Previously, 1 CL was queried first, followed by all remaining ones
in parallel. The purpose was to ensure that valid authentication
token is present. However, it still required one call to Rietveld
or Gerrit, needlessly increasing latency.
This CL first loops over all CLs and ensures that credentials are
present, refreshing refresh tokens for Rietveld if necessary. Then,
all CLs are queried in parallel.
R=sergiyb@chromium.org,clemensh@chromium.org
BUG=681704
Change-Id: Ic125ac7c2a684d6f3c34e4e8b899192abbed01bb
Reviewed-on: https://chromium-review.googlesource.com/431033
Reviewed-by: Sergiy Byelozyorov <sergiyb@chromium.org>
Commit-Queue: Andrii Shyshkalov <tandrii@chromium.org>
We have to clean up the title, but if the user hasn't expressed any
input about the title (i.e. it's one we autogenerated), then don't
bother them by pointing it out.
BUG=684079
Change-Id: I8215e0a30f786466697fe1df178ca90e1980d9b0
Reviewed-on: https://chromium-review.googlesource.com/431162
Commit-Queue: Aaron Gable <agable@chromium.org>
Reviewed-by: Andrew Bonventre <andybons@chromium.org>
Reviewed-by: Andrii Shyshkalov <tandrii@chromium.org>
Git cl decides if git-numberer is enabled on a repository by writing
Gerrit's project.config from refs/meta/config into a tempfile, which is
then queried using `git config -f tempfile --get ...`. The file itself
is only flushed, but not closed after writing because Python's
tempfile.NamedTemporaryFile is deleted on closing. This worked fine on
Linix/Mac, but not on Windows, where `git config` apparently doesn't see
file or its contents.
This CL rewrites the above using yet another contexmanager temp
directory into which a file is written and closed before git config is
ran.
R=machenbach@chromium.org,grt@chromium.org
BUG=683202
Change-Id: I7974d66b1b2b0478ab4b6f7ac04e547a4981c46c
Reviewed-on: https://chromium-review.googlesource.com/430719
Commit-Queue: Andrii Shyshkalov <tandrii@chromium.org>
Reviewed-by: Vadim Shtayura <vadimsh@chromium.org>
This CL adds .js to the list of file extensions that get clang-formatted.
BUG=567770
Change-Id: Ieed162a84ca68877e5107da1824b1f49c71d51ed
Reviewed-on: https://chromium-review.googlesource.com/429551
Reviewed-by: Andrii Shyshkalov <tandrii@chromium.org>
Commit-Queue: Dan Beam <dbeam@chromium.org>
It looks like this regressed when we switched from specifying CCs
via the refspec arguments to using the API.
BUG=680605
Change-Id: Iabd397b639989f050932188b1a1aa488639ffbbe
Reviewed-on: https://chromium-review.googlesource.com/427344
Reviewed-by: Andrew Bonventre <andybons@chromium.org>
Reviewed-by: Andrii Shyshkalov <tandrii@chromium.org>
Commit-Queue: Aaron Gable <agable@chromium.org>
Also set the line wrap limit to 72 + indent from 70 columns total. The
latter was limiting the CL description to 68 columns total before
wrapping kicked in.
BUG=none
Change-Id: I93c984c7b121d4bb042d0dc81a662352f77df4d1
Reviewed-on: https://chromium-review.googlesource.com/420243
Reviewed-by: Dirk Pranke <dpranke@chromium.org>
Commit-Queue: Dirk Pranke <dpranke@chromium.org>
This affects a bunch of files, but only changes comments,
and shouldn't make any difference to behavior.
The purpose is to slightly improve readability of pylint
disable comments.
Change-Id: Ic6cd0f8de792b31d91c6125f6da2616450b30f11
Reviewed-on: https://chromium-review.googlesource.com/420412
Reviewed-by: Aaron Gable <agable@chromium.org>
Commit-Queue: Quinten Yearsley <qyearsley@chromium.org>
Specifically, this CL was made by running codespell
(https://github.com/lucasdemarchi/codespell), manually filtering
for changes in non-third-party files that appear correct.
Change-Id: Ia16c1b29483d777744450d7bea45a178cf877a25
Reviewed-on: https://chromium-review.googlesource.com/420871
Commit-Queue: Quinten Yearsley <qyearsley@chromium.org>
Reviewed-by: Aaron Gable <agable@chromium.org>
This is a reland of 947f2ee808,
which was reverted in a5a1eea537
BUG=672332
Change-Id: If33c54e500fbeac11f60d81a19549880506c63d8
Reviewed-on: https://chromium-review.googlesource.com/419737
Reviewed-by: Andrii Shyshkalov <tandrii@chromium.org>
Commit-Queue: Aaron Gable <agable@chromium.org>
This is a fix to 32978d969c,
which introduced a "Landed as <gerrit link>" feature to git-cl, but
which at the last minute introduced a typo causing the whole feature
to not actually work.
BUG=661187
Change-Id: Ifef3379a51f035973bc5f3842862528f90bfdf84
Reviewed-on: https://chromium-review.googlesource.com/419782
Reviewed-by: Stephen Martinis <martiniss@chromium.org>
Commit-Queue: Aaron Gable <agable@chromium.org>
In other words, end 2 end test was a really awesome idea.
R=machenbach@chromium.org,sergiyb@chromium.org
BUG=642493
TEST=git cl land of https://codereview.chromium.org/2575043003 succeeded
Change-Id: I568ce79baf109b2aa556e4343527b63f39c10d00
Reviewed-on: https://chromium-review.googlesource.com/419478
Commit-Queue: Andrii Shyshkalov <tandrii@chromium.org>
Reviewed-by: Aaron Gable <agable@chromium.org>
The new class is unused and doesn't change any existing functionality.
BUG=chromium:642493
R=machenbach@chromium.org,iannucci@chromium.org
Change-Id: Id3fe71b07b694339f0a620b427816e52560069d8
Reviewed-on: https://chromium-review.googlesource.com/416430
Reviewed-by: Michael Achenbach <machenbach@chromium.org>
Commit-Queue: Andrii Shyshkalov <tandrii@chromium.org>
This is preperatory work for Cr- footers generation in git cl instead
of gnumbd.
BUG=642493
Change-Id: I4cfdd882fe6caa7972e51ffa81d335104ddb56dd
Reviewed-on: https://chromium-review.googlesource.com/414464
Commit-Queue: Andrii Shyshkalov <tandrii@chromium.org>
Reviewed-by: Michael Achenbach <machenbach@chromium.org>
This causes other automated tooling to think that the entire
upload has failed, when really only a small part of it has.
BUG=666160
Change-Id: I0d646e77c54330d642097aebe70ec145ae291267
Reviewed-on: https://chromium-review.googlesource.com/412281
Commit-Queue: Andrii Shyshkalov <tandrii@chromium.org>
Reviewed-by: Andrii Shyshkalov <tandrii@chromium.org>
Before, presubmit_support would fail with not very useful stacktrace if
Gerrit returns 404, which is usually due to missing/invalid credentials.
This CL fixes that and improves the exception message, and also improves
logic in git_cl.
R=agable@chromium.org
BUG=
Change-Id: Iae8f0c24422c46af70929c7d5d71993164887511
Reviewed-on: https://chromium-review.googlesource.com/409650
Commit-Queue: Aaron Gable <agable@chromium.org>
Reviewed-by: Aaron Gable <agable@chromium.org>
This is still important, but at least now it won't result
in a stacktrace.
Change-Id: If6e7e15a98b0a03df7e978201762fea670312ede
Reviewed-on: https://chromium-review.googlesource.com/408660
Reviewed-by: Andrii Shyshkalov <tandrii@chromium.org>
Commit-Queue: Aaron Gable <agable@chromium.org>
Note: This CL originally just removed a deprecated use of Remove use of DoGetTrySlaves, suggested in http://crrev.com/2442153002, then was expanded to remove DoGetTrySlaves, GetPreferredTrySlaves and GetTrySlavesExecuter since these are all deprecated and unused.
BUG=660453
Review-Url: https://codereview.chromium.org/2453823002
Assuming that assuming bots are always just builder names, and they're never specified along with tests in git cl --bot options, then some logic used for git cl try can be removed.
Review-Url: https://codereview.chromium.org/2448193006
The purpose of this change is to prepare for modifying git cl try
so that builders on multiple masters can be triggered in one invocation
(http://crbug.com/640740).
This should not affect behavior; this CL makes several non-essential
changes to formatting and comments.
Review-Url: https://codereview.chromium.org/2442153002
Andrii Shyshkalov
Aaron Gable
Christopher Lam
Daniel Cheng
Asanka Herath
Quinten Yearsley
Sergiy Byelozyorov
Ravi Mistry
scottmg
nodir
qyearsley