OSDN Git Service

git-core/git.git
11 years agoMerge branch 'lt/commit-tree-guess-utf-8'
Junio C Hamano [Fri, 7 Sep 2012 18:08:38 +0000 (11:08 -0700)]
Merge branch 'lt/commit-tree-guess-utf-8'

Teach "git commit" and "git commit-tree" the "we are told to use
utf-8 in log message, but this does not look like utf-8---attempt to
pass it through convert-from-latin1-to-utf8 and see if it makes
sense" heuristics "git mailinfo" already uses.

* lt/commit-tree-guess-utf-8:
  commit/commit-tree: correct latin1 to utf-8

11 years agoMerge branch 'pw/p4-use-client-spec-branch-detection'
Junio C Hamano [Fri, 7 Sep 2012 18:08:27 +0000 (11:08 -0700)]
Merge branch 'pw/p4-use-client-spec-branch-detection'

Fix "git p4" when "--use-client-spec" and "--detect-branches" are
used together (the command used to misdetect branches).

* pw/p4-use-client-spec-branch-detection:
  git p4: make branch detection work with --use-client-spec
  git p4: do wildcard decoding in stripRepoPath
  git p4: set self.branchPrefixes in initialization
  git p4 test: add broken --use-client-spec --detect-branches tests
  git p4 test: move client_view() function to library

11 years agoMerge branch 'tr/maint-parseopt-avoid-empty'
Junio C Hamano [Fri, 7 Sep 2012 18:08:10 +0000 (11:08 -0700)]
Merge branch 'tr/maint-parseopt-avoid-empty'

A workaround to avoid doing _(""), which translates to unwanted
magic string in the .po files.

* tr/maint-parseopt-avoid-empty:
  gettext: do not translate empty string

11 years agoMerge branch 'nd/i18n-poison-test-updates'
Junio C Hamano [Fri, 7 Sep 2012 18:07:59 +0000 (11:07 -0700)]
Merge branch 'nd/i18n-poison-test-updates'

Update tests that can be broken with gettext-poison builds.

* nd/i18n-poison-test-updates:
  Fix tests under GETTEXT_POISON on parseopt
  Fix tests under GETTEXT_POISON on git-remote
  Fix tests under GETTEXT_POISON on pack-object
  Fix tests under GETTEXT_POISON on git-apply
  Fix tests under GETTEXT_POISON on diffstat
  Fix tests under GETTEXT_POISON on git-stash
  Fix tests under GETTEXT_POISON on relative dates

11 years agoUpdate mailmap for a handful of folks
Junio C Hamano [Mon, 3 Sep 2012 23:24:23 +0000 (16:24 -0700)]
Update mailmap for a handful of folks

Signed-off-by: Junio C Hamano <gitster@pobox.com>
11 years agoThe third batch for 1.8.0
Junio C Hamano [Mon, 3 Sep 2012 23:23:44 +0000 (16:23 -0700)]
The third batch for 1.8.0

Signed-off-by: Junio C Hamano <gitster@pobox.com>
11 years agoMerge branch 'js/use-sc-open-max'
Junio C Hamano [Mon, 3 Sep 2012 22:54:41 +0000 (15:54 -0700)]
Merge branch 'js/use-sc-open-max'

Introduce get_max_fd_limit() to absorb platforms that do not have
getrlimit(RLIMIT_NOFILE) and/or sysconf(_SC_OPEN_MAX).

* js/use-sc-open-max:
  sha1_file.c: introduce get_max_fd_limit() helper

11 years agoMerge branch 'js/compat-mkdir'
Junio C Hamano [Mon, 3 Sep 2012 22:54:37 +0000 (15:54 -0700)]
Merge branch 'js/compat-mkdir'

Some mkdir(2) implementations do not want to see trailing slash in
its parameter.

* js/compat-mkdir:
  compat: some mkdir() do not like a slash at the end

11 years agoMerge branch 'js/no-curl-easy-strerror-on-old-curl'
Junio C Hamano [Mon, 3 Sep 2012 22:54:30 +0000 (15:54 -0700)]
Merge branch 'js/no-curl-easy-strerror-on-old-curl'

Done to support compilation on __TANDEM, but is independently useful
for people with older version of libcURL.

* js/no-curl-easy-strerror-on-old-curl:
  http.c: don't use curl_easy_strerror prior to curl-7.12.0

11 years agoMerge branch 'jc/maint-mergetool-style-fix'
Junio C Hamano [Mon, 3 Sep 2012 22:54:25 +0000 (15:54 -0700)]
Merge branch 'jc/maint-mergetool-style-fix'

* jc/maint-mergetool-style-fix:
  mergetool: style fixes

11 years agoMerge branch 'sz/submodule-force-update'
Junio C Hamano [Mon, 3 Sep 2012 22:54:17 +0000 (15:54 -0700)]
Merge branch 'sz/submodule-force-update'

"git submodule update --force" used to leave the working tree of the
submodule intact when there were local changes.  It is more intiutive
to make "--force" a sign to run "checkout -f" to overwrite them.

* sz/submodule-force-update:
  Make 'git submodule update --force' always check out submodules.

11 years agoMerge branch 'ph/stash-rerere'
Junio C Hamano [Mon, 3 Sep 2012 22:54:12 +0000 (15:54 -0700)]
Merge branch 'ph/stash-rerere'

"git stash" internally used "git merge-recursive" backend, which did
not trigger "rerere" upon conflicts unlike other mergy operations.

* ph/stash-rerere:
  stash: invoke rerere in case of conflict
  test: git-stash conflict sets up rerere

11 years agoMerge branch 'jc/daemon-access-hook'
Junio C Hamano [Mon, 3 Sep 2012 22:54:03 +0000 (15:54 -0700)]
Merge branch 'jc/daemon-access-hook'

Allow an external command to tell git-daemon to decline service
based on the client address, repository path, etc.

