git.git
11 years agolog: grep author/committer using mailmap
Antoine Pelisse [Sat, 5 Jan 2013 21:26:45 +0000 (22:26 +0100)]
log: grep author/committer using mailmap

Currently you can use mailmap to display log authors and committers
but you can't use the mailmap to find commits with mapped values.

This commit allows you to run:

    git log --use-mailmap --author mapped_name_or_email
    git log --use-mailmap --committer mapped_name_or_email

Of course it only works if the --use-mailmap option is used.

The new name and email are copied only when necessary.

Signed-off-by: Antoine Pelisse <apelisse@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
11 years agotest: add test for --use-mailmap option
Antoine Pelisse [Sat, 5 Jan 2013 21:26:44 +0000 (22:26 +0100)]
test: add test for --use-mailmap option

The new option '--use-mailmap' can be used to make sure that mailmap
file is used to convert name when running log commands.

The test is simple and checks that the Author line
is correctly replaced when running log.

Signed-off-by: Antoine Pelisse <apelisse@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
11 years agolog: add --use-mailmap option
Antoine Pelisse [Sat, 5 Jan 2013 21:26:43 +0000 (22:26 +0100)]
log: add --use-mailmap option

Add the --use-mailmap option to log commands. It allows to display
names from mailmap file when displaying logs, whatever the format
used.

Signed-off-by: Antoine Pelisse <apelisse@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
11 years agopretty: use mailmap to display username and email
Antoine Pelisse [Sat, 5 Jan 2013 21:26:42 +0000 (22:26 +0100)]
pretty: use mailmap to display username and email

Use the mailmap information to display the rewritten
username and email address in all log commands.

Signed-off-by: Antoine Pelisse <apelisse@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
11 years agomailmap: add mailmap structure to rev_info and pp
Antoine Pelisse [Sat, 5 Jan 2013 21:26:41 +0000 (22:26 +0100)]
mailmap: add mailmap structure to rev_info and pp

Pass a mailmap from rev_info to pretty_print_context to so that the
pretty printer can use rewritten name and email address when showing
commits.

Signed-off-by: Antoine Pelisse <apelisse@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
11 years agomailmap: simplify map_user() interface
Antoine Pelisse [Sat, 5 Jan 2013 21:26:40 +0000 (22:26 +0100)]
mailmap: simplify map_user() interface

Simplify map_user(), mostly to avoid copies of string buffers. It
also simplifies caller functions.

map_user() directly receive pointers and length from the commit buffer
as mail and name. If mapping of the user and mail can be done, the
pointer is updated to a new location. Lengths are also updated if
necessary.

The caller of map_user() can then copy the new email and name if
necessary.

Signed-off-by: Antoine Pelisse <apelisse@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
11 years agomailmap: remove email copy and length limitation
Junio C Hamano [Sat, 5 Jan 2013 21:26:39 +0000 (22:26 +0100)]
mailmap: remove email copy and length limitation

In map_user(), we have email pointer that points at the beginning of
an e-mail address, but the buffer is not terminated with a NUL after
the e-mail address.  It typically has ">" after the address, and it
could have even more if it comes from author/committer line in a
commit object.  Or it may not have ">" after it.

We used to copy the e-mail address proper into a temporary buffer
before asking the string-list API to find the e-mail address in the
mailmap, because string_list_lookup() function only takes a NUL
terminated full string.

Introduce a helper function lookup_prefix that takes the email
pointer and the length, and finds a matching entry in the string
list used for the mailmap, by doing the following:

 - First ask string_list_find_insert_index() where in its sorted
   list the e-mail address we have (including the possible trailing
   junk ">...") would be inserted.

 - It could find an exact match (e.g. we had a clean e-mail address
   without any trailing junk).  We can return the item in that case.

 - Or it could return the index of an item that sorts after the
   e-mail address we have.

 - If we did not find an exact match against a clean e-mail address,
   then the record we are looking for in the mailmap has to exist
   before the index returned by the function (i.e. "email>junk"
   always sorts later than "email").  Iterate, starting from that
   index, down the map->items[] array until we find the exact record
   we are looking for, or we see a record with a key that definitely
   sorts earlier than the e-mail we are looking for (i.e. when we
   are looking for "email" in "email>junk", a record in the mailmap
   that begins with "emaik" strictly sorts before "email", if such a
   key existed in the mailmap).

This, together with the earlier enhancement to support
case-insensitive sorting, allow us to remove an extra copy of email
buffer to downcase it.

A part of this is based on Antoine Pelisse's previous work.

Signed-off-by: Junio C Hamano <gitster@pobox.com>
11 years agoUse split_ident_line to parse author and committer
Antoine Pelisse [Sat, 5 Jan 2013 21:26:38 +0000 (22:26 +0100)]
Use split_ident_line to parse author and committer

Currently blame.c::get_acline(), pretty.c::pp_user_info() and
shortlog.c::insert_one_record() are parsing author name, email, time
and tz themselves.

Use ident.c::split_ident_line() for better code reuse.

Signed-off-by: Antoine Pelisse <apelisse@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
11 years agostring-list: allow case-insensitive string list
Junio C Hamano [Mon, 7 Jan 2013 20:24:55 +0000 (12:24 -0800)]
string-list: allow case-insensitive string list

Some string list needs to be searched case insensitively, and for
that to work correctly, the string needs to be sorted case
insensitively from the beginning.

Allow a custom comparison function to be defined on a string list
instance and use it throughout in place of strcmp().

Signed-off-by: Junio C Hamano <gitster@pobox.com>
12 years agoSync with 1.8.0.2
Junio C Hamano [Mon, 10 Dec 2012 21:07:12 +0000 (13:07 -0800)]
Sync with 1.8.0.2

* maint:
  Git 1.8.0.2
  Documentation/git-stash.txt: add a missing verb
  git(1): remove a defunct link to "list of authors"

Signed-off-by: Junio C Hamano <gitster@pobox.com>
12 years agoGit 1.8.0.2
Junio C Hamano [Mon, 10 Dec 2012 21:05:47 +0000 (13:05 -0800)]
Git 1.8.0.2

Signed-off-by: Junio C Hamano <gitster@pobox.com>
12 years agoMakefile: whitespace style fixes in macro definitions
Stefano Lattarini [Sun, 9 Dec 2012 10:36:17 +0000 (11:36 +0100)]
Makefile: whitespace style fixes in macro definitions

Consistently use a single space before and after the "=" (or ":=", "+=",
etc.) in assignments to make macros.  Granted, this was not a big deal,
but I did find the needless inconsistency quite distracting.

