Git footers doesn't seem to support v8, let alone know that trunk actually maps
to the candidate branch.  This should fix it.

BUG=
TBR=agable

Review URL: https://codereview.chromium.org/645183002

git-svn-id: svn://svn.chromium.org/chrome/trunk/tools/depot_tools@292445 0039d316-1c4b-4281-b951-d872f2087c98

Don't discard stderr from failed rebase operations
I had an issue where stdout of the failed rebase was
empty but stderr contained:
First, rewinding head to replay your work on top of it...
Dirty index: cannot apply patches (dirty: internal_gyp
third_party/html_office).

Also, in my case the second rebase was actually succeeding
for some reason, which is clearly no expected, so assert
in this case.

BUG=410339

Review URL: https://codereview.chromium.org/645763002

git-svn-id: svn://svn.chromium.org/chrome/trunk/tools/depot_tools@292444 0039d316-1c4b-4281-b951-d872f2087c98

This is just making the same change to the copy of upload.py that is in depot_tools.

BUG=419156
R=iannucci@chromium.org

Original Review URL: https://codereview.chromium.org/624513002

Review URL: https://codereview.chromium.org/639503002

git-svn-id: svn://svn.chromium.org/chrome/trunk/tools/depot_tools@292432 0039d316-1c4b-4281-b951-d872f2087c98

If presubmit.py for a directory (such as net/) fails due to not being clang formatted, then 'git cl format' needs to be run for that directory, not the entire patch.

BUG=422332

Review URL: https://codereview.chromium.org/648713003

git-svn-id: svn://svn.chromium.org/chrome/trunk/tools/depot_tools@292407 0039d316-1c4b-4281-b951-d872f2087c98

R=iannucci@chromium.org

Review URL: https://codereview.chromium.org/621783002

git-svn-id: svn://svn.chromium.org/chrome/trunk/tools/depot_tools@292379 0039d316-1c4b-4281-b951-d872f2087c98

This check never triggered before, since cpplint.py was looking for
override not OVERRIDE. Since Chromium prefers override now, the
linter is now (correctly) warning. However, the old Chromium convention
(and the one implemented by the clang plugin) is to explicitly annotate
all overrides with both virtual and override. The clang plugin code has
been updated, but new binaries have not yet been built. Until clang
rolls, suppress the warning.

BUG=417463
TBR=stip@chromium.org

Review URL: https://codereview.chromium.org/627163002

git-svn-id: svn://svn.chromium.org/chrome/trunk/tools/depot_tools@292319 0039d316-1c4b-4281-b951-d872f2087c98

BUG=
R=maruel@chromium.org

Review URL: https://codereview.chromium.org/614413003

git-svn-id: svn://svn.chromium.org/chrome/trunk/tools/depot_tools@292291 0039d316-1c4b-4281-b951-d872f2087c98

This tool can automatically set up the necessary git-svn metadata for our
repos that live in SVN but are mirrored to Chrome.

R=iannucci@chromium.org, tandrii@chromium.org
BUG=418973

Review URL: https://codereview.chromium.org/611253003

git-svn-id: svn://svn.chromium.org/chrome/trunk/tools/depot_tools@292241 0039d316-1c4b-4281-b951-d872f2087c98

This concerns the "committed" message posted to Rietveld. Prior to
34504a15, this used the committer's own core.abbrev. Now it's using
a hard-coded 7. Per recent discussions on chromium-dev, 7 is
insufficient to uniquely identify commits in large repositories such as
ours. If any abbreviation is to be done, a 12-digit minimum should be
used, but there's no reason not to use the entire hash.

Review URL: https://codereview.chromium.org/592293002

git-svn-id: svn://svn.chromium.org/chrome/trunk/tools/depot_tools@292164 0039d316-1c4b-4281-b951-d872f2087c98

TEST="roll_dep.py breakpad/src 10000" should print ~1387, not 4.