* jc/daemon-access-hook:
  daemon: --access-hook option

11 years agoMerge branch 'jc/send-email-reconfirm'
Junio C Hamano [Mon, 3 Sep 2012 22:53:54 +0000 (15:53 -0700)]
Merge branch 'jc/send-email-reconfirm'

Validate interactive input to "git send-email" to avoid common
mistakes such as saying "y<RETURN>" to sender mail address whose
prompt is given with a correctly guessed default.

* jc/send-email-reconfirm:
  send-email: validate & reconfirm interactive responses

11 years agoMerge branch 'cw/cherry-pick-allow-empty-message'
Junio C Hamano [Mon, 3 Sep 2012 22:53:37 +0000 (15:53 -0700)]
Merge branch 'cw/cherry-pick-allow-empty-message'

"git cherry-pick" by default stops when it sees a commit without any
log message.  The "--allow-empty-message" option can be used to
silently proceed.

* cw/cherry-pick-allow-empty-message:
  cherry-pick: add --allow-empty-message option

11 years agoMerge branch 'jc/maint-sane-execvp-notdir'
Junio C Hamano [Mon, 3 Sep 2012 22:53:26 +0000 (15:53 -0700)]
Merge branch 'jc/maint-sane-execvp-notdir'

"git foo" errored out with "Not a directory" when the user had a non
directory on $PATH, and worse yet it masked an alias "foo" to run.

* jc/maint-sane-execvp-notdir:
  sane_execvp(): ignore non-directory on $PATH

11 years agoMerge branch 'jc/maint-config-exit-status'
Junio C Hamano [Mon, 3 Sep 2012 22:53:06 +0000 (15:53 -0700)]
Merge branch 'jc/maint-config-exit-status'

The exit status code from "git config" was way overspecified while
being incorrect.  Update the implementation to give the documented
status for a case that was documented, and introduce a new code for
"all other errors".

* jc/maint-config-exit-status:
  config: "git config baa" should exit with status 1

11 years agoLatter half of the second batch for 1.8.0
Junio C Hamano [Wed, 29 Aug 2012 22:00:30 +0000 (15:00 -0700)]
Latter half of the second batch for 1.8.0

Signed-off-by: Junio C Hamano <gitster@pobox.com>
11 years agoMerge branch 'rr/precompose-utf8-cleanup'
Junio C Hamano [Wed, 29 Aug 2012 21:50:35 +0000 (14:50 -0700)]
Merge branch 'rr/precompose-utf8-cleanup'

* rr/precompose-utf8-cleanup:
  precompose-utf8: do not call checks for non-ascii "utf8"
  cleanup precompose_utf8

11 years agoMerge branch 'mh/maint-config-doc-proxy-command'
Junio C Hamano [Wed, 29 Aug 2012 21:50:30 +0000 (14:50 -0700)]
Merge branch 'mh/maint-config-doc-proxy-command'

* mh/maint-config-doc-proxy-command:
  git-config doc: unconfuse an example
  git-config.txt: fix example

11 years agoMerge branch 'ef/win32-cred-helper'
Junio C Hamano [Wed, 29 Aug 2012 21:50:24 +0000 (14:50 -0700)]
Merge branch 'ef/win32-cred-helper'

Credential helper for Win32 to allow access to the keychain of
the logged-in user.

* ef/win32-cred-helper:
  contrib: add win32 credential-helper

11 years agoMerge branch 'hv/submodule-path-unmatch'
Junio C Hamano [Wed, 29 Aug 2012 21:50:15 +0000 (14:50 -0700)]
Merge branch 'hv/submodule-path-unmatch'

* hv/submodule-path-unmatch:
  Let submodule command exit with error status if path does not exist

11 years agoMerge branch 'jc/capabilities'
Junio C Hamano [Wed, 29 Aug 2012 21:50:07 +0000 (14:50 -0700)]
Merge branch 'jc/capabilities'

Some capabilities were asked by fetch-pack even when upload-pack did
not advertise that they are available.  Fix fetch-pack not to do so.

* jc/capabilities:
  fetch-pack: mention server version with verbose output
  parse_feature_request: make it easier to see feature values
  fetch-pack: do not ask for unadvertised capabilities
  do not send client agent unless server does first
  send-pack: fix capability-sending logic
  include agent identifier in capability string

11 years agoMerge branch 'mg/rebase-i-onto-reflog-in-full'
Junio C Hamano [Wed, 29 Aug 2012 21:50:01 +0000 (14:50 -0700)]
Merge branch 'mg/rebase-i-onto-reflog-in-full'

The reflog entries left by "git rebase" and "git rebase -i" were
inconsistent.

* mg/rebase-i-onto-reflog-in-full:
  rebase -i: use full onto sha1 in reflog

11 years agoMerge branch 'mz/empty-rebase-test'
Junio C Hamano [Wed, 29 Aug 2012 21:49:57 +0000 (14:49 -0700)]
Merge branch 'mz/empty-rebase-test'

* mz/empty-rebase-test:
  add tests for 'git rebase --keep-empty'

11 years agoMerge branch 'jc/maint-protect-sh-from-ifs'
Junio C Hamano [Wed, 29 Aug 2012 21:49:45 +0000 (14:49 -0700)]
Merge branch 'jc/maint-protect-sh-from-ifs'

When the user exports a non-default IFS without HT, scripts that
rely on being able to parse "ls-files -s | while read a b c..."
start to fail.  Protect them from such a misconfiguration.

* jc/maint-protect-sh-from-ifs:
  sh-setup: protect from exported IFS

11 years agoMerge branch 'bc/prune-info'
Junio C Hamano [Wed, 29 Aug 2012 21:49:39 +0000 (14:49 -0700)]
Merge branch 'bc/prune-info'

Teach "git prune" without "-v" to be silent about leftover temporary
files.

* bc/prune-info:
  prune.c: only print informational message in show_only or verbose mode

11 years agoMerge branch 'mz/cherry-code-cleanup'
Junio C Hamano [Wed, 29 Aug 2012 21:49:29 +0000 (14:49 -0700)]
Merge branch 'mz/cherry-code-cleanup'

Minor code clean-up on the cherry-pick codepath.

