1253713 - [e10s] Make widget/tests/test_plugin_scroll_invalidation.html work under e10s

Status: RESOLVED FIXED

(Core Graveyard :: Plug-ins, defect)

Description

[Drew Willcoxon :adw]

Attachment: patch

[e10s] Make widget/tests/test_plugin_scroll_invalidation.html work under e10s.

Andrew, this patch is based off your patch in bug 1237799.  It moves the test to dom/plugins/test/mochitest/ and calls setTestPluginEnabledState, which fixes the test locally for me on e10s and also it looks like on try (but I don't understand why :-))

https://treeherder.mozilla.org/#/jobs?repo=try&revision=e3d415ca2221

Comment 1

[Andrew McCreight [:mccr8]]

The reason the patch works is that the implementation of setTestPluginEnabledState in the widget directory is broken in e10s (it tries to set the property directly, but that can't be done in the content process):
  http://mxr.mozilla.org/mozilla-central/source/widget/tests/utils.js

but the one in plugin-utils has been fixed (it uses a special powers function to send a message to the parent process that actually does the change, or something like that):
  http://mxr.mozilla.org/mozilla-central/source/dom/plugins/test/mochitest/plugin-utils.js#28

The basic problem is that people copied and pasted this broken setTestPluginEnableState thing all over the place.

I guess the widget utils.js one can't just be fixed because there are still other Mochitest Chrome tests that use it.

Ideally, SpecialPowers.testPluginState() would be fixed to not require a wrapper like this and then we wouldn't need plugin-utils.js but nobody has gotten around to it.

Comment 2

[Andrew McCreight [:mccr8]]

Anyways, I filed bug 1253742 for making SpecialPowers nicer but I guess this is okay for now.

Comment 3

[Drew Willcoxon :adw]

https://hg.mozilla.org/integration/fx-team/rev/85e94038875f

Comment 4

[Drew Willcoxon :adw]

Shoulda pushed that to inbound....

Comment 5

[Ryan VanderMeulen [:RyanVM]]

https://hg.mozilla.org/mozilla-central/rev/85e94038875f