Signed-off-by: Stefano Lattarini <stefano.lattarini@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
12 years agoDocumentation/git-stash.txt: add a missing verb
Sébastien Loriot [Mon, 10 Dec 2012 07:22:34 +0000 (08:22 +0100)]
Documentation/git-stash.txt: add a missing verb

Signed-off-by: Sébastien Loriot <sloriot.ml@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
12 years agogit(1): remove a defunct link to "list of authors"
Junio C Hamano [Fri, 7 Dec 2012 17:54:50 +0000 (09:54 -0800)]
git(1): remove a defunct link to "list of authors"

The linked page has not been showing the promised "more complete
list" for more than 6 months by now, and nobody has resurrected
the list there nor elsewhere since then.

Signed-off-by: Junio C Hamano <gitster@pobox.com>
12 years agoGit 1.8.1-rc1
Junio C Hamano [Fri, 7 Dec 2012 22:18:55 +0000 (14:18 -0800)]
Git 1.8.1-rc1

Signed-off-by: Junio C Hamano <gitster@pobox.com>
12 years agoDocumentation/diff-config: work around AsciiDoc misfortune
Junio C Hamano [Fri, 7 Dec 2012 23:15:59 +0000 (15:15 -0800)]
Documentation/diff-config: work around AsciiDoc misfortune

The line that happens to begin with indent followed by "3. " was
interpreted as if it was an enumerated list; just wrap the lines
differently to work it around for now.

Signed-off-by: Junio C Hamano <gitster@pobox.com>
12 years agoMerge branch 'maint'
Junio C Hamano [Fri, 7 Dec 2012 22:16:52 +0000 (14:16 -0800)]
Merge branch 'maint'

* maint:
  Update draft release notes to 1.8.0.2

12 years agoUpdate draft release notes to 1.8.0.2
Junio C Hamano [Fri, 7 Dec 2012 22:16:38 +0000 (14:16 -0800)]
Update draft release notes to 1.8.0.2

Signed-off-by: Junio C Hamano <gitster@pobox.com>
12 years agoMerge branch 'jc/doc-push-satellite' into maint
Junio C Hamano [Fri, 7 Dec 2012 22:11:21 +0000 (14:11 -0800)]
Merge branch 'jc/doc-push-satellite' into maint

* jc/doc-push-satellite:
  Documentation/git-push.txt: clarify the "push from satellite" workflow

12 years agoMerge branch 'jc/same-encoding' into maint
Junio C Hamano [Fri, 7 Dec 2012 22:10:56 +0000 (14:10 -0800)]
Merge branch 'jc/same-encoding' into maint

Various codepaths checked if two encoding names are the same using
ad-hoc code and some of them ended up asking iconv() to convert
between "utf8" and "UTF-8".  The former is not a valid way to spell
the encoding name, but often people use it by mistake, and we
equated them in some but not all codepaths. Introduce a new helper
function to make these codepaths consistent.

* jc/same-encoding:
  reencode_string(): introduce and use same_encoding()

12 years agoMerge branch 'lt/diff-stat-show-0-lines' into maint
Junio C Hamano [Fri, 7 Dec 2012 22:10:17 +0000 (14:10 -0800)]
Merge branch 'lt/diff-stat-show-0-lines' into maint

"git diff --stat" miscounted the total number of changed lines when
binary files were involved and hidden beyond --stat-count.  It also
miscounted the total number of changed files when there were
unmerged paths.

* lt/diff-stat-show-0-lines:
  t4049: refocus tests
  diff --shortstat: do not count "unmerged" entries
  diff --stat: do not count "unmerged" entries
  diff --stat: move the "total count" logic to the last loop
  diff --stat: use "file" temporary variable to refer to data->files[i]
  diff --stat: status of unmodified pair in diff-q is not zero
  test: add failing tests for "diff --stat" to t4049
  Fix "git diff --stat" for interesting - but empty - file changes

12 years agoMerge branch 'master' of git://github.com/git-l10n/git-po
Junio C Hamano [Fri, 7 Dec 2012 18:32:22 +0000 (10:32 -0800)]
Merge branch 'master' of git://github.com/git-l10n/git-po

* 'master' of git://github.com/git-l10n/git-po:
  l10n: de.po: translate 22 new messages
  l10n: de.po: translate 825 new messages
  l10n: Update Swedish translation (1979t0f0u)
  l10n: vi.po: update to git-v1.8.0.1-347-gf94c3
  l10n: Update git.pot (5 new, 1 removed messages)

12 years agoMerge branch 'rr/t4041-cleanup'
Junio C Hamano [Fri, 7 Dec 2012 18:31:19 +0000 (10:31 -0800)]
Merge branch 'rr/t4041-cleanup'

* rr/t4041-cleanup:
  t4041 (diff-submodule-option): modernize style
  t4041 (diff-submodule-option): rewrite add_file() routine
  t4041 (diff-submodule-option): parse digests sensibly
  t4041 (diff-submodule-option): don't hardcode SHA-1 in expected outputs

12 years agoMerge branch git://github.com/ralfth/git-po-de
Jiang Xin [Fri, 7 Dec 2012 09:28:58 +0000 (17:28 +0800)]
Merge branch git://github.com/ralfth/git-po-de

* 'rt/de-l10n-updates-for-1.8.1' of git://github.com/ralfth/git-po-de:
  l10n: de.po: translate 22 new messages
  l10n: de.po: translate 825 new messages

Signed-off-by: Jiang Xin <worldhello.net@gmail.com>
12 years agol10n: de.po: translate 22 new messages
Ralf Thielow [Sun, 2 Dec 2012 15:57:38 +0000 (16:57 +0100)]
l10n: de.po: translate 22 new messages

Translate 22 new messages came from git.pot
updates in 9306b5b (l10n: Update git.pot (3 new,
6 removed messages)), fe52cd6 (l10n: Update git.pot
(14 new, 3 removed messages)) and f9472e3
(l10n: Update git.pot (5 new, 1 removed messages)).

