David Teller [:Yoric] - still alive but not very active Reporter
	Description • 10 years ago

Once we land bug 976205, the following errors will cause oranges.

TEST-UNEXPECTED-FAIL | A promise chain failed to handle a rejection.
04:59:01     INFO -  Date: Wed Mar 19 2014 04:59:00 GMT-0700 (PDT)
04:59:01     INFO -  Full Message: [Exception... "Component returned failure code: 0x80004005 (NS_ERROR_FAILURE) [nsIDOMWindowUtils.sendKeyEvent]"  nsresult: "0x80004005 (NS_ERROR_FAILURE)"  location: "JS frame :: chrome://mochikit/content/tests/SimpleTest/specialpowersAPI.js :: doApply :: line 95"  data: no]
04:59:01     INFO -  Full Stack: JS frame :: resource://gre/modules/Promise.jsm -> resource://gre/modules/Promise-backend.js :: PendingErrors.register :: line 154
04:59:01     INFO -  JS frame :: resource://gre/modules/Promise.jsm -> resource://gre/modules/Promise-backend.js :: this.PromiseWalker.completePromise :: line 616
04:59:01     INFO -  JS frame :: resource://gre/modules/Promise.jsm -> resource://gre/modules/Promise-backend.js :: Handler.prototype.process :: line 845
04:59:01     INFO -  JS frame :: resource://gre/modules/Promise.jsm -> resource://gre/modules/Promise-backend.js :: this.PromiseWalker.walkerLoop :: line 688
04:59:01     INFO -  JS frame :: resource://gre/modules/commonjs/toolkit/loader.js -> resource://gre/modules/devtools/server/main.js -> resource://gre/modules/devtools/server/actors/script.js :: EventLoop.prototype.enter :: line 387
04:59:01     INFO -  JS frame :: resource://gre/modules/commonjs/toolkit/loader.js -> resource://gre/modules/devtools/server/main.js -> resource://gre/modules/devtools/server/actors/script.js :: ThreadActor.prototype._pushThreadPause :: line 557
04:59:01     INFO -  JS frame :: resource://gre/modules/commonjs/toolkit/loader.js -> resource://gre/modules/devtools/server/main.js -> resource://gre/modules/devtools/server/actors/script.js :: ThreadActor.prototype._pauseAndRespond :: line 840
04:59:01     INFO -  JS frame :: resource://gre/modules/commonjs/toolkit/loader.js -> resource://gre/modules/devtools/server/main.js -> resource://gre/modules/devtools/server/actors/script.js :: ThreadActor.prototype.onDebuggerStatement :: line 2169
04:59:01     INFO -  JS frame :: http://example.com/browser/browser/devtools/webconsole/test/test-closures.html :: injectPerson :: line 18
04:59:01     INFO -  JS frame :: http://example.com/browser/browser/devtools/webconsole/test/test-closures.html :: onclick :: line 1
04:59:01     INFO -  JS frame :: chrome://mochitests/content/browser/browser/devtools/webconsole/test/browser_webconsole_closure_inspection.js :: test/onLoad/</< :: line 40
04:59:01     INFO -  JS frame :: resource://gre/modules/Promise.jsm -> resource://gre/modules/Promise-backend.js :: Handler.prototype.process :: line 809
For testing, you may wish to import the patch of bug 976205.

	Panos Astithas (he/him) [:past] (please ni?) Assignee
	Comment 1 • 10 years ago

This looks like it has something to do with bug 1020609. It looks like GetWidget() returns null in SendKeyEvent for some reason.

	Bobby Holley (:bholley)
	Comment 2 • 10 years ago

Given that this is a browser-chrome test, SpecialPowers should not be involved at all.

Given that SpecialPowers exists in chrome tests, we probably need to rejigger the ordering here:
http://mxr.mozilla.org/mozilla-central/source/testing/mochitest/tests/SimpleTest/EventUtils.js#841

	Panos Astithas (he/him) [:past] (please ni?) Assignee
	Comment 4 • 10 years ago

https://hg.mozilla.org/integration/fx-team/rev/5a6721fbc07a

	Bobby Holley (:bholley)
	Comment 5 • 10 years ago

(In reply to Bobby Holley (:bholley) from comment #2)
> Given that this is a browser-chrome test, SpecialPowers should not be
> involved at all.
> 
> Given that SpecialPowers exists in chrome tests, we probably need to
> rejigger the ordering here:
> http://mxr.mozilla.org/mozilla-central/source/testing/mochitest/tests/
> SimpleTest/EventUtils.js#841

Did this turn out to be unrelated? Can we at least get a bug on file for it?

	Panos Astithas (he/him) [:past] (please ni?) Assignee
	Comment 6 • 10 years ago

Changing the order of the conditionals didn't help as one or the other would always match. Looking around at the implementation of sendMouseEvent() it appears that SpecialPowers is explicitly used, and combined with what I read on bug 674323, bug 694976 and bug 649360 it seems to me that (some?) browser mochitests depend on SpecialPowers.

We should probably get a bug on file for the key event on a particular element can't be sent, but I don't know how to frame the issue in terms of SpecialPowers and EventUtils.

	Bobby Holley (:bholley)
	Comment 7 • 10 years ago

(In reply to Panos Astithas [:past] from comment #6)
> Changing the order of the conditionals didn't help as one or the other would
> always match. Looking around at the implementation of sendMouseEvent() it
> appears that SpecialPowers is explicitly used, and combined with what I read
> on bug 674323, bug 694976 and bug 649360 it seems to me that (some?) browser
> mochitests depend on SpecialPowers.
> 
> We should probably get a bug on file for the key event on a particular
> element can't be sent, but I don't know how to frame the issue in terms of
> SpecialPowers and EventUtils.

Using SpecialPowers in general is fine - the issue is that we shouldn't be going through SpecialPowers.wrap in privileged code, since we're basically just proxying things over to another privileged scope for no reason.

	Wes Kocher (:KWierso) (Not reading bugmail; email directly if needed)
	Comment 8 • 10 years ago

https://hg.mozilla.org/mozilla-central/rev/5a6721fbc07a

	Panos Astithas (he/him) [:past] (please ni?) Assignee
	Comment 9 • 10 years ago

(In reply to Bobby Holley (:bholley) (PTO 6/25-6/29) from comment #5)
> Did this turn out to be unrelated? Can we at least get a bug on file for it?

Filed bug 1030657 and CCed you.