git
6 years agoGit 2.16.2 v2.16.2
Junio C Hamano [Thu, 15 Feb 2018 23:21:23 +0000 (15:21 -0800)] 
Git 2.16.2

Signed-off-by: Junio C Hamano <gitster@pobox.com>
6 years agoMerge branch 'ab/doc-cat-file-e-still-shows-errors' into maint
Junio C Hamano [Thu, 15 Feb 2018 23:18:15 +0000 (15:18 -0800)] 
Merge branch 'ab/doc-cat-file-e-still-shows-errors' into maint

Doc update.

* ab/doc-cat-file-e-still-shows-errors:
  cat-file doc: document that -e will return some output

6 years agoMerge branch 'as/read-tree-prefix-doc-fix' into maint
Junio C Hamano [Thu, 15 Feb 2018 23:18:14 +0000 (15:18 -0800)] 
Merge branch 'as/read-tree-prefix-doc-fix' into maint

Doc update.

* as/read-tree-prefix-doc-fix:
  doc/read-tree: remove obsolete remark

6 years agoMerge branch 'nd/add-i-ignore-submodules' into maint
Junio C Hamano [Thu, 15 Feb 2018 23:18:13 +0000 (15:18 -0800)] 
Merge branch 'nd/add-i-ignore-submodules' into maint

"git add -p" was taught to ignore local changes to submodules as
they do not interfere with the partial addition of regular changes
anyway.

* nd/add-i-ignore-submodules:
  add--interactive: ignore submodule changes except HEAD

6 years agoMerge branch 'tg/stash-with-pathspec-fix' into maint
Junio C Hamano [Thu, 15 Feb 2018 23:18:13 +0000 (15:18 -0800)] 
Merge branch 'tg/stash-with-pathspec-fix' into maint

"git stash -- <pathspec>" incorrectly blew away untracked files in
the directory that matched the pathspec, which has been corrected.

* tg/stash-with-pathspec-fix:
  stash: don't delete untracked files that match pathspec

6 years agoMerge branch 'jk/abort-clone-with-existing-dest' into maint
Junio C Hamano [Thu, 15 Feb 2018 23:18:12 +0000 (15:18 -0800)] 
Merge branch 'jk/abort-clone-with-existing-dest' into maint

"git clone $there $here" is allowed even when here directory exists
as long as it is an empty directory, but the command incorrectly
removed it upon a failure of the operation.

* jk/abort-clone-with-existing-dest:
  clone: do not clean up directories we didn't create
  clone: factor out dir_exists() helper
  t5600: modernize style
  t5600: fix outdated comment about unborn HEAD

6 years agoMerge branch 'jc/merge-symlink-ours-theirs' into maint
Junio C Hamano [Thu, 15 Feb 2018 23:18:12 +0000 (15:18 -0800)] 
Merge branch 'jc/merge-symlink-ours-theirs' into maint

"git merge -Xours/-Xtheirs" learned to use our/their version when
resolving a conflicting updates to a symbolic link.

* jc/merge-symlink-ours-theirs:
  merge: teach -Xours/-Xtheirs to symbolic link merge

6 years agoMerge branch 'rs/lose-leak-pending' into maint
Junio C Hamano [Thu, 15 Feb 2018 23:18:11 +0000 (15:18 -0800)] 
Merge branch 'rs/lose-leak-pending' into maint

API clean-up around revision traversal.

* rs/lose-leak-pending:
  commit: remove unused function clear_commit_marks_for_object_array()
  revision: remove the unused flag leak_pending
  checkout: avoid using the rev_info flag leak_pending
  bundle: avoid using the rev_info flag leak_pending
  bisect: avoid using the rev_info flag leak_pending
  object: add clear_commit_marks_all()
  ref-filter: use clear_commit_marks_many() in do_merge_filter()
  commit: use clear_commit_marks_many() in remove_redundant()
  commit: avoid allocation in clear_commit_marks_many()

6 years agoMerge branch 'jm/svn-pushmergeinfo-fix' into maint
Junio C Hamano [Thu, 15 Feb 2018 23:18:11 +0000 (15:18 -0800)] 
Merge branch 'jm/svn-pushmergeinfo-fix' into maint

"git svn dcommit" did not take into account the fact that a
svn+ssh:// URL with a username@ (typically used for pushing) refers
to the same SVN repository without the username@ and failed when
svn.pushmergeinfo option is set.

* jm/svn-pushmergeinfo-fix:
  git-svn: fix svn.pushmergeinfo handling of svn+ssh usernames.

6 years agoMerge branch 'dk/describe-all-output-fix' into maint
Junio C Hamano [Thu, 15 Feb 2018 23:18:10 +0000 (15:18 -0800)] 
Merge branch 'dk/describe-all-output-fix' into maint

An old regression in "git describe --all $annotated_tag^0" has been
fixed.

* dk/describe-all-output-fix:
  describe: prepend "tags/" when describing tags with embedded name

6 years agoMerge branch 'ab/perf-grep-threads' into maint
Junio C Hamano [Thu, 15 Feb 2018 23:18:09 +0000 (15:18 -0800)] 
Merge branch 'ab/perf-grep-threads' into maint

More perf tests for threaded grep

* ab/perf-grep-threads:
  perf: amend the grep tests to test grep.threads

6 years agoGit 2.16.1 v2.16.1
Junio C Hamano [Mon, 22 Jan 2018 05:14:25 +0000 (21:14 -0800)] 
Git 2.16.1

Signed-off-by: Junio C Hamano <gitster@pobox.com>
6 years agoMerge branch 'bc/hash-algo' into maint
Junio C Hamano [Mon, 22 Jan 2018 05:12:37 +0000 (21:12 -0800)] 
Merge branch 'bc/hash-algo' into maint

* bc/hash-algo:
  t5601-clone: test case-conflicting files on case-insensitive filesystem
  repository: pre-initialize hash algo pointer

6 years agot5601-clone: test case-conflicting files on case-insensitive filesystem
Eric Sunshine [Sun, 21 Jan 2018 08:07:28 +0000 (03:07 -0500)] 
t5601-clone: test case-conflicting files on case-insensitive filesystem

A recently introduced regression caused a segfault at clone time on
case-insensitive filesystems when filenames differing only in case are
present. This bug has already been fixed (repository: pre-initialize
hash algo pointer, 2018-01-18), but it's not the first time similar
problems have arisen. Therefore, introduce a test to catch this case and
protect against future regressions.

Signed-off-by: Eric Sunshine <sunshine@sunshineco.com>
Signed-off-by: brian m. carlson <sandals@crustytoothpaste.net>
Signed-off-by: Eric Sunshine <sunshine@sunshineco.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
6 years agorepository: pre-initialize hash algo pointer
brian m. carlson [Fri, 19 Jan 2018 04:18:25 +0000 (04:18 +0000)] 
repository: pre-initialize hash algo pointer

There are various git subcommands (among them, clone) which don't set up
the repository (that is, they lack RUN_SETUP or RUN_SETUP_GENTLY) but
end up needing to have information about the hash algorithm in use.
Because the hash algorithm is part of struct repository and it's only
initialized in repository setup, we can end up dereferencing a NULL
pointer in some cases if we call one of these subcommands and look up
the empty blob or empty tree values.

A "git clone" of a project that has two paths that differ only in
case suffers from this if it is run on a case insensitive platform.
When the command attempts to check out one of these two paths after
checking out the other one, the checkout codepath needs to see if
the version that is already on the filesystem (which should not
happen if the FS were case sensitive) is dirty, and it needs to
exercise the hashing code at that point.

In the future, we can add a command line option for this or read it
from the configuration, but until we're ready to expose that
functionality to the user, simply initialize the repository
structure to use the current hash algorithm, SHA-1.

Signed-off-by: brian m. carlson <sandals@crustytoothpaste.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
6 years agoGit 2.16 v2.16.0
Junio C Hamano [Wed, 17 Jan 2018 21:06:51 +0000 (13:06 -0800)] 
Git 2.16

Signed-off-by: Junio C Hamano <gitster@pobox.com>
6 years agoMerge tag 'l10n-2.16.0-rnd2' of git://github.com/git-l10n/git-po
Junio C Hamano [Tue, 16 Jan 2018 22:49:58 +0000 (14:49 -0800)] 
Merge tag 'l10n-2.16.0-rnd2' of git://github.com/git-l10n/git-po

l10n for Git 2.16.0 round 2

* tag 'l10n-2.16.0-rnd2' of git://github.com/git-l10n/git-po: (24 commits)
  l10n: de.po: translate 72 new messages
  l10n: de.po: improve messages when a branch starts to track another ref
  l10n: bg.po: Updated Bulgarian translation (3288t)
  l10n: TEAMS: add zh_CN team members
  l10n: zh_CN: for git v2.16.0 l10n round 2
  l10n: sv.po: Update Swedish translation (3288t0f0u)
  l10n: ru.po: update Russian translation
  l10n: TEAMS: Add ko team members
  l10n: ko.po: Update Korean translation
  l10n: fr.po 2.16 round 2
  l10n: es.po: Spanish translation 2.16.0 round 2
  l10n: vi.po(3288t): Updated Vietnamese translation for v2.16.0 round 2
  l10n: git.pot: v2.16.0 round 2 (8 new, 4 removed)
  l10n: es.po: Update Spanish Translation v2.16.0
  l10n: fr.po v2.16.0 round 1
  l10n: bg.po: Updated Bulgarian translation (3284t)
  l10n: sv.po: Update Swedish translation (3284t0f0u)
  l10n: fr.po: "worktree list" mistranslated as prune
  l10n: git.pot: v2.16.0 round 1 (64 new, 25 removed)
  l10n: fixes to German translation
  ...

6 years agoadd--interactive: ignore submodule changes except HEAD
Nguyễn Thái Ngọc Duy [Sat, 13 Jan 2018 12:10:38 +0000 (19:10 +0700)] 
add--interactive: ignore submodule changes except HEAD

For 'add -i' and 'add -p', the only action we can take on a dirty
submodule entry is update the index with a new value from its HEAD. The
content changes inside (from its own index, untracked files...) do not
matter, at least until 'git add -i' learns about launching a new
interactive add session inside a submodule.

Ignore all other submodules changes except HEAD. This reduces the number
of entries the user has to check through in 'git add -i', and the number
of 'no' they have to answer to 'git add -p' when dirty submodules are
present.

Signed-off-by: Nguyễn Thái Ngọc Duy <pclouds@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
6 years agol10n: de.po: translate 72 new messages
Ralf Thielow [Tue, 2 Jan 2018 09:14:40 +0000 (10:14 +0100)] 
l10n: de.po: translate 72 new messages

Translate 72 new messages came from git.pot update in 18a907225 (l10n:
git.pot: v2.16.0 round 1 (64 new, 25 removed)) and 005c62fe4 (l10n:
git.pot: v2.16.0 round 2 (8 new, 4 removed)).

Signed-off-by: Ralf Thielow <ralf.thielow@gmail.com>
Acked-by: Matthias Rüster <matthias.ruester@gmail.com>
6 years agol10n: de.po: improve messages when a branch starts to track another ref
Ralf Thielow [Sat, 16 Dec 2017 19:32:16 +0000 (20:32 +0100)] 
l10n: de.po: improve messages when a branch starts to track another ref

Signed-off-by: Ralf Thielow <ralf.thielow@gmail.com>
6 years agoRelNotes: minor typofix
SZEDER Gábor [Fri, 12 Jan 2018 10:47:36 +0000 (11:47 +0100)] 
RelNotes: minor typofix

Signed-off-by: SZEDER Gábor <szeder.dev@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
6 years agoGit 2.16-rc2 v2.16.0-rc2
Junio C Hamano [Thu, 11 Jan 2018 21:20:41 +0000 (13:20 -0800)] 
Git 2.16-rc2

Signed-off-by: Junio C Hamano <gitster@pobox.com>
6 years agoMerge branch 'jh/object-filtering'
Junio C Hamano [Thu, 11 Jan 2018 21:16:37 +0000 (13:16 -0800)] 
Merge branch 'jh/object-filtering'

Hotfix for a topic already in 'master'.

* jh/object-filtering:
  oidset: don't return value from oidset_init

6 years agoMerge branch 'tg/worktree-create-tracking'
Junio C Hamano [Thu, 11 Jan 2018 21:16:36 +0000 (13:16 -0800)] 
Merge branch 'tg/worktree-create-tracking'

Doc hotfix.

* tg/worktree-create-tracking:
  Documentation/git-worktree.txt: add missing `

6 years agoMerge branch 'js/test-with-ws-in-path'
Junio C Hamano [Thu, 11 Jan 2018 21:16:36 +0000 (13:16 -0800)] 
Merge branch 'js/test-with-ws-in-path'

Hot fix to a test.

* js/test-with-ws-in-path:
  t3900: add some more quotes

6 years agol10n: bg.po: Updated Bulgarian translation (3288t)
Alexander Shopov [Thu, 11 Jan 2018 20:59:37 +0000 (21:59 +0100)] 
l10n: bg.po: Updated Bulgarian translation (3288t)

Signed-off-by: Alexander Shopov <ash@kambanaria.org>
6 years agoDocumentation/git-worktree.txt: add missing `
Ralf Thielow [Thu, 11 Jan 2018 18:18:21 +0000 (19:18 +0100)] 
Documentation/git-worktree.txt: add missing `

Signed-off-by: Ralf Thielow <ralf.thielow@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
6 years agocat-file doc: document that -e will return some output
Ævar Arnfjörð Bjarmason [Wed, 10 Jan 2018 12:55:53 +0000 (12:55 +0000)] 
cat-file doc: document that -e will return some output

The -e option added in 7950571ad7 ("A few more options for
git-cat-file", 2005-12-03) has always errored out with message on
stderr saying that the provided object is malformed, like this:

    $ git cat-file -e malformed; echo $?
    fatal: Not a valid object name malformed
    128

A reader of this documentation may be misled into thinking that

    if ! git cat-file -e "$object" [...]

as opposed to:

    if ! git cat-file -e "$object" 2>/dev/null [...]

is sufficient to implement a truly silent test that checks whether
some arbitrary $object string was both valid, and pointed to an
object that exists.

