Add some tests to make sure message ports properly queue up message.

Add some tests to make sure message ports properly queue up message.

In particular this adds tests for the case were a message port is being
transfered to a service woker that isn't currently running.

Depends on https://codereview.chromium.org/695643002/

BUG=432678
Committed: https://src.chromium.org/viewvc/blink?view=rev&revision=186485

[falken, 2014-11-20 04:09:28]

falken@chromium.org changed reviewers:
+ falken@chromium.org

[falken, 2014-11-20 04:09:29]

let me revisit this after the patch it depends
<https://codereview.chromium.org/688413002/> on lands first

https://codereview.chromium.org/729923004/diff/1/LayoutTests/http/tests/servi...
File
LayoutTests/http/tests/serviceworker/chromium/postmessage-after-terminate.html
(right):

https://codereview.chromium.org/729923004/diff/1/LayoutTests/http/tests/servi...
LayoutTests/http/tests/serviceworker/chromium/postmessage-after-terminate.html:68:
return get_newest_worker(registration);
registration.active

https://codereview.chromium.org/729923004/diff/1/LayoutTests/http/tests/servi...
LayoutTests/http/tests/serviceworker/chromium/postmessage-after-terminate.html:83:
return delay(1);
can you comment what this delay is for?

https://codereview.chromium.org/729923004/diff/1/LayoutTests/http/tests/servi...
LayoutTests/http/tests/serviceworker/chromium/postmessage-after-terminate.html:100:
console.log(message);
debugging code?

https://codereview.chromium.org/729923004/diff/1/LayoutTests/http/tests/servi...
LayoutTests/http/tests/serviceworker/chromium/postmessage-after-terminate.html:109:
}, 'postMessage a port to a not running service worker');
hard to parse this... "on a port"?

https://codereview.chromium.org/729923004/diff/1/LayoutTests/http/tests/servi...
LayoutTests/http/tests/serviceworker/chromium/postmessage-after-terminate.html:159:
console.log(message);
debugging code?

https://codereview.chromium.org/729923004/diff/1/LayoutTests/http/tests/servi...
LayoutTests/http/tests/serviceworker/chromium/postmessage-after-terminate.html:168:
}, 'postMessage a port to a port to a not running service worker');
I can't parse this

https://codereview.chromium.org/729923004/diff/1/LayoutTests/http/tests/servi...
File LayoutTests/http/tests/serviceworker/chromium/resources/ping-worker.js
(right):

https://codereview.chromium.org/729923004/diff/1/LayoutTests/http/tests/servi...
LayoutTests/http/tests/serviceworker/chromium/resources/ping-worker.js:1: var
counter = 1;
I'm a bit lost, why is this change needed? Feels like cheating to start at 1
before you get a ping.

https://codereview.chromium.org/729923004/diff/1/LayoutTests/http/tests/servi...
LayoutTests/http/tests/serviceworker/chromium/resources/ping-worker.js:2:
self.onmessage = function(event) {
Seems cleaner do to counter++; postMessage(counter); here?

[Marijn Kruisselbrink, 2014-11-20 21:37:09]

Patchset #2 (id:20001) has been deleted

[Marijn Kruisselbrink, 2014-11-20 21:40:45]

Patchset #2 (id:40001) has been deleted

[Marijn Kruisselbrink, 2014-11-20 21:42:38]

https://codereview.chromium.org/729923004/diff/1/LayoutTests/http/tests/servi...
File
LayoutTests/http/tests/serviceworker/chromium/postmessage-after-terminate.html
(right):

https://codereview.chromium.org/729923004/diff/1/LayoutTests/http/tests/servi...
LayoutTests/http/tests/serviceworker/chromium/postmessage-after-terminate.html:68:
return get_newest_worker(registration);
On 2014/11/20 04:09:28, falken wrote:
> registration.active

Done.

https://codereview.chromium.org/729923004/diff/1/LayoutTests/http/tests/servi...
LayoutTests/http/tests/serviceworker/chromium/postmessage-after-terminate.html:83:
return delay(1);
On 2014/11/20 04:09:29, falken wrote:
> can you comment what this delay is for?

Done.

https://codereview.chromium.org/729923004/diff/1/LayoutTests/http/tests/servi...
LayoutTests/http/tests/serviceworker/chromium/postmessage-after-terminate.html:100:
console.log(message);
On 2014/11/20 04:09:28, falken wrote:
> debugging code?

Done.

https://codereview.chromium.org/729923004/diff/1/LayoutTests/http/tests/servi...
LayoutTests/http/tests/serviceworker/chromium/postmessage-after-terminate.html:109:
}, 'postMessage a port to a not running service worker');
On 2014/11/20 04:09:28, falken wrote:
> hard to parse this... "on a port"?

Done.

https://codereview.chromium.org/729923004/diff/1/LayoutTests/http/tests/servi...
LayoutTests/http/tests/serviceworker/chromium/postmessage-after-terminate.html:159:
console.log(message);
On 2014/11/20 04:09:29, falken wrote:
> debugging code?

Done.

https://codereview.chromium.org/729923004/diff/1/LayoutTests/http/tests/servi...
LayoutTests/http/tests/serviceworker/chromium/postmessage-after-terminate.html:168:
}, 'postMessage a port to a port to a not running service worker');
On 2014/11/20 04:09:28, falken wrote:
> I can't parse this
I tried to rephrase it. If you have suggestions on how to make it clearer,
they're welcome. It's trying to say that this test tests that sending a message
over a message ports works correctly, when that message port itself was send to
a serviceworker using postmessage, where the service worker wasn't running at
the time of the post message.
(this is an interesting case since it shows that not only ports transferred
directly to a service worker need to be queued, but also all ports send on those
ports (recursively)).

https://codereview.chromium.org/729923004/diff/1/LayoutTests/http/tests/servi...
File LayoutTests/http/tests/serviceworker/chromium/resources/ping-worker.js
(right):

https://codereview.chromium.org/729923004/diff/1/LayoutTests/http/tests/servi...
LayoutTests/http/tests/serviceworker/chromium/resources/ping-worker.js:1: var
counter = 1;
On 2014/11/20 04:09:29, falken wrote:
> I'm a bit lost, why is this change needed? Feels like cheating to start at 1
> before you get a ping.
Sorry, these changes were just an incorrect merge between my various local
branches. This CL is not supposed to change this file.

[Marijn Kruisselbrink, 2014-12-03 13:44:02]

PTAL, all the dependencies of this CL have landed

[falken, 2014-12-04 06:21:56]

lgtm!

[Marijn Kruisselbrink, 2014-12-04 08:22:01]

The CQ bit was checked by mek@chromium.org

[commit-bot: I haz the power, 2014-12-04 08:23:02]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/729923004/100001

[commit-bot: I haz the power, 2014-12-04 08:25:26]

Committed patchset #4 (id:100001) as
https://src.chromium.org/viewvc/blink?view=rev&revision=186485