Code

git.git
15 years agoMerge branch 'lt/pack-object-memuse'
Junio C Hamano [Sat, 18 Apr 2009 21:46:17 +0000 (14:46 -0700)]
Merge branch 'lt/pack-object-memuse'

* lt/pack-object-memuse:
  show_object(): push path_name() call further down
  process_{tree,blob}: show objects without buffering

Conflicts:
builtin-pack-objects.c
builtin-rev-list.c
list-objects.c
list-objects.h
upload-pack.c

15 years agoMerge branch 'nd/archive-attribute'
Junio C Hamano [Sat, 18 Apr 2009 21:46:08 +0000 (14:46 -0700)]
Merge branch 'nd/archive-attribute'

* nd/archive-attribute:
  archive test: attributes
  archive: do not read .gitattributes in working directory
  unpack-trees: do not muck with attributes when we are not checking out
  attr: add GIT_ATTR_INDEX "direction"
  archive tests: do not use .gitattributes in working directory

15 years agoMerge branch 'maint'
Junio C Hamano [Sat, 18 Apr 2009 21:45:59 +0000 (14:45 -0700)]
Merge branch 'maint'

* maint:
  Describe fixes since 1.6.2.3
  doc/git-daemon: add missing arguments to max-connections option
  doc/git-daemon: add missing arguments to options
  init: Do not segfault on big GIT_TEMPLATE_DIR environment variable
  imap-send: use correct configuration variable in documentation

15 years agoDescribe fixes since 1.6.2.3
Junio C Hamano [Sat, 18 Apr 2009 21:01:50 +0000 (14:01 -0700)]
Describe fixes since 1.6.2.3

Signed-off-by: Junio C Hamano <gitster@pobox.com>
15 years agoMerge branch 'maint-1.6.1' into maint
Junio C Hamano [Sat, 18 Apr 2009 21:43:39 +0000 (14:43 -0700)]
Merge branch 'maint-1.6.1' into maint

* maint-1.6.1:

15 years agoMerge branch 'maint-1.6.0' into maint-1.6.1
Junio C Hamano [Sat, 18 Apr 2009 21:43:24 +0000 (14:43 -0700)]
Merge branch 'maint-1.6.0' into maint-1.6.1

* maint-1.6.0:
  doc/git-daemon: add missing arguments to options
  init: Do not segfault on big GIT_TEMPLATE_DIR environment variable

15 years agoMerge branch 'bs/maint-1.6.0-tree-walk-prefix' into maint
Junio C Hamano [Sat, 18 Apr 2009 21:18:32 +0000 (14:18 -0700)]
Merge branch 'bs/maint-1.6.0-tree-walk-prefix' into maint

* bs/maint-1.6.0-tree-walk-prefix:
  match_tree_entry(): a pathspec only matches at directory boundaries
  tree_entry_interesting: a pathspec only matches at directory boundary

15 years agoMerge branch 'js/maint-submodule-checkout' into maint
Junio C Hamano [Sat, 18 Apr 2009 21:18:29 +0000 (14:18 -0700)]
Merge branch 'js/maint-submodule-checkout' into maint

* js/maint-submodule-checkout:
  Fix 'git checkout <submodule>' to update the index

15 years agoMerge branch 'cb/maint-merge-recursive-submodule-fix' into maint
Junio C Hamano [Sat, 18 Apr 2009 21:18:25 +0000 (14:18 -0700)]
Merge branch 'cb/maint-merge-recursive-submodule-fix' into maint

* cb/maint-merge-recursive-submodule-fix:
  simplify output of conflicting merge
  update cache for conflicting submodule entries
  add tests for merging with submodules

15 years agodoc/git-daemon: add missing arguments to max-connections option
Markus Heidelberg [Sat, 18 Apr 2009 09:46:06 +0000 (11:46 +0200)]
doc/git-daemon: add missing arguments to max-connections option

Signed-off-by: Markus Heidelberg <markus.heidelberg@web.de>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
15 years agoMerge branch 'maint-1.6.0' into maint
Junio C Hamano [Sat, 18 Apr 2009 20:39:52 +0000 (13:39 -0700)]
Merge branch 'maint-1.6.0' into maint

* maint-1.6.0:
  doc/git-daemon: add missing arguments to options
  init: Do not segfault on big GIT_TEMPLATE_DIR environment variable

15 years agodoc/git-daemon: add missing arguments to options
Markus Heidelberg [Sat, 18 Apr 2009 09:46:06 +0000 (11:46 +0200)]
doc/git-daemon: add missing arguments to options

Also fix some spellings and typos.

Signed-off-by: Markus Heidelberg <markus.heidelberg@web.de>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
15 years agoinit: Do not segfault on big GIT_TEMPLATE_DIR environment variable
Frank Lichtenheld [Sat, 18 Apr 2009 14:14:02 +0000 (16:14 +0200)]
init: Do not segfault on big GIT_TEMPLATE_DIR environment variable

Signed-off-by: Frank Lichtenheld <flichtenheld@astaro.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
15 years agoimap-send: use correct configuration variable in documentation
Paul Bolle [Sat, 18 Apr 2009 11:26:42 +0000 (13:26 +0200)]
imap-send: use correct configuration variable in documentation

It's imap.pass (not imap.password).

Signed-off-by: Paul Bolle <pebolle@tiscali.nl>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
15 years agoMerge branch 'ac/color-graph'
Junio C Hamano [Sat, 18 Apr 2009 04:42:13 +0000 (21:42 -0700)]
Merge branch 'ac/color-graph'

* ac/color-graph:
  graph API: Added logic for colored edges

15 years agoMerge branch 'jk/cobdoc'
Junio C Hamano [Sat, 18 Apr 2009 04:42:13 +0000 (21:42 -0700)]
Merge branch 'jk/cobdoc'

* jk/cobdoc:
  docs/checkout: clarify what "non-branch" means
  doc/checkout: split checkout and branch creation in synopsis
  doc/checkout: refer to git-branch(1) as appropriate
  doc: refer to tracking configuration as "upstream"
  doc: clarify --no-track option

15 years agoMerge branch 'mk/apply-swap'
Junio C Hamano [Sat, 18 Apr 2009 04:42:13 +0000 (21:42 -0700)]
Merge branch 'mk/apply-swap'

* mk/apply-swap:
  builtin-apply: keep information about files to be deleted
  tests: test applying criss-cross rename patch

15 years agoMerge branch 'mm/add-p-quit'
Junio C Hamano [Sat, 18 Apr 2009 04:42:12 +0000 (21:42 -0700)]
Merge branch 'mm/add-p-quit'

* mm/add-p-quit:
  Update git-add.txt according to the new possibilities of 'git add -p'.
  add-interactive: refactor mode hunk handling
  git add -p: new "quit" command at the prompt.

