Add a no-op wrapper library for Git's shell scripts. To split up the
gettext series I'm first submitting patches to gettextize the source
tree before I add any of the Makefile and Shell library changes needed
to actually use them.
Signed-off-by: NÆvar Arnfjörð Bjarmason <avarab@gmail.com>
Signed-off-by: NJunio C Hamano <gitster@pobox.com>

Add a git-sh-i18n--envsubst program which is a stripped-down version
of the GNU envsubst(1) program that comes with GNU gettext for use in
the eval_gettext() fallback.

We need a C helper program because implementing eval_gettext() purely
in shell turned out to be unworkable. Digging through the Git mailing
list archives will reveal two shell implementations of eval_gettext
that are almost good enough, but fail on an edge case which is tested
for in the tests which are part of this patch.

These are the modifications I made to envsubst.c as I turned it into
sh-i18n--envsubst.c:

 * Added our git-compat-util.h header for xrealloc() and friends.

 * Removed inclusion of gettext-specific headers.

 * Removed most of main() and replaced it with my own. The modified
   version only does option parsing for --variables. That's all it
   needs.

 * Modified error() invocations to use our error() instead of
   error(3).

 * Replaced the gettext XNMALLOC(n, size) macro with just
   xmalloc(n). Since XNMALLOC() only allocated char's.

 * Removed the string_list_destroy function. It's redundant (also in
   the upstream code).

 * Replaced the use of stdbool.h (a C99 header) by doing the following
   replacements on the code:

    * s/bool/unsigned short int/g
    * s/true/1/g
    * s/false/0/g
Reported-by: NJohannes Sixt <j.sixt@viscovery.net>
Signed-off-by: NÆvar Arnfjörð Bjarmason <avarab@gmail.com>
Signed-off-by: NJunio C Hamano <gitster@pobox.com>

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

* jn/run-command-error-failure:
  run-command: handle short writes and EINTR in die_child
  tests: check error message from run_command

* js/info-man-path:
  Documentation: clarify meaning of --html-path, --man-path, and --info-path
  git: add --info-path and --man-path options

Conflicts:
	Makefile

* im/hashcmp-optim:
  hashcmp(): inline memcmp() by hand to optimize

* jk/merge-one-file-working-tree:
  merge-one-file: fix broken merges with alternate work trees
  add tests for merge-index / merge-one-file

* jc/fix-diff-files-unmerged:
  diff-files: show unmerged entries correctly
  diff: remove often unused parameters from diff_unmerge()
  diff.c: return filepair from diff_unmerge()
  test: use $_z40 from test-lib

* cj/p4merge:
  Pass empty file to p4merge where no base is suitable.

* gr/cvsimport-alternative-cvspass-location:
  Look for password in both CVS and CVSNT password files.

* sg/completion-cleanup:
  completion: remove unnecessary _get_comp_words_by_ref() invocations
  completion: don't modify the $cur variable in completion functions

* js/blame-parsename:
  t/annotate-tests: Use echo & cat instead of sed
  blame: tolerate bogus e-mail addresses a bit better

* jk/format-patch-quote-special-in-from:
  pretty: quote rfc822 specials in email addresses

Conflicts:
	pretty.c
	t/t4014-format-patch.sh

* nd/struct-pathspec:
  pathspec: rename per-item field has_wildcard to use_wildcard
  Improve tree_entry_interesting() handling code
  Convert read_tree{,_recursive} to support struct pathspec
  Reimplement read_tree_recursive() using tree_entry_interesting()

* aw/maint-rebase-i-p-no-ff:
  git-rebase--interactive.sh: preserve-merges fails on merges created with no-ff

Conflicts:
	git-rebase--interactive.sh

* mg/alias-expose-prefix:
  handle_alias: provide GIT_PREFIX to !alias
  t1020: test !alias in subdirectory

* mg/diff-uiconfig-doc:
  config.txt,diff-options.txt: porcelain vs. plumbing for color.diff

* maint:
  Remove duplicated "is a"

Signed-off-by: NJoão Britto <jabcalves@gmail.com>
Signed-off-by: NJunio C Hamano <gitster@pobox.com>

* maint:
  t3701: fix here document
  git-fast-import.txt: --relative-marks takes no parameter
  shell: add missing initialization of argv0_path

A broken here-document was not caught because end of file is taken by
an implicit end of the here document (POSIX does not seem to say it is
an error to lack the delimiter), and everything in the test just turned
into a single "cat into a file".
Noticed-by: NKacper Kornet <draenog@pld-linux.org>
Signed-off-by: NJunio C Hamano <gitster@pobox.com>

Remove spurious "=" after --relative-marks.
Signed-off-by: NMichael J Gruber <git@drmicha.warpmail.net>
Signed-off-by: NJunio C Hamano <gitster@pobox.com>

According to c6dfb399 (remote-curl: add missing initialization of
argv0_path, 2009-10-13), stand-alone programs (non-builtins)
must call git_extract_argv0_path(argv[0]) in order to help builds
that derive the installation prefix at runtime. Without this call,
the program segfaults (or raises an assertion failure).
Signed-off-by: NDima Sharov <git.avalakvista@gmail.com>
Acked-by: NJohannes Sixt <j6t@kdbg.org>
Signed-off-by: NJunio C Hamano <gitster@pobox.com>

The use of the sed command "1i No robots allowed" caused the version
of sed in OS X to die with

   sed: 1: "1i "No robots allowed"\n": command i expects \ followed by
   text

Since this command was just trying to add a single line to the
beginning of the file, do the same with "echo" followed by "cat".

Unbreaks t8001 and t8002 on OS X 10.6.7
Signed-off-by: NBrian Gernhardt <brian@gernhardtsoftware.com>
Signed-off-by: NJunio C Hamano <gitster@pobox.com>

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

* ml/test-readme:
  t/README: unify documentation of test function args

* ft/gitweb-tar-with-gzip-n:
  gitweb: supply '-n' to gzip for identical output

* mg/rev-list-count-cherry:
  rev-list --count: separate count for --cherry-mark

* ss/doc-svn:
  remove noise and inaccuracies from git-svn docs

* jn/format-patch-doc:
  Documentation/format-patch: suggest Toggle Word Wrap add-on for Thunderbird
  Documentation: publicize hints for sending patches with GMail
  Documentation: publicize KMail hints for sending patches inline
  Documentation: hints for sending patches inline with Thunderbird
  Documentation: explain how to check for patch corruption

* vr/merge-base-doc:
  Restructure documentation for git-merge-base.
  Documentation: update to git-merge-base --octopus

* ss/cherry-pick-x-doc:
  doc: Clarify that "cherry-pick -x" does not use "git notes"

* jn/maint-format-patch-doc:
  Documentation: describe the format of messages with inline patches

* jk/format-patch-multiline-header:
  format-patch: wrap email addresses after long names

* jm/mergetool-submodules:
  mergetool: Teach about submodules

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

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

* jh/dirstat:
  --dirstat: In case of renames, use target filename instead of source filename
  Teach --dirstat not to completely ignore rearranged lines within a file
  --dirstat-by-file: Make it faster and more correct
  --dirstat: Describe non-obvious differences relative to --stat or regular diff

* sp/maint-clear-postfields:
  http: clear POSTFIELDS when initializing a slot

* jk/maint-stash-oob:
  stash: fix false positive in the invalid ref test.
  stash: fix accidental apply of non-existent stashes

Conflicts:
	t/t3903-stash.sh