Signed-off-by: Ralf Thielow <ralf.thielow@gmail.com>
Reviewed-by: Michael J Gruber <git@drmicha.warpmail.net>
12 years agol10n: de.po: translate 825 new messages
Ralf Thielow [Sun, 16 Sep 2012 11:15:34 +0000 (13:15 +0200)]
l10n: de.po: translate 825 new messages

Translate 825 new messages came from git.pot update in
cc76011 ("l10n: Update git.pot (825 new, 24 removed messages)").

Signed-off-by: Ralf Thielow <ralf.thielow@gmail.com>
Reviewed-by: Thomas Rast <trast@student.ethz.ch>
Helped-by: Michael J Gruber <git@drmicha.warpmail.net>
12 years agoMerge branch 'mm/status-push-pull-advise'
Junio C Hamano [Tue, 4 Dec 2012 21:34:10 +0000 (13:34 -0800)]
Merge branch 'mm/status-push-pull-advise'

* mm/status-push-pull-advise:
  document that statusHints affects git checkout

12 years agodocument that statusHints affects git checkout
Matthieu Moy [Tue, 4 Dec 2012 09:15:03 +0000 (10:15 +0100)]
document that statusHints affects git checkout

Signed-off-by: Matthieu Moy <Matthieu.Moy@imag.fr>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
12 years agoGit 1.8.1-rc0
Junio C Hamano [Mon, 3 Dec 2012 17:49:51 +0000 (09:49 -0800)]
Git 1.8.1-rc0

Signed-off-by: Junio C Hamano <gitster@pobox.com>
12 years agoMerge branch 'mm/status-push-pull-advise'
Junio C Hamano [Mon, 3 Dec 2012 17:27:29 +0000 (09:27 -0800)]
Merge branch 'mm/status-push-pull-advise'

Finishing touch to allow the new advice message squelched
with an advice.* configuration variable.

* mm/status-push-pull-advise:
  status: respect advice.statusHints for ahead/behind advice

12 years agostatus: respect advice.statusHints for ahead/behind advice
Jeff King [Mon, 3 Dec 2012 06:16:57 +0000 (01:16 -0500)]
status: respect advice.statusHints for ahead/behind advice

If the user has unset advice.statusHints, we already
suppress the "use git reset to..." hints in each stanza. The
new "use git push to publish..." hint is the same type of
hint. Let's respect statusHints for it, rather than making
the user set yet another advice flag.

Signed-off-by: Jeff King <peff@peff.net>
Acked-by: Matthieu Moy <Matthieu.Moy@grenoble-inp.fr>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
12 years agot4041 (diff-submodule-option): modernize style
Ramkumar Ramachandra [Fri, 30 Nov 2012 11:37:36 +0000 (17:07 +0530)]
t4041 (diff-submodule-option): modernize style

- Enclose tests in single quotes as opposed to double quotes.  This is
  the prevalent style in other tests.
- Remove the unused variable $head4_full.
- Indent the expected output so that it lines up with the rest of the
  test text.

Signed-off-by: Ramkumar Ramachandra <artagnon@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
12 years agot4041 (diff-submodule-option): rewrite add_file() routine
Ramkumar Ramachandra [Fri, 30 Nov 2012 11:37:35 +0000 (17:07 +0530)]
t4041 (diff-submodule-option): rewrite add_file() routine

Instead of "cd there and then come back", use the "cd there in a
subshell" pattern.  Also fix '&&' chaining in one place.

Suggested-by: Junio C Hamano <gitster@pobox.com>
Signed-off-by: Ramkumar Ramachandra <artagnon@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
12 years agot4041 (diff-submodule-option): parse digests sensibly
Ramkumar Ramachandra [Fri, 30 Nov 2012 11:37:34 +0000 (17:07 +0530)]
t4041 (diff-submodule-option): parse digests sensibly

`git rev-list --max-count=1 HEAD` is a roundabout way of saying `git
rev-parse --verify HEAD`; replace a bunch of instances of the former
with the latter.  Also, don't unnecessarily `cut -c1-7` the rev-parse
output when the `--short` option is available.

Signed-off-by: Ramkumar Ramachandra <artagnon@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
12 years agol10n: Update Swedish translation (1979t0f0u)
Peter Krefting [Fri, 30 Nov 2012 09:51:14 +0000 (10:51 +0100)]
l10n: Update Swedish translation (1979t0f0u)

Signed-off-by: Peter Krefting <peter@softwolves.pp.se>
12 years agol10n: vi.po: update to git-v1.8.0.1-347-gf94c3
Tran Ngoc Quan [Fri, 30 Nov 2012 06:43:11 +0000 (13:43 +0700)]
l10n: vi.po: update to git-v1.8.0.1-347-gf94c3

 * updated all new messages (1979t0f0u)

Signed-off-by: Tran Ngoc Quan <vnwildman@gmail.com>
12 years agol10n: Update git.pot (5 new, 1 removed messages)
Jiang Xin [Fri, 30 Nov 2012 04:41:47 +0000 (12:41 +0800)]
l10n: Update git.pot (5 new, 1 removed messages)

L10n for git 1.8.1 round 2: Generate po/git.pot from v1.8.0.1-347-gf94c3.

Signed-off-by: Jiang Xin <worldhello.net@gmail.com>
12 years agoUpdate draft release notes to 1.8.1
Junio C Hamano [Thu, 29 Nov 2012 21:57:09 +0000 (13:57 -0800)]
Update draft release notes to 1.8.1

12 years agoMerge branch 'pw/p4-various-fixes'
Junio C Hamano [Thu, 29 Nov 2012 21:44:28 +0000 (13:44 -0800)]
Merge branch 'pw/p4-various-fixes'

* pw/p4-various-fixes:
  git p4: remove unneeded cmd initialization
  git p4: fix labelDetails typo in exception
  git p4 test: display unresolvable host error
  git p4: catch p4 errors when streaming file contents
  git p4: handle servers without move support
  git p4: catch p4 describe errors

12 years agoMerge branch 'lt/diff-stat-show-0-lines'
Junio C Hamano [Thu, 29 Nov 2012 20:53:54 +0000 (12:53 -0800)]
Merge branch 'lt/diff-stat-show-0-lines'

"git diff --stat" miscounted the total number of changed lines when
binary files were involved and hidden beyond --stat-count.  It also
miscounted the total number of changed files when there were
unmerged paths.

