ServiceWorker: Add layout test for worker script MIME type.

ServiceWorker: Add layout test for worker script MIME type.

A worker script with invalid MIME type should not be registered.

BUG=400604
Committed: https://src.chromium.org/viewvc/blink?view=rev&revision=181006

[xiang, 2014-08-07 05:20:07]

PTAL, thanks!
BTW, I found the exception code is not match the spec(some AbortError should be
SecurityError/NetworkError), do we need to change that?

[falken, 2014-08-07 05:34:34]

Let me see if I understand. Fetching 'resources/plain-text-worker.php' returns
an empty "service worker" with content-type 'text/plain'. Since it's not a
JavaScript content-type, register() should fail.

Do we already pass this test? If the spec says it should throw SecurityError, we
should. However, I don't see where it's spec'd, looks like it's just a GitHub
issue at this moment?

[xiang, 2014-08-07 06:19:50]

On 2014/08/07 05:34:34, falken wrote:
> Let me see if I understand. Fetching 'resources/plain-text-worker.php' returns
> an empty "service worker" with content-type 'text/plain'. Since it's not a
> JavaScript content-type, register() should fail.
> 

Right.

> Do we already pass this test? If the spec says it should throw SecurityError,
we
> should. However, I don't see where it's spec'd, looks like it's just a GitHub
> issue at this moment?

No, the test depends on https://codereview.chromium.org/446323002/. Sorry, I
forget to point to it in the message.

Seems the spec hasn't updated according to this issue ATM. The exception code
mismatch also happens when script load fail or redirected, the spec says
NetworkError/SecurityError should be thrown, but we throw AbortError in these
cases.

Should I wait for the spec update? Or send another CL to fix the error code
first? What's your suggestion?

[falken, 2014-08-07 07:12:09]

On 2014/08/07 06:19:50, xiang wrote:
> On 2014/08/07 05:34:34, falken wrote:
> > Let me see if I understand. Fetching 'resources/plain-text-worker.php'
returns
> > an empty "service worker" with content-type 'text/plain'. Since it's not a
> > JavaScript content-type, register() should fail.
> > 
> 
> Right.
> 
> > Do we already pass this test? If the spec says it should throw
SecurityError,
> we
> > should. However, I don't see where it's spec'd, looks like it's just a
GitHub
> > issue at this moment?
> 
> No, the test depends on https://codereview.chromium.org/446323002/. Sorry, I
> forget to point to it in the message.
> 
> Seems the spec hasn't updated according to this issue ATM. The exception code
> mismatch also happens when script load fail or redirected, the spec says
> NetworkError/SecurityError should be thrown, but we throw AbortError in these
> cases.
> 
> Should I wait for the spec update? Or send another CL to fix the error code
> first? What's your suggestion?

I see, thanks. We should fix the error names but can do it after this patch.
This patch lgtm.

[jsbell, 2014-08-26 19:22:13]

What's the status of this patch? Can we get this test landed? Anything I can do
to help?

(We should file a new bug for the incorrect error type.)

[xiang, 2014-08-28 01:16:53]

The CQ bit was checked by xiang.long@intel.com

[xiang, 2014-08-28 01:17:24]

The CQ bit was unchecked by xiang.long@intel.com

[xiang, 2014-08-28 01:18:13]

The CQ bit was checked by xiang.long@intel.com

[commit-bot: I haz the power, 2014-08-28 01:18:34]

CQ is trying da patch. Follow status at
 https://chromium-status.appspot.com/cq/xiang.long@intel.com/448063002/20001

[commit-bot: I haz the power, 2014-08-28 02:16:12]

Committed patchset #2 (id:20001) as 181006