WebKit Bugzilla
New
Browse
Log In
×
Sign in with GitHub
or
Remember my login
Create Account
·
Forgot Password
Forgotten password account recovery
RESOLVED FIXED
98810
webkit-patch rebaseline-expectations doesn't work w/o failures specified
https://bugs.webkit.org/show_bug.cgi?id=98810
Summary
webkit-patch rebaseline-expectations doesn't work w/o failures specified
Dirk Pranke
Reported
2012-10-09 12:00:38 PDT
webkit-patch rebaseline-expectations doesn't work w/o failures specified
Attachments
Patch
(2.89 KB, patch)
2012-10-09 12:01 PDT
,
Dirk Pranke
no flags
Details
Formatted Diff
Diff
just get all of the tests marked for rebaselining
(2.80 KB, patch)
2012-10-09 14:09 PDT
,
Dirk Pranke
no flags
Details
Formatted Diff
Diff
Show Obsolete
(1)
View All
Add attachment
proposed patch, testcase, etc.
Dirk Pranke
Comment 1
2012-10-09 12:01:21 PDT
Created
attachment 167807
[details]
Patch
Ojan Vafai
Comment 2
2012-10-09 12:21:06 PDT
Comment on
attachment 167807
[details]
Patch View in context:
https://bugs.webkit.org/attachment.cgi?id=167807&action=review
> Tools/Scripts/webkitpy/layout_tests/models/test_expectations.py:872 > + return self._model.get_test_set(REBASELINE, IMAGE) | self._model.get_test_set(REBASELINE, FAIL) | self._model.get_test_set(REBASELINE, PASS)
Can you just use self._model.get_test_set("REBASELINE")? In addition to the fixme to get rid of get_test_set, I worry that we'll eventually do the cleanup to not mark skip/rebaseline tests as pass, but just have the expectation be skip/rebaseline. Then this code will break again.
Dirk Pranke
Comment 3
2012-10-09 12:26:53 PDT
(In reply to
comment #2
)
> (From update of
attachment 167807
[details]
) > View in context:
https://bugs.webkit.org/attachment.cgi?id=167807&action=review
> > > Tools/Scripts/webkitpy/layout_tests/models/test_expectations.py:872 > > + return self._model.get_test_set(REBASELINE, IMAGE) | self._model.get_test_set(REBASELINE, FAIL) | self._model.get_test_set(REBASELINE, PASS) > > Can you just use self._model.get_test_set("REBASELINE")? >
Oh, yeah, I guess you can. I was thinking (a) that you had to have expectations, and (b) that we wouldn't want to rebaseline something if it was marked Crash or Timeout, but perhaps that's overly paranoid.
> In addition to the fixme to get rid of get_test_set, I worry that we'll eventually do the cleanup to not mark skip/rebaseline tests as pass, but just have the expectation be skip/rebaseline. Then this code will break again.
But the test should catch it if it breaks again, right?
Ojan Vafai
Comment 4
2012-10-09 12:28:00 PDT
Comment on
attachment 167807
[details]
Patch View in context:
https://bugs.webkit.org/attachment.cgi?id=167807&action=review
>>> Tools/Scripts/webkitpy/layout_tests/models/test_expectations.py:872 >>> + return self._model.get_test_set(REBASELINE, IMAGE) | self._model.get_test_set(REBASELINE, FAIL) | self._model.get_test_set(REBASELINE, PASS) >> >> Can you just use self._model.get_test_set("REBASELINE")? >> >> In addition to the fixme to get rid of get_test_set, I worry that we'll eventually do the cleanup to not mark skip/rebaseline tests as pass, but just have the expectation be skip/rebaseline. Then this code will break again. > > Oh, yeah, I guess you can. I was thinking (a) that you had to have expectations, and (b) that we wouldn't want to rebaseline something if it was marked Crash or Timeout, but perhaps that's overly paranoid.
Yeah, I think that's overly paranoid. If you add rebaseline, you're asking for it I think.
Dirk Pranke
Comment 5
2012-10-09 14:09:18 PDT
Created
attachment 167845
[details]
just get all of the tests marked for rebaselining
WebKit Review Bot
Comment 6
2012-10-09 15:19:34 PDT
Comment on
attachment 167845
[details]
just get all of the tests marked for rebaselining Clearing flags on attachment: 167845 Committed
r130819
: <
http://trac.webkit.org/changeset/130819
>
WebKit Review Bot
Comment 7
2012-10-09 15:19:38 PDT
All reviewed patches have been landed. Closing bug.
Note
You need to
log in
before you can comment on or make changes to this bug.
Top of Page
Format For Printing
XML
Clone This Bug