Signed-off-by: Ævar Arnfjörð Bjarmason <avarab@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
6 years agot3900: add some more quotes
Beat Bolli [Wed, 10 Jan 2018 09:58:32 +0000 (10:58 +0100)] 
t3900: add some more quotes

In 89a70b80 ("t0302 & t3900: add forgotten quotes", 2018-01-03), quotes
were added to protect against spaces in $HOME. In the test_when_finished
command, two files are deleted which must be quoted individually.

[jc: with \$HOME in the test_when_finished command quoted, as
pointed out by j6t].

Signed-off-by: Beat Bolli <dev+git@drbeat.li>
Helped-by: Johannes Sixt <j6t@kdbg.org>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
6 years agoRelNotes update before -rc2
Junio C Hamano [Wed, 10 Jan 2018 22:01:50 +0000 (14:01 -0800)] 
RelNotes update before -rc2

Signed-off-by: Junio C Hamano <gitster@pobox.com>
6 years agoMerge branch 'js/perl-path-workaround-in-tests'
Junio C Hamano [Wed, 10 Jan 2018 22:01:31 +0000 (14:01 -0800)] 
Merge branch 'js/perl-path-workaround-in-tests'

* js/perl-path-workaround-in-tests:
  mingw: handle GITPERLLIB in t0021 in a Windows-compatible way

6 years agoMerge branch 'ew/empty-merge-with-dirty-index'
Junio C Hamano [Wed, 10 Jan 2018 22:01:25 +0000 (14:01 -0800)] 
Merge branch 'ew/empty-merge-with-dirty-index'

"git merge -s recursive" did not correctly abort when the index is
dirty, if the merged tree happened to be the same as the current
HEAD, which has been fixed.

* ew/empty-merge-with-dirty-index:
  merge-recursive: do not look at the index during recursive merge

6 years agoMerge branch 'ma/bisect-leakfix'
Junio C Hamano [Wed, 10 Jan 2018 22:01:25 +0000 (14:01 -0800)] 
Merge branch 'ma/bisect-leakfix'

A hotfix for a recent update that broke 'git bisect'.

* ma/bisect-leakfix:
  bisect: fix a regression causing a segfault

6 years agoMerge branch 'js/fix-merge-arg-quoting-in-rebase-p'
Junio C Hamano [Wed, 10 Jan 2018 22:01:24 +0000 (14:01 -0800)] 
Merge branch 'js/fix-merge-arg-quoting-in-rebase-p'

"git rebase -p -X<option>" did not propagate the option properly
down to underlying merge strategy backend.

* js/fix-merge-arg-quoting-in-rebase-p:
  rebase -p: fix quoting when calling `git merge`

6 years agomingw: handle GITPERLLIB in t0021 in a Windows-compatible way
Johannes Schindelin [Sat, 6 Jan 2018 22:01:22 +0000 (23:01 +0100)] 
mingw: handle GITPERLLIB in t0021 in a Windows-compatible way

Git's assumption that all path lists are colon-separated is not only
wrong on Windows, it is not even an assumption that is compatible with
POSIX.

In the interest of time, let's not try to fix this properly but simply
work around the obvious breakage on Windows, where the MSYS2 Bash used
by Git for Windows to interpret the Git's Unix shell scripts will
automagically convert path lists in the environment to
semicolon-separated lists of Windows paths (with drive letter and the
corresponding colon and all that jazz).

In other words, we simply look whether there is a semicolon in
GITPERLLIB and split by semicolons if found instead of colons. This is
not fool-proof, of course, as the path list could consist of a single
path. But that is not the case in Git for Windows' test suite, there are
always two paths in GITPERLLIB.

Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
6 years agol10n: TEAMS: add zh_CN team members
Jiang Xin [Tue, 9 Jan 2018 01:55:12 +0000 (09:55 +0800)] 
l10n: TEAMS: add zh_CN team members

Add Fangyi Zhou to zh_CN l10n team members.

Signed-off-by: Jiang Xin <worldhello.net@gmail.com>
6 years agol10n: zh_CN: for git v2.16.0 l10n round 2
Jiang Xin [Sun, 31 Dec 2017 02:50:17 +0000 (10:50 +0800)] 
l10n: zh_CN: for git v2.16.0 l10n round 2

Translate 72 messages (3288t0f0u) for git v2.16.0-rc1.

Signed-off-by: Jiang Xin <worldhello.net@gmail.com>
Reviewed-by: 依云 <lilydjwg@gmail.com>
Reviewed-by: Fangyi Zhou <fangyi.zhou@yuriko.moe>
6 years agoMerge branch 'master' of git://github.com/nafmo/git-l10n-sv
Jiang Xin [Wed, 10 Jan 2018 03:30:04 +0000 (11:30 +0800)] 
Merge branch 'master' of git://github.com/nafmo/git-l10n-sv

* 'master' of git://github.com/nafmo/git-l10n-sv:
  l10n: sv.po: Update Swedish translation (3288t0f0u)

6 years agoMerge branch 'russian-l10n' of https://github.com/DJm00n/git-po-ru
Jiang Xin [Wed, 10 Jan 2018 03:28:56 +0000 (11:28 +0800)] 
Merge branch 'russian-l10n' of https://github.com/DJm00n/git-po-ru

* 'russian-l10n' of https://github.com/DJm00n/git-po-ru:
  l10n: ru.po: update Russian translation

6 years agoMerge branch 'jk/doc-diff-options'
Junio C Hamano [Tue, 9 Jan 2018 22:32:57 +0000 (14:32 -0800)] 
Merge branch 'jk/doc-diff-options'

Doc update.

* jk/doc-diff-options:
  docs/diff-options: clarify scope of diff-filter types

6 years agoMerge branch 'bw/protocol-v1'
Junio C Hamano [Tue, 9 Jan 2018 22:32:56 +0000 (14:32 -0800)] 
Merge branch 'bw/protocol-v1'

Test fix for a topic already in 'master'.

* bw/protocol-v1:
  http: fix v1 protocol tests with apache httpd < 2.4

6 years agoMerge branch 'sg/travis-check-untracked'
Junio C Hamano [Tue, 9 Jan 2018 22:32:55 +0000 (14:32 -0800)] 
Merge branch 'sg/travis-check-untracked'

* sg/travis-check-untracked:
  travis-ci: check that all build artifacts are .gitignore-d
  travis-ci: don't store P4 and Git LFS in the working tree

6 years agoMerge branch 'js/test-with-ws-in-path'
Junio C Hamano [Tue, 9 Jan 2018 22:32:55 +0000 (14:32 -0800)] 
Merge branch 'js/test-with-ws-in-path'

Test fixes.

* js/test-with-ws-in-path:
  t0302 & t3900: add forgotten quotes
  Allow the test suite to pass in a directory whose name contains spaces

6 years agoMerge branch 'bc/submitting-patches-in-asciidoc'
Junio C Hamano [Tue, 9 Jan 2018 22:32:54 +0000 (14:32 -0800)] 
Merge branch 'bc/submitting-patches-in-asciidoc'

Doc readability update.

* bc/submitting-patches-in-asciidoc:
  doc/SubmittingPatches: improve text formatting

6 years agoMerge branch 'sg/travis-skip-identical-test'
Junio C Hamano [Tue, 9 Jan 2018 22:32:54 +0000 (14:32 -0800)] 
Merge branch 'sg/travis-skip-identical-test'

Avoid repeatedly testing the same tree in TravisCI that have been
tested successfully already.

* sg/travis-skip-identical-test:
  travis-ci: record and skip successfully built trees
  travis-ci: create the cache directory early in the build process
  travis-ci: print the "tip of branch is exactly at tag" message in color

6 years agoMerge branch 'ab/dc-sha1-loose-ends'
Junio C Hamano [Tue, 9 Jan 2018 22:32:53 +0000 (14:32 -0800)] 
Merge branch 'ab/dc-sha1-loose-ends'

Tying loose ends for the recent integration work of
collision-detecting SHA-1 implementation.

* ab/dc-sha1-loose-ends:
  Makefile: NO_OPENSSL=1 should no longer imply BLK_SHA1=1

6 years agoMerge branch 'sg/travis-fixes'
Junio C Hamano [Tue, 9 Jan 2018 22:32:53 +0000 (14:32 -0800)] 
Merge branch 'sg/travis-fixes'

Assorted updates for TravisCI integration.

* sg/travis-fixes:
  travis-ci: only print test failures if there are test results available
  travis-ci: save prove state for the 32 bit Linux build
  travis-ci: don't install default addon packages for the 32 bit Linux build
  travis-ci: fine tune the use of 'set -x' in 'ci/*' scripts

6 years agodoc/read-tree: remove obsolete remark
Andreas G. Schacker [Tue, 9 Jan 2018 15:30:34 +0000 (16:30 +0100)] 
doc/read-tree: remove obsolete remark

Earlier versions of `git read-tree` required the `--prefix` option value
to end with a slash. This restriction was eventually lifted without a
corresponding amendment to the documentation.

Signed-off-by: Andreas G. Schacker <andreas.schacker@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
6 years agol10n: sv.po: Update Swedish translation (3288t0f0u)
Peter Krefting [Tue, 9 Jan 2018 19:10:14 +0000 (20:10 +0100)] 
l10n: sv.po: Update Swedish translation (3288t0f0u)

Signed-off-by: Peter Krefting <peter@softwolves.pp.se>
6 years agoMerge branch 'js/misc-git-gui-stuff' of ../git-gui
Junio C Hamano [Tue, 9 Jan 2018 19:07:03 +0000 (11:07 -0800)] 
Merge branch 'js/misc-git-gui-stuff' of ../git-gui

* 'js/misc-git-gui-stuff' of ../git-gui:
  git-gui: allow Ctrl+T to toggle multiple paths
  git-gui: fix exception when trying to stage with empty file list
  git-gui: avoid exception upon Ctrl+T in an empty list
  git gui: fix staging a second line to a 1-line file

6 years agogit-gui: allow Ctrl+T to toggle multiple paths
Johannes Schindelin [Tue, 9 Jan 2018 14:33:04 +0000 (15:33 +0100)] 
git-gui: allow Ctrl+T to toggle multiple paths

It is possible to select multiple files in the "Unstaged Changes" and
the "Staged Changes" lists. But when hitting Ctrl+T, surprisingly only
one entry is handled, not all selected ones.

Let's just use the same code path as for the "Stage To Commit" and the
"Unstage From Commit" menu items.

This fixes https://github.com/git-for-windows/git/issues/1012

Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
6 years agogit-gui: fix exception when trying to stage with empty file list
Johannes Schindelin [Tue, 9 Jan 2018 14:33:01 +0000 (15:33 +0100)] 
git-gui: fix exception when trying to stage with empty file list

If there is nothing to stage, there is nothing to stage. Let's not try
to, even if the file list contains nothing at all.

This fixes https://github.com/git-for-windows/git/issues/1075

Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
6 years agogit-gui: avoid exception upon Ctrl+T in an empty list
Johannes Schindelin [Tue, 9 Jan 2018 14:32:58 +0000 (15:32 +0100)] 
git-gui: avoid exception upon Ctrl+T in an empty list

Previously unstaged files can be staged by clicking on them and then
pressing Ctrl+T. Conveniently, the next unstaged file is selected
automatically so that the unstaged files can be staged by repeatedly
pressing Ctrl+T.

When a user hits Ctrl+T one time too many, though, Git GUI used to throw
this exception:

expected number but got ""
expected number but got ""
    while executing
"expr {int([lindex [$w tag ranges in_diff] 0])}"
    (procedure "toggle_or_diff" line 13)
    invoked from within
"toggle_or_diff toggle .vpane.files.workdir.list "
    (command bound to event)

Let's just avoid that by skipping the operation when there are no more
files to stage.

This fixes https://github.com/git-for-windows/git/issues/1060

Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
6 years agogit gui: fix staging a second line to a 1-line file
Johannes Schindelin [Tue, 9 Jan 2018 14:32:54 +0000 (15:32 +0100)] 
git gui: fix staging a second line to a 1-line file

When a 1-line file is augmented by a second line, and the user tries to
stage that single line via the "Stage Line" context menu item, we do not
want to see "apply: corrupt patch at line 5".

The reason for this error was that the hunk header looks like this:

@@ -1 +1,2 @@

but the existing code expects the original range always to contain a
comma. This problem is easily fixed by cutting the string "1 +1,2"
(that Git GUI formerly mistook for the starting line) at the space.

This fixes https://github.com/git-for-windows/git/issues/515

Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
6 years agoMerge branch 'ew/empty-merge-with-dirty-index-maint' into ew/empty-merge-with-dirty...
Junio C Hamano [Tue, 9 Jan 2018 18:41:37 +0000 (10:41 -0800)] 
Merge branch 'ew/empty-merge-with-dirty-index-maint' into ew/empty-merge-with-dirty-index

* ew/empty-merge-with-dirty-index-maint:
  merge-recursive: do not look at the index during recursive merge

6 years agomerge-recursive: do not look at the index during recursive merge
Junio C Hamano [Tue, 9 Jan 2018 18:05:51 +0000 (10:05 -0800)] 
merge-recursive: do not look at the index during recursive merge

When merging another branch into ours, if their tree is the same as
the common ancestor's, we can declare that our tree represents the
result of three-way merge.  In such a case, the recursive merge
backend incorrectly used to create a commit out of our index, even
when the index has changes.