Review URL: https://codereview.chromium.org/604463002

git-svn-id: svn://svn.chromium.org/chrome/trunk/tools/depot_tools@292129 0039d316-1c4b-4281-b951-d872f2087c98

R=szager@chromium.org
BUG=413391

Review URL: https://codereview.chromium.org/598153003

git-svn-id: svn://svn.chromium.org/chrome/trunk/tools/depot_tools@292123 0039d316-1c4b-4281-b951-d872f2087c98

BUG=413391
R=iannucci@chromium.org,agable@chromium.org

Review URL: https://codereview.chromium.org/567873002

git-svn-id: svn://svn.chromium.org/chrome/trunk/tools/depot_tools@292117 0039d316-1c4b-4281-b951-d872f2087c98

BUG=371012

Review URL: https://codereview.chromium.org/562953002

git-svn-id: svn://svn.chromium.org/chrome/trunk/tools/depot_tools@292088 0039d316-1c4b-4281-b951-d872f2087c98

After switching to a pure git workflow, most of the patchset we upload
to rietveld have git metadata. Apparently, however, the scripts here in
depot tools, which are used by the CQ and trybots, are ignoring that
and applying patches without taking advantage of git metadata.
In practice this causes people to rebase and reupload patches more than
necessary, even in the cases when it could be handled automatically by
means of a three way merge.
This change updates the GitCheckout class of depot_tools to use that.

BUG=416003

Review URL: https://codereview.chromium.org/577203002

git-svn-id: svn://svn.chromium.org/chrome/trunk/tools/depot_tools@292087 0039d316-1c4b-4281-b951-d872f2087c98

This CL fixes some issues with map-branches:
* Branches with no upstream were not being shown.
* -vv from a detached HEAD would crash
* GONE upstreams would crash when git cleaned up in a way that caused
  hash_one to fail

This CL also adds a blue coloring to branches that start with
'branch-heads' for Chromium release branches.

BUG=416530

Review URL: https://codereview.chromium.org/576423002

git-svn-id: svn://svn.chromium.org/chrome/trunk/tools/depot_tools@292083 0039d316-1c4b-4281-b951-d872f2087c98

R=dpranke@chromium.org
BUG=none
TEST=no whining about <include src="long/path/name.js">

Review URL: https://codereview.chromium.org/579383004

git-svn-id: svn://svn.chromium.org/chrome/trunk/tools/depot_tools@292080 0039d316-1c4b-4281-b951-d872f2087c98

Then this was changed to CheckCallAndFilter it broke
the expectations of code in the buildbot tree:
 scripts/slave/chromium_commands.py:extract_revisions

Alternatively we could update the expectations
of chromium_commands.py.

BUG= https://code.google.com/p/chromium/issues/detail?id=412818

Review URL: https://codereview.chromium.org/587253005

git-svn-id: svn://svn.chromium.org/chrome/trunk/tools/depot_tools@292072 0039d316-1c4b-4281-b951-d872f2087c98

According to Python's doc (
https://docs.python.org/2/library/subprocess.html#subprocess.check_output
):
if check_output raises exception CalledProcessError, the exception object
should contain stdout data as `output` attribute. Before this commit,
subprocess2.CalledProcessError had `output` always None, and used `stdout`
instead. This commit fixes this problem storing the same data in both `stdout`
and `output`.

BUG=NONE

Review URL: https://codereview.chromium.org/582933002

git-svn-id: svn://svn.chromium.org/chrome/trunk/tools/depot_tools@292036 0039d316-1c4b-4281-b951-d872f2087c98

This is analogous to the change made in
https://codereview.chromium.org/559243002/.  I want to change ARB to use the
commit message generated by roll-dep.

BUG=408700

Review URL: https://codereview.chromium.org/564363003

git-svn-id: svn://svn.chromium.org/chrome/trunk/tools/depot_tools@291975 0039d316-1c4b-4281-b951-d872f2087c98

This reverts commit 931b6c6a.

This was causing our try bots to fail on DEPS changes.

After applying the patch successfully, apply_issue.py tries to run gclient sync to get the updated DEPS, resulting in this error:
src (ERROR)
----------------------------------------
[0:00:00] Started.
----------------------------------------
Error: 
____ src at origin/master
	Your index contains uncommitted changes
	Please commit, stash, or reset.

BUG=413326
TESTED=Manually: https://paste.googleplex.com/6375653263802368

Review URL: https://codereview.chromium.org/563873002

git-svn-id: svn://svn.chromium.org/chrome/trunk/tools/depot_tools@291932 0039d316-1c4b-4281-b951-d872f2087c98

Pulls the logic for computing an issue's "status" - has LGTM, is in
CQ, etc. - out of the "git cl status" command itself, so that other
tools can get the status of a Changelist.