* mz/cherry-code-cleanup:
  cherry: remove redundant check for merge commit
  cherry: don't set ignored rev_info options
  remove unnecessary parameter from get_patch_ids()

11 years agoFirst half of the second batch for 1.8.0
Junio C Hamano [Mon, 27 Aug 2012 19:10:07 +0000 (12:10 -0700)]
First half of the second batch for 1.8.0

Signed-off-by: Junio C Hamano <gitster@pobox.com>
11 years agoMerge branch 'jk/docs-docbook-monospace-display'
Junio C Hamano [Mon, 27 Aug 2012 18:55:28 +0000 (11:55 -0700)]
Merge branch 'jk/docs-docbook-monospace-display'

The documentation in the TeXinfo format was using indented output
for materials meant to be examples that are better typeset in
monospace.

* jk/docs-docbook-monospace-display:
  docs: monospace listings in docbook output

11 years agoMerge branch 'da/difftool-updates'
Junio C Hamano [Mon, 27 Aug 2012 18:55:16 +0000 (11:55 -0700)]
Merge branch 'da/difftool-updates'

"git difftool --dir-diff" learned to use symbolic links to prepare
temporary copy of the working tree when available.

* da/difftool-updates:
  difftool: silence warning
  Add Code Compare v2.80.4 as a merge / diff tool for Windows
  mergetool,difftool: Document --tool-help consistently
  difftool: Disable --symlinks on cygwin
  difftool: Handle compare() returning -1
  difftool: Wrap long lines for readability
  difftool: Check all return codes from compare()
  difftool: Handle finding mergetools/ in a path with spaces
  difftool: Use symlinks when diffing against the worktree
  difftool: Call the temp directory "git-difftool"
  difftool: Move option values into a hash
  difftool: Eliminate global variables
  difftool: Simplify print_tool_help()

11 years agoMerge branch 'js/grep-patterntype-config'
Junio C Hamano [Mon, 27 Aug 2012 18:55:09 +0000 (11:55 -0700)]
Merge branch 'js/grep-patterntype-config'

"grep" learned to use a non-standard pattern type by default if a
configuration variable tells it to.

* js/grep-patterntype-config:
  grep: add a grep.patternType configuration setting

11 years agoMerge branch 'bc/receive-pack-stdout-protection'
Junio C Hamano [Mon, 27 Aug 2012 18:55:00 +0000 (11:55 -0700)]
Merge branch 'bc/receive-pack-stdout-protection'

When "git push" triggered the automatic gc on the receiving end, a
message from "git prune" that said it was removing cruft leaked to
the standard output, breaking the communication protocol.

* bc/receive-pack-stdout-protection:
  receive-pack: do not leak output from auto-gc to standard output
  t/t5400: demonstrate breakage caused by informational message from prune

11 years agoMerge branch 'ab/diff-write-incomplete-line'
Junio C Hamano [Mon, 27 Aug 2012 18:54:46 +0000 (11:54 -0700)]
Merge branch 'ab/diff-write-incomplete-line'

The output from "git diff -B" for a file that ends with an
incomplete line did not put "\ No newline..." on a line of its own.

* ab/diff-write-incomplete-line:
  Fix '\ No newline...' annotation in rewrite diffs

11 years agoMerge branch 'jc/maint-t7406-rev-parse-max-count-huh'
Junio C Hamano [Mon, 27 Aug 2012 18:54:36 +0000 (11:54 -0700)]
Merge branch 'jc/maint-t7406-rev-parse-max-count-huh'

Test clean-up, with no behaviour change.

* jc/maint-t7406-rev-parse-max-count-huh:
  t7406: fix misleading "rev-parse --max-count=1 HEAD"

11 years agoMerge branch 'jk/maint-null-in-trees'
Junio C Hamano [Mon, 27 Aug 2012 18:54:28 +0000 (11:54 -0700)]
Merge branch 'jk/maint-null-in-trees'

We do not want a link to 0{40} object stored anywhere in our objects.

* jk/maint-null-in-trees:
  fsck: detect null sha1 in tree entries
  do not write null sha1s to on-disk index
  diff: do not use null sha1 as a sentinel value

11 years agoMerge branch 'mm/push-default-switch-warning'
Junio C Hamano [Mon, 27 Aug 2012 18:54:04 +0000 (11:54 -0700)]
Merge branch 'mm/push-default-switch-warning'

In the next major release, we will switch "git push [$there]" that
does not say what to push from the traditional "matching" to the
updated "simple" semantics, that pushes the current branch to the
branch with the same name only when the current branch is set to
integrate with that remote branch (all other cases will error out).

* mm/push-default-switch-warning:
  push: start warning upcoming default change for push.default

11 years agoFix tests under GETTEXT_POISON on parseopt
Jiang Xin [Mon, 27 Aug 2012 05:36:55 +0000 (13:36 +0800)]
Fix tests under GETTEXT_POISON on parseopt

Use the i18n-specific test functions in test scripts for parseopt tests.
This issue was was introduced in v1.7.10.1-488-g54e6d:

    54e6d i18n: parseopt: lookup help and argument translations when showing usage

and been broken under GETTEXT_POISON=YesPlease since.

Signed-off-by: Jiang Xin <worldhello.net@gmail.com>
Signed-off-by: Nguyễn Thái Ngọc Duy <pclouds@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
11 years agoFix tests under GETTEXT_POISON on git-remote
Jiang Xin [Mon, 27 Aug 2012 05:36:54 +0000 (13:36 +0800)]
Fix tests under GETTEXT_POISON on git-remote

Use the i18n-specific test functions in test scripts for git-remote.
This issue was was introduced in v1.7.10-233-gbb16d5:

    bb16d5 i18n: remote: mark strings for translation

and been broken under GETTEXT_POISON=YesPlease since.