A recent fix attempted to prevent this by adding a comparison
between "our" tree and the index, but forgot that this check must be
restricted only to the outermost merge.  Inner merges performed by
the recursive backend across merge bases are by definition made from
scratch without having any local changes added to the index.  The
call to index_has_changes() during an inner merge is working on the
index that has no relation to the merge being performed, preventing
legitimate merges from getting carried out.

Fix it by limiting the check to the outermost merge.

Signed-off-by: Junio C Hamano <gitster@pobox.com>
6 years agol10n: ru.po: update Russian translation
Dimitriy Ryazantcev [Tue, 9 Jan 2018 10:21:52 +0000 (12:21 +0200)] 
l10n: ru.po: update Russian translation

Signed-off-by: Dimitriy Ryazantcev <dimitriy.ryazantcev@gmail.com>
6 years agol10n: TEAMS: Add ko team members
Changwoo Ryu [Tue, 9 Jan 2018 02:32:47 +0000 (11:32 +0900)] 
l10n: TEAMS: Add ko team members

Add Gwan-gyeong Mun and Sihyeon Jang.

Signed-off-by: Changwoo Ryu <cwryu@debian.org>
6 years agoMerge branch 'ko/merge-l10n' of https://github.com/git-l10n-ko/git-l10n-ko
Jiang Xin [Tue, 9 Jan 2018 01:47:11 +0000 (09:47 +0800)] 
Merge branch 'ko/merge-l10n' of https://github.com/git-l10n-ko/git-l10n-ko

* 'ko/merge-l10n' of https://github.com/git-l10n-ko/git-l10n-ko:
  l10n: ko.po: Update Korean translation

6 years agooidset: don't return value from oidset_init
Thomas Gummerer [Sun, 7 Jan 2018 23:04:04 +0000 (23:04 +0000)] 
oidset: don't return value from oidset_init

c3a9ad3117 ("oidset: add iterator methods to oidset", 2017-11-21)
introduced a 'oidset_init()' function in oidset.h, which has void as
return type, but returns an expression.

This makes the solaris compiler fail with:

    "oidset.h", line 30: void function cannot return value

As the return type is void, and even the return type of the expression
we're trying to return (oidmap_init) is void just remove the return
statement to fix the compiler error.

Signed-off-by: Thomas Gummerer <t.gummerer@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
6 years agostash: don't delete untracked files that match pathspec
Thomas Gummerer [Sat, 6 Jan 2018 00:24:20 +0000 (00:24 +0000)] 
stash: don't delete untracked files that match pathspec

Currently when 'git stash push -- <pathspec>' is used, untracked files
that match the pathspec will be deleted, even though they do not end up
in a stash anywhere.

This is because the original commit introducing the pathspec feature in
git stash push (df6bba0937 ("stash: teach 'push' (and 'create_stash') to
honor pathspec", 2017-02-28)) used the sequence of 'git reset <pathspec>
&& git ls-files --modified <pathspec> | git checkout-index && git clean
<pathspec>'.

The intention was to emulate what 'git reset --hard -- <pathspec>' would
do.  The call to 'git clean' was supposed to clean up the files that
were unstaged by 'git reset'.  This would work fine if the pathspec
doesn't match any files that were untracked before 'git stash push --
<pathspec>'.  However if <pathspec> matches a file that was untracked
before invoking the 'stash' command, all untracked files matching the
pathspec would inadvertently be deleted as well, even though they
wouldn't end up in the stash, and are therefore lost.

This behaviour was never what was intended, only blobs that also end up
in the stash should be reset to their state in HEAD, previously
untracked files should be left alone.

To achieve this, first match what's in the index and what's in the
working tree by adding all changes to the index, ask diff-index what
changed between HEAD and the current index, and then apply that patch in
reverse to get rid of the changes, which includes removal of added
files and resurrection of removed files.

Reported-by: Reid Price <reid.price@gmail.com>
Helped-by: Junio C Hamano <gitster@pobox.com>
Signed-off-by: Thomas Gummerer <t.gummerer@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
6 years agol10n: ko.po: Update Korean translation
Changwoo Ryu [Mon, 8 Jan 2018 01:25:27 +0000 (10:25 +0900)] 
l10n: ko.po: Update Korean translation

Signed-off-by: Changwoo Ryu <cwryu@debian.org>
Signed-off-by: Sihyeon Jang <uneedsihyeon@gmail.com>
Signed-off-by: Gwan-gyeong Mun <elongbug@gmail.com>
Reviewed-by: Changwoo Ryu <cwryu@debian.org>
6 years agoMerge branch '2.16' of https://github.com/ChrisADR/git-po
Jiang Xin [Mon, 8 Jan 2018 02:59:24 +0000 (10:59 +0800)] 
Merge branch '2.16' of https://github.com/ChrisADR/git-po

* '2.16' of https://github.com/ChrisADR/git-po:
  l10n: es.po: Spanish translation 2.16.0 round 2

6 years agoMerge branch 'fr_2.16-rc1' of git://github.com/jnavila/git
Jiang Xin [Mon, 8 Jan 2018 01:17:24 +0000 (09:17 +0800)] 
Merge branch 'fr_2.16-rc1' of git://github.com/jnavila/git

* 'fr_2.16-rc1' of git://github.com/jnavila/git:
  l10n: fr.po 2.16 round 2

6 years agol10n: fr.po 2.16 round 2
Jean-Noel Avila [Sun, 7 Jan 2018 17:57:48 +0000 (18:57 +0100)] 
l10n: fr.po 2.16 round 2

Signed-off-by: Jean-Noel Avila <jn.avila@free.fr>
6 years agol10n: es.po: Spanish translation 2.16.0 round 2
Christopher Díaz Riveros [Sun, 7 Jan 2018 17:15:35 +0000 (12:15 -0500)] 
l10n: es.po: Spanish translation 2.16.0 round 2

Signed-off-by: Christopher Díaz Riveros <chrisadr@gentoo.org>
6 years agol10n: vi.po(3288t): Updated Vietnamese translation for v2.16.0 round 2
Tran Ngoc Quan [Sun, 7 Jan 2018 01:20:27 +0000 (08:20 +0700)] 
l10n: vi.po(3288t): Updated Vietnamese translation for v2.16.0 round 2

Signed-off-by: Tran Ngoc Quan <vnwildman@gmail.com>
6 years agol10n: git.pot: v2.16.0 round 2 (8 new, 4 removed)
Jiang Xin [Sat, 6 Jan 2018 23:50:31 +0000 (07:50 +0800)] 
l10n: git.pot: v2.16.0 round 2 (8 new, 4 removed)

Generate po/git.pot from v2.16.0-rc1 for git v2.16.0 l10n round 2.

