Rewrite layout_browsertests to use content_shell --dump-render-tree to execute layout tests

Rewrite layout_browsertests to use content_shell --dump-render-tree to execute layout tests

break down of changes
- refactor WebKitTestResultPrinter to print to a stream
- allow for swapping the result printer
- add testRunner.workerThreadCount implementation and testRunner.overridePreference stub
- