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

Issue 827367 link

Starred by 2 users

Issue metadata

Status: Fixed
Owner:
Last visit > 30 days ago
Closed: Apr 2018
Cc:
Components:
EstimatedDays: ----
NextAction: ----
OS: ----
Pri: 3
Type: Bug



Sign in to add a comment

WPT import breaks webaudio IDL tests

Project Member Reported by rtoy@chromium.org, Mar 29 2018

Issue description

The WPT import in https://chromium-review.googlesource.com/c/chromium/src/+/986281 added a bunch of expected results files for webaudio IDL tests.  These tests used to pass but now fail, hence the new expected results files.


webaudio/idl-harness:
FAIL Test driver promise_test: Unhandled rejection with value: object "TrackEventInit inherits EventInit, but TrackEventInit is not an interface."

delaynode:
FAIL DelayNode IDL Test Uncaught DelayOptions inherits AudioNodeOptions, but DelayOptions is not an interface.
3	
Harness: the test ran to completion.

gain:
FAIL GainNode IDL Test Uncaught GainOptions inherits AudioNodeOptions, but GainOptions is not an interface.
3	
Harness: the test ran to completion.
 

Comment 1 by jsb...@chromium.org, Mar 29 2018

Cc: robertma@chromium.org

Comment 2 by jsb...@chromium.org, Mar 29 2018

It "broke" basically all the idlharness tests. Something changed in idlharness.js and testharness.js 


Comment 3 by rtoy@chromium.org, Mar 29 2018

Yeah. I just wanted to capture this change before I forgot, because the
expectations will pass and I won't remember these tests are actually broken.
Looks like another IDL breakage caused by the large-scale IDL
overhaul/update in the upstream by Luke and Philip, similar to what
happened last week (SVGElement, if you still remember). I'm on holiday
today and replying from my cellphone. Could you assign to me and CC foolip?
I'll take a look next week.

By the way, import notifier successfully captured the new failures and
filed bugs to both storage and webaudio. If such incidents happened earlier
last year, it would indeed go unnoticed. But thanks to the notification we
should be able to catch them by triaging the bugs now.

Comment 5 by rtoy@chromium.org, Mar 30 2018

Cc: -robertma@chromium.org foolip@chromium.org
Owner: robertma@chromium.org
Status: Assigned (was: Available)
Assigned.  Thanks for your help on this!

I'm going to dup 827513 to this. I am surprised there's no bug for idl-harness and delay node failures.

Comment 6 by rtoy@chromium.org, Mar 30 2018

Cc: robertma@chromium.org rtoy@chromium.org hongchan@chromium.org
 Issue 827513  has been merged into this issue.

Comment 7 by foolip@chromium.org, Mar 30 2018

Owner: lukebjerring@chromium.org
This is due to https://github.com/w3c/web-platform-tests/pull/10240 and, I think, assuming that inheritance implies interfaces, when dictionaries can also be inherited.

Comment 8 by jsb...@chromium.org, Mar 30 2018

Yeah, I just identified the same root cause. Testing a local fix now...

Comment 9 by jsb...@chromium.org, Mar 30 2018

For the record: foolip@ is reverting the upstream change. 

Revert https://github.com/w3c/web-platform-tests/pull/10259 now landed upstream (by lukebjerring@, we were both sending reverts)
Cc: michaeln@chromium.org mek@chromium.org
 Issue 827511  has been merged into this issue.
Status: Fixed (was: Assigned)
The revert has been imported and all affected IDL tests are back to normal.

Thanks for taking care of this during the weekend!

Sign in to add a comment