* lt/diff-stat-show-0-lines:
  t4049: refocus tests
  diff --shortstat: do not count "unmerged" entries
  diff --stat: do not count "unmerged" entries
  diff --stat: move the "total count" logic to the last loop
  diff --stat: use "file" temporary variable to refer to data->files[i]
  diff --stat: status of unmodified pair in diff-q is not zero
  test: add failing tests for "diff --stat" to t4049

12 years agoMerge branch 'fc/remote-hg'
Junio C Hamano [Thu, 29 Nov 2012 20:53:50 +0000 (12:53 -0800)]
Merge branch 'fc/remote-hg'

New remote helper for hg.

* fc/remote-hg: (22 commits)
  remote-hg: fix for older versions of python
  remote-hg: fix for files with spaces
  remote-hg: avoid bad refs
  remote-hg: try the 'tip' if no checkout present
  remote-hg: fix compatibility with older versions of hg
  remote-hg: add missing config for basic tests
  remote-hg: the author email can be null
  remote-hg: add option to not track branches
  remote-hg: add extra author test
  remote-hg: add tests to compare with hg-git
  remote-hg: add bidirectional tests
  test-lib: avoid full path to store test results
  remote-hg: add basic tests
  remote-hg: fake bookmark when there's none
  remote-hg: add compat for hg-git author fixes
  remote-hg: add support for hg-git compat mode
  remote-hg: match hg merge behavior
  remote-hg: make sure the encoding is correct
  remote-hg: add support to push URLs
  remote-hg: add support for remote pushing
  ...

12 years agoMerge branch 'mk/complete-tcsh'
Junio C Hamano [Thu, 29 Nov 2012 20:53:38 +0000 (12:53 -0800)]
Merge branch 'mk/complete-tcsh'

Finishing touches for tcsh completion.

* mk/complete-tcsh:
  Support for git aliasing for tcsh completion

12 years agoMerge branch 'jc/doc-push-satellite'
Junio C Hamano [Thu, 29 Nov 2012 20:52:54 +0000 (12:52 -0800)]
Merge branch 'jc/doc-push-satellite'

* jc/doc-push-satellite:
  Documentation/git-push.txt: clarify the "push from satellite" workflow

12 years agoMerge branch 'km/send-email-remove-cruft-in-address'
Junio C Hamano [Thu, 29 Nov 2012 20:52:49 +0000 (12:52 -0800)]
Merge branch 'km/send-email-remove-cruft-in-address'

* km/send-email-remove-cruft-in-address:
  git-send-email: allow edit invalid email address
  git-send-email: ask what to do with an invalid email address
  git-send-email: remove invalid addresses earlier
  git-send-email: fix fallback code in extract_valid_address()
  git-send-email: remove garbage after email address

12 years agoMerge branch 'jk/send-email-sender-prompt'
Junio C Hamano [Thu, 29 Nov 2012 20:52:45 +0000 (12:52 -0800)]
Merge branch 'jk/send-email-sender-prompt'

General clean-ups in various areas, originally written to support a
patch that later turned out to be unneeded.

* jk/send-email-sender-prompt:
  t9001: check send-email behavior with implicit sender
  t: add tests for "git var"
  ident: keep separate "explicit" flags for author and committer
  ident: make user_ident_explicitly_given static
  t7502: factor out autoident prerequisite
  test-lib: allow negation of prerequisites

12 years agoMerge branch 'fc/send-email-no-sender-prompt'
Junio C Hamano [Thu, 29 Nov 2012 20:52:42 +0000 (12:52 -0800)]
Merge branch 'fc/send-email-no-sender-prompt'

* fc/send-email-no-sender-prompt:
  send-email: avoid questions when user has an ident

12 years agoMerge branch 'er/doc-add-new-commands'
Junio C Hamano [Thu, 29 Nov 2012 20:52:36 +0000 (12:52 -0800)]
Merge branch 'er/doc-add-new-commands'

* er/doc-add-new-commands:
  Documentation: how to add a new command

12 years agoMerge branch 'jl/submodule-rm'
Junio C Hamano [Thu, 29 Nov 2012 20:52:30 +0000 (12:52 -0800)]
Merge branch 'jl/submodule-rm'

Finishing touches to "git rm $submodule" that removes the working
tree of a submodule.

* jl/submodule-rm:
  Teach rm to remove submodules when given with a trailing '/'

12 years agoMerge branch 'pp/gitweb-config-underscore'
Junio C Hamano [Thu, 29 Nov 2012 20:52:16 +0000 (12:52 -0800)]
Merge branch 'pp/gitweb-config-underscore'

The key "gitweb.remote_heads" is not legal git config; this maps it to
"gitweb.remoteheads".

* pp/gitweb-config-underscore:
  gitweb: make remote_heads config setting work

12 years agoMerge branch 'fc/completion-test-simplification'
Junio C Hamano [Thu, 29 Nov 2012 20:52:10 +0000 (12:52 -0800)]
Merge branch 'fc/completion-test-simplification'

Clean up completion tests.  Use of conslidated helper may make
instrumenting one particular test during debugging of the test
itself, but I think that issue should be addressed in some other
way (e.g. making sure individual tests in 9902 can be skipped).