15 years agoMerge branch 'eb/upload-archive-from-git-shell'
Junio C Hamano [Sat, 18 Apr 2009 04:42:12 +0000 (21:42 -0700)]
Merge branch 'eb/upload-archive-from-git-shell'

* eb/upload-archive-from-git-shell:
  git-shell: Add 'git-upload-archive' to allowed commands.

15 years agoMerge branch 'jc/shared-literally'
Junio C Hamano [Sat, 18 Apr 2009 04:42:12 +0000 (21:42 -0700)]
Merge branch 'jc/shared-literally'

* jc/shared-literally:
  t1301-shared-repo: fix forced modes test

15 years agoMerge branch 'ns/am-to-empty'
Junio C Hamano [Sat, 18 Apr 2009 04:42:12 +0000 (21:42 -0700)]
Merge branch 'ns/am-to-empty'

* ns/am-to-empty:
  git-am: teach git-am to apply a patch to an unborn branch

15 years agoMerge branch 'bw/short-ref-strict'
Junio C Hamano [Sat, 18 Apr 2009 04:42:12 +0000 (21:42 -0700)]
Merge branch 'bw/short-ref-strict'

* bw/short-ref-strict:
  remote.c: use shorten_unambiguous_ref
  rev-parse: --abbrev-ref option to shorten ref name
  for-each-ref: utilize core.warnAmbiguousRefs for :short-format
  shorten_unambiguous_ref(): add strict mode

15 years agoMerge branch 'da/difftool'
Junio C Hamano [Sat, 18 Apr 2009 04:42:12 +0000 (21:42 -0700)]
Merge branch 'da/difftool'

* da/difftool:
  mergetool--lib: simplify API usage by removing more global variables
  Fix misspelled mergetool.keepBackup
  difftool/mergetool: refactor commands to use git-mergetool--lib
  mergetool: use $( ... ) instead of `backticks`
  bash completion: add git-difftool
  difftool: add support for a difftool.prompt config variable
  difftool: add various git-difftool tests
  difftool: move 'git-difftool' out of contrib
  difftool/mergetool: add diffuse as merge and diff tool
  difftool: add a -y shortcut for --no-prompt
  difftool: use perl built-ins when testing for msys
  difftool: remove the backup file feature
  difftool: remove merge options for opendiff, tkdiff, kdiff3 and xxdiff
  git-mergetool: add new merge tool TortoiseMerge
  git-mergetool/difftool: make (g)vimdiff workable under Windows
  doc/merge-config: list ecmerge as a built-in merge tool

15 years agoMerge branch 'maint'
Junio C Hamano [Sat, 18 Apr 2009 04:29:15 +0000 (21:29 -0700)]
Merge branch 'maint'

* maint:
  doc/gitattributes: clarify location of config text
  Fix buffer overflow in config parser
  git-apply: fix option description

15 years agodoc/gitattributes: clarify location of config text
Jeff King [Thu, 16 Apr 2009 07:17:12 +0000 (03:17 -0400)]
doc/gitattributes: clarify location of config text

The gitattributes documentation has a section on the "diff"
attribute, with subsections for each of the things you might
want to configure in your diff config section (external
diff, hunk headers, etc). The first such subsection
specifically notes that the definition of the diff driver
should go into $GIT_DIR/config, but subsequent sections do
not.

This location is implied if you are reading the
documentation sequentially, but it is not uncommon for a new
user to jump to (or be referred to) a specific section. For
a new user who does not know git well enough to recognize
the config syntax, it is not clear that those directives
don't also go into the gitattributes file.

This patch just mentions the config file in each subsection,
similar to the way it is mentioned in the first.

Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
15 years agohook/update: example of how to prevent branch creation
Pierre Habouzit [Thu, 16 Apr 2009 20:00:44 +0000 (22:00 +0200)]
hook/update: example of how to prevent branch creation

Since git doesn't provide a receive.denyBranchCreation or similar, here is
an example of how to be sure users cannot create branches remotely by
pushing a new reference.

This setup has been proven useful to prevent creation of spurious branches
because of users having their remote.origin.push set to HEAD, when they
use `git push` while being on a local topic branch of theirs instead of
the proper one.

Signed-off-by: Pierre Habouzit <madcoder@debian.org>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
15 years agoMerge branch 'maint-1.6.1' into maint
Junio C Hamano [Sat, 18 Apr 2009 04:20:58 +0000 (21:20 -0700)]
Merge branch 'maint-1.6.1' into maint

* maint-1.6.1:
  Fix buffer overflow in config parser

15 years agoMerge branch 'maint-1.6.0' into maint-1.6.1
Junio C Hamano [Sat, 18 Apr 2009 04:06:11 +0000 (21:06 -0700)]
Merge branch 'maint-1.6.0' into maint-1.6.1

* maint-1.6.0:
  Fix buffer overflow in config parser

15 years agoarchive test: attributes
René Scharfe [Fri, 17 Apr 2009 22:18:10 +0000 (00:18 +0200)]
archive test: attributes

Add a test script for all archive attributes and their handling in
normal and bare repositories.  export-ignore and export-subst are
tested, as well as the effect of the option --worktree-attributes.

Signed-off-by: Rene Scharfe <rene.scharfe@lsrfire.ath.cx>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
15 years agoarchive: do not read .gitattributes in working directory
Nguyễn Thái Ngọc Duy [Fri, 17 Apr 2009 22:18:05 +0000 (00:18 +0200)]
archive: do not read .gitattributes in working directory

The old behaviour still remains with --worktree-attributes, and it is
always on for the legacy "git tar-tree".

Signed-off-by: Nguyễn Thái Ngọc Duy <pclouds@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
15 years agounpack-trees: do not muck with attributes when we are not checking out
Junio C Hamano [Fri, 17 Apr 2009 22:18:01 +0000 (00:18 +0200)]
unpack-trees: do not muck with attributes when we are not checking out

Signed-off-by: Junio C Hamano <gitster@pobox.com>
15 years agoattr: add GIT_ATTR_INDEX "direction"
Nguyễn Thái Ngọc Duy [Fri, 17 Apr 2009 22:17:58 +0000 (00:17 +0200)]
attr: add GIT_ATTR_INDEX "direction"

This instructs attr mechanism, not to look into working .gitattributes
at all. Needed by tools that does not handle working directory, such
as "git archive".

Signed-off-by: Nguyễn Thái Ngọc Duy <pclouds@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
15 years agoarchive tests: do not use .gitattributes in working directory
René Scharfe [Fri, 17 Apr 2009 22:17:49 +0000 (00:17 +0200)]
archive tests: do not use .gitattributes in working directory

We are interested in using archive mostly from a bare repository, so it
should not add .gitattributes to the work tree.

Signed-off-by: Rene Scharfe <rene.scharfe@lsrfire.ath.cx>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
15 years agoFix buffer overflow in config parser
Thomas Jarosch [Fri, 17 Apr 2009 12:05:11 +0000 (14:05 +0200)]
Fix buffer overflow in config parser

When interpreting a config value, the config parser reads in 1+ space
character(s) and puts -one- space character in the buffer as soon as
the first non-space character is encountered (if not inside quotes).

Unfortunately the buffer size check lacks the extra space character
which gets inserted at the next non-space character, resulting in
a crash with a specially crafted config entry.

The unit test now uses Java to compile a platform independent
.NET framework to output the test string in C# :o)

    Read: Thanks to Johannes Sixt for the correct printf call
    which replaces the perl invocation.

Signed-off-by: Thomas Jarosch <thomas.jarosch@intra2net.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
15 years agoremote.c: use shorten_unambiguous_ref
Michael J Gruber [Thu, 16 Apr 2009 08:20:44 +0000 (10:20 +0200)]
remote.c: use shorten_unambiguous_ref

Use the new shorten_unambiguous_ref() for simplifying the output of
upstream branch names. This affects status and checkout.

Signed-off-by: Michael J Gruber <git@drmicha.warpmail.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
15 years agoget_local_heads(): do not return random pointer if there is no head
Nguyễn Thái Ngọc Duy [Thu, 16 Apr 2009 22:16:23 +0000 (08:16 +1000)]
get_local_heads(): do not return random pointer if there is no head

it's silly to do this:

mkdir foo && cd foo && git init && git push somewhere.git

but segfault should not happen even in that case.

Signed-off-by: Nguyễn Thái Ngọc Duy <pclouds@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
15 years agogit-apply: fix option description
Ulrich Windl [Thu, 9 Apr 2009 09:12:58 +0000 (11:12 +0200)]
git-apply: fix option description

Do not use non ASCII single quote.

Signed-off-by: Junio C Hamano <gitster@pobox.com>
15 years agoUpdate git-add.txt according to the new possibilities of 'git add -p'.
Matthieu Moy [Thu, 16 Apr 2009 16:46:23 +0000 (18:46 +0200)]
Update git-add.txt according to the new possibilities of 'git add -p'.

The text is merely cut-and-pasted from git-add--interactive.perl. The
cut-and-paste also fixes a typo.

Signed-off-by: Matthieu Moy <Matthieu.Moy@imag.fr>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
15 years agoadd-interactive: refactor mode hunk handling
Jeff King [Thu, 16 Apr 2009 07:14:15 +0000 (03:14 -0400)]
add-interactive: refactor mode hunk handling

The original implementation considered the mode separately
from the rest of the hunks, asking about it outside the main
hunk-selection loop. This patch instead places a mode change
as the first hunk in the loop. This has two advantages:

  1. less duplicated code (since we use the main selection
     loop). This also cleans up an inconsistency, which is
     that the main selection loop separates options with a
     comma, whereas the mode prompt used slashes.

  2. users can now skip the mode change and come back to it,
     search for it (via "/mode"), etc, as they can with other
     hunks.

To facilitate this, each hunk is now marked with a "type".
Mode hunks are not considered for splitting (which would
make no sense, and also confuses the split_hunk function),
nor are they editable. In theory, one could edit the mode
lines and change to a new mode. In practice, there are only
two modes that git cares about (0644 and 0755), so either
you want to move from one to the other or not (and you can
do that by staging or not staging).

Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
15 years agogit add -p: new "quit" command at the prompt.
Matthieu Moy [Fri, 10 Apr 2009 14:57:01 +0000 (16:57 +0200)]
git add -p: new "quit" command at the prompt.

There's already 'd' to stop staging hunks in a file, but no explicit
command to stop the interactive staging (for the current files and the
remaining ones).  Of course you can do 'd' and then ^C, but it would be
more intuitive to allow 'quit' action.

Signed-off-by: Junio C Hamano <gitster@pobox.com>
15 years agobuiltin-apply: keep information about files to be deleted
Michał Kiedrowicz [Sat, 11 Apr 2009 19:31:00 +0000 (21:31 +0200)]
builtin-apply: keep information about files to be deleted

Example correct diff generated by `diff -M -B' might look like this:

diff --git a/file1 b/file2
similarity index 100%
rename from file1
rename to file2
diff --git a/file2 b/file1
similarity index 100%
rename from file2
rename to file1

Information about removing `file2' comes after information about creation
of new `file2' (renamed from `file1'). Existing implementation isn't able to
apply such patch, because it has to know in advance which files will be
removed.

This patch populates fn_table with information about removal of files
before calling check_patch() for each patch to be applied.

Signed-off-by: Michał Kiedrowicz <michal.kiedrowicz@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
15 years agograph API: Added logic for colored edges
Allan Caffee [Mon, 13 Apr 2009 19:53:41 +0000 (15:53 -0400)]
graph API: Added logic for colored edges

Modified the graph drawing logic to colorize edges based on parent-child
relationships similiarly to gitk.

