Closed
Bug 886589
Opened 11 years ago
Closed 11 years ago
Change - Add infrastructure for tapping center of elements rather than relying on hard-coded offsets
Categories
(Firefox for Metro Graveyard :: Tests, defect, P3)
Tracking
(Not tracked)
RESOLVED
FIXED
People
(Reporter: jwilde, Assigned: jwilde)
References
Details
(Whiteboard: feature=change c=tbd u=tbd p=2)
Attachments
(1 file, 1 obsolete file)
5.74 KB,
patch
|
jimm
:
review+
|
Details | Diff | Splinter Review |
Some of the tests turn out to be really fragile in certain environments due to the use of magic numbers to tap more or less in the center of elements. Sometimes the magic number offsets don't end up hitting in bounds of the element that we want to hit.
Updated•11 years ago
|
Blocks: metrov1defect&change
Whiteboard: feature=change c=tbd u=tbd p=0
Updated•11 years ago
|
Priority: -- → P3
Updated•11 years ago
|
QA Contact: jbecerra
Whiteboard: feature=change c=tbd u=tbd p=0 → feature=change c=tbd u=tbd p=2
Assignee | ||
Comment 1•11 years ago
|
||
Attachment #767464 -
Flags: review?(jmathies)
Assignee | ||
Comment 2•11 years ago
|
||
Attachment #767464 -
Attachment is obsolete: true
Attachment #767464 -
Flags: review?(jmathies)
Attachment #767466 -
Flags: review?(jmathies)
Updated•11 years ago
|
Attachment #767466 -
Flags: review?(jmathies) → review+
Assignee | ||
Comment 3•11 years ago
|
||
https://hg.mozilla.org/integration/mozilla-inbound/rev/be5046383861
Comment 4•11 years ago
|
||
https://hg.mozilla.org/mozilla-central/rev/be5046383861
Status: ASSIGNED → RESOLVED
Closed: 11 years ago
Resolution: --- → FIXED
Comment 5•11 years ago
|
||
Jonathan, how would I go about verifying this? I tried going through the list of bugs for intermittent failures, but I can't see if there's a decline. Should there be?
Flags: needinfo?(jwilde)
Assignee | ||
Comment 6•11 years ago
|
||
These changes won't really help with the intermittent issues. They mainly help in ensuring that a developer won't break a test by changing the size of an element. I think everybody now runs the tests when developing/reviewing and catches test failures before landing. I'm really not sure how one would quantitatively verify this.
Flags: needinfo?(jwilde)
Comment 7•11 years ago
|
||
Should this be excluded from the regular QA regression testing?
Assignee | ||
Comment 9•11 years ago
|
||
Given that this is exceedingly hard to support/refute whether this has indeed been fixed by the patches, I'd say that's probably a good plan.
Flags: needinfo?(jwilde)
Comment 10•11 years ago
|
||
Thanks. I've removed it from the Metro testing sheet.
You need to log in
before you can comment on or make changes to this bug.
Description
•