print an SHA1.

Signed-off-by: Jim Meyering <meyering@redhat.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
15 years agotests: Add tests for missing format-patch long options
Stephen Boyd [Sat, 16 May 2009 09:24:45 +0000 (02:24 -0700)]
tests: Add tests for missing format-patch long options

Exercise format-patch's --signoff, --in-reply-to and --start-number long
options.

Signed-off-by: Stephen Boyd <bebarino@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
15 years agoapi-parse-options.txt: use 'func' instead of 'funct'
Stephen Boyd [Sat, 16 May 2009 09:24:44 +0000 (02:24 -0700)]
api-parse-options.txt: use 'func' instead of 'funct'

Signed-off-by: Stephen Boyd <bebarino@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
15 years agoTurn on USE_ST_TIMESPEC for OpenBSD
Tony Kemp [Thu, 14 May 2009 06:47:41 +0000 (16:47 +1000)]
Turn on USE_ST_TIMESPEC for OpenBSD

Like Darwin, OpenBSD's stat struct uses st_ctimespec and st_mtimestruct
rather than st_ctim and st_mtim.

Signed-off-by: Tony Kemp <tony.kemp@newcastle.edu.au>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
15 years agoMerge branch 'maint-1.6.2' into maint
Junio C Hamano [Sat, 16 May 2009 07:12:58 +0000 (00:12 -0700)]
Merge branch 'maint-1.6.2' into maint

* maint-1.6.2:
  ls-tree manpage: output of ls-tree is compatible with update-index
  ls-tree manpage: use "unless" instead of "when ... is not"

15 years agoMerge branch 'maint-1.6.1' into maint-1.6.2
Junio C Hamano [Thu, 14 May 2009 04:06:11 +0000 (21:06 -0700)]
Merge branch 'maint-1.6.1' into maint-1.6.2

* maint-1.6.1:
  ls-tree manpage: output of ls-tree is compatible with update-index
  ls-tree manpage: use "unless" instead of "when ... is not"

15 years agoMerge branch 'maint-1.6.0' into maint-1.6.1
Junio C Hamano [Thu, 14 May 2009 04:05:59 +0000 (21:05 -0700)]
Merge branch 'maint-1.6.0' into maint-1.6.1

* maint-1.6.0:
  ls-tree manpage: output of ls-tree is compatible with update-index
  ls-tree manpage: use "unless" instead of "when ... is not"

15 years agoQuote LF in urls git fetch saves in FETCH_HEAD
Alex Riesen [Wed, 13 May 2009 18:08:53 +0000 (20:08 +0200)]
Quote LF in urls git fetch saves in FETCH_HEAD

The fmt-merge-msg does a strong syntax checking of its input and fails
with if it is incorrect. The LF character is the only character
important for fmt-merge-msg. As the url in FETCH_HEAD plays only
informational role, a quoted representation of the url should be good
and true enough.
The url often comes from either user-editable config or command line,
so it is reasonable to expect all kinds of characters in it, including
the characters which the format of FETCH_HEAD considers special (line
separator in this case).

Noticed and reported by Hugo Mildenberger.

Signed-off-by: Alex Riesen <raa.lkml@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
15 years agoDocumentation: clarify / requirement in 'git check-ref-format'
Michael J Gruber [Wed, 13 May 2009 15:43:06 +0000 (17:43 +0200)]
Documentation: clarify / requirement in 'git check-ref-format'

'git check-ref-format' checks for the presence of at least one '/', the
idea being that there should be no refs directly below 'refs/', so there
should be a category like 'heads/' or 'tags/' in a refname.

Try and make this clearer in the man page.

Signed-off-by: Michael J Gruber <git@drmicha.warpmail.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
15 years agobash completion: complete variable names for "git config" with options
Stephen Boyd [Sat, 9 May 2009 01:23:32 +0000 (18:23 -0700)]
bash completion: complete variable names for "git config" with options

This makes it easier for users to get and unset their configuration
variables without having to open documentation or dig through their
configuration file.

__git_config_get_set_variables() retrieves the set configuration
variables from the appropriate configuration file. For example, if
the user has previously specified --global only the global variables
are returned. The same applies for --system, and --file. If no
location has been specified, all set variables are returned.

Signed-off-by: Stephen Boyd <bebarino@gmail.com>
Acked-by: Shawn O. Pearce <spearce@spearce.org>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
15 years agoMerge branch 'maint'
Junio C Hamano [Wed, 13 May 2009 05:51:09 +0000 (22:51 -0700)]
Merge branch 'maint'

* maint:
  GIT 1.6.3.1
  Revert "checkout branch: prime cache-tree fully"

15 years agoGIT 1.6.3.1 v1.6.3.1
Junio C Hamano [Wed, 13 May 2009 05:28:22 +0000 (22:28 -0700)]
GIT 1.6.3.1

Signed-off-by: Junio C Hamano <gitster@pobox.com>
15 years agoMerge branch 'maint-1.6.2' into maint
Junio C Hamano [Tue, 12 May 2009 16:58:34 +0000 (09:58 -0700)]
Merge branch 'maint-1.6.2' into maint

* maint-1.6.2:
  Revert "checkout branch: prime cache-tree fully"

15 years agoMerge branch 'jc/maint-read-tree-multi' into maint-1.6.2
Junio C Hamano [Tue, 12 May 2009 16:58:21 +0000 (09:58 -0700)]
Merge branch 'jc/maint-read-tree-multi' into maint-1.6.2

* jc/maint-read-tree-multi:
  Revert "checkout branch: prime cache-tree fully"

15 years agoRevert "checkout branch: prime cache-tree fully"
Junio C Hamano [Tue, 12 May 2009 16:41:28 +0000 (09:41 -0700)]
Revert "checkout branch: prime cache-tree fully"

The logic in 83ae209 (checkout branch: prime cache-tree fully,
2009-04-20) is bogus; checkout can switch branches with a dirty
index and in such a case the tree won't match HEAD.

Add t2014-switch to catch this breakage.

Signed-off-by: Junio C Hamano <gitster@pobox.com>
15 years agot3900: ISO-2022-JP has more than one popular variants
Junio C Hamano [Tue, 12 May 2009 09:01:51 +0000 (02:01 -0700)]
t3900: ISO-2022-JP has more than one popular variants

When converting from other encodings (e.g. EUC-JP or UTF-8), there are
subtly different variants of ISO-2022-JP, all of which are valid.  At the
end of line or when a run of string switches to 1-byte sequence, ESC ( B
can be used to switch to ASCII or ESC ( J can be used to switch to ISO
646:JP (JIS X 0201) but they essentially are the same character set and
are used interchangeably.  Similarly the set ESC $ @ switches to (JIS X
0208-1978) and ESC $ B switches to (JIS X 0208-1983) are in practice used
interchangeably.

Depending on the iconv library and the locale definition on the system, a
program that converts from another encoding to ISO-2022-JP can produce
different byte sequence, and GIT_TEST_CMP (aka "diff -u") will report the
difference as a failure.

Fix this by converting the expected and the actual output to UTF-8 before
comparing when the end result is ISO-2022-JP.  The test vector string in
t3900/ISO-2022-JP.txt is expressed with ASCII and JIS X 0208-1983, but it
can be expressed with any other possible variant, and when converted back
to UTF-8, these variants produce identical byte sequences.

Signed-off-by: Junio C Hamano <gitster@pobox.com>
15 years agobisect: make "git bisect" use new "--next-all" bisect-helper function
Christian Couder [Sat, 9 May 2009 15:55:47 +0000 (17:55 +0200)]
bisect: make "git bisect" use new "--next-all" bisect-helper function

This patch replace the "--next-exit" option of "git bisect--helper"
with a "--next-all" option that does merge base checking using
the "check_good_are_ancestors_of_bad" function implemented in
"bisect.c" in a former patch.

The new "--next-all" option is then used in "git-bisect.sh" instead
of the "--next-exit" option, and all the shell functions in
"git-bisect.sh" that are now unused are removed.

Signed-off-by: Christian Couder <chriscool@tuxfamily.org>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
15 years agobisect: add "check_good_are_ancestors_of_bad" function
Christian Couder [Sat, 9 May 2009 15:55:46 +0000 (17:55 +0200)]
bisect: add "check_good_are_ancestors_of_bad" function

This is a port of the function with the same name that is in
"git-bisect.sh". The new function is not used yet but will be in
a later patch.

We also implement an helper "check_ancestors" function that use
"start_command" and "finish_command" to launch
"git rev-list $good ^$bad".

Signed-off-by: Christian Couder <chriscool@tuxfamily.org>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
15 years agobisect: implement the "check_merge_bases" function
Christian Couder [Sat, 9 May 2009 15:55:45 +0000 (17:55 +0200)]
bisect: implement the "check_merge_bases" function

And all functions needed to make it work.

This is a port from the shell function with the same name
"git-bisect.sh". This function is not used yet but it will be used
later.

Signed-off-by: Christian Couder <chriscool@tuxfamily.org>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
15 years agobisect: automatically sort sha1_array if needed when looking it up
Christian Couder [Sat, 9 May 2009 15:55:44 +0000 (17:55 +0200)]
bisect: automatically sort sha1_array if needed when looking it up

This makes sha1_array easier to use, so later patches will be simpler.

Signed-off-by: Christian Couder <chriscool@tuxfamily.org>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
15 years agobisect: make skipped array functions more generic
Christian Couder [Sat, 9 May 2009 15:55:43 +0000 (17:55 +0200)]
bisect: make skipped array functions more generic

So they can be used on the good array too.

This is done by renaming many functions and some variables to
remove "skip" in the name, and by adding a
"struct sha1_array *array" argument where needed.

While at it, make the second argument to "lookup_sha1_array"
const. It becomes "const unsigned char *sha1".

Signed-off-by: Christian Couder <chriscool@tuxfamily.org>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
15 years agobisect: remove too much function nesting
Christian Couder [Sat, 9 May 2009 15:55:42 +0000 (17:55 +0200)]
bisect: remove too much function nesting

This patch moves some function calls into "bisect_next_exit" so
that functions are nesting less.

The call to "bisect_rev_setup" is moved from "bisect_common" into
"bisect_next_exit" and the call to "read_bisect_refs" from
"bisect_rev_setup" into "bisect_next_exit".

While at it, "rev_argv" is moved into "bisect_rev_setup".

This will make it easier and cleaner to implement checking merge
bases.

Signed-off-by: Christian Couder <chriscool@tuxfamily.org>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
15 years agobisect: use new "struct argv_array" to prepare argv for "setup_revisions"
Christian Couder [Sat, 9 May 2009 15:55:41 +0000 (17:55 +0200)]
bisect: use new "struct argv_array" to prepare argv for "setup_revisions"

Because we will use other instances of this struct.

The "rev_argv_push" function is changed into 2 functions
"argv_array_push" and "argv_array_push_sha1" that take a "struct
argv_array *" as first argument. And these functions are used to
simplify "bisect_rev_setup".

Signed-off-by: Christian Couder <chriscool@tuxfamily.org>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
15 years agobisect: store good revisions in a "sha1_array"
Christian Couder [Sat, 9 May 2009 15:55:40 +0000 (17:55 +0200)]
bisect: store good revisions in a "sha1_array"

This will make it easier to use good revisions for checking merge
bases later.

To simplify the code, a new "sha1_array_push" function is also
introduced.

And while at it we move the earlier part of the code to fill the
argv that is passed to "setup_revisions", so that all this code is
now completely after "read_bisect_refs".