Signed-off-by: Allan Caffee <allan.caffee@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
15 years agotests: test applying criss-cross rename patch
Michał Kiedrowicz [Sat, 11 Apr 2009 15:26:24 +0000 (17:26 +0200)]
tests: test applying criss-cross rename patch

Originally reported by Linus in $gmane/116198

Signed-off-by: Michał Kiedrowicz <michal.kiedrowicz@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
15 years agorev-parse: --abbrev-ref option to shorten ref name
Bert Wesarg [Mon, 13 Apr 2009 11:20:26 +0000 (13:20 +0200)]
rev-parse: --abbrev-ref option to shorten ref name

This applies the shorten_unambiguous_ref function to the object name.
Default mode is controlled by core.warnAmbiguousRefs. Else it is given as
optional argument to --abbrev-ref={strict|loose}.

This should be faster than 'git for-each-ref --format="%(refname:short)" <ref>'
for single refs.

Signed-off-by: Bert Wesarg <bert.wesarg@googlemail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
15 years agofor-each-ref: utilize core.warnAmbiguousRefs for :short-format
Bert Wesarg [Mon, 13 Apr 2009 10:25:47 +0000 (12:25 +0200)]
for-each-ref: utilize core.warnAmbiguousRefs for :short-format

core.warnAmbiguousRefs is used to select strict mode for the
abbreviation for the ":short" format specifier of "refname" and "upstream".

In strict mode, the abbreviated ref will never trigger the
'warn_ambiguous_refs' warning. I.e. for these refs:

  refs/heads/xyzzy
  refs/tags/xyzzy

the abbreviated forms are:

  heads/xyzzy
  tags/xyzzy

Signed-off-by: Bert Wesarg <bert.wesarg@googlemail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
15 years agoshorten_unambiguous_ref(): add strict mode
Bert Wesarg [Mon, 13 Apr 2009 10:25:46 +0000 (12:25 +0200)]
shorten_unambiguous_ref(): add strict mode

Add the strict mode of abbreviation to shorten_unambiguous_ref(), i.e. the
resulting ref won't trigger the ambiguous ref warning.

All users of shorten_unambiguous_ref() still use the loose mode.

Signed-off-by: Bert Wesarg <bert.wesarg@googlemail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
15 years agodocs/checkout: clarify what "non-branch" means
Jeff King [Mon, 13 Apr 2009 11:21:04 +0000 (07:21 -0400)]
docs/checkout: clarify what "non-branch" means

In the code we literally stick "refs/heads/" on the front
and see if it resolves, so that is probably the best
explanation.

Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
15 years agodoc/checkout: split checkout and branch creation in synopsis
Jeff King [Mon, 13 Apr 2009 11:19:33 +0000 (07:19 -0400)]
doc/checkout: split checkout and branch creation in synopsis

These can really be thought of as two different modes, since
the "<branch>" parameter is treated differently in the two
(in one it is the branch to be checked out, but in the other
it is really a start-point for branch creation).

Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
15 years agodoc/checkout: refer to git-branch(1) as appropriate
Jeff King [Mon, 13 Apr 2009 11:18:52 +0000 (07:18 -0400)]
doc/checkout: refer to git-branch(1) as appropriate

Most of description for the branch creation options is
simply cut and paste from git-branch. There are two reasons
to fix this:

  1. It can grow stale with respect to what's in "git
     branch" (which it is now is).

  2. It is not just an implementation detail, but rather the
     desired mental model for the command that we are using
     "git branch" here. Being explicit about that can help
     the user understand what is going on.

It also makes sense to strip the branch creation options
from the synopsis, as they are making it a long,
hard-to-read line. They are still easily discovered by
reading the options list, and --track is explicitly
referenced when branch creation is described.

Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
15 years agodoc: refer to tracking configuration as "upstream"
Jeff King [Mon, 13 Apr 2009 11:11:56 +0000 (07:11 -0400)]
doc: refer to tracking configuration as "upstream"

The term "tracking" often creates confusion between remote
tracking branches and local branches which track a remote
branch. The term "upstream" captures more clearly the idea
of "branch A is based on branch B in some way", so it makes
sense to mention it.

At the same time, upstream branches are used for more
than just git-pull these days; let's mention that here.

Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
15 years agodoc: clarify --no-track option
Jeff King [Mon, 13 Apr 2009 11:11:16 +0000 (07:11 -0400)]
doc: clarify --no-track option

It is not really about ignoring the config option; it is
about turning off tracking, _even if_ the config option is
set.

Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
15 years agogitignore git-bisect--helper
Junio C Hamano [Mon, 13 Apr 2009 04:15:59 +0000 (21:15 -0700)]
gitignore git-bisect--helper

Signed-off-by: Junio C Hamano <gitster@pobox.com>
15 years agogit-am: teach git-am to apply a patch to an unborn branch
Nanako Shiraishi [Fri, 10 Apr 2009 00:34:42 +0000 (09:34 +0900)]
git-am: teach git-am to apply a patch to an unborn branch

People sometimes wonder why they cannot apply a patch that only
creates new files to an unborn branch.

Signed-off-by: Nanako Shiraishi <nanako3@lavabit.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
15 years agoMark t1301 permission test to depend on POSIXPERM
Junio C Hamano [Mon, 13 Apr 2009 00:55:18 +0000 (17:55 -0700)]
Mark t1301 permission test to depend on POSIXPERM

This prepares the topic for inclusion to master.

15 years agoshow_object(): push path_name() call further down
Linus Torvalds [Sat, 11 Apr 2009 01:15:26 +0000 (18:15 -0700)]
show_object(): push path_name() call further down

In particular, pushing the "path_name()" call _into_ the show() function
would seem to allow

 - more clarity into who "owns" the name (ie now when we free the name in
   the show_object callback, it's because we generated it ourselves by
   calling path_name())

 - not calling path_name() at all, either because we don't care about the
   name in the first place, or because we are actually happy walking the
   linked list of "struct name_path *" and the last component.

Now, I didn't do that latter optimization, because it would require some
more coding, but especially looking at "builtin-pack-objects.c", we really
don't even want the whole pathname, we really would be better off with the
list of path components.

Why? We use that name for two things:
 - add_preferred_base_object(), which actually _wants_ to traverse the
   path, and now does it by looking for '/' characters!
 - for 'name_hash()', which only cares about the last 16 characters of a
   name, so again, generating the full name seems to be just unnecessary
   work.

Anyway, so I didn't look any closer at those things, but it did convince
me that the "show_object()" calling convention was crazy, and we're
actually better off doing _less_ in list-objects.c, and giving people
access to the internal data structures so that they can decide whether
they want to generate a path-name or not.

