Make rebaseline-from-try-jobs actually download baselines.

Make rebaseline-from-try-jobs actually download baselines.

After this CL, `webkit-patch rebaseline-from-try-jobs` should actually
be able to be used for downloading new baselines.

I plan to use http://crrev.com/2106873003 as a sample CL to try it out.

BUG=474273
Committed: https://crrev.com/25eb9f6e1c7a1a18e5496030d4eea82f16805749
Cr-Commit-Position: refs/heads/master@{#404820}

[qyearsley, 2016-06-30 17:50:38]

qyearsley@chromium.org changed reviewers:
+ dpranke@chromium.org, wkorman@chromium.org

[wkorman, 2016-06-30 19:24:01]

lgtm

https://codereview.chromium.org/2118513002/diff/1/third_party/WebKit/Tools/Sc...
File
third_party/WebKit/Tools/Scripts/webkitpy/tool/commands/rebaseline_from_try_jobs.py
(right):

https://codereview.chromium.org/2118513002/diff/1/third_party/WebKit/Tools/Sc...
third_party/WebKit/Tools/Scripts/webkitpy/tool/commands/rebaseline_from_try_jobs.py:47:
"""Returns a listing of tests to rebaseline to pass to |self._rebaseline|."""
Could be better to avoid explicitly mentioning control flow via method names in
docstrings since they're prone to change.

https://codereview.chromium.org/2118513002/diff/1/third_party/WebKit/Tools/Sc...
third_party/WebKit/Tools/Scripts/webkitpy/tool/commands/rebaseline_from_try_jobs.py:61:
assert try_jobs, try_jobs
What is this assertion doing?

[qyearsley, 2016-06-30 20:59:48]

Now updated; also, I've now noticed another thing that needs to be changed
before this works; AbstractParallelRebaselineCommand._builders_to_fetch_from has
to be changed so that it will consider try bot builders.

https://codereview.chromium.org/2118513002/diff/1/third_party/WebKit/Tools/Sc...
File
third_party/WebKit/Tools/Scripts/webkitpy/tool/commands/rebaseline_from_try_jobs.py
(right):

https://codereview.chromium.org/2118513002/diff/1/third_party/WebKit/Tools/Sc...
third_party/WebKit/Tools/Scripts/webkitpy/tool/commands/rebaseline_from_try_jobs.py:47:
"""Returns a listing of tests to rebaseline to pass to |self._rebaseline|."""
On 2016/06/30 at 19:24:01, wkorman wrote:
> Could be better to avoid explicitly mentioning control flow via method names
in docstrings since they're prone to change.

Makes sense. I think I've seen out-of-date and misleading docstrings that say
something is used in some way but is not actually any more :-)

https://codereview.chromium.org/2118513002/diff/1/third_party/WebKit/Tools/Sc...
third_party/WebKit/Tools/Scripts/webkitpy/tool/commands/rebaseline_from_try_jobs.py:61:
assert try_jobs, try_jobs
On 2016/06/30 at 19:24:01, wkorman wrote:
> What is this assertion doing?

This was for debugging, accidentally left in -- thanks for catching it.

[qyearsley, 2016-07-11 18:33:29]

The CQ bit was checked by qyearsley@chromium.org to run a CQ dry run

[commit-bot: I haz the power, 2016-07-11 18:33:44]

Dry run: CQ is trying da patch. Follow status at
 
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.chromium.or...

[commit-bot: I haz the power, 2016-07-11 19:35:30]

The CQ bit was unchecked by commit-bot@chromium.org

[commit-bot: I haz the power, 2016-07-11 19:35:31]

Dry run: Try jobs failed on following builders:
  mac_chromium_rel_ng on master.tryserver.chromium.mac (JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.mac/builders/mac_chromium_rel_...)

[qyearsley, 2016-07-12 17:10:16]

This CL has now been updated so that it uses a temporary workaround to allow
_rebaseline_commands to generate commands without trying to access the results
from builder_data() for the builders -- for rebaselining from try jobs we don't
to use builder_data() as-is (since it always fetches latest results, rather than
fetching results for a particular build).

Would it be OK to submit this as-is, or would it be better to refactor so that
this workaround is unnecessary first?

[wkorman, 2016-07-12 17:42:45]

lgtm

Workaround is ok with me.

[qyearsley, 2016-07-12 17:47:11]

The CQ bit was checked by qyearsley@chromium.org

[commit-bot: I haz the power, 2016-07-12 17:48:56]

CQ is trying da patch. Follow status at
 
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.chromium.or...

[commit-bot: I haz the power, 2016-07-12 19:29:05]

Committed patchset #5 (id:80001)

[commit-bot: I haz the power, 2016-07-12 19:30:19]

CQ bit was unchecked.

[commit-bot: I haz the power, 2016-07-12 19:31:08]

Description was changed from

==========
Make rebaseline-from-try-jobs actually download baselines.

After this CL, `webkit-patch rebaseline-from-try-jobs` should actually
be able to be used for downloading new baselines.

I plan to use http://crrev.com/2106873003 as a sample CL to try it out.

BUG=474273
==========

to

==========
Make rebaseline-from-try-jobs actually download baselines.

After this CL, `webkit-patch rebaseline-from-try-jobs` should actually
be able to be used for downloading new baselines.

I plan to use http://crrev.com/2106873003 as a sample CL to try it out.

BUG=474273

Committed: https://crrev.com/25eb9f6e1c7a1a18e5496030d4eea82f16805749
Cr-Commit-Position: refs/heads/master@{#404820}
==========

[commit-bot: I haz the power, 2016-07-12 19:31:10]

Patchset 5 (id:??) landed as
https://crrev.com/25eb9f6e1c7a1a18e5496030d4eea82f16805749
Cr-Commit-Position: refs/heads/master@{#404820}

[Dirk Pranke, 2016-08-09 04:01:18]

lgtm (very belatedly).