Signed-off-by: Jiang Xin <worldhello.net@gmail.com>
6 years agoMerge branch 'master' of git://github.com/git-l10n/git-po
Jiang Xin [Sat, 6 Jan 2018 23:49:43 +0000 (07:49 +0800)] 
Merge branch 'master' of git://github.com/git-l10n/git-po

* 'master' of git://github.com/git-l10n/git-po:
  l10n: es.po: Update Spanish Translation v2.16.0
  l10n: fr.po v2.16.0 round 1
  l10n: bg.po: Updated Bulgarian translation (3284t)
  l10n: sv.po: Update Swedish translation (3284t0f0u)
  l10n: fr.po: "worktree list" mistranslated as prune
  l10n: git.pot: v2.16.0 round 1 (64 new, 25 removed)
  l10n: fixes to German translation
  l10n: Update Spanish translation
  l10n: zh_CN translate parameter name
  l10n: zh_CN Fix typo
  l10n: Fixes to Catalan translation

6 years agoMerge branch '2.16' of https://github.com/ChrisADR/git-po
Jiang Xin [Sat, 6 Jan 2018 02:26:30 +0000 (10:26 +0800)] 
Merge branch '2.16' of https://github.com/ChrisADR/git-po

* '2.16' of https://github.com/ChrisADR/git-po:
  l10n: es.po: Update Spanish Translation v2.16.0

6 years agoMerge branch 'fr_2.16' of git://github.com/jnavila/git
Jiang Xin [Sat, 6 Jan 2018 02:24:52 +0000 (10:24 +0800)] 
Merge branch 'fr_2.16' of git://github.com/jnavila/git

* 'fr_2.16' of git://github.com/jnavila/git:
  l10n: fr.po v2.16.0 round 1
  l10n: fr.po: "worktree list" mistranslated as prune

6 years agoGit 2.16-rc1 v2.16.0-rc1
Junio C Hamano [Fri, 5 Jan 2018 21:45:17 +0000 (13:45 -0800)] 
Git 2.16-rc1

Signed-off-by: Junio C Hamano <gitster@pobox.com>
6 years agoMerge branch 'js/sequencer-cleanups'
Junio C Hamano [Fri, 5 Jan 2018 21:28:11 +0000 (13:28 -0800)] 
Merge branch 'js/sequencer-cleanups'

Code cleanup.

* js/sequencer-cleanups:
  sequencer: do not invent whitespace when transforming OIDs
  sequencer: report when noop has an argument
  sequencer: remove superfluous conditional
  sequencer: strip bogus LF at end of error messages
  rebase: do not continue when the todo list generation failed

6 years agoMerge branch 'jh/memihash-opt'
Junio C Hamano [Fri, 5 Jan 2018 21:28:11 +0000 (13:28 -0800)] 
Merge branch 'jh/memihash-opt'

Squelch compiler warning.

* jh/memihash-opt:
  t/helper/test-lazy-name-hash: fix compilation

6 years agoMerge branch 'tb/test-lint-wc-l'
Junio C Hamano [Fri, 5 Jan 2018 21:28:11 +0000 (13:28 -0800)] 
Merge branch 'tb/test-lint-wc-l'

Test update.

* tb/test-lint-wc-l:
  check-non-portable-shell.pl: `wc -l` may have leading WS

6 years agoMerge branch 'rs/use-argv-array-in-child-process'
Junio C Hamano [Fri, 5 Jan 2018 21:28:10 +0000 (13:28 -0800)] 
Merge branch 'rs/use-argv-array-in-child-process'

Code cleanup.

* rs/use-argv-array-in-child-process:
  send-pack: use internal argv_array of struct child_process
  http: use internal argv_array of struct child_process

6 years agoMerge branch 'ld/p4-multiple-shelves'
Junio C Hamano [Fri, 5 Jan 2018 21:28:10 +0000 (13:28 -0800)] 
Merge branch 'ld/p4-multiple-shelves'

"git p4" update.

* ld/p4-multiple-shelves:
  git-p4: update multiple shelved change lists

6 years agoMerge branch 'jd/fix-strbuf-add-urlencode-bytes'
Junio C Hamano [Fri, 5 Jan 2018 21:28:10 +0000 (13:28 -0800)] 
Merge branch 'jd/fix-strbuf-add-urlencode-bytes'

Bytes with high-bit set were encoded incorrectly and made
credential helper fail.

* jd/fix-strbuf-add-urlencode-bytes:
  strbuf: fix urlencode format string on signed char

6 years agoMerge branch 'ew/empty-merge-with-dirty-index'
Junio C Hamano [Fri, 5 Jan 2018 21:28:09 +0000 (13:28 -0800)] 
Merge branch 'ew/empty-merge-with-dirty-index'

"git merge -s recursive" did not correctly abort when the index is
dirty, if the merged tree happened to be the same as the current
HEAD, which has been fixed.

* ew/empty-merge-with-dirty-index:
  merge-recursive: avoid incorporating uncommitted changes in a merge
  move index_has_changes() from builtin/am.c to merge.c for reuse
  t6044: recursive can silently incorporate dirty changes in a merge

6 years agoMerge branch 'db/doc-config-section-names-with-bs'
Junio C Hamano [Fri, 5 Jan 2018 21:28:09 +0000 (13:28 -0800)] 
Merge branch 'db/doc-config-section-names-with-bs'

Doc update.

* db/doc-config-section-names-with-bs:
  config.txt: document behavior of backslashes in subsections

6 years agoMerge branch 'jk/test-suite-tracing'
Junio C Hamano [Fri, 5 Jan 2018 21:28:09 +0000 (13:28 -0800)] 
Merge branch 'jk/test-suite-tracing'

Assorted fixes around running tests with "-x" tracing option.

* jk/test-suite-tracing:
  t/Makefile: introduce TEST_SHELL_PATH
  test-lib: make "-x" work with "--verbose-log"
  t5615: avoid re-using descriptor 4
  test-lib: silence "-x" cleanup under bash

6 years agorebase -p: fix quoting when calling `git merge`
Johannes Schindelin [Thu, 4 Jan 2018 21:31:52 +0000 (22:31 +0100)] 
rebase -p: fix quoting when calling `git merge`

It has been reported that strategy arguments are not passed to `git
merge` correctly when rebasing interactively, preserving merges.

The reason is that the strategy arguments are already quoted, and then
quoted again.

This fixes https://github.com/git-for-windows/git/issues/1321

Original-patch-by: Kim Gybels <kgybels@infogroep.be>
Also-reported-by: Matwey V. Kornilov <matwey.kornilov@gmail.com>
Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
6 years agol10n: es.po: Update Spanish Translation v2.16.0
Christopher Díaz Riveros [Thu, 4 Jan 2018 21:06:40 +0000 (16:06 -0500)] 
l10n: es.po: Update Spanish Translation v2.16.0

Signed-off-by: Christopher Díaz Riveros <chrisadr@gentoo.org>
6 years agodocs/diff-options: clarify scope of diff-filter types
Jeff King [Thu, 4 Jan 2018 16:49:42 +0000 (11:49 -0500)] 
docs/diff-options: clarify scope of diff-filter types

The same document for "--diff-filter" is included by many
programs in the diff family. Because it mentions all
possible types (added, removed, etc), this may imply to the
reader that all types can be generated by a particular
command. But this isn't necessarily the case; "diff-files"
cannot generally produce an "Added" entry, since the diff is
limited to what is already in the index.