This patch does that, and then for people who did use the name (even if
they might do something more clever in the future), it just does the
straightforward "name = path_name(path, component); .. free(name);" thing.

Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
15 years agoprocess_{tree,blob}: show objects without buffering
Linus Torvalds [Sat, 11 Apr 2009 00:27:58 +0000 (17:27 -0700)]
process_{tree,blob}: show objects without buffering

Here's a less trivial thing, and slightly more dubious one.

I was looking at that "struct object_array objects", and wondering why we
do that. I have honestly totally forgotten. Why not just call the "show()"
function as we encounter the objects? Rather than add the objects to the
object_array, and then at the very end going through the array and doing a
'show' on all, just do things more incrementally.

Now, there are possible downsides to this:

 - the "buffer using object_array" _can_ in theory result in at least
   better I-cache usage (two tight loops rather than one more spread out
   one). I don't think this is a real issue, but in theory..

 - this _does_ change the order of the objects printed. Instead of doing a
   "process_tree(revs, commit->tree, &objects, NULL, "");" in the loop
   over the commits (which puts all the root trees _first_ in the object
   list, this patch just adds them to the list of pending objects, and
   then we'll traverse them in that order (and thus show each root tree
   object together with the objects we discover under it)

   I _think_ the new ordering actually makes more sense, but the object
   ordering is actually a subtle thing when it comes to packing
   efficiency, so any change in order is going to have implications for
   packing. Good or bad, I dunno.

 - There may be some reason why we did it that odd way with the object
   array, that I have simply forgotten.

Anyway, now that we don't buffer up the objects before showing them
that may actually result in lower memory usage during that whole
traverse_commit_list() phase.

This is seriously not very deeply tested. It makes sense to me, it seems
to pass all the tests, it looks ok, but...

Does anybody remember why we did that "object_array" thing? It used to be
an "object_list" a long long time ago, but got changed into the array due
to better memory usage patterns (those linked lists of obejcts are
horrible from a memory allocation standpoint). But I wonder why we didn't
do this back then. Maybe there's a reason for it.

Or maybe there _used_ to be a reason, and no longer is.

Signed-off-by: Junio C Hamano <gitster@pobox.com>
15 years agoGIT 1.6.3-rc0 v1.6.3-rc0
Junio C Hamano [Mon, 13 Apr 2009 00:05:55 +0000 (17:05 -0700)]
GIT 1.6.3-rc0

Signed-off-by: Junio C Hamano <gitster@pobox.com>
15 years agoDocumentation/git.txt: GIT 1.6.2.2 has been out for a while
Nanako Shiraishi [Fri, 10 Apr 2009 00:34:40 +0000 (09:34 +0900)]
Documentation/git.txt: GIT 1.6.2.2 has been out for a while

These links inside "stalenotes" section need to be updated on the master
branch every time a new stable or maintenance release is made.

Signed-off-by: Nanako Shiraishi <nanako3@lavabit.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
15 years agoMerge branch 'jk/no-perl'
Junio C Hamano [Sun, 12 Apr 2009 23:46:42 +0000 (16:46 -0700)]
Merge branch 'jk/no-perl'

* jk/no-perl:
  tests: skip perl tests if NO_PERL is defined
  Makefile: allow building without perl

15 years agoMerge branch 'sb/doc-upstream-branch'
Junio C Hamano [Sun, 12 Apr 2009 23:46:42 +0000 (16:46 -0700)]
Merge branch 'sb/doc-upstream-branch'

* sb/doc-upstream-branch:
  Documentation: Introduce "upstream branch"

15 years agoMerge branch 'jk/show-upstream'
Junio C Hamano [Sun, 12 Apr 2009 23:46:42 +0000 (16:46 -0700)]
Merge branch 'jk/show-upstream'

* jk/show-upstream:
  branch: show upstream branch when double verbose
  make get_short_ref a public function
  for-each-ref: add "upstream" format field
  for-each-ref: refactor refname handling
  for-each-ref: refactor get_short_ref function

15 years agoMerge branch 'fg/remote-prune'
Junio C Hamano [Sun, 12 Apr 2009 23:46:41 +0000 (16:46 -0700)]
Merge branch 'fg/remote-prune'

* fg/remote-prune:
  add tests for remote groups
  git remote update: Fallback to remote if group does not exist
  remote: New function remote_is_configured()
  git remote update: Report error for non-existing groups
  git remote update: New option --prune
  builtin-remote.c: Split out prune_remote as a separate function.

15 years agoMerge branch 'cc/sha1-bsearch'
Junio C Hamano [Sun, 12 Apr 2009 23:46:41 +0000 (16:46 -0700)]
Merge branch 'cc/sha1-bsearch'

* cc/sha1-bsearch:
  sha1-lookup: fix up the assertion message

15 years agoMerge branch 'cc/bisect-filter'
Junio C Hamano [Sun, 12 Apr 2009 23:46:40 +0000 (16:46 -0700)]
Merge branch 'cc/bisect-filter'

* cc/bisect-filter: (21 commits)
  rev-list: add "int bisect_show_flags" in "struct rev_list_info"
  rev-list: remove last static vars used in "show_commit"
  list-objects: add "void *data" parameter to show functions
  bisect--helper: string output variables together with "&&"
  rev-list: pass "int flags" as last argument of "show_bisect_vars"
  t6030: test bisecting with paths
  bisect: use "bisect--helper" and remove "filter_skipped" function
  bisect: implement "read_bisect_paths" to read paths in "$GIT_DIR/BISECT_NAMES"
  bisect--helper: implement "git bisect--helper"
  bisect: use the new generic "sha1_pos" function to lookup sha1
  rev-list: call new "filter_skip" function
  patch-ids: use the new generic "sha1_pos" function to lookup sha1
  sha1-lookup: add new "sha1_pos" function to efficiently lookup sha1
  rev-list: pass "revs" to "show_bisect_vars"
  rev-list: make "show_bisect_vars" non static
  rev-list: move code to show bisect vars into its own function
  rev-list: move bisect related code into its own file
  rev-list: make "bisect_list" variable local to "cmd_rev_list"
  refs: add "for_each_ref_in" function to refactor "for_each_*_ref" functions
  quote: add "sq_dequote_to_argv" to put unwrapped args in an argv array
  ...