Signed-off-by: Christian Couder <chriscool@tuxfamily.org>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
15 years agobisect: implement "rev_argv_push" to fill an argv with revs
Christian Couder [Sat, 9 May 2009 15:55:39 +0000 (17:55 +0200)]
bisect: implement "rev_argv_push" to fill an argv with revs

This patch is a minor clean up right now, but the new function
will evolve and be used more later.

Signed-off-by: Christian Couder <chriscool@tuxfamily.org>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
15 years agobisect: use "sha1_array" to store skipped revisions
Christian Couder [Sat, 9 May 2009 15:55:38 +0000 (17:55 +0200)]
bisect: use "sha1_array" to store skipped revisions

This patch creates a "struct sha1_array" to store skipped revisions,
so that the same struct can be reused in a later patch for good
revisions.

Signed-off-by: Christian Couder <chriscool@tuxfamily.org>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
15 years agols-tree manpage: output of ls-tree is compatible with update-index
Alex Riesen [Sun, 10 May 2009 16:14:49 +0000 (18:14 +0200)]
ls-tree manpage: output of ls-tree is compatible with update-index

Such format relationships are very useful things to remember for
script writers.

Signed-off-by: Alex Riesen <raa.lkml@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
15 years agols-tree manpage: use "unless" instead of "when ... is not"
Alex Riesen [Sun, 10 May 2009 16:13:45 +0000 (18:13 +0200)]
ls-tree manpage: use "unless" instead of "when ... is not"

Delayed negation in a statement is harder to spot and keep in mind.

Signed-off-by: Alex Riesen <raa.lkml@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
15 years agoam: simplify "sq" function by using "git rev-parse --sq-quote"
Christian Couder [Fri, 24 Apr 2009 06:29:01 +0000 (08:29 +0200)]
am: simplify "sq" function by using "git rev-parse --sq-quote"

Signed-off-by: Christian Couder <chriscool@tuxfamily.org>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
15 years agobisect: use "git rev-parse --sq-quote" instead of a custom "sq" function
Christian Couder [Fri, 24 Apr 2009 06:29:00 +0000 (08:29 +0200)]
bisect: use "git rev-parse --sq-quote" instead of a custom "sq" function

As the "sq" function was the only place using Perl in "git-bisect.sh",
this removes the Perl dependency in this script.

While at it, we also remove the sed instruction in the Makefile that
substituted @@PERL@@ with the Perl path in shell scripts, as this is
not needed anymore. (It is now only needed in "git-instaweb.sh" but
this command is dealt with separately in the Makefile.)

Signed-off-by: Christian Couder <chriscool@tuxfamily.org>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
15 years agorev-parse: add --sq-quote to shell quote arguments
Christian Couder [Sat, 25 Apr 2009 04:55:26 +0000 (06:55 +0200)]
rev-parse: add --sq-quote to shell quote arguments

Signed-off-by: Christian Couder <chriscool@tuxfamily.org>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
15 years agorev-list: remove stringed output flag from "show_bisect_vars"
Christian Couder [Tue, 21 Apr 2009 05:54:10 +0000 (07:54 +0200)]
rev-list: remove stringed output flag from "show_bisect_vars"

Because it was used only by "git bisect--helper --next-vars" but
the "--next-vars" option has been removed.

Signed-off-by: Christian Couder <chriscool@tuxfamily.org>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
15 years agobisect--helper: remove "--next-vars" option as it is now useless
Christian Couder [Tue, 21 Apr 2009 05:54:09 +0000 (07:54 +0200)]
bisect--helper: remove "--next-vars" option as it is now useless

Because it has been replaced by "--next-exit".

Signed-off-by: Christian Couder <chriscool@tuxfamily.org>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
15 years agobisect: use "git bisect--helper --next-exit" in "git-bisect.sh"
Christian Couder [Sun, 19 Apr 2009 09:56:16 +0000 (11:56 +0200)]
bisect: use "git bisect--helper --next-exit" in "git-bisect.sh"

instead of "git bisect--helper --next-vars".