BUG=379849
R=iannucci@chromium.org

Review URL: https://codereview.chromium.org/555973005

git-svn-id: svn://svn.chromium.org/chrome/trunk/tools/depot_tools@291928 0039d316-1c4b-4281-b951-d872f2087c98

This was done because a line ending with a '\' will not be printed properly.
Completely escape the string.

R=stip@chromium.org
BUG=

Review URL: https://codereview.chromium.org/557143002

git-svn-id: svn://svn.chromium.org/chrome/trunk/tools/depot_tools@291906 0039d316-1c4b-4281-b951-d872f2087c98

BUG=https://code.google.com/p/chromium/issues/detail?id=409185
TEST=local
R=iannucci@chromium.org

Review URL: https://codereview.chromium.org/555383002

git-svn-id: svn://svn.chromium.org/chrome/trunk/tools/depot_tools@291900 0039d316-1c4b-4281-b951-d872f2087c98

R=luqui@chromium.org, prasadv@chromium.org
BUG=

Review URL: https://codereview.chromium.org/559773002

git-svn-id: svn://svn.chromium.org/chrome/trunk/tools/depot_tools@291897 0039d316-1c4b-4281-b951-d872f2087c98

This is to support non-chrome repos in the bisect scripts.

R=hinoka@chromium.org, luqui@chromium.org
BUG=

Review URL: https://codereview.chromium.org/552403003

git-svn-id: svn://svn.chromium.org/chrome/trunk/tools/depot_tools@291895 0039d316-1c4b-4281-b951-d872f2087c98

R=luqui@chromium.org

Review URL: https://codereview.chromium.org/558783003

git-svn-id: svn://svn.chromium.org/chrome/trunk/tools/depot_tools@291893 0039d316-1c4b-4281-b951-d872f2087c98

TBR=iannucci@chromium.org

Review URL: https://codereview.chromium.org/561433002

git-svn-id: svn://svn.chromium.org/chrome/trunk/tools/depot_tools@291889 0039d316-1c4b-4281-b951-d872f2087c98

R=iannucci@chromium.org, maruel@chromium.org
BUG=410959

Review URL: https://codereview.chromium.org/549733002

git-svn-id: svn://svn.chromium.org/chrome/trunk/tools/depot_tools@291883 0039d316-1c4b-4281-b951-d872f2087c98

Without --exit-code, diff-index always returns 0 and thus never throws
the exception for a dirty index.

R=iannucci@chromium.org

Review URL: https://codereview.chromium.org/548553003

git-svn-id: svn://svn.chromium.org/chrome/trunk/tools/depot_tools@291881 0039d316-1c4b-4281-b951-d872f2087c98

Because everyone needs humour.

Fix a small bug in subcommands.py with undocumented commands.

R=stip@chromium.org
BUG=

Review URL: https://codereview.chromium.org/554123003

git-svn-id: svn://svn.chromium.org/chrome/trunk/tools/depot_tools@291867 0039d316-1c4b-4281-b951-d872f2087c98

BUG=411236

Review URL: https://codereview.chromium.org/545213002

git-svn-id: svn://svn.chromium.org/chrome/trunk/tools/depot_tools@291865 0039d316-1c4b-4281-b951-d872f2087c98

This incorporates:

https://chromium-review.googlesource.com/#/c/216850

R=davidjames@chromium.org,petermayo@chromium.org
BUG=

Review URL: https://codereview.chromium.org/554723003

git-svn-id: svn://svn.chromium.org/chrome/trunk/tools/depot_tools@291857 0039d316-1c4b-4281-b951-d872f2087c98

improvements.

  * No longer crash on HEAD
  * Always correctly highlight lines based on hash-matching-ness
  * Only import git_cl if you want it (it can be slow)
  * Print a message when NO branches are found (instead of blank output)

R=calamity@chromium.org, jsbell@chromium.org
BUG=410353

Review URL: https://codereview.chromium.org/554523003

git-svn-id: svn://svn.chromium.org/chrome/trunk/tools/depot_tools@291847 0039d316-1c4b-4281-b951-d872f2087c98

This CL prevents map-branches from loading the tracking status when
the -v flag is not supplied. This prevents the vanilla map-branches
from taking a potentially long time to retrieve the tracking status.