15 years agoMerge branch 'maint'
Junio C Hamano [Sun, 12 Apr 2009 23:01:25 +0000 (16:01 -0700)]
Merge branch 'maint'

* maint:
  GIT 1.6.2.3
  State the effect of filter-branch on graft explicitly
  process_{tree,blob}: Remove useless xstrdup calls

Conflicts:
GIT-VERSION-GEN

15 years agoGIT 1.6.2.3 v1.6.2.3
Junio C Hamano [Sun, 12 Apr 2009 22:43:24 +0000 (15:43 -0700)]
GIT 1.6.2.3

Signed-off-by: Junio C Hamano <gitster@pobox.com>
15 years agoMerge branch 'maint-1.6.1' into maint
Junio C Hamano [Sun, 12 Apr 2009 22:34:53 +0000 (15:34 -0700)]
Merge branch 'maint-1.6.1' into maint

* maint-1.6.1:
  State the effect of filter-branch on graft explicitly
  process_{tree,blob}: Remove useless xstrdup calls

15 years agoMerge branch 'maint-1.6.0' into maint-1.6.1
Junio C Hamano [Sun, 12 Apr 2009 22:20:29 +0000 (15:20 -0700)]
Merge branch 'maint-1.6.0' into maint-1.6.1

* maint-1.6.0:
  State the effect of filter-branch on graft explicitly
  process_{tree,blob}: Remove useless xstrdup calls

15 years agomergetool--lib: simplify API usage by removing more global variables
David Aguilar [Sun, 12 Apr 2009 03:41:56 +0000 (20:41 -0700)]
mergetool--lib: simplify API usage by removing more global variables

The mergetool--lib scriplet was tricky to use because it relied upon
the existance of several global shell variables.  This removes more
global variables so that things are simpler for callers.

A side effect is that some variables are recomputed each time
run_merge_tool() is called, but the overhead for recomputing
them is justified by the simpler implementation.

Signed-off-by: David Aguilar <davvid@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
15 years agoState the effect of filter-branch on graft explicitly
Daniel Cheng (aka SDiZ) [Fri, 10 Apr 2009 06:26:49 +0000 (14:26 +0800)]
State the effect of filter-branch on graft explicitly

Signed-off-by: Daniel Cheng (aka SDiZ) <j16sdiz+freenet@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
15 years agoprocess_{tree,blob}: Remove useless xstrdup calls
Linus Torvalds [Fri, 10 Apr 2009 22:20:18 +0000 (15:20 -0700)]
process_{tree,blob}: Remove useless xstrdup calls

On Wed, 8 Apr 2009, Björn Steinbrink wrote:
>
> The name of the processed object was duplicated for passing it to
> add_object(), but that already calls path_name, which allocates a new
> string anyway. So the memory allocated by the xstrdup calls just went
> nowhere, leaking memory.

Ack, ack.

There's another easy 5% or so for the built-in object walker: once we've
created the hash from the name, the name isn't interesting any more, and
so something trivial like this can help a bit.

Does it matter? Probably not on its own. But a few more memory saving
tricks and it might all make a difference.

Linus

Signed-off-by: Junio C Hamano <gitster@pobox.com>
15 years agot1301-shared-repo: fix forced modes test
Johannes Sixt [Sun, 12 Apr 2009 19:22:02 +0000 (21:22 +0200)]
t1301-shared-repo: fix forced modes test

This test was added recently (5a688fe, "core.sharedrepository = 0mode"
should set, not loosen; 2009-03-28). It checked the result of a sed
invocation for emptyness, but in some cases it forgot to print anything
at all, so that those checks would never be false.

Due to this mistake, it went unnoticed that the files in objects/info are
not necessarily 0440, but can also be 0660.  Because the 0mode setting
tries to guarantee that the files are accessible only to the people they
are meant to be used by, we should only make sure that they are readable
by the user and the group when the configuration is set to 0660.  It is a
separate matter from the core.shredrepository settings that w-bit from
immutable object files under objects/[0-9a-f][0-9a-f] directories should
be dropped.

COMMIT_EDITMSG is still world-readable, but it (and any transient files
that are meant for repositories with a work tree) does not matter.  If you
are working on a shared machine and on a sekrit stuff, the root of the
work tree would be with mode 0700 (or 0750 to allow peeking by other
people in the group), and that would mean that .git/COMMIT_EDITMSG in such
a repository would not be readable by the strangers anyway.

Also, in the real-world use case, .git/COMMIT_EDITMSG will be given to an
arbitrary editor the user happens to use, and we have no guarantee what it
does (e.g. it may create a new file with umask and replace, it may rewrite
in place, it may leave an editor backup file but use umask to create it,
etc.), and the protection of the file lies majorly on the protection of
the root of the work tree.

This test cannot be run on Windows; it requires POSIXPERM when merged to
'master'.

Signed-off-by: Johannes Sixt <j6t@kdbg.org>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
15 years agodocumentation: Makefile accounts for SHELL_PATH setting
Ben Walton [Sun, 22 Mar 2009 13:20:44 +0000 (09:20 -0400)]
documentation: Makefile accounts for SHELL_PATH setting

Ensure that the Makefile that generates and installs the Documentation is
aware of any SHELL_PATH setting.  Use this value if found or the current
setting for SHELL if not.  This is an accommodation for systems where sh
is not POSIX enough.

Signed-off-by: Ben Walton <bwalton@artsci.utoronto.ca>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
15 years agoAllow users to un-configure rename detection
Linus Torvalds [Thu, 9 Apr 2009 18:46:15 +0000 (11:46 -0700)]
Allow users to un-configure rename detection

I told people on the kernel mailing list to please use "-M" when sending
me rename patches, so that I can see what they do while reading email
rather than having to apply the patch and then look at the end result.

I also told them that if they want to make it the default, they can just
add

[diff]
renames

to their ~/.gitconfig file. And while I was thinking about that, I wanted
to also check whether you can then mark individual projects to _not_ have
that default in the per-repository .git/config file.

And you can't. Currently you cannot have a global "enable renames by
default" and then a local ".. but not for _this_ project". Why? Because if
somebody writes

[diff]
renames = no

we simply ignore it, rather than resetting "diff_detect_rename_default"
back to zero.

Fixed thusly.

Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
15 years agoReplace ",<,>,& with their respective XML entities in DAV requests
Mike Hommey [Thu, 9 Apr 2009 22:25:37 +0000 (00:25 +0200)]
Replace ",<,>,& with their respective XML entities in DAV requests

If the repo url or the user email contain XML special characters, the
remote DAV server is likely to reject the LOCK requests because the XML
is then malformed.

Signed-off-by: Mike Hommey <mh@glandium.org>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
15 years agogit-svn: always initialize with core.autocrlf=false
Johannes Schindelin [Thu, 9 Apr 2009 11:29:57 +0000 (13:29 +0200)]
git-svn: always initialize with core.autocrlf=false

It has been reported time and time again in relation to msysGit that
git-svn does not work well when core.autocrlf has any value other than
'false'.  So let's make it so by default.

Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de>
Acked-by: Eric Wong <normalperson@yhbt.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
15 years agoUpdate delta compression message to be less misleading
Dan McGee [Thu, 9 Apr 2009 15:45:39 +0000 (10:45 -0500)]
Update delta compression message to be less misleading

In the case of a small repository, pack-objects is smart enough to not
start more threads than necessary. However, the output to the user always
reports the value of the pack.threads configuration and not the real
number of threads to be used.

Signed-off-by: Dan McGee <dpmcgee@gmail.com>
Acked-by: Nicolas Pitre <nico@cam.org>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
15 years agogit-svn: Save init/clone --ignore-paths in config
Ben Jackson [Sat, 11 Apr 2009 17:46:18 +0000 (10:46 -0700)]
git-svn: Save init/clone --ignore-paths in config

The --ignored-paths argument is now stored as
"svn-remote.$REMOTE_NAME.ignore-paths" in the config file.

[ew: edited subject and message]
Signed-off-by: Ben Jackson <ben@ben.com>
Acked-by: Eric Wong <normalperson@yhbt.net>
15 years agogit-svn: Add per-svn-remote ignore-paths config
Ben Jackson [Sat, 11 Apr 2009 17:46:17 +0000 (10:46 -0700)]
git-svn: Add per-svn-remote ignore-paths config

The --ignore-paths option to fetch is very useful for working on a subset
of a SVN repository.  For proper operation, every command that causes a
fetch (explicit or implied) must include a matching --ignore-paths option.

This patch adds a persistent svn-remote.$repo_id.ignore-paths config by
promoting Fetcher::is_path_ignored to a member function and initializing
$self->{ignore_regex} in Fetcher::new.  Command line --ignore-paths is
still recognized and acts in addition to the config value.

Signed-off-by: Ben Jackson <ben@ben.com>
Acked-by: Eric Wong <normalperson@yhbt.net>
15 years agogit-svn: speed up blame command
Boris Byk [Fri, 10 Apr 2009 20:32:41 +0000 (00:32 +0400)]
git-svn: speed up blame command

'git svn blame' now uses the 'git cat-file --batch' command to
speed up resolving SVN revision number out of commit SHA by
removing fork+exec overhead.

[ew: enforced 80-column line wrap]

Signed-off-by: Boris Byk <boris.byk@gmail.com>
Acked-by: Eric Wong <normalperson@yhbt.net>
15 years agogit-svn: add fetch --parent option
Jason Merrill [Mon, 6 Apr 2009 20:37:59 +0000 (16:37 -0400)]
git-svn: add fetch --parent option

Signed-off-by: Jason Merrill <jason@redhat.com>
Acked-By: Eric Wong <normalperson@yhbt.net>
15 years agoFix misspelled mergetool.keepBackup
Ferry Huberts [Fri, 10 Apr 2009 19:33:57 +0000 (21:33 +0200)]
Fix misspelled mergetool.keepBackup

In several places mergetool.keepBackup was misspelled as merge.keepBackup.

Signed-off-by: Ferry Huberts <ferry.huberts@pelagic.nl>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
15 years agogit-shell: Add 'git-upload-archive' to allowed commands.
Erik Broes [Thu, 9 Apr 2009 19:58:52 +0000 (21:58 +0200)]
git-shell: Add 'git-upload-archive' to allowed commands.

This allows for example gitosis to allow use of 'git archive --remote' in a
controlled environment.

Signed-off-by: Erik Broes <erikbroes@ripe.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
15 years agoadd tests for remote groups
Jeff King [Mon, 6 Apr 2009 20:18:23 +0000 (16:18 -0400)]
add tests for remote groups

This tries to systematically cover existing behavior, and
also mark some expect_failure cases for desired behavior.

Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
15 years agoUpdate draft release notes to 1.6.3
Junio C Hamano [Thu, 9 Apr 2009 07:04:17 +0000 (00:04 -0700)]
Update draft release notes to 1.6.3

Signed-off-by: Junio C Hamano <gitster@pobox.com>
15 years agoMerge branch 'maint'
Junio C Hamano [Thu, 9 Apr 2009 06:41:27 +0000 (23:41 -0700)]
Merge branch 'maint'

* maint:
  Start 1.6.2.3 preparation
  process_{tree,blob}: Remove useless xstrdup calls
  git-pull.sh: better warning message for "git pull" on detached head.

Conflicts:
RelNotes

15 years agoStart 1.6.2.3 preparation
Junio C Hamano [Thu, 9 Apr 2009 06:40:33 +0000 (23:40 -0700)]
Start 1.6.2.3 preparation

Signed-off-by: Junio C Hamano <gitster@pobox.com>
15 years agoMerge branch 'jc/shared-literally' into maint
Junio C Hamano [Thu, 9 Apr 2009 06:23:41 +0000 (23:23 -0700)]
Merge branch 'jc/shared-literally' into maint

* jc/shared-literally:
  t1301: loosen test for forced modes
  set_shared_perm(): sometimes we know what the final mode bits should look like
  move_temp_to_file(): do not forget to chmod() in "Coda hack" codepath
  Move chmod(foo, 0444) into move_temp_to_file()
  "core.sharedrepository = 0mode" should set, not loosen

15 years agoMerge branch 'jc/maint-1.6.0-diff-borrow-carefully' into maint
Junio C Hamano [Thu, 9 Apr 2009 06:23:17 +0000 (23:23 -0700)]
Merge branch 'jc/maint-1.6.0-diff-borrow-carefully' into maint

* jc/maint-1.6.0-diff-borrow-carefully:
  diff --cached: do not borrow from a work tree when a path is marked as assume-unchanged