Signed-off-by: Christian Couder <chriscool@tuxfamily.org>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
15 years agobisect--helper: add "--next-exit" to output bisect results
Christian Couder [Sun, 19 Apr 2009 09:56:07 +0000 (11:56 +0200)]
bisect--helper: add "--next-exit" to output bisect results

The goal of this patch is to port more shell code from the "bisect_next"
function in "git-bisect.sh" to C code in "builtin-bisect--helper.c".

So we port the code that interprets the bisection result and stops or
continues (by checking out the next revision) the bisection process.

Signed-off-by: Christian Couder <chriscool@tuxfamily.org>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
15 years agobisect: move common bisect functionality to "bisect_common"
Christian Couder [Sun, 19 Apr 2009 09:55:57 +0000 (11:55 +0200)]
bisect: move common bisect functionality to "bisect_common"

So we can easily reuse the code in a later patch.

Signed-off-by: Christian Couder <chriscool@tuxfamily.org>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
15 years agorev-list: refactor printing bisect vars
Christian Couder [Sun, 19 Apr 2009 09:55:43 +0000 (11:55 +0200)]
rev-list: refactor printing bisect vars

This simplifies the code, and while at it we create the
"print_commit_list" function that will be reused later.

Signed-off-by: Christian Couder <chriscool@tuxfamily.org>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
15 years agorev-list: make "estimate_bisect_steps" non static
Christian Couder [Sun, 19 Apr 2009 09:55:38 +0000 (11:55 +0200)]
rev-list: make "estimate_bisect_steps" non static

Because it will be used from "bisect.c" too.

Signed-off-by: Christian Couder <chriscool@tuxfamily.org>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
15 years agomerge-recursive: never leave index unmerged while recursing
Dave Olszewski [Sat, 9 May 2009 21:49:59 +0000 (14:49 -0700)]
merge-recursive: never leave index unmerged while recursing

When you are trying to come up with the final result (i.e. depth=0), you
want to record how the conflict arose by registering the state of the
common ancestor, your branch and the other branch in the index, hence you
want to do update_stages().

When you are merging with positive depth, that is because of a criss-cross
merge situation.  In such a case, you would need to record the tentative
result, with conflict markers and all, as if the merge went cleanly, even
if there are conflicts, in order to write it out as a tree object later to
be used as a common ancestor tree.

update_file() calls update_file_flags() with update_cache=1 to signal that
the result needs to be written to the index at stage #0 (i.e. merged), and
the code should not clobber the index further by calling update_stages().

The codepath to deal with rename/delete conflict in a recursive merge
however left the index unmerged.

Signed-off-by: Dave Olszewski <cxreg@pobox.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
15 years agoTeach 'git checkout' to preload the index contents
Linus Torvalds [Sat, 9 May 2009 22:11:17 +0000 (15:11 -0700)]
Teach 'git checkout' to preload the index contents

This makes git checkout know to use the threaded index preloading if it
is enabled in the config file. You need to have

[core]
preloadindex = true

in your config file to see it, and for that feature to make sense your
filesystem needs to be able to do concurrent 'lstat()' lookups, but when
that is the case (especially NFS over a high-latency network), this can
be a noticeable performance win.

But with a low-latency network and at least older Linux NFS clients, this
will clearly potentially cause a lot of lock contention. It may still
speed up the uncached case, but the threading and locking overhead will
result in the cached case likely slowing down.

That was almost certainly fixed by Linux commit fc0f684c2 ("NFS: Remove
BKL from NFS lookup code"), but that one got merged into 2.6.27-rc1, so
older kernel versions than 2.6.27 will not scale very well.

But regardless, it's the right thing to do. If your filesystem doesn't
scale, don't enable index preloading.

Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
15 years agoAvoid unnecessary 'lstat()' calls in 'get_stat_data()'
Linus Torvalds [Sat, 9 May 2009 21:57:30 +0000 (14:57 -0700)]
Avoid unnecessary 'lstat()' calls in 'get_stat_data()'

When we ask get_stat_data() to get the mode and size of an index entry,
we can avoid the lstat() call if we have marked the index entry as being
uptodate due to earlier lstat() calls.

This avoids a lot of unnecessary lstat() calls in eg 'git checkout',
where the last phase shows the differences to the working tree
(requiring a diff), but earlier phases have already verified the index.

On the kernel repo (with a fast machine and everything cached), this
changes timings of a nul 'git checkout' from

 - Before (best of ten):

0.14user 0.05system 0:00.19elapsed 100%CPU (0avgtext+0avgdata 0maxresident)k
0inputs+0outputs (0major+13237minor)pagefaults 0swaps

 - After
0.11user 0.03system 0:00.15elapsed 98%CPU (0avgtext+0avgdata 0maxresident)k
0inputs+0outputs (0major+13235minor)pagefaults 0swaps

so it can obviously be noticeable, although equally obviously it's not a
show-stopper on this particular machine. The difference is likely larger
on slower machines, or with operating systems that don't do as good a job
of name caching.

Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
15 years agoMerge branch 'maint'
Junio C Hamano [Sat, 9 May 2009 16:27:43 +0000 (09:27 -0700)]
Merge branch 'maint'

* maint:
  Documentation: cloning to empty directory is allowed
  Clarify kind of conflict in merge-one-file helper
  git config: clarify --add and --get-color
  archive-tar.c: squelch a type mismatch warning

15 years agoallow -t abbreviation for --track in git branch
Sitaram Chamarty [Tue, 28 Apr 2009 15:21:20 +0000 (20:51 +0530)]
allow -t abbreviation for --track in git branch

also makes it consistent with git-checkout

Signed-off-by: Junio C Hamano <gitster@pobox.com>
15 years agoAdd --reference option to git submodule.
Michael S. Tsirkin [Mon, 4 May 2009 19:30:01 +0000 (22:30 +0300)]
Add --reference option to git submodule.

This adds --reference option to git submodule add and
git submodule update commands, which is passed to git clone.

Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
15 years agoadd: don't complain when adding empty project root
Jeff King [Wed, 29 Apr 2009 03:21:01 +0000 (23:21 -0400)]
add: don't complain when adding empty project root

We try to warn the user if one of their pathspecs caused no
matches, as it may have been a typo. However, we disable the
warning if the pathspec points to an existing file, since
that means it is not a typo but simply an empty directory.

Unfortunately, the file_exists() test was broken for one
special case: the pathspec of the project root is just "".
This patch detects this special case and acts as if the file
exists (which it must, since it is the project root).

The user-visible effect is that this:

  $ mkdir repo && cd repo && git init && git add .

used to complain like:

  fatal: pathspec '' did not match any files

but now is a silent no-op.

Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
15 years agoGit.pm: Always set Repository to absolute path if autodetecting
Frank Lichtenheld [Thu, 7 May 2009 13:41:28 +0000 (15:41 +0200)]
Git.pm: Always set Repository to absolute path if autodetecting

So far we only set it to absolute paths in some cases which lead
to problems like wc_chdir not working.

Signed-off-by: Frank Lichtenheld <flichtenheld@astaro.com>
Acked-by: Petr Baudis <pasky@suse.cz>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
15 years agoGit.pm: Set GIT_WORK_TREE if we set GIT_DIR
Frank Lichtenheld [Thu, 7 May 2009 13:41:27 +0000 (15:41 +0200)]
Git.pm: Set GIT_WORK_TREE if we set GIT_DIR

Otherwise git will use the current directory as work tree which will
lead to unexpected results if we operate in sub directory of the
work tree.

Signed-off-by: Frank Lichtenheld <flichtenheld@astaro.com>
Acked-by: Petr Baudis <pasky@suse.cz>
Signed-off-by: Junio C Hamano <gitster@pobox.com>