BUG=410137

Review URL: https://codereview.chromium.org/536793002

git-svn-id: svn://svn.chromium.org/chrome/trunk/tools/depot_tools@291846 0039d316-1c4b-4281-b951-d872f2087c98

TBR=iannucci@chromium.org

Review URL: https://codereview.chromium.org/549643002

git-svn-id: svn://svn.chromium.org/chrome/trunk/tools/depot_tools@291840 0039d316-1c4b-4281-b951-d872f2087c98

BUG=
TBR=iannucci

Review URL: https://codereview.chromium.org/549613002

git-svn-id: svn://svn.chromium.org/chrome/trunk/tools/depot_tools@291835 0039d316-1c4b-4281-b951-d872f2087c98

For me, PATH was a unicode instead of a str for some reason, so things
failed with

D:\src\chromefetch\src>git cl format
Traceback (most recent call last):
  File "d:\src\depot_tools\git_cl.py", line 2821, in <module>
    sys.exit(main(sys.argv[1:]))
  File "d:\src\depot_tools\git_cl.py", line 2807, in main
    return dispatcher.execute(OptionParser(), argv)
  File "d:\src\depot_tools\subcommand.py", line 245, in execute
    return command(parser, args[1:])
  File "d:\src\depot_tools\git_cl.py", line 2768, in CMDformat
    stdout = RunCommand(cmd, stdin=diff_output, cwd=top_dir, env=env)
  File "d:\src\depot_tools\git_cl.py", line 81, in RunCommand
    return subprocess2.check_output(args, shell=False, **kwargs)
  File "d:\src\depot_tools\subprocess2.py", line 515, in check_output
    return check_call_out(args, stdout=PIPE, **kwargs)[0]
  File "d:\src\depot_tools\subprocess2.py", line 475, in check_call_out
    out, returncode = communicate(args, **kwargs)
  File "d:\src\depot_tools\subprocess2.py", line 449, in communicate
    proc = Popen(args, **kwargs)
  File "d:\src\depot_tools\subprocess2.py", line 237, in __init__
    super(Popen, self).__init__(args, **kwargs)
  File "d:\src\depot_tools\python276_bin\lib\subprocess.py", line 709, in __init__
    errread, errwrite)
  File "d:\src\depot_tools\python276_bin\lib\subprocess.py", line 957, in _execute_child
    startupinfo)
TypeError: environment can only contain strings
Sending crash report ...
  args: ['d:\\src\\depot_tools\\git_cl.py', 'format']
  cwd: D:\src\chromefetch\src
  exception: environment can only contain strings
  host: THAKIS1-W.ad.corp.google.com
  stack:   File "d:\src\depot_tools\git_cl.py", line 2821,
  user: thakis
  version: 2.7.6 (default, Nov 10 2013, 19:24:18)

This fixes the glitch.

Review URL: https://codereview.chromium.org/545333002

git-svn-id: svn://svn.chromium.org/chrome/trunk/tools/depot_tools@291825 0039d316-1c4b-4281-b951-d872f2087c98

R=agable@chromium.org, mmoss@chromium.org
BUG=

Review URL: https://codereview.chromium.org/549513002

git-svn-id: svn://svn.chromium.org/chrome/trunk/tools/depot_tools@291823 0039d316-1c4b-4281-b951-d872f2087c98

In https://codereview.chromium.org/341533006/ a change
was made so that gn.py is not looking for the .gn file
to identify the root of the checkout. This breaks
GN functionality for projects that uses gclient but
have a top directory named something else than 'src'.

This change adds support for arbitrarily named primary (the first)
solutions in the .gclient file.
It also adds a check for the generated GN path so a friendly
error message can be printed if the GN executable cannot be found.


BUG=389883
TESTED=Various cases of Chromium, WebRTC and custom checkouts
with .gclient containing empty solution list, solution missing the
'name' key and so on.

Review URL: https://codereview.chromium.org/538393002

git-svn-id: svn://svn.chromium.org/chrome/trunk/tools/depot_tools@291819 0039d316-1c4b-4281-b951-d872f2087c98

So the original intention of moving it to a different directory to be deleted later
was to (1) save it for diagnosis (2) be a single inode swap rather than a long
rmtree delete.

1. No one is actually looking at these directories
2. The tmpdir sometimes end up on a different partition, so it ends up being
   a copy + delete instead.  Since we're not getting timeouts from that, its
   probably actually better to just straight up delete it.

BUG=410727

Review URL: https://codereview.chromium.org/538993002

git-svn-id: svn://svn.chromium.org/chrome/trunk/tools/depot_tools@291818 0039d316-1c4b-4281-b951-d872f2087c98