Let's make it clear that the list here is the full one, and
does not imply anything about what a particular invocation
may produce.

Note that conditionally including items (e.g., omitting
"Added" in the git-diff-files manpage) isn't the right
solution here for two reasons:

  - The problem isn't diff-files, but doing an index to
    working tree diff. "git diff" can do the same diff, but
    also has other modes where "Added" does show up.

  - The direction of the diff matters. Doing "diff-files -R"
    can get you Added entries (but not Deleted ones).

So it's best just to explain that the set of available types
depends on the specific diff invocation.

Reported-by: John Cheng <johnlicheng@gmail.com>
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
6 years agohttp: fix v1 protocol tests with apache httpd < 2.4
Todd Zullinger [Sun, 31 Dec 2017 02:32:34 +0000 (21:32 -0500)] 
http: fix v1 protocol tests with apache httpd < 2.4

The apache config used by tests was updated to use the SetEnvIf
directive to set the Git-Protocol header in 19113a26b6 ("http: tell
server that the client understands v1", 2017-10-16).

Setting the Git-Protocol header is restricted to httpd >= 2.4, but
mod_setenvif and the SetEnvIf directive work with lower versions, at
least as far back as 2.0, according to the httpd documentation:

    https://httpd.apache.org/docs/2.0/mod/mod_setenvif.html

Drop the restriction.  Tested with httpd 2.2 and 2.4.

Signed-off-by: Todd Zullinger <tmz@pobox.com>
Acked-by: Brandon Williams <bmwill@google.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
6 years agoperf: amend the grep tests to test grep.threads
Ævar Arnfjörð Bjarmason [Fri, 29 Dec 2017 22:59:03 +0000 (22:59 +0000)] 
perf: amend the grep tests to test grep.threads

Ever since 5b594f457a ("Threaded grep", 2010-01-25) the number of
threads git-grep uses under PTHREADS has been hardcoded to 8, but
there's no performance test to check whether this is an optimal
setting.

Amend the existing tests for the grep engines to support a mode where
this can be tested, e.g.:

    GIT_PERF_GREP_THREADS='1 8 16' GIT_PERF_LARGE_REPO=~/g/linux ./run p782*

Signed-off-by: Ævar Arnfjörð Bjarmason <avarab@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
6 years agot0302 & t3900: add forgotten quotes
Johannes Schindelin [Wed, 3 Jan 2018 16:54:54 +0000 (17:54 +0100)] 
t0302 & t3900: add forgotten quotes

When cleaning up files in the $HOME directory, it really makes sense to
quote the path, especially in Git's test suite, where the HOME directory
is *guaranteed* to contain spaces in its name.

It would appear that those two tests pass even without cleaning up the
files, but really more by pure chance than by design (the cleanup seems
not actually to be necessary).

However, if anybody would have a left-over `trash/` directory in Git's
`t/` directory, these tests would fail, because they would all of a
sudden try to delete that directory, but without the `-r` (recursive)
flag. That is how this issue was found.

Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de>
Reviewed-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
6 years agoAllow the test suite to pass in a directory whose name contains spaces
Johannes Schindelin [Wed, 3 Jan 2018 16:54:50 +0000 (17:54 +0100)] 
Allow the test suite to pass in a directory whose name contains spaces

It is totally legitimate to clone Git's source code anywhere, including
into, say, directories whose name (or the name of its absolute path)
contains spaces.

However, a couple of tests failed to anticipate this, for lack of
quoting (or in one instance, for failure to expect more than one space
in the absolute path of the TEST_DIRECTORY). This can be easily verified
by calling these commands in your current clone:

git clone . with\ spaces
cd with\ spaces
make -j15 test

Let's fix this.

Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de>
Reviewed-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
6 years agobisect: fix a regression causing a segfault
Ævar Arnfjörð Bjarmason [Wed, 3 Jan 2018 18:48:52 +0000 (18:48 +0000)] 
bisect: fix a regression causing a segfault

In 7c117184d7 ("bisect: fix off-by-one error in
`best_bisection_sorted()`", 2017-11-05) the more careful logic dealing
with freeing p->next in 50e62a8e70 ("rev-list: implement
--bisect-all", 2007-10-22) was removed.

Restore the more careful check to avoid segfaulting. Ideally this
would come with a test case, but we don't have steps to reproduce
this, only a backtrace from gdb pointing to this being the issue.

Reported-by: Yasushi SHOJI <yasushi.shoji@gmail.com>
Signed-off-by: Ævar Arnfjörð Bjarmason <avarab@gmail.com>
Acked-by: Martin Ågren <martin.agren@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
6 years agodoc/SubmittingPatches: improve text formatting
Todd Zullinger [Tue, 2 Jan 2018 15:33:50 +0000 (10:33 -0500)] 
doc/SubmittingPatches: improve text formatting

049e64aa50 ("Documentation: convert SubmittingPatches to AsciiDoc",
2017-11-12) changed the `git blame` and `git shortlog` examples given in
the section on sending your patches.

In order to italicize the `$path` argument the commands are enclosed in
plus characters as opposed to backticks.  The difference between the
quoting methods is that backtick enclosed text is not subject to further
expansion.  This formatting makes reading SubmittingPatches in a git
clone a little more difficult.  In addition to the underscores around
`$path` the `--` chars in `git shortlog --no-merges` must be replaced
with `{litdd}`.

Use backticks to quote these commands.  The italicized `$path` is lost
from the html version but the commands can be read (and copied) more
easily by users reading the text version.  These readers are more likely
to use the commands while submitting patches.  Make it easier for them.

Signed-off-by: Todd Zullinger <tmz@pobox.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
6 years agoclone: do not clean up directories we didn't create
Jeff King [Tue, 2 Jan 2018 21:11:39 +0000 (16:11 -0500)] 
clone: do not clean up directories we didn't create

Once upon a time, git-clone would refuse to write into a
directory that it did not itself create. The cleanup
routines for a failed clone could therefore just remove the
git and worktree dirs completely.

In 55892d2398 (Allow cloning to an existing empty directory,
2009-01-11), we learned to write into an existing directory.
Which means that doing:

  mkdir foo
  git clone will-fail foo

ends up deleting foo. This isn't a huge catastrophe, since
by definition foo must be empty. But it's somewhat
confusing; we should leave the filesystem as we found it.

Because we know that the only directory we'll write into is
an empty one, we can handle this case by just passing the
KEEP_TOPLEVEL flag to our recursive delete (if we could
write into populated directories, we'd have to keep track of
what we wrote and what we did not, which would be much
harder).

Note that we need to handle the work-tree and git-dir
separately, though, as only one might exist (and the new
tests in t5600 cover all cases).