* fc/completion-test-simplification:
  completion: simplify __gitcomp() test helper
  completion: refactor __gitcomp related tests
  completion: consolidate test_completion*() tests
  completion: simplify tests using test_completion_long()
  completion: standardize final space marker in tests
  completion: add comment for test_completion()

12 years agoMerge branch 'maint'
Junio C Hamano [Thu, 29 Nov 2012 20:21:17 +0000 (12:21 -0800)]
Merge branch 'maint'

* maint:
  git-fast-import.txt: improve documentation for quoted paths
  git-remote-mediawiki: escape ", \, and LF in file names

12 years agogit-fast-import.txt: improve documentation for quoted paths
Matthieu Moy [Thu, 29 Nov 2012 19:11:32 +0000 (20:11 +0100)]
git-fast-import.txt: improve documentation for quoted paths

The documentation mentioned only newlines and double quotes as
characters needing escaping, but the backslash also needs it. Also, the
documentation was not clearly saying that double quotes around the file
name were required (double quotes in the examples could be interpreted as
part of the sentence, not part of the actual string).

Signed-off-by: Matthieu Moy <Matthieu.Moy@imag.fr>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
12 years agogit-remote-mediawiki: escape ", \, and LF in file names
Matthieu Moy [Thu, 29 Nov 2012 17:00:55 +0000 (18:00 +0100)]
git-remote-mediawiki: escape ", \, and LF in file names

A mediawiki page can contain, and even start with a " character, we have
to escape it when generating the fast-export stream, as well as \
character. While we're there, also escape newlines, but I don't think we
can get them from MediaWiki pages.

Signed-off-by: Matthieu Moy <Matthieu.Moy@imag.fr>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
12 years agoMerge branch 'master' of git://github.com/git-l10n/git-po
Junio C Hamano [Thu, 29 Nov 2012 18:05:51 +0000 (10:05 -0800)]
Merge branch 'master' of git://github.com/git-l10n/git-po

Further l10n updates.

* 'master' of git://github.com/git-l10n/git-po:
  l10n: vi.po: Update follow git-v1.8.0-273-g2d242

12 years agot4049: refocus tests
Junio C Hamano [Thu, 29 Nov 2012 17:46:30 +0000 (09:46 -0800)]
t4049: refocus tests

The primary thing Linus's patch wanted to change was to make sure
that 0-line change appears for a mode-only change.  Update the
first test to chmod a file that we can see in the output (limited
by --stat-count) to demonstrate it.  Also make sure to use test_chmod
and compare the index and the tree, so that we can run this test
even on a filesystem without permission bits.

Later two tests are about fixes to separate issues that were
introduced and/or uncovered by Linus's patch as a side effect, but
the issues are not related to mode-only changes.  Remove chmod from
the tests.

Signed-off-by: Junio C Hamano <gitster@pobox.com>
12 years agocompletion: fix warning for zsh
Felipe Contreras [Thu, 29 Nov 2012 08:20:57 +0000 (09:20 +0100)]
completion: fix warning for zsh

Otherwise the user might get something like:

  git-completion.sh:2466: command not found: compdef

If this script is loaded before compinit. The script would work either
way, but let's not be more annoying to the user.

Signed-off-by: Felipe Contreras <felipe.contreras@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
12 years agoMerge git://github.com/vnwildman/git
Jiang Xin [Thu, 29 Nov 2012 08:25:40 +0000 (16:25 +0800)]
Merge git://github.com/vnwildman/git

* git://github.com/vnwildman/git:
  l10n: vi.po: Update follow git-v1.8.0-273-g2d242

12 years agoMerge branch 'master' of git://github.com/git-l10n/git-po
Junio C Hamano [Thu, 29 Nov 2012 05:58:27 +0000 (21:58 -0800)]
Merge branch 'master' of git://github.com/git-l10n/git-po

Update the localization string up to 2d242fb (Update draft release
notes for 1.8.1, 2012-11-21)

* 'master' of git://github.com/git-l10n/git-po:
  l10n: Update Swedish translation (1975t0f0u)
  l10n: vi.po: update to git-v1.7.12-437-g1084f
  l10n: Update git.pot (14 new, 3 removed messages)

12 years agoMerge branch 'maint'
Junio C Hamano [Wed, 28 Nov 2012 21:49:33 +0000 (13:49 -0800)]
Merge branch 'maint'

12 years agoUpdate draft release notes to 1.8.1
Junio C Hamano [Wed, 28 Nov 2012 21:49:10 +0000 (13:49 -0800)]
Update draft release notes to 1.8.1

Signed-off-by: Junio C Hamano <gitster@pobox.com>
12 years agoMerge branch 'fc/zsh-completion'
Junio C Hamano [Wed, 28 Nov 2012 21:42:36 +0000 (13:42 -0800)]
Merge branch 'fc/zsh-completion'

* fc/zsh-completion:
  completion: start moving to the new zsh completion
  completion: add new zsh completion

12 years agoMerge branch 'mm/status-push-pull-advise'
Junio C Hamano [Wed, 28 Nov 2012 21:42:30 +0000 (13:42 -0800)]
Merge branch 'mm/status-push-pull-advise'

* mm/status-push-pull-advise:
  status: add advice on how to push/pull to tracking branch

12 years agoMerge branch 'jk/pickaxe-textconv'
Junio C Hamano [Wed, 28 Nov 2012 21:42:24 +0000 (13:42 -0800)]
Merge branch 'jk/pickaxe-textconv'

Use textconv filters when searching with "log -S".

* jk/pickaxe-textconv:
  pickaxe: use textconv for -S counting
  pickaxe: hoist empty needle check

12 years agoStart preparing for 1.8.0.2
Junio C Hamano [Wed, 28 Nov 2012 21:40:02 +0000 (13:40 -0800)]
Start preparing for 1.8.0.2

Signed-off-by: Junio C Hamano <gitster@pobox.com>
12 years agot9001: check send-email behavior with implicit sender
Jeff King [Wed, 28 Nov 2012 20:06:26 +0000 (15:06 -0500)]
t9001: check send-email behavior with implicit sender

We allow send-email to use an implicitly-defined identity
for the sender (because there is still a confirmation step),
but we abort when we cannot generate such an identity. Let's
make sure that we test this.

Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
12 years agoMerge branch 'rh/maint-gitweb-highlight-ext' into maint
Junio C Hamano [Wed, 28 Nov 2012 20:05:30 +0000 (12:05 -0800)]
Merge branch 'rh/maint-gitweb-highlight-ext' into maint

Syntax highlighting in "gitweb" was not quite working.

