New issue
Advanced search Search tips
Note: Color blocks (like █ or █) mean that a user may not be available. Tooltip shows the reason.

Issue 815284 link

Starred by 1 user
Status: Fixed
Owner:
shend@chromium.org
Closed: Feb 2018
Cc:
█ nainar@chromium.org
foolip@chromium.org
style-bugs@google.com
Components:
EstimatedDays: ----
NextAction: ----
OS: ----
Pri: 1
Type: Bug



Sign in to add a comment

Broken WPT export after Rejig CSSResourceValue hierarchy.

Project Member Reported by danyao@chromium.org, Feb 23 2018 
Hi - I'm filing this bug as Web Platform Test sheriff.

The upstream WPT PR for https://chromium-review.googlesource.com/c/chromium/src/+/934001/3 failed to build for firefox:nightly, safari:11.0, and MicrosoftEdge:16.16299.

The firefox job log contains the following errors:
PID 7501 | JavaScript error: http://web-platform.test:8000/css/css-typed-om/stylevalue-serialization/cssTransformValue.tentative.html, line 13: ReferenceError: CSSTranslate is not defined
PID 7887 | JavaScript error: http://web-platform.test:8000/css/css-typed-om/stylevalue-subclasses/cssPerspective.tentative.html, line 16: TypeError: CSS.number is not a function
PID 8013 | JavaScript error: http://web-platform.test:8000/css/css-typed-om/stylevalue-subclasses/cssPositionValue.html, line 17: TypeError: CSS.px is not a function
...

These seem that they could be related to your change. Please help investigate.

Reference:
PR build status on GitHub: https://github.com/w3c/web-platform-tests/pull/9641
Firefox nightly build logs for this PR: https://travis-ci.org/w3c/web-platform-tests/jobs/345157778

Comment 1 by shend@chromium.org, Feb 24 2018 

Sorry I'm a bit confused, the tests I added are for a new spec which is only shipped by Chrome so far. Wouldn't it make sense that those tests fail? (e.g. CSSTranslate is part of the new spec, other browsers don't have this yet).

foolip@, do you know what might be happening here?

Comment 2 by foolip@chromium.org, Feb 24 2018 

Merely failing tests should not cause any Travis checks to fail, as all we *should* be checking for is flaky tests. https://travis-ci.org/w3c/web-platform-tests/jobs/345157778 ran for 39 minutes, so I was thinking a timeout, but the log ends with:

```
Log reached capacity (3145728 bytes); output disabled.

No output has been received in the last 10m0s, this potentially indicates a stalled build or something wrong with the build itself.
Check the details on how to adjust your build configuration on: https://docs.travis-ci.com/user/common-build-problems/#Build-times-out-because-no-output-was-received

The build has been terminated
```

I don't think this is a timeout since those logs end differently, and the time limit is 50 minutes. I'll restart the Travis job.

If the problem persist and there's still no reason to think it's because of the tests, I usually file another issue and then force merge the PR.

Comment 3 by foolip@chromium.org, Feb 24 2018 

Something similar is happening in https://github.com/w3c/web-platform-tests/pull/9238 too.

Comment 4 by foolip@chromium.org, Feb 26 2018

Status: Fixed (was: Untriaged)
I've merged those two PRs now.
►

Sign in to add a comment