15 years agoMerge branch 'maint-1.6.1' into maint
Junio C Hamano [Thu, 9 Apr 2009 06:22:09 +0000 (23:22 -0700)]
Merge branch 'maint-1.6.1' into maint

* maint-1.6.1:
  process_{tree,blob}: Remove useless xstrdup calls

15 years agoMerge branch 'maint-1.6.0' into maint-1.6.1
Junio C Hamano [Thu, 9 Apr 2009 06:22:05 +0000 (23:22 -0700)]
Merge branch 'maint-1.6.0' into maint-1.6.1

* maint-1.6.0:
  process_{tree,blob}: Remove useless xstrdup calls

15 years agoMerge branch 'jc/maint-1.6.0-keep-pack' into maint
Junio C Hamano [Thu, 9 Apr 2009 06:21:10 +0000 (23:21 -0700)]
Merge branch 'jc/maint-1.6.0-keep-pack' into maint

* jc/maint-1.6.0-keep-pack:
  pack-objects: don't loosen objects available in alternate or kept packs
  t7700: demonstrate repack flaw which may loosen objects unnecessarily
  Remove --kept-pack-only option and associated infrastructure
  pack-objects: only repack or loosen objects residing in "local" packs
  git-repack.sh: don't use --kept-pack-only option to pack-objects
  t7700-repack: add two new tests demonstrating repacking flaws
  is_kept_pack(): final clean-up
  Simplify is_kept_pack()
  Consolidate ignore_packed logic more
  has_sha1_kept_pack(): take "struct rev_info"
  has_sha1_pack(): refactor "pretend these packs do not exist" interface
  git-repack: resist stray environment variable

Conflicts:
t/t7700-repack.sh

15 years agoMerge branch 'jc/maint-1.6.0-blame-s' into maint-1.6.1
Junio C Hamano [Thu, 9 Apr 2009 06:02:17 +0000 (23:02 -0700)]
Merge branch 'jc/maint-1.6.0-blame-s' into maint-1.6.1

* jc/maint-1.6.0-blame-s:
  blame: read custom grafts given by -S before calling setup_revisions()

Conflicts:
builtin-blame.c

15 years agoMerge branch 'bc/maint-1.6.1-branch-deleted-was' into maint-1.6.1
Junio C Hamano [Thu, 9 Apr 2009 06:01:15 +0000 (23:01 -0700)]
Merge branch 'bc/maint-1.6.1-branch-deleted-was' into maint-1.6.1

* bc/maint-1.6.1-branch-deleted-was:
  git-branch: display "was sha1" on branch deletion rather than just "sha1"

15 years agoMerge branch 'js/maint-1.6.0-exec-path-env' into maint-1.6.1
Junio C Hamano [Thu, 9 Apr 2009 06:01:10 +0000 (23:01 -0700)]
Merge branch 'js/maint-1.6.0-exec-path-env' into maint-1.6.1

* js/maint-1.6.0-exec-path-env:
  Propagate --exec-path setting to external commands via GIT_EXEC_PATH

15 years agoMerge branch 'tr/maint-1.6.1-doc-format-patch--root' into maint-1.6.1
Junio C Hamano [Thu, 9 Apr 2009 06:00:21 +0000 (23:00 -0700)]
Merge branch 'tr/maint-1.6.1-doc-format-patch--root' into maint-1.6.1

* tr/maint-1.6.1-doc-format-patch--root:
  Documentation: format-patch --root clarifications

15 years agoprocess_{tree,blob}: Remove useless xstrdup calls
Björn Steinbrink [Wed, 8 Apr 2009 11:28:54 +0000 (13:28 +0200)]
process_{tree,blob}: Remove useless xstrdup calls

The name of the processed object was duplicated for passing it to
add_object(), but that already calls path_name, which allocates a new
string anyway. So the memory allocated by the xstrdup calls just went
nowhere, leaking memory.

This reduces the RSS usage for a "rev-list --all --objects" by about 10% on
the gentoo repo (fully packed) as well as linux-2.6.git:

    gentoo:
                    | old           | new
    ----------------|-------------------------------
    RSS             |       1537284 |       1388408
    VSZ             |       1816852 |       1667952
    time elapsed    |       1:49.62 |       1:48.99
    min. page faults|        417178 |        379919

    linux-2.6.git:
                    | old           | new
    ----------------|-------------------------------
    RSS             |        324452 |        292996
    VSZ             |        491792 |        460376
    time elapsed    |       0:14.53 |       0:14.28
    min. page faults|         89360 |         81613

Signed-off-by: Björn Steinbrink <B.Steinbrink@gmx.de>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
15 years agotests: skip perl tests if NO_PERL is defined
Jeff King [Fri, 3 Apr 2009 19:33:59 +0000 (15:33 -0400)]
tests: skip perl tests if NO_PERL is defined

These scripts all test git programs that are written in
perl, and thus obviously won't work if NO_PERL is defined.
We pass NO_PERL to the scripts from the building Makefile
via the GIT-BUILD-OPTIONS file.

Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
15 years agoMakefile: allow building without perl
Jeff King [Fri, 3 Apr 2009 19:32:20 +0000 (15:32 -0400)]
Makefile: allow building without perl

For systems with a missing or broken perl, it is nicer to
explicitly say "we don't want perl" because:

  1. The Makefile knows not to bother with Perl-ish things
     like Git.pm.

  2. We can print a more user-friendly error message
     than "foo is not a git command" or whatever the broken
     perl might barf

  3. Test scripts that require perl can mark themselves and
     such and be skipped

This patch implements parts (1) and (2). The perl/
subdirectory is skipped entirely, gitweb is not built, and
any git commands which rely on perl will print a
human-readable message and exit with an error code.

This patch is based on one from Robin H. Johnson.

Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
15 years agogit-pull.sh: better warning message for "git pull" on detached head.
Matthieu Moy [Wed, 8 Apr 2009 07:24:03 +0000 (09:24 +0200)]
git-pull.sh: better warning message for "git pull" on detached head.

Otherwise, git complains about not finding a branch to pull from in
'branch..merge', which is hardly understandable. While we're there,
reword the sentences slightly.

Signed-off-by: Matthieu Moy <Matthieu.Moy@imag.fr>
Signed-off-by: Junio C Hamano <gitster@pobox.com>