* rh/maint-gitweb-highlight-ext:
  gitweb.perl: fix %highlight_ext mappings

12 years agoMerge branch 'pw/maint-p4-rcs-expansion-newline' into maint
Junio C Hamano [Wed, 28 Nov 2012 20:04:32 +0000 (12:04 -0800)]
Merge branch 'pw/maint-p4-rcs-expansion-newline' into maint

"git p4" used to try expanding malformed "$keyword$" that spans
across multiple lines.

* pw/maint-p4-rcs-expansion-newline:
  git p4: RCS expansion should not span newlines

12 years agocompletion: add options --single-branch and --branch to "git clone"
Ralf Thielow [Wed, 28 Nov 2012 18:27:02 +0000 (19:27 +0100)]
completion: add options --single-branch and --branch to "git clone"

Signed-off-by: Ralf Thielow <ralf.thielow@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
12 years agoMerge branch 'fc/send-email-no-sender-prompt' into jk/send-email-sender-prompt
Junio C Hamano [Wed, 28 Nov 2012 18:50:20 +0000 (10:50 -0800)]
Merge branch 'fc/send-email-no-sender-prompt' into jk/send-email-sender-prompt

* fc/send-email-no-sender-prompt:
  send-email: avoid questions when user has an ident

12 years agot: add tests for "git var"
Jeff King [Wed, 28 Nov 2012 18:26:43 +0000 (13:26 -0500)]
t: add tests for "git var"

We do not currently have any explicit tests for "git var" at
all (though we do exercise it to some degree as a part of
other tests). Let's add a few basic sanity checks.

Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
12 years agoDocumentation/git-push.txt: clarify the "push from satellite" workflow
Junio C Hamano [Tue, 27 Nov 2012 23:52:27 +0000 (15:52 -0800)]
Documentation/git-push.txt: clarify the "push from satellite" workflow

The context of the example to push into refs/remotes/satellite/
hierarchy of the other repository needs to be spelled out explicitly
for the value of this example to be fully appreciated.  Make it so.

Signed-off-by: Junio C Hamano <gitster@pobox.com>
12 years agoremote-hg: fix for older versions of python
Felipe Contreras [Wed, 28 Nov 2012 01:01:33 +0000 (02:01 +0100)]
remote-hg: fix for older versions of python

As Amit Bakshi reported, older versions of python (< 2.7) don't have
subprocess.check_output, so let's use subprocess.Popen directly as
suggested.

Suggested-by: Amit Bakshi <ambakshi@gmail.com>
Signed-off-by: Felipe Contreras <felipe.contreras@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
12 years agoremote-hg: fix for files with spaces
Felipe Contreras [Wed, 28 Nov 2012 01:01:32 +0000 (02:01 +0100)]
remote-hg: fix for files with spaces

Signed-off-by: Felipe Contreras <felipe.contreras@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
12 years agodiff --shortstat: do not count "unmerged" entries
Junio C Hamano [Tue, 27 Nov 2012 22:19:36 +0000 (14:19 -0800)]
diff --shortstat: do not count "unmerged" entries

Fix the same issue as the previous one for "git diff --stat";
unmerged entries was doubly-counted.

Signed-off-by: Junio C Hamano <gitster@pobox.com>
12 years agoMerge branch 'maint'
Junio C Hamano [Tue, 27 Nov 2012 21:29:08 +0000 (13:29 -0800)]
Merge branch 'maint'

12 years agoMerge branch 'nd/maint-compat-fnmatch-fix' into maint
Junio C Hamano [Tue, 27 Nov 2012 21:29:00 +0000 (13:29 -0800)]
Merge branch 'nd/maint-compat-fnmatch-fix' into maint

* nd/maint-compat-fnmatch-fix:
  compat/fnmatch: fix off-by-one character class's length check

12 years agoMerge branch 'jh/update-ref-d-through-symref' into maint
Junio C Hamano [Tue, 27 Nov 2012 21:28:45 +0000 (13:28 -0800)]
Merge branch 'jh/update-ref-d-through-symref' into maint

* jh/update-ref-d-through-symref:
  Fix failure to delete a packed ref through a symref
  t1400-update-ref: Add test verifying bug with symrefs in delete_ref()

12 years agoMerge branch 'esr/maint-doc-fast-import' into maint
Junio C Hamano [Tue, 27 Nov 2012 21:28:31 +0000 (13:28 -0800)]
Merge branch 'esr/maint-doc-fast-import' into maint

* esr/maint-doc-fast-import:
  doc/fast-import: clarify how content states are built

12 years agoMerge branch 'wtk/submodule-doc-fixup' into maint
Junio C Hamano [Tue, 27 Nov 2012 21:28:18 +0000 (13:28 -0800)]
Merge branch 'wtk/submodule-doc-fixup' into maint

* wtk/submodule-doc-fixup:
  git-submodule: wrap branch option with "<>" in usage strings.

12 years agodiff --stat: do not count "unmerged" entries
Junio C Hamano [Tue, 27 Nov 2012 20:05:10 +0000 (12:05 -0800)]
diff --stat: do not count "unmerged" entries

Even though we show a separate *UNMERGED* entry in the patch and
diffstat output (or in the --raw format, for that matter) in
addition to and separately from the diff against the specified stage
(defaulting to #2) for unmerged paths, they should not be counted in
the total number of files affected---that would lead to counting the
same path twice.

The separation done by the previous step makes this fix simple and
straightforward.  Among the filepairs in diff_queue, paths that
weren't modified, and the extra "unmerged" entries do not count as
total number of files.

Signed-off-by: Junio C Hamano <gitster@pobox.com>
12 years agodiff --stat: move the "total count" logic to the last loop
Junio C Hamano [Tue, 27 Nov 2012 19:47:46 +0000 (11:47 -0800)]
diff --stat: move the "total count" logic to the last loop

The diffstat generation logic, with --stat-count limit, is
implemented as three loops.

 - The first counts the width necessary to show stats up to
   specified number of entries, and notes up to how many entries in
   the data we need to iterate to show the graph;

 - The second iterates that many times to draw the graph, adjusts
   the number of "total modified files", and counts the total
   added/deleted lines for the part that was shown in the graph;

 - The third iterates over the remainder and only does the part to
   count "total added/deleted lines" and to adjust "total modified
   files" without drawing anything.

Move the logic to count added/deleted lines and modified files from
the second loop to the third loop.

This incidentally fixes a bug.  The third loop was not filtering
binary changes (counted in bytes) from the total added/deleted as it
should.  The second loop implemented this correctly, so if a binary
change appeared earlier than the --stat-count cutoff, the code
counted number of added/deleted lines correctly, but if it appeared
beyond the cutoff, the number of lines would have mixed with the
byte count in the buggy third loop.

Signed-off-by: Junio C Hamano <gitster@pobox.com>
12 years agodiff --stat: use "file" temporary variable to refer to data->files[i]
Junio C Hamano [Tue, 27 Nov 2012 19:24:54 +0000 (11:24 -0800)]
diff --stat: use "file" temporary variable to refer to data->files[i]

The generated code shouldn't change but it is easier to read.

Signed-off-by: Junio C Hamano <gitster@pobox.com>
12 years agodiff --stat: status of unmodified pair in diff-q is not zero
Junio C Hamano [Tue, 27 Nov 2012 19:17:14 +0000 (11:17 -0800)]
diff --stat: status of unmodified pair in diff-q is not zero

It is spelled DIFF_STATUS_UNKNOWN these days, and is different from zero.

Signed-off-by: Junio C Hamano <gitster@pobox.com>
12 years agotest: add failing tests for "diff --stat" to t4049
Junio C Hamano [Tue, 27 Nov 2012 20:55:00 +0000 (12:55 -0800)]
test: add failing tests for "diff --stat" to t4049

There are a few problems in diff.c around --stat area, partially
caused by the recent 74faaa1 (Fix "git diff --stat" for interesting
- but empty - file changes, 2012-10-17), and largely caused by the
earlier change that introduced when --stat-count was added.

Add a few test pieces to t4049 to expose the issues.

Signed-off-by: Junio C Hamano <gitster@pobox.com>
12 years agot4041 (diff-submodule-option): don't hardcode SHA-1 in expected outputs
Ramkumar Ramachandra [Mon, 26 Nov 2012 13:54:28 +0000 (19:24 +0530)]
t4041 (diff-submodule-option): don't hardcode SHA-1 in expected outputs

The expected SHA-1 digests are always available in variables.  Use
them instead of hardcoding.

Signed-off-by: Ramkumar Ramachandra <artagnon@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
12 years agoSupport for git aliasing for tcsh completion
Marc Khouzam [Tue, 27 Nov 2012 04:13:41 +0000 (23:13 -0500)]
Support for git aliasing for tcsh completion

tcsh users sometimes alias the 'git' command to another name.  In
this case, the user expects to only have to issue a new 'complete'
command using the alias name.

However, the tcsh script currently uses the command typed by the
user to call the appropriate function in git-completion.bash, either
_git() or _gitk().  When using an alias, this technique no longer
works.

This change specifies the real name of the command (either 'git' or
'gitk') as a parameter to the script handling tcsh completion.  This
allows the user to use any alias for the 'git' or 'gitk' commands,
while still getting completion to work.

A check for the presence of ${HOME}/.git-completion.bash is also
added to help the user make use of the script properly.

Signed-off-by: Marc Khouzam <marc.khouzam@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
12 years agoDocumentation: improve phrasing in git-push.txt
Mark Szepieniec [Tue, 27 Nov 2012 01:37:34 +0000 (01:37 +0000)]
Documentation: improve phrasing in git-push.txt

The current version contains the sentence:

Further suppose that the other person already pushed changes leading to
A back to the original repository you two obtained the original commit
X.

which doesn't parse for me; I've changed it to

Further suppose that the other person already pushed changes leading to
A back to the original repository from which you two obtained the
original commit X.

Signed-off-by: Mark Szepieniec <mszepien@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
12 years agogit-send-email: allow edit invalid email address
Krzysztof Mazur [Thu, 22 Nov 2012 18:12:12 +0000 (19:12 +0100)]
git-send-email: allow edit invalid email address

In some cases the user may want to send email with "Cc:" line with
email address we cannot extract. Now we allow user to extract
such email address for us.

Signed-off-by: Krzysztof Mazur <krzysiek@podlesie.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
12 years agogit-send-email: ask what to do with an invalid email address
Krzysztof Mazur [Thu, 22 Nov 2012 18:12:11 +0000 (19:12 +0100)]
git-send-email: ask what to do with an invalid email address

We used to warn about invalid emails and just drop them. Such warnings
can be unnoticed by user or noticed after sending email when we are not
giving the "final sanity check [Y/n]?"

Now we quit by default.

Signed-off-by: Krzysztof Mazur <krzysiek@podlesie.net>
Suggested-by: Junio C Hamano <gitster@pobox.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
12 years agogit-send-email: remove invalid addresses earlier
Krzysztof Mazur [Thu, 22 Nov 2012 18:12:10 +0000 (19:12 +0100)]
git-send-email: remove invalid addresses earlier

Some addresses are passed twice to unique_email_list() and invalid addresses
may be reported twice per send_message. Now we warn about them earlier
and we also remove invalid addresses.

This also removes using of undefined values for string comparison
for invalid addresses in cc list processing.

Signed-off-by: Krzysztof Mazur <krzysiek@podlesie.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
12 years agoMerge branch 'maint'
Junio C Hamano [Mon, 26 Nov 2012 22:12:07 +0000 (14:12 -0800)]
Merge branch 'maint'

* maint:
  Fix typo in remote set-head usage
  Makefile: hide stderr of curl-config test

12 years agoDocumentation: how to add a new command
Eric S. Raymond [Mon, 26 Nov 2012 05:35:57 +0000 (00:35 -0500)]
Documentation: how to add a new command

This document contains no new policies or proposals; it attempts to
document established practices and interface requirements.

Signed-off-by: Eric S. Raymond <esr@thyrsus.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
12 years agoFix typo in remote set-head usage
Antoine Pelisse [Mon, 26 Nov 2012 19:21:54 +0000 (20:21 +0100)]
Fix typo in remote set-head usage

parenthesis are not matching in `builtin_remote_sethead_usage`
as a square bracket is closing something never opened.

Signed-off-by: Antoine Pelisse <apelisse@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
12 years agosend-email: avoid questions when user has an ident
Felipe Contreras [Sat, 24 Nov 2012 11:16:19 +0000 (12:16 +0100)]
send-email: avoid questions when user has an ident

Currently we keep getting questions even when the user has properly
configured his full name and password:

  Who should the emails appear to be from?
  [Felipe Contreras <felipe.contreras@gmail.com>]

And once a question pops up, other questions are turned on. This is
annoying.

The reason it's safe to avoid this question is because currently the
script fails completely when the author (or committer) is not correct,
so we won't even be reaching this point in the code.

The scenarios, and the current situation:

1) No information at all, no fully qualified domain name

  fatal: empty ident name (for <felipec@nysa.(none)>) not allowed

2) Only full name

  fatal: unable to auto-detect email address (got 'felipec@nysa.(none)')

3) Full name + fqdm

  Who should the emails appear to be from?
  [Felipe Contreras <felipec@nysa.felipec.org>]

4) Full name + EMAIL

  Who should the emails appear to be from?
  [Felipe Contreras <felipe.contreras@gmail.com>]

5) User configured
6) GIT_COMMITTER
7) GIT_AUTHOR

All these are the same as 4)

After this patch:

1) 2) won't change: git send-email would still die

4) 5) 6) 7) will change: git send-email won't ask the user

This is good, that's what we would expect, because the identity is
explicit.

3) will change: git send-email won't ask the user

This is bad, because we will try with an address such as
'felipec@nysa.felipec.org', which is most likely not what the user
wants, but the user will get warned by default (confirm=auto), and if
not, most likely the sending won't work, which the user would readily
note and fix.

The worst possible scenario is that such mail address does work, and the
user sends an email from that address unintentionally, when in fact the
user expected to correct that address in the prompt. This is a very,
very, very unlikely scenario, with many dependencies:

1) No configured user.name/user.email
2) No specified $EMAIL
3) No configured sendemail.from
4) No specified --from argument
5) A fully qualified domain name
6) A full name in the geckos field
7) A sendmail configuration that allows sending from this domain name
8) confirm=never, or
8.1) confirm configuration not hitting, or
8.2) Getting the error, not being aware of it
9) The user expecting to correct this address in the prompt

In a more likely scenario where 7) is not the case (can't send from
nysa.felipec.org), the user will simply see the mail was not sent
properly, and fix the problem.

The much more likely scenario though, is where 5) is not the case
(nysa.(none)), and git send-email will fail right away like it does now.

So the likelihood of this affecting anybody seriously is very very slim,
and the chances of this affecting somebody slightly are still very
small. The vast majority, if not all, of git users won't be affected
negatively, and a lot will benefit from this.

Tests-by: Jeff King <peff@peff.net>
Signed-off-by: Felipe Contreras <felipe.contreras@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
12 years agogit p4: remove unneeded cmd initialization
Pete Wyckoff [Fri, 23 Nov 2012 22:35:39 +0000 (17:35 -0500)]
git p4: remove unneeded cmd initialization

It confuses pylint, and is never needed.

Signed-off-by: Pete Wyckoff <pw@padd.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
12 years agogit p4: fix labelDetails typo in exception
Pete Wyckoff [Fri, 23 Nov 2012 22:35:38 +0000 (17:35 -0500)]
git p4: fix labelDetails typo in exception

Signed-off-by: Pete Wyckoff <pw@padd.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
12 years agogit p4 test: display unresolvable host error
Pete Wyckoff [Fri, 23 Nov 2012 22:35:37 +0000 (17:35 -0500)]
git p4 test: display unresolvable host error

This test passes already.  Make sure p4 diagnostic errors are displayed.

Signed-off-by: Pete Wyckoff <pw@padd.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
12 years agogit p4: catch p4 errors when streaming file contents
Pete Wyckoff [Fri, 23 Nov 2012 22:35:36 +0000 (17:35 -0500)]
git p4: catch p4 errors when streaming file contents

Error messages that arise during the "p4 print" phase of
generating commits were silently ignored.  Catch them,
abort the fast-import, and exit.

Without this fix, the sync/clone appears to work, but files that
are inaccessible by the p4d server will still be imported to git,
although without the proper contents.  Instead the errant files
will contain a p4 error message, such as "Librarian checkout
//depot/path failed".

Signed-off-by: Pete Wyckoff <pw@padd.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
12 years agogit p4: handle servers without move support
Pete Wyckoff [Fri, 23 Nov 2012 22:35:35 +0000 (17:35 -0500)]
git p4: handle servers without move support

Support for the "p4 move" command was added in 8e9497c (git p4:
add support for 'p4 move' in P4Submit, 2012-07-12), which checks
to make sure that the client and server support the command.

But older versions of p4d may not handle the "-k" argument, and
newer p4d allow disabling "p4 move" with a configuration setting.
Check for both these cases by testing a p4 move command on bogus
filenames and looking for strings in the error messages.

Reported-by: Vitor Antunes <vitor.hda@gmail.com>
Signed-off-by: Pete Wyckoff <pw@padd.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
12 years agogit p4: catch p4 describe errors
Pete Wyckoff [Fri, 23 Nov 2012 22:35:34 +0000 (17:35 -0500)]
git p4: catch p4 describe errors

Group the two calls to "p4 describe" into a new helper function,
and try to validate the p4 results.  The current behavior when p4
describe fails is to die with a python backtrace.  The new behavior
will print the full response.

This does not solve any particular problem, but adds more
checking in hopes of narrowing down odd behavior seen on
at least two occasions.

Based-on-patch-by: Matt Arsenault <arsenm2@gmail.com>
Reported-by: Arthur <a.foulon@amesys.fr>
Signed-off-by: Pete Wyckoff <pw@padd.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>