Reported-by: Stephan Janssen <sjanssen@you-get.com>
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
6 years agoclone: factor out dir_exists() helper
Jeff King [Tue, 2 Jan 2018 21:10:14 +0000 (16:10 -0500)] 
clone: factor out dir_exists() helper

Two parts of git-clone's setup logic check whether a
directory exists, and they both call stat directly with the
same scratch "struct stat" buffer. Let's pull that into a
helper, which has a few advantages:

  - it makes the purpose of the stat calls more obvious

  - it makes it clear that we don't care about the
    information in "buf" remaining valid

  - if we later decide to make the check more robust (e.g.,
    complaining about non-directories), we can do it in one
    place

Note that we could just use file_exists() for this, which
has identical code. But we specifically care about
directories, so this future-proofs us against that function
later getting more picky about seeing actual files.

Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
6 years agot5600: modernize style
Jeff King [Tue, 2 Jan 2018 21:09:00 +0000 (16:09 -0500)] 
t5600: modernize style

This is an old script which could use some updating before
we add to it:

  - use the standard line-breaking:

      test_expect_success 'title' '
              body
      '

  - run all code inside test_expect blocks to catch
    unexpected failures in setup steps

  - use "test_commit -C" instead of manually entering
    sub-repo

  - use test_when_finished for cleanup steps

  - test_path_is_* as appropriate

Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
6 years agot5600: fix outdated comment about unborn HEAD
Jeff King [Tue, 2 Jan 2018 21:08:33 +0000 (16:08 -0500)] 
t5600: fix outdated comment about unborn HEAD

Back when this test was written, git-clone could not handle
a repository without any commits. These days it works fine,
and this comment is out of date.

At first glance it seems like we could just drop this code
entirely now, but it's necessary for the final test, which
was added later. That test corrupts the repository by
temporarily removing its objects, which means we need to
have some objects to move.

Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
6 years agotravis-ci: check that all build artifacts are .gitignore-d
SZEDER Gábor [Sun, 31 Dec 2017 16:02:06 +0000 (17:02 +0100)] 
travis-ci: check that all build artifacts are .gitignore-d

Every once in a while our explicit .gitignore files get out of sync
when our build process learns to create new artifacts, like test
helper executables, but the .gitignore files are not updated
accordingly.

Use Travis CI to help catch such issues earlier: check that there are
no untracked files at the end of any build jobs building Git (i.e. the
64 bit Clang and GCC Linux and OSX build jobs, plus the GETTEXT_POISON
and 32 bit Linux build jobs) or its documentation, and fail the build
job if there are any present.

Signed-off-by: SZEDER Gábor <szeder.dev@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
6 years agotravis-ci: don't store P4 and Git LFS in the working tree
SZEDER Gábor [Sun, 31 Dec 2017 16:02:05 +0000 (17:02 +0100)] 
travis-ci: don't store P4 and Git LFS in the working tree

The Clang and GCC 64 bit Linux build jobs download and store the P4
and Git LFS executables under the current directory, which is the
working tree that we are about to build and test.  This means that Git
commands like 'status' or 'ls-files' would list these files as
untracked.  The next commit is about to make sure that there are no
untracked files present after the build, and the downloaded
executables in the working tree are interfering with those upcoming
checks.

Therefore, let's download P4 and Git LFS in the home directory,
outside of the working tree.

Signed-off-by: SZEDER Gábor <szeder.dev@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
6 years agomerge: teach -Xours/-Xtheirs to symbolic link merge
Junio C Hamano [Tue, 26 Sep 2017 02:40:42 +0000 (11:40 +0900)] 
merge: teach -Xours/-Xtheirs to symbolic link merge

The -Xours/-Xtheirs merge options were originally defined as a way
to "force" the resolution of 3way textual merge conflicts to take
one side without using your editor, hence did not even trigger in
situations where you would normally not get the <<< === >>> conflict
markers.

This was improved for binary files back in 2012 with a944af1d
("merge: teach -Xours/-Xtheirs to binary ll-merge driver",
2012-09-08).

Teach a similar trick to the codepath that deals with merging two
conflicting changes to symbolic links.

Signed-off-by: Junio C Hamano <gitster@pobox.com>
Tested-by: Yaroslav Halchenko <yoh@onerussian.com>
Reviewed-by: Elijah Newren <newren@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
6 years agol10n: fr.po v2.16.0 round 1
Jean-Noel Avila [Sun, 31 Dec 2017 16:35:53 +0000 (17:35 +0100)] 
l10n: fr.po v2.16.0 round 1

Signed-off-by: Jean-Noel Avila <jn.avila@free.fr>
6 years agotravis-ci: record and skip successfully built trees
SZEDER Gábor [Sun, 31 Dec 2017 10:12:05 +0000 (11:12 +0100)] 
travis-ci: record and skip successfully built trees

Travis CI dutifully builds and tests each new branch tip, even if its
tree has previously been successfully built and tested.  This happens
often enough in contributors' workflows, when a work-in-progress
branch is rebased changing e.g. only commit messages or the order or
number of commits while leaving the resulting code intact, and is then
pushed to a Travis CI-enabled GitHub fork.

This is wasting Travis CI's resources and is sometimes scary-annoying
when the new tip commit with a tree identical to the previous,
successfully tested one is suddenly reported in red, because one of
the OSX build jobs happened to exceed the time limit yet again.

So extend our Travis CI build scripts to skip building commits whose
trees have previously been successfully built and tested.  Use the
Travis CI cache feature to keep a record of the object names of trees
that tested successfully, in a plain and simple flat text file, one
line per tree object name.  Append the current tree's object name at
the end of every successful build job to this file, along with a bit
of additional info about the build job (commit object name, Travis CI
job number and id).  Limit the size of this file to 1000 records, to
prevent it from growing too large for git/git's forever living
integration branches.  Check, using a simple grep invocation, in each
build job whether the current commit's tree is already in there, and
skip the build if it is.  Include a message in the skipped build job's
trace log, containing the URL to the build job successfully testing
that tree for the first time and instructions on how to force a
re-build.  Catch the case when a build job, which successfully built
and tested a particular tree for the first time, is restarted and omit
the URL of the previous build job's trace log, as in this case it's
the same build job and the trace log has just been overwritten.

Note: this won't kick in if two identical trees are on two different
branches, because Travis CI caches are not shared between build jobs
of different branches.

Signed-off-by: SZEDER Gábor <szeder.dev@gmail.com>
Reviewed-by: Lars Schneider <larsxschneider@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
6 years agotravis-ci: create the cache directory early in the build process
SZEDER Gábor [Sun, 31 Dec 2017 10:12:04 +0000 (11:12 +0100)] 
travis-ci: create the cache directory early in the build process

It seems that Travis CI creates the cache directory for us anyway,
even when a previous cache doesn't exist for the current build job.
Alas, this behavior is not explicitly documented, therefore we don't
rely on it and create the cache directory ourselves in those build
jobs that read/write cached data (currently only the prove state).

In the following commit we'll start to cache additional data in every
build job, and will access the cache much earlier in the build
process.

Therefore move creating the cache directory to 'ci/lib-travisci.sh' to
make sure that it exists at the very beginning of every build job.

Signed-off-by: SZEDER Gábor <szeder.dev@gmail.com>
Reviewed-by: Lars Schneider <larsxschneider@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>