764232 - Disable oop browser element tests on Windows in a smarter way

Status: RESOLVED FIXED

(Firefox OS Graveyard :: General, defect)

Description

[Justin Lebar (not reading bugmail)]

The way I disabled oop browser element tests on Windows was pretty dumb: We still run the *_oop_* tests, but we run them in process.

Problem is, some tests will only work out of process.  It would be convenient to have a _oop_ test file and no corresponding _inproc_ file, except that doesn't work under the current scheme, because on Windows, the _oop_ file will be run in process!

I should make this smarter.  The simplest thing I can do would probably be to disable the oop tests in the build.

Comment 1

[Justin Lebar (not reading bugmail)]

Attachment: Patch v1

Mounir, who last reviewed this code, is out for the week.  Vivien, do you feel comfortable reviewing this change?

Comment 2

[Vivien Nicolas (:vingtetun) (:21) - (NOT reading bugmails, needinfo? please)]

Comment on attachment 632494 [details] [diff] [review]
Patch v1

Review of attachment 632494 [details] [diff] [review]:
-----------------------------------------------------------------

::: dom/browser-element/mochitest/browserElementTestHelpers.js
@@ +124,5 @@
>  
> +// Enable or disable OOP-by-default depending on the test's filename.  You can
> +// still force OOP on or off with <iframe mozbrowser remote=true/false>, at
> +// least until bug 756376 lands.
> +var oop = location.pathname.indexOf('_inproc_') == -1;

Why don't you test |location.pathname.indexOf('_oop_') != -1|. Does all files that does not contains _inproc_ are oop files?

Comment 3

[Justin Lebar (not reading bugmail)]

> Why don't you test |location.pathname.indexOf('_oop_') != -1|. Does all files that does not contains 
> _inproc_ are oop files?

Vice versa; there are some tests which contain neither _inproc_ nor _oop_ which we should run in-process, so long as oop doesn't work on Windows.  We could of course change their names.

Comment 4

[Justin Lebar (not reading bugmail)]

Inbound for FF16: https://hg.mozilla.org/integration/mozilla-inbound/rev/ee9516c7fad7

Comment 5

[Ed Morley [:emorley]]

https://hg.mozilla.org/mozilla-central/rev/ee9516c7fad7