Signed-off-by: Jiang Xin <worldhello.net@gmail.com>
Signed-off-by: Nguyễn Thái Ngọc Duy <pclouds@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
11 years agoFix tests under GETTEXT_POISON on pack-object
Jiang Xin [Mon, 27 Aug 2012 05:36:53 +0000 (13:36 +0800)]
Fix tests under GETTEXT_POISON on pack-object

Use the i18n-specific test functions in test scripts for pack-object.
This issue was was introduced in v1.7.10.2-556-g46140:

    46140 index-pack: use streaming interface for collision test on large blobs
    cf2ba pack-objects: use streaming interface for reading large loose blobs

and been broken under GETTEXT_POISON=YesPlease since.

Signed-off-by: Jiang Xin <worldhello.net@gmail.com>
Signed-off-by: Nguyễn Thái Ngọc Duy <pclouds@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
11 years agoFix tests under GETTEXT_POISON on git-apply
Jiang Xin [Mon, 27 Aug 2012 05:36:52 +0000 (13:36 +0800)]
Fix tests under GETTEXT_POISON on git-apply

Use the i18n-specific test functions in test scripts for git-apply.
This issue was was introduced in the following commits:

    de373 i18n: apply: mark parseopt strings for translation
    3638e i18n: apply: mark strings for translation

and been broken under GETTEXT_POISON=YesPlease since.

Signed-off-by: Jiang Xin <worldhello.net@gmail.com>
Signed-off-by: Nguyễn Thái Ngọc Duy <pclouds@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
11 years agoFix tests under GETTEXT_POISON on diffstat
Jiang Xin [Mon, 27 Aug 2012 05:36:51 +0000 (13:36 +0800)]
Fix tests under GETTEXT_POISON on diffstat

Use the i18n-specific test functions in test scripts for diffstat.
This issue was was introduced in v1.7.9-1-g7f814:

    7f814 Use correct grammar in diffstat summary line

and been broken under GETTEXT_POISON=YesPlease since.

Signed-off-by: Jiang Xin <worldhello.net@gmail.com>
Signed-off-by: Nguyễn Thái Ngọc Duy <pclouds@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
11 years agoFix tests under GETTEXT_POISON on git-stash
Jiang Xin [Mon, 27 Aug 2012 05:36:50 +0000 (13:36 +0800)]
Fix tests under GETTEXT_POISON on git-stash

Use i18n-specific test functions in test scripts for git-stash.
This issue was was introduced in v1.7.4.1-119-g355ec:

    355ec i18n: git-status basic messages

and been broken under GETTEXT_POISON=YesPlease since.

Signed-off-by: Jiang Xin <worldhello.net@gmail.com>
Signed-off-by: Nguyễn Thái Ngọc Duy <pclouds@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
11 years agoFix tests under GETTEXT_POISON on relative dates
Jiang Xin [Mon, 27 Aug 2012 05:36:49 +0000 (13:36 +0800)]
Fix tests under GETTEXT_POISON on relative dates

Use the i18n-specific test_i18ncmp in t/t0006-date.sh for relative dates
tests. This issue was was introduced in v1.7.10-230-g7d29a:

    7d29a i18n: mark relative dates for translation

and been broken under GETTEXT_POISON=YesPlease since.

Signed-off-by: Jiang Xin <worldhello.net@gmail.com>
Signed-off-by: Nguyễn Thái Ngọc Duy <pclouds@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
11 years agoMerge branch 'maint'
Junio C Hamano [Fri, 24 Aug 2012 20:13:53 +0000 (13:13 -0700)]
Merge branch 'maint'

* maint:
  Prepare for 1.7.11.6
  Make the ciabot scripts completely self-configuring in the normal case.
  Improved documentation for the ciabot scripts.

11 years agoMerge branch 'maint-1.7.11' into maint
Junio C Hamano [Fri, 24 Aug 2012 19:34:19 +0000 (12:34 -0700)]
Merge branch 'maint-1.7.11' into maint

* maint-1.7.11:
  Prepare for 1.7.11.6
  Make the ciabot scripts completely self-configuring in the normal case.
  Improved documentation for the ciabot scripts.
  man: git pull -r is a short for --rebase
  gitcli: describe abbreviation of long options
  rev-list docs: clarify --topo-order description
  Documentation/CodingGuidelines: spell out more shell guidelines
  Documentation: do not mention .git/refs/* directories
  tests: Introduce test_seq

11 years agoPrepare for 1.7.11.6
Junio C Hamano [Fri, 24 Aug 2012 19:33:31 +0000 (12:33 -0700)]
Prepare for 1.7.11.6

Signed-off-by: Junio C Hamano <gitster@pobox.com>
11 years agoMerge branch 'mv/pull-r-for-rebase' into maint-1.7.11
Junio C Hamano [Fri, 24 Aug 2012 19:05:47 +0000 (12:05 -0700)]
Merge branch 'mv/pull-r-for-rebase' into maint-1.7.11

A minor documentation update.

* mv/pull-r-for-rebase:
  man: git pull -r is a short for --rebase

11 years agoMerge branch 'jc/maint-abbrev-option-cli' into maint-1.7.11
Junio C Hamano [Fri, 24 Aug 2012 19:05:44 +0000 (12:05 -0700)]
Merge branch 'jc/maint-abbrev-option-cli' into maint-1.7.11

We did not document that many commands take unique prefix
abbreviations of long options (e.g. "--option" may be the only flag
that the command accepts that begin with "--opt", in which case you
can give "--opt") anywhere easy to find for new people.

* jc/maint-abbrev-option-cli:
  gitcli: describe abbreviation of long options

11 years agoMerge branch 'jc/maint-rev-list-topo-doc' into maint-1.7.11
Junio C Hamano [Fri, 24 Aug 2012 19:05:40 +0000 (12:05 -0700)]
Merge branch 'jc/maint-rev-list-topo-doc' into maint-1.7.11

It was unclear what "--topo-order" was really about in the
documentation. It is not just about "children before parent", but
also about "don't mix lineages".

* jc/maint-rev-list-topo-doc:
  rev-list docs: clarify --topo-order description

11 years agoMerge branch 'hv/coding-guidelines' into maint-1.7.11
Junio C Hamano [Fri, 24 Aug 2012 19:05:35 +0000 (12:05 -0700)]
Merge branch 'hv/coding-guidelines' into maint-1.7.11

In earlier days, "imitate the style in the neibouring code" was
sufficient to keep the coherent style, but over time some parts of
the codebase have drifted enough to make it ineffective.

* hv/coding-guidelines:
  Documentation/CodingGuidelines: spell out more shell guidelines

11 years agoMerge branch 'jc/tag-doc' into maint-1.7.11
Junio C Hamano [Fri, 24 Aug 2012 19:05:30 +0000 (12:05 -0700)]
Merge branch 'jc/tag-doc' into maint-1.7.11

Our documentation used to assume having files in .git/refs/*
directories was the only to have branches and tags, but that is not
true for quite some time.

* jc/tag-doc:
  Documentation: do not mention .git/refs/* directories

11 years agoMerge branch 'mk/test-seq' into maint-1.7.11
Junio C Hamano [Fri, 24 Aug 2012 19:05:24 +0000 (12:05 -0700)]
Merge branch 'mk/test-seq' into maint-1.7.11

Add a compatibility/utility function to the test framework.

* mk/test-seq:
  tests: Introduce test_seq

11 years agoMerge branch 'lp/no-cmd-http-fetch' into maint-1.7.11
Junio C Hamano [Fri, 24 Aug 2012 19:05:19 +0000 (12:05 -0700)]
Merge branch 'lp/no-cmd-http-fetch' into maint-1.7.11

* lp/no-cmd-http-fetch:
  builtin.h: remove unused cmd_<foo> declarations

11 years agoMerge branch 'bw/maint-1.7.9-solaris-getpass' into maint-1.7.11
Junio C Hamano [Fri, 24 Aug 2012 19:05:11 +0000 (12:05 -0700)]
Merge branch 'bw/maint-1.7.9-solaris-getpass' into maint-1.7.11

* bw/maint-1.7.9-solaris-getpass:
  Enable HAVE_DEV_TTY for Solaris
  terminal: seek when switching between reading and writing

11 years agoMerge branch 'jk/maint-commit-check-committer-early' into maint-1.7.11
Junio C Hamano [Fri, 24 Aug 2012 19:05:08 +0000 (12:05 -0700)]
Merge branch 'jk/maint-commit-check-committer-early' into maint-1.7.11

* jk/maint-commit-check-committer-early:
  commit: check committer identity more strictly

11 years agocompat: some mkdir() do not like a slash at the end
Joachim Schmitz [Fri, 24 Aug 2012 10:31:03 +0000 (12:31 +0200)]
compat: some mkdir() do not like a slash at the end

Introduce a compatibility helper for platforms with such a mkdir().

Signed-off-by: Joachim Schmitz <jojo@schmitz-digital.de>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
11 years agosha1_file.c: introduce get_max_fd_limit() helper
Joachim Schmitz [Fri, 24 Aug 2012 09:52:22 +0000 (11:52 +0200)]
sha1_file.c: introduce get_max_fd_limit() helper

Not all platforms have getrlimit(), but there are other ways to see
the maximum number of files that a process can have open.  If
getrlimit() is unavailable, fall back to sysconf(_SC_OPEN_MAX) if
available, and use OPEN_MAX from <limits.h>.

Signed-off-by: Joachim Schmitz <jojo@schmitz-digital.de>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
11 years agoMake 'git submodule update --force' always check out submodules.
Stefan Zager [Wed, 25 Jul 2012 17:41:54 +0000 (10:41 -0700)]
Make 'git submodule update --force' always check out submodules.

Currently, it will only do a checkout if the sha1 registered in the containing
repository doesn't match the HEAD of the submodule, regardless of whether the
submodule is dirty.  As discussed on the mailing list, the '--force' flag is a
strong indicator that the state of the submodule is suspect, and should be reset
to HEAD.

Signed-off-by: Stefan Zager <szager@google.com>
Acked-by: Jens Lehmann <Jens.Lehmann@web.de>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
11 years agomergetool: style fixes
Junio C Hamano [Thu, 23 Aug 2012 05:33:15 +0000 (22:33 -0700)]
mergetool: style fixes

This script is one of the sizeable ones that tempted people to copy
its "neibouring style" in their new code, but was littered with
styles incompatible with our style guide.

 - use one tab, not four spaces, per indent level;

 - long lines can be wrapped after '|', '&&', or '||' for
   readability.

 - structures like "if .. then .. else .. fi", "while .. do .. done"
   are split into lines in such a way that does not require
   unnecessary semicolon.

 - case, esac and case-arms align at the same column.

Signed-off-by: Junio C Hamano <gitster@pobox.com>
11 years agoMake the ciabot scripts completely self-configuring in the normal case.
Eric S. Raymond [Thu, 23 Aug 2012 05:21:53 +0000 (01:21 -0400)]
Make the ciabot scripts completely self-configuring in the normal case.

Signed-off-by: Eric S. Raymond <esr@thyrsus.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
11 years agoImproved documentation for the ciabot scripts.
Eric S. Raymond [Thu, 23 Aug 2012 04:10:53 +0000 (00:10 -0400)]
Improved documentation for the ciabot scripts.

Signed-off-by: Eric S. Raymond <esr@thyrsus.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
11 years agohttp.c: don't use curl_easy_strerror prior to curl-7.12.0
Joachim Schmitz [Thu, 23 Aug 2012 07:33:55 +0000 (09:33 +0200)]
http.c: don't use curl_easy_strerror prior to curl-7.12.0

Reverts be22d92 (http: avoid empty error messages for some curl
errors, 2011-09-05) on platforms with older versions of libcURL
where the function is not available.

Signed-off-by: Joachim Schmitz <jojo@schmitz-digital.de>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
11 years agoKick off cycle towards 1.8.0
Junio C Hamano [Wed, 22 Aug 2012 18:49:26 +0000 (11:49 -0700)]
Kick off cycle towards 1.8.0

Signed-off-by: Junio C Hamano <gitster@pobox.com>
11 years agoMerge branch 'maint'
Junio C Hamano [Wed, 22 Aug 2012 18:53:58 +0000 (11:53 -0700)]
Merge branch 'maint'

* maint:
  contrib/ciabot: Get ciabot configuration from git variables

11 years agoMerge branch 'jc/doc-git-updates'
Junio C Hamano [Wed, 22 Aug 2012 18:53:36 +0000 (11:53 -0700)]
Merge branch 'jc/doc-git-updates'

A minor documentation update.

* jc/doc-git-updates:
  Documentation: update the introductory section

11 years agoMerge branch 'mv/pull-r-for-rebase'
Junio C Hamano [Wed, 22 Aug 2012 18:53:31 +0000 (11:53 -0700)]
Merge branch 'mv/pull-r-for-rebase'

A minor documentation update.

* mv/pull-r-for-rebase:
  man: git pull -r is a short for --rebase

11 years agoMerge branch 'jc/maint-abbrev-option-cli'
Junio C Hamano [Wed, 22 Aug 2012 18:53:25 +0000 (11:53 -0700)]
Merge branch 'jc/maint-abbrev-option-cli'

We did not document that many commands take unique prefix
abbreviations of long options (e.g. "--option" may be the only flag
that the command accepts that begin with "--opt", in which case you
can give "--opt") anywhere easy to find for new people.

* jc/maint-abbrev-option-cli:
  gitcli: describe abbreviation of long options

11 years agoMerge branch 'jc/maint-rev-list-topo-doc'
Junio C Hamano [Wed, 22 Aug 2012 18:53:20 +0000 (11:53 -0700)]
Merge branch 'jc/maint-rev-list-topo-doc'

It was unclear what "--topo-order" was really about in the
documentation. It is not just about "children before parent", but
also about "don't mix lineages".

* jc/maint-rev-list-topo-doc:
  rev-list docs: clarify --topo-order description

11 years agoMerge branch 'hv/coding-guidelines'
Junio C Hamano [Wed, 22 Aug 2012 18:53:07 +0000 (11:53 -0700)]
Merge branch 'hv/coding-guidelines'

In earlier days, "imitate the style in the neibouring code" was
sufficient to keep the coherent style, but over time some parts of
the codebase have drifted enough to make it ineffective.

* hv/coding-guidelines:
  Documentation/CodingGuidelines: spell out more shell guidelines

11 years agoMerge branch 'jk/check-docs-update'
Junio C Hamano [Wed, 22 Aug 2012 18:53:00 +0000 (11:53 -0700)]
Merge branch 'jk/check-docs-update'

Simplify "make check-docs" implementation and update its coverage.

* jk/check-docs-update:
  check-docs: get documented command list from Makefile
  check-docs: drop git-help special-case
  check-docs: list git-gui as a command
  check-docs: factor out command-list
  command-list: mention git-credential-* helpers
  command-list: add git-sh-i18n
  check-docs: update non-command documentation list
  check-docs: mention gitweb specially

11 years agoMerge branch 'jc/tag-doc'
Junio C Hamano [Wed, 22 Aug 2012 18:52:55 +0000 (11:52 -0700)]
Merge branch 'jc/tag-doc'

Our documentation used to assume having files in .git/refs/*
directories was the only to have branches and tags, but that is not
true for quite some time.

* jc/tag-doc:
  Documentation: do not mention .git/refs/* directories

11 years agoMerge branch 'js/gitweb-path-info-unquote'
Junio C Hamano [Wed, 22 Aug 2012 18:52:32 +0000 (11:52 -0700)]
Merge branch 'js/gitweb-path-info-unquote'

"gitweb" when used with PATH_INFO failed to notice directories with
SP (and other characters that need URL-style quoting) in them.

* js/gitweb-path-info-unquote:
  gitweb: URL-decode $my_url/$my_uri when stripping PATH_INFO

11 years agoMerge branch 'tr/void-diff-setup-done'
Junio C Hamano [Wed, 22 Aug 2012 18:52:26 +0000 (11:52 -0700)]
Merge branch 'tr/void-diff-setup-done'

Remove unnecessary code.

* tr/void-diff-setup-done:
  diff_setup_done(): return void

11 years agoMerge branch 'tr/merge-recursive-flush'
Junio C Hamano [Wed, 22 Aug 2012 18:52:19 +0000 (11:52 -0700)]
Merge branch 'tr/merge-recursive-flush'

Remove unnecessary code.

* tr/merge-recursive-flush:
  merge-recursive: eliminate flush_buffer() in favor of write_in_full()

11 years agoMerge branch 'mm/die-with-dashdash-help'
Junio C Hamano [Wed, 22 Aug 2012 18:51:53 +0000 (11:51 -0700)]
Merge branch 'mm/die-with-dashdash-help'

When the user gives an argument that can be taken as both a revision
name and a pathname without disambiguating with "--", we used to
give a help message "Use '--' to separate".  The message has been
clarified to show where that '--' goes on the command line.

* mm/die-with-dashdash-help:
  setup: clarify error messages for file/revisions ambiguity

11 years agoMerge branch 'tr/maint-send-email-2047'
Junio C Hamano [Wed, 22 Aug 2012 18:51:47 +0000 (11:51 -0700)]
Merge branch 'tr/maint-send-email-2047'

"git send-email" did not unquote encoded words that appear on the
header correctly, and lost "_" from strings.

* tr/maint-send-email-2047:
  send-email: improve RFC2047 quote parsing

11 years agoMerge branch 'nd/index-errno'
Junio C Hamano [Wed, 22 Aug 2012 18:51:42 +0000 (11:51 -0700)]
Merge branch 'nd/index-errno'

Assignments to errno before calling system functions that used to
matter in the old code were left behind after the code structure
changed sufficiently to make them useless.

* nd/index-errno:
  read_index_from: remove bogus errno assignments

11 years agoMerge branch 'mk/test-seq'
Junio C Hamano [Wed, 22 Aug 2012 18:51:37 +0000 (11:51 -0700)]
Merge branch 'mk/test-seq'

Add a compatibility/utility function to the test framework.

* mk/test-seq:
  tests: Introduce test_seq

11 years agoMerge branch 'pg/maint-apply-remove-unused-variable'
Junio C Hamano [Wed, 22 Aug 2012 18:51:33 +0000 (11:51 -0700)]
Merge branch 'pg/maint-apply-remove-unused-variable'

Remove an unused field.

* pg/maint-apply-remove-unused-variable:
  apply: delete unused deflate_origlen from patch struct

11 years agoMerge branch 'jc/test-prereq'
Junio C Hamano [Wed, 22 Aug 2012 18:51:26 +0000 (11:51 -0700)]
Merge branch 'jc/test-prereq'

Teaches the test framework to probe rarely used prerequistes lazily,
and make use of it for detecting SYMLINKS, CASE_INSENSITIVE_FS and
NKD/NKC MacOS x gotcha.

* jc/test-prereq:
  t3910: use the UTF8_NFD_TO_NFC test prereq
  test-lib: provide UTF8 behaviour as a prerequisite
  t0050: use the SYMLINKS test prereq
  t0050: use the CASE_INSENSITIVE_FS test prereq
  test-lib: provide case insensitivity as a prerequisite
  test: allow prerequisite to be evaluated lazily
  test: rename $satisfied to $satisfied_prereq

11 years agoMerge branch 'ms/git-svn-1.7'
Junio C Hamano [Wed, 22 Aug 2012 18:51:20 +0000 (11:51 -0700)]
Merge branch 'ms/git-svn-1.7'

A series by Michael Schwern via Eric to update git-svn to revamp the
way URLs are internally passed around, to make it work with SVN 1.7.

* ms/git-svn-1.7:
  git-svn: remove ad-hoc canonicalizations
  git-svn: canonicalize newly-minted URLs
  git-svn: introduce add_path_to_url function
  git-svn: canonicalize earlier
  git-svn: replace URL escapes with canonicalization
  git-svn: attempt to mimic SVN 1.7 URL canonicalization
  t9107: fix typo
  t9118: workaround inconsistency between SVN versions
  Git::SVN{,::Ra}: canonicalize earlier
  git-svn: path canonicalization uses SVN API
  Git::SVN::Utils: remove irrelevant comment
  git-svn: add join_paths() to safely concatenate paths
  git-svn: factor out _collapse_dotdot function
  git-svn: use SVN 1.7 to canonicalize when possible
  git-svn: move canonicalization to Git::SVN::Utils
  use Git::SVN{,::RA}->url accessor globally
  use Git::SVN->path accessor globally
  Git::SVN::Ra: use accessor for URLs
  Git::SVN: use accessor for URLs internally
  Git::SVN: use accessors internally for path

11 years agot3910: use the UTF8_NFD_TO_NFC test prereq
Michael J Gruber [Mon, 30 Jul 2012 09:57:18 +0000 (11:57 +0200)]
t3910: use the UTF8_NFD_TO_NFC test prereq

Besides reusing the new test prerequisite, this fixes also the issue
that the current output is not TAP compliant and produces the output "no
reason given" [for skipping].

Signed-off-by: Michael J Gruber <git@drmicha.warpmail.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
11 years agoMerge branch 'maint-1.7.11' into maint
Junio C Hamano [Wed, 22 Aug 2012 18:27:30 +0000 (11:27 -0700)]
Merge branch 'maint-1.7.11' into maint

* maint-1.7.11:
  contrib/ciabot: Get ciabot configuration from git variables

11 years agocontrib/ciabot: Get ciabot configuration from git variables
Eric S. Raymond [Wed, 22 Aug 2012 10:52:30 +0000 (06:52 -0400)]
contrib/ciabot: Get ciabot configuration from git variables

These changes remove all need to modify the ciabot scripts for installation.
Instead, per-project configuration can be dome via variables in a [ciabot]
section of the config file.

Also, correct for the new server address.

Signed-off-by: Eric S. Raymond <esr@thyrsus.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
11 years agocommit/commit-tree: correct latin1 to utf-8
Linus Torvalds [Thu, 28 Jun 2012 18:24:14 +0000 (11:24 -0700)]
commit/commit-tree: correct latin1 to utf-8

When a line in the message is not a valid utf-8, "git mailinfo"
attempts to convert it to utf-8 assuming the input is latin1 (and
punt if it does not convert cleanly).  Using the same heuristics in
"git commit" and "git commit-tree" lets the editor output be in
latin1 to make the overall system more consistent.

Signed-off-by: Junio C Hamano <gitster@pobox.com>
11 years agodifftool: silence warning
Ross Lagerwall [Tue, 21 Aug 2012 10:21:40 +0000 (12:21 +0200)]
difftool: silence warning

Silence a warning given when running git difftool --dir-diff and
there are no changes.

This is because command_oneline returns undef when the command has no
output, not ''.

Signed-off-by: Ross Lagerwall <rosslagerwall@gmail.com>
Acked-by: David Aguilar <davvid@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
11 years agogettext: do not translate empty string
Thomas Rast [Mon, 20 Aug 2012 18:24:56 +0000 (20:24 +0200)]
gettext: do not translate empty string

The gettext .po files have a header, but it looks like the
translation specification for an empty string.  This results in
_("") actually returning that header.

Check the input to _() and do not call gettext() on an empty string;
in some places, we run _(opts->help) where opts->help may be empty.

Signed-off-by: Thomas Rast <trast@student.ethz.ch>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
11 years agogit p4: make branch detection work with --use-client-spec
Pete Wyckoff [Sat, 11 Aug 2012 16:55:04 +0000 (12:55 -0400)]
git p4: make branch detection work with --use-client-spec

The bug report in http://stackoverflow.com/questions/11893688
observes that files are mapped into the wrong locations in
git when both --use-client-spec and --branch-detection are enabled.

Fix this by changing the relative path prefix to match discovered
branches when using a client spec.

The problem was likely introduced with ecb7cf9 (git-p4: rewrite view
handling, 2012-01-02).

Signed-off-by: Pete Wyckoff <pw@padd.com>
Tested-by: Matthew Korich <matthew@korich.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
11 years agoprecompose-utf8: do not call checks for non-ascii "utf8"
Junio C Hamano [Mon, 20 Aug 2012 18:12:58 +0000 (11:12 -0700)]
precompose-utf8: do not call checks for non-ascii "utf8"

As suggested by Linus, this function is not checking UTF-8-ness of the
string; it only is seeing if it is pure US-ASCII.

Signed-off-by: Junio C Hamano <gitster@pobox.com>
11 years agoGit 1.7.12 v1.7.12
Junio C Hamano [Mon, 20 Aug 2012 00:02:11 +0000 (17:02 -0700)]
Git 1.7.12

Signed-off-by: Junio C Hamano <gitster@pobox.com>
12 years agogit-config doc: unconfuse an example
Junio C Hamano [Sat, 18 Aug 2012 23:35:09 +0000 (16:35 -0700)]
git-config doc: unconfuse an example

One fictitious command "proxy-command" is enclosed inside a double
quote pair, while another fictitious command "default-proxy" is not
in the example, but the quoting does not change anything in the pair
of examples.  Remove the quotes to avoid unnecessary confusion.

Noticed by Michael Haggerty.

Signed-off-by: Junio C Hamano <gitster@pobox.com>
12 years agogit-config.txt: fix example
Michael Haggerty [Sat, 18 Aug 2012 17:32:10 +0000 (19:32 +0200)]
git-config.txt: fix example

The "--add" option is required to add a new value to a multivalued
configuration entry.

Signed-off-by: Michael Haggerty <mhagger@alum.mit.edu>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
12 years agoMerge branch 'jc/doc-git-updates' (early part)
Junio C Hamano [Fri, 17 Aug 2012 20:27:10 +0000 (13:27 -0700)]
Merge branch 'jc/doc-git-updates' (early part)

* 'jc/doc-git-updates' (early part):
  Documentation: update URL for formatted pages

12 years agoDocumentation: update the introductory section
Junio C Hamano [Fri, 17 Aug 2012 19:48:52 +0000 (12:48 -0700)]
Documentation: update the introductory section

The second paragraph in the git(1) description section were meant to
guide people who are not ready to dive into this page away from here.
Referring migrating CVS users to another page before they get
acquainted with Git was somewhat out of place.  Move the reference to
the "FURTHER DOCUMENTATION" section and push that section down.

Signed-off-by: Junio C Hamano <gitster@pobox.com>
12 years agoDocumentation: update URL for formatted pages
Junio C Hamano [Fri, 17 Aug 2012 19:14:57 +0000 (12:14 -0700)]
Documentation: update URL for formatted pages

The one at kernel.org has not been updated for quite a while and
can no longer be called "the latest".

Signed-off-by: Junio C Hamano <gitster@pobox.com>
12 years agostash: invoke rerere in case of conflict
Phil Hord [Tue, 10 Jul 2012 22:52:28 +0000 (18:52 -0400)]
stash: invoke rerere in case of conflict

"stash apply" directly calls a backend merge function which does not
automatically invoke rerere.  This confuses mergetool when leftover
rerere state is left behind from previous merges.

Invoke rerere explicitly when we encounter a conflict during stash
apply.  This turns the test introduced by the previous commit to
succeed.

Signed-off-by: Phil Hord <hordp@cisco.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
12 years agotest: git-stash conflict sets up rerere
Phil Hord [Tue, 10 Jul 2012 22:52:27 +0000 (18:52 -0400)]
test: git-stash conflict sets up rerere

Add a test to make sure that a conflicted "stash apply" invokes
rerere to record the conflicts and resolve the the files it can
(the current code doesn't, so the test is marked as failing).

Without correct state recorded for rerere, mergetool may be
confused, causing it to think no files have conflicts even though
they do.  This condition is not verified by this test since a
subsequent commit will change the behavior to enable rerere for
stash conflicts.

Also, the next test expected us to finish up with a reset, which is
impossible to do if we fail (as we must) and it's an unreasonable
expectation anyway.  Begin the next test with a reset of its own
instead.

Signed-off-by: Phil Hord <hordp@cisco.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
12 years agocleanup precompose_utf8
Robin Rosenberg [Fri, 17 Aug 2012 14:53:10 +0000 (16:53 +0200)]
cleanup precompose_utf8

 - Remove extraneous parentheses and braces;

 - Remove redundant NUL-termination before strcpy();

 - Check result of unlink when probing for decomposed file names;

 - Adjust for the coding style by adding missing whitespaces;

 - Move storage class "static" at the beginning of the decl.

Signed-off-by: Robin Rosenberg <robin.rosenberg@dewire.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
12 years agoman: git pull -r is a short for --rebase
Miklos Vajna [Thu, 16 Aug 2012 09:50:18 +0000 (11:50 +0200)]
man: git pull -r is a short for --rebase

Letting the "--rebase" option squat on the short-and-sweet single
letter option "-r" was an unintended accident and was not even
documented, but the short option seems to be already used in the
wild. Let's document it so that other options that begin with "r"
would not be tempted to steal it.

Signed-off-by: Miklos Vajna <vmiklos@suse.cz>
Signed-off-by: Junio C Hamano <gitster@pobox.com>