[Findit] Updating extractors.py to account for new compile failure stdio format

[Findit] Updating extractors.py to account for new compile failure stdio format

BUG=607635
Committed: https://chromium.googlesource.com/infra/infra/+/f8549ed97a8bde4ecc58b8707ae5f3f855a80cf7

[lijeffrey, 2016-04-29 02:45:42]

Description was changed from

==========
[Findit] Updating extractors.py to account for new compile failure stdio format

BUG=
==========

to

==========
[Findit] Updating extractors.py to account for new compile failure stdio format

BUG=607635
==========

[lijeffrey, 2016-04-29 02:45:43]

lijeffrey@chromium.org changed reviewers:
+ chanli@chromium.org, stgao@chromium.org

[lijeffrey, 2016-04-29 02:46:24]

ptal

[stgao, 2016-04-29 16:38:24]

lgtm with nits.

https://codereview.chromium.org/1927423002/diff/1/appengine/findit/waterfall/...
File appengine/findit/waterfall/extractors.py (right):

https://codereview.chromium.org/1927423002/diff/1/appengine/findit/waterfall/...
appengine/findit/waterfall/extractors.py:246: continue  # pragma: no cover
Hm, I think this should be executed in a common testcase. Is there anything
wrong?

[chanli, 2016-04-29 18:26:34]

https://codereview.chromium.org/1927423002/diff/1/appengine/findit/waterfall/...
File appengine/findit/waterfall/extractors.py (right):

https://codereview.chromium.org/1927423002/diff/1/appengine/findit/waterfall/...
appengine/findit/waterfall/extractors.py:129: ERROR_LINE_END_PATTERN2 =
re.compile(r'failed with exit code \d+')
Question: in the stdio file of the
example(https://build.chromium.org/p/chromium.webkit/builders/WebKit%20Linux/builds/63875)
that was mentioned in bug: I couldn't find 'failed with exit code' but 'step
returned non-zero exit code: 1'. Is this what expected or I just looked into
wrong place?

[lijeffrey, 2016-04-29 19:23:35]

https://codereview.chromium.org/1927423002/diff/1/appengine/findit/waterfall/...
File appengine/findit/waterfall/extractors.py (right):

https://codereview.chromium.org/1927423002/diff/1/appengine/findit/waterfall/...
appengine/findit/waterfall/extractors.py:129: ERROR_LINE_END_PATTERN2 =
re.compile(r'failed with exit code \d+')
On 2016/04/29 18:26:34, chanli wrote:
> Question: in the stdio file of the
>
example(https://build.chromium.org/p/chromium.webkit/builders/WebKit%20Linux/builds/63875)
> that was mentioned in bug: I couldn't find 'failed with exit code' but 'step
> returned non-zero exit code: 1'. Is this what expected or I just looked into
> wrong place?

The example for "failed with exit code" was from
https://findit-for-me.appspot.com/waterfall/failure-log?url=https://build.chr...

"returned non-zero exit code: 1" is outside wher we want to be looking. We
should only be looking between

FAILED: (target that failed) <-- this is part of the new format
command that failed <-- extract targets from the first line
... <-- extract files
"n error(s) generated" or "linker command failed with exit code 1" <-- stop here

If neither of those lines is encountered while scanning a FAILED block but we
encounter something like [123/456], then we also stop.

https://codereview.chromium.org/1927423002/diff/1/appengine/findit/waterfall/...
appengine/findit/waterfall/extractors.py:246: continue  # pragma: no cover
On 2016/04/29 16:38:24, stgao wrote:
> Hm, I think this should be executed in a common testcase. Is there anything
> wrong?

It seems there's a bug in coverage

a = False 
for i in range(1, 4): 
  if i == 2: 
    if not a:
      a = True 
    # b = 3 
    continue 
  a = False

as it is, with code coverage the "continue" line will be marked as not executed
but with "# b = 3" uncommented, the continue line is executed.

To work around this I'll put is_build_command_line = True after the if statement
instead of before

[lijeffrey, 2016-04-29 20:08:39]

The CQ bit was checked by lijeffrey@chromium.org

[lijeffrey, 2016-04-29 20:08:39]

The patchset sent to the CQ was uploaded after l-g-t-m from stgao@chromium.org
Link to the patchset: https://codereview.chromium.org/1927423002/#ps40001
(title: "Fixing unintentional whitespace change")

[commit-bot: I haz the power, 2016-04-29 20:08:44]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1927423002/40001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1927423002/40001

[commit-bot: I haz the power, 2016-04-29 20:13:02]

Description was changed from

==========
[Findit] Updating extractors.py to account for new compile failure stdio format

BUG=607635
==========

to

==========
[Findit] Updating extractors.py to account for new compile failure stdio format

BUG=607635

Committed:
https://chromium.googlesource.com/infra/infra/+/f8549ed97a8bde4ecc58b8707ae5f...
==========

[commit-bot: I haz the power, 2016-04-29 20:13:03]

Committed patchset #3 (id:40001) as
https://chromium.googlesource.com/infra/infra/+/f8549ed97a8bde4ecc58b8707ae5f...