1086678 - Guard against subsuite being None

Status: RESOLVED FIXED

(Testing :: Mozbase, defect)

Description

[Jonathan Griffin (:jgriffin)]

Mozmill is hitting an error when invoked with -t, because the subsuite ends up being None:

This fails:
> mozmill -t %test% -b %binary%
with:
> Traceback (most recent call last):
>   File "/Users/andrei.eftimie/work/mozilla/env/mozmill/bin/mozmill", line 9, in <module>
>     load_entry_point('mozmill==2.1-dev', 'console_scripts', 'mozmill')()
>   File "/Users/andrei.eftimie/work/mozilla/mozmill/src/mozmill/mozmill/__init__.py", line 938, in cli
>     CLI(args).run()
>   File "/Users/andrei.eftimie/work/mozilla/mozmill/src/mozmill/mozmill/__init__.py", line 918, in run
>     tests = self.manifest.active_tests(**mozinfo.info)
>   File "/Users/andrei.eftimie/work/mozilla/env/mozmill/lib/python2.7/site-packages/manifestparser-0.7-py2.7.egg/manifestparser/manifestparser.py", line 1160, in active_tests
>     if ',' in subsuite:
> TypeError: argument of type 'NoneType' is not iterable

Comment 1

[Jonathan Griffin (:jgriffin)]

Attachment: Provide a default value for subsuite,

Comment 2

[Joel Maher ( :jmaher ) (UTC -8)]

Comment on attachment 8508798 [details] [diff] [review]
Provide a default value for subsuite,

Review of attachment 8508798 [details] [diff] [review]:
-----------------------------------------------------------------

there is a block after this that looks for subsuite:
http://dxr.mozilla.org/mozilla-central/source/testing/mozbase/manifestparser/manifestparser/manifestparser.py#1175

we should guard against it there as well.

Comment 3

[Jonathan Griffin (:jgriffin)]

Attachment: Provide a default value for subsuite,

Thanks; updated patch.

Comment 4

[Joel Maher ( :jmaher ) (UTC -8)]

Comment on attachment 8508895 [details] [diff] [review]
Provide a default value for subsuite,

Review of attachment 8508895 [details] [diff] [review]:
-----------------------------------------------------------------

thanks!

Comment 5

[Jonathan Griffin (:jgriffin)]

https://hg.mozilla.org/integration/mozilla-inbound/rev/325d41a714e3

Comment 6

[Wes Kocher (:KWierso) (Not reading bugmail; email directly if needed)]

Backed out for lots of bustage in https://hg.mozilla.org/integration/mozilla-inbound/rev/8d62499aacc2

https://treeherder.mozilla.org/ui/logviewer.html#?job_id=3179486&repo=mozilla-inbound
https://treeherder.mozilla.org/ui/logviewer.html#?job_id=3181336&repo=mozilla-inbound

Comment 7

[Jonathan Griffin (:jgriffin)]

(In reply to Wes Kocher (:KWierso) from comment #6)
> Backed out for lots of bustage in
> https://hg.mozilla.org/integration/mozilla-inbound/rev/8d62499aacc2
> 
> https://treeherder.mozilla.org/ui/logviewer.
> html#?job_id=3179486&repo=mozilla-inbound
> https://treeherder.mozilla.org/ui/logviewer.
> html#?job_id=3181336&repo=mozilla-inbound

Oops, that had a syntax error.  Sorry about that!  I've fixed and pushed to try for good measure:

https://treeherder.mozilla.org/ui/#/jobs?repo=try&revision=bc27fd947130

Comment 8

[Jonathan Griffin (:jgriffin)]

try looks good: https://hg.mozilla.org/integration/mozilla-inbound/rev/8a3ff43fc878

Comment 9

[Henrik Skupin [:whimboo][⌚️UTC+2]]

Shouldn't we have added a unit test for this regression, which seems to be major enough?

Comment 10

[Jonathan Griffin (:jgriffin)]

Yes, we should add a test for this.  I didn't yesterday since it was unclear how you were hitting this, but you've explained it in bug 1056045.  I'll file a separate bug for adding a test for this, since I won't have time to do it myself today.

Comment 11

[Ryan VanderMeulen [:RyanVM]]

https://hg.mozilla.org/mozilla-central/rev/8a3ff43fc878