Make sure the test finishes after EventSource loading has completed.

Make sure the test finishes after EventSource loading has completed.

With --site-per-process, test was finishing before MixedContentChecker
had a chance to process the EventSource from the test.  After the changes
the test finishes only after EventSource-specific event fires.

I have checked and AFAICT frame-with-insecure-eventsource.html is only
used in this specific test (i.e. changes to the timing of the postMessage
will only impact the insecure-eventsource-in-main-frame.html test).

BUG=477150
Committed: https://crrev.com/013afc75fb8e42c5bd50cfec0007f90b84eb423a
Cr-Commit-Position: refs/heads/master@{#391247}

[Łukasz Anforowicz, 2016-04-29 21:09:46]

lukasza@chromium.org changed reviewers:
+ dcheng@chromium.org

[Łukasz Anforowicz, 2016-04-29 21:09:46]

Daniel, PTAL?  This is another case where the test finishing was racing with the
behavior under test.

[dcheng, 2016-04-29 22:41:22]

Was this flaky at all outside --site-per-process? Seems like it ought to be
flaky either way...

https://codereview.chromium.org/1940593002/diff/20001/third_party/WebKit/Layo...
File
third_party/WebKit/LayoutTests/http/tests/security/mixedContent/insecure-eventsource-in-main-frame-expected.txt
(right):

https://codereview.chromium.org/1940593002/diff/20001/third_party/WebKit/Layo...
third_party/WebKit/LayoutTests/http/tests/security/mixedContent/insecure-eventsource-in-main-frame-expected.txt:2:
CONSOLE ERROR: EventSource cannot load
http://127.0.0.1:8000/eventsource/resources/simple-event-stream.asis. No
'Access-Control-Allow-Origin' header is present on the requested resource.
Origin 'https://127.0.0.1:8443' is therefore not allowed access.
Hmm, this doesn't really seem like part of the test. Maybe we should be setting
this header?

[Łukasz Anforowicz, 2016-04-29 23:54:49]

PTAL?

On 2016/04/29 22:41:22, dcheng wrote:
> Was this flaky at all outside --site-per-process? Seems like it ought to be
> flaky either way...

Hmmm... so when I run this locally, on Goobuntu, with:
dcheck_always_on = true
is_asan = true
is_component_build = true
is_debug = false
use_goma = true

then:
- with --site-per-process it consistently fails (and according to my debugging,
MixedContentChecker *is* getting involved, but *after* TestFinished is already
called).
- without --site-per-process the test failed 17/20 times (which is surprising
because nobody is complaining / there is no test expectation outside of
site-per-process...)

https://codereview.chromium.org/1940593002/diff/20001/third_party/WebKit/Layo...
File
third_party/WebKit/LayoutTests/http/tests/security/mixedContent/insecure-eventsource-in-main-frame-expected.txt
(right):

https://codereview.chromium.org/1940593002/diff/20001/third_party/WebKit/Layo...
third_party/WebKit/LayoutTests/http/tests/security/mixedContent/insecure-eventsource-in-main-frame-expected.txt:2:
CONSOLE ERROR: EventSource cannot load
http://127.0.0.1:8000/eventsource/resources/simple-event-stream.asis. No
'Access-Control-Allow-Origin' header is present on the requested resource.
Origin 'https://127.0.0.1:8443' is therefore not allowed access.
On 2016/04/29 22:41:22, dcheng wrote:
> Hmm, this doesn't really seem like part of the test. Maybe we should be
setting
> this header?

Done.

[dcheng, 2016-05-03 07:00:36]

lgtm

[Łukasz Anforowicz, 2016-05-03 15:01:44]

The CQ bit was checked by lukasza@chromium.org

[commit-bot: I haz the power, 2016-05-03 15:02:08]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1940593002/40001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1940593002/40001

[commit-bot: I haz the power, 2016-05-03 16:15:53]

Committed patchset #3 (id:40001)

[commit-bot: I haz the power, 2016-05-03 16:17:58]

Description was changed from

==========
Make sure the test finishes after EventSource loading has completed.

With --site-per-process, test was finishing before MixedContentChecker
had a chance to process the EventSource from the test.  After the changes
the test finishes only after EventSource-specific event fires.

I have checked and AFAICT frame-with-insecure-eventsource.html is only
used in this specific test (i.e. changes to the timing of the postMessage
will only impact the insecure-eventsource-in-main-frame.html test).

BUG=477150
==========

to

==========
Make sure the test finishes after EventSource loading has completed.

With --site-per-process, test was finishing before MixedContentChecker
had a chance to process the EventSource from the test.  After the changes
the test finishes only after EventSource-specific event fires.

I have checked and AFAICT frame-with-insecure-eventsource.html is only
used in this specific test (i.e. changes to the timing of the postMessage
will only impact the insecure-eventsource-in-main-frame.html test).

BUG=477150

Committed: https://crrev.com/013afc75fb8e42c5bd50cfec0007f90b84eb423a
Cr-Commit-Position: refs/heads/master@{#391247}
==========

[commit-bot: I haz the power, 2016-05-03 16:17:59]

Patchset 3 (id:??) landed as
https://crrev.com/013afc75fb8e42c5bd50cfec0007f90b84eb423a
Cr-Commit-Position: refs/heads/master@{#391247}