Invalid MediaSource duration value should throw TyperError instead of InvalidStateError

[Nael Ouedraogo]

Invalid MediaSource duration value should throw TyperError instead of InvalidStateError.

[Nael Ouedraogo]

Created attachment 309037 [details]
Patch

[Nael Ouedraogo]

I am going to fix the remaining failing tests in  bug 171668. I think the timeout for mac ports is not related to this patch since it happens also without this patch.

[youenn fablet]

Comment on attachment 309037 [details]
Patch

View in context: https://bugs.webkit.org/attachment.cgi?id=309037&action=review

> LayoutTests/ChangeLog:16
> +        * platform/mac/imported/w3c/web-platform-tests/media-source/mediasource-duration-boundaryconditions-expected.txt: Added.

Why are we not rebasing LayoutTests/imported/w3c/web-platform-tests/media-source/mediasource-duration-boundaryconditions-expected.txt?

[Nael Ouedraogo]

Thanks for the review.

> Why are we not rebasing
> LayoutTests/imported/w3c/web-platform-tests/media-source/mediasource-
> duration-boundaryconditions-expected.txt?

Currently, failure are different for mac and gtk ports. Mac port has a timeout issue when setting duration to zero. Gtk port fails to set duration to POSITIVE_INFINITY. I am going to fix remaining failures for gtk port (bug 171668 and bug 171860 ) and then remove the expected results for gtk platform.

[WebKit Commit Bot]

Comment on attachment 309037 [details]
Patch

Clearing flags on attachment: 309037

Committed r216508: <http://trac.webkit.org/changeset/216508>

[WebKit Commit Bot]

All reviewed patches have been landed.  Closing bug.

[Matt Lewis]

Unsure if this is expected but Sierra is failing but El Capitan is passing. 

https://build.webkit.org/results/Apple%20Sierra%20Release%20WK2%20(Tests)/r216515%20(1315)/results.html

https://webkit-test-results.webkit.org/dashboards/flakiness_dashboard.html#showAllRuns=true&tests=imported%2Fw3c%2Fweb-platform-tests%2Fmedia-source%2Fmediasource-duration-boundaryconditions.html

--- /Volumes/Data/slave/sierra-release-tests-wk2/build/layout-test-results/imported/w3c/web-platform-tests/media-source/mediasource-duration-boundaryconditions-expected.txt
+++ /Volumes/Data/slave/sierra-release-tests-wk2/build/layout-test-results/imported/w3c/web-platform-tests/media-source/mediasource-duration-boundaryconditions-actual.txt
@@ -1,5 +1,3 @@
-
-Harness Error (TIMEOUT), message = null
 
 PASS Set duration to 2^31 - 1 
 PASS Set duration to 1 
@@ -12,6 +10,6 @@
 PASS Set duration to lowest value. 
 PASS Set duration to a negative double. 
 PASS Set duration to a positive double. 
-TIMEOUT Set duration to zero Test timed out
+PASS Set duration to zero 
 PASS Set duration to NaN

[Matt Lewis]

Reverted r216508 for reason:

The layout test enabled in this change is failing on Sierra.

Committed r216532: <http://trac.webkit.org/changeset/216532>

[Nael Ouedraogo]

Created attachment 309709 [details]
Patch

[Nael Ouedraogo]

> Reverted r216508 for reason:
> 
> The layout test enabled in this change is failing on Sierra.
Sorry for the trouble.

In uploaded patch, I rebased LayoutTests/imported/w3c/web-platform-tests/media-source/mediasource-duration-boundaryconditions-expected.txt so that it passes on Sierra. I added new expected results for El Capitan and I updated Gtk's results.

[Chris Dumez]

Comment on attachment 309709 [details]
Patch

r=me

[Nael Ouedraogo]

Thanks for the review.

[WebKit Commit Bot]

Comment on attachment 309709 [details]
Patch

Clearing flags on attachment: 309709

Committed r216860: <http://trac.webkit.org/changeset/216860>

[WebKit Commit Bot]

All reviewed patches have been landed.  Closing bug.