Add 'self.onerror' to Worker tests.

Add 'self.onerror' to Worker tests.

We're currently not testing 'self.onerror' for same-domain or cross-domain
calls to 'importScript'. This patch adds tests for this functionality, but makes
no functional changes.

R=ch.dumez@sisa.samsung.com

Committed: https://src.chromium.org/viewvc/blink?view=rev&revision=154914

[Mike West, 2013-07-24 19:00:56]

Hi Christophe, do you have a burning desire to review MOAR Workers changes? :)

-mike

[abarth-chromium, 2013-07-24 22:43:36]

https://codereview.chromium.org/19605005/diff/2002/Source/bindings/v8/WorkerS...
File Source/bindings/v8/WorkerScriptController.cpp (right):

https://codereview.chromium.org/19605005/diff/2002/Source/bindings/v8/WorkerS...
Source/bindings/v8/WorkerScriptController.cpp:170: 
Can we add an ASSERT that the invariants you wrote about in the description
remain true?

[do-not-use, 2013-07-25 07:25:28]

I get the following error when trying to look at the changes:
error: old chunk mismatch

[Mike West, 2013-07-25 07:45:08]

On 2013/07/25 07:25:28, Christophe Dumez wrote:
> I get the following error when trying to look at the changes:
> error: old chunk mismatch

That turns out to be a good thing; I need to throw this CL away, as I've
apparently pulled the CORS requirement out of thin air. Apologies for my
confusion.

[Mike West, 2013-07-25 07:55:17]

On 2013/07/25 07:45:08, Mike West wrote:
> On 2013/07/25 07:25:28, Christophe Dumez wrote:
> > I get the following error when trying to look at the changes:
> > error: old chunk mismatch
> 
> That turns out to be a good thing; I need to throw this CL away, as I've
> apparently pulled the CORS requirement out of thin air. Apologies for my
> confusion.

Ok, I've dropped the functional change, and kept the changes to the tests (added
simple checks for `self.onerror`). Would you mind taking a look at those? I'll
come back to the sanitization problem once I have a reasonable solution.

[do-not-use, 2013-07-25 08:08:13]

https://codereview.chromium.org/19605005/diff/20001/LayoutTests/http/tests/wo...
File
LayoutTests/http/tests/workers/resources/worker-importscripts-onerror-crossorigin.js
(right):

https://codereview.chromium.org/19605005/diff/20001/LayoutTests/http/tests/wo...
LayoutTests/http/tests/workers/resources/worker-importscripts-onerror-crossorigin.js:2:
postMessage({ 'message': message, 'url': url, 'line': line, 'column': column });
Might be nice to use the same naming as in ErrorEvent for consistency, what do
you think?
(http://www.whatwg.org/specs/web-apps/current-work/multipage/webappapis.html#e...)

https://codereview.chromium.org/19605005/diff/20001/LayoutTests/http/tests/wo...
File
LayoutTests/http/tests/workers/worker-importScripts-onerror-crossorigin-expected.txt
(right):

https://codereview.chromium.org/19605005/diff/20001/LayoutTests/http/tests/wo...
LayoutTests/http/tests/workers/worker-importScripts-onerror-crossorigin-expected.txt:12:
FAIL onerrorLine should be 1. Was 5.
Why is this failing?

https://codereview.chromium.org/19605005/diff/20001/LayoutTests/http/tests/wo...
File
LayoutTests/http/tests/workers/worker-importScripts-onerror-redirect-to-crossorigin-expected.txt
(right):

https://codereview.chromium.org/19605005/diff/20001/LayoutTests/http/tests/wo...
LayoutTests/http/tests/workers/worker-importScripts-onerror-redirect-to-crossorigin-expected.txt:12:
FAIL onerrorLine should be 2. Was 6.
Why is this failing now?

https://codereview.chromium.org/19605005/diff/20001/LayoutTests/http/tests/wo...
File
LayoutTests/http/tests/workers/worker-importScripts-onerror-sameorigin-expected.txt
(right):

https://codereview.chromium.org/19605005/diff/20001/LayoutTests/http/tests/wo...
LayoutTests/http/tests/workers/worker-importScripts-onerror-sameorigin-expected.txt:12:
FAIL onerrorLine should be 1. Was 5.
failing?

[Mike West, 2013-07-25 08:29:00]

One more try.

https://codereview.chromium.org/19605005/diff/20001/LayoutTests/http/tests/wo...
File
LayoutTests/http/tests/workers/resources/worker-importscripts-onerror-crossorigin.js
(right):

https://codereview.chromium.org/19605005/diff/20001/LayoutTests/http/tests/wo...
LayoutTests/http/tests/workers/resources/worker-importscripts-onerror-crossorigin.js:2:
postMessage({ 'message': message, 'url': url, 'line': line, 'column': column });
On 2013/07/25 08:08:14, Christophe Dumez wrote:
> Might be nice to use the same naming as in ErrorEvent for consistency, what do
> you think?
>
(http://www.whatwg.org/specs/web-apps/current-work/multipage/webappapis.html#e...)

I can do that.

Honestly, I'd prefer to rename the ErrorEvent properties, as they're already
internally inconsistent (lineno (!!) vs column, for instance). :)

https://codereview.chromium.org/19605005/diff/20001/LayoutTests/http/tests/wo...
File
LayoutTests/http/tests/workers/worker-importScripts-onerror-crossorigin-expected.txt
(right):

https://codereview.chromium.org/19605005/diff/20001/LayoutTests/http/tests/wo...
LayoutTests/http/tests/workers/worker-importScripts-onerror-crossorigin-expected.txt:12:
FAIL onerrorLine should be 1. Was 5.
On 2013/07/25 08:08:14, Christophe Dumez wrote:
> Why is this failing?

I added the 'self.onerror' handler, which moved everything down by 4 lines, but
I failed to update the expectations in the test. I've fixed that.

https://codereview.chromium.org/19605005/diff/20001/LayoutTests/http/tests/wo...
File
LayoutTests/http/tests/workers/worker-importScripts-onerror-redirect-to-crossorigin-expected.txt
(right):

https://codereview.chromium.org/19605005/diff/20001/LayoutTests/http/tests/wo...
LayoutTests/http/tests/workers/worker-importScripts-onerror-redirect-to-crossorigin-expected.txt:12:
FAIL onerrorLine should be 2. Was 6.
On 2013/07/25 08:08:14, Christophe Dumez wrote:
> Why is this failing now?

Done.

https://codereview.chromium.org/19605005/diff/20001/LayoutTests/http/tests/wo...
File
LayoutTests/http/tests/workers/worker-importScripts-onerror-sameorigin-expected.txt
(right):

https://codereview.chromium.org/19605005/diff/20001/LayoutTests/http/tests/wo...
LayoutTests/http/tests/workers/worker-importScripts-onerror-sameorigin-expected.txt:12:
FAIL onerrorLine should be 1. Was 5.
On 2013/07/25 08:08:14, Christophe Dumez wrote:
> failing?

Ditto.

[do-not-use, 2013-07-25 08:30:55]

LGTM, thanks.

[commit-bot: I haz the power, 2013-07-25 08:33:08]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/mkwst@chromium.org/19605005/31001

[Mike West, 2013-07-25 13:13:26]

Committed patchset #4 manually as r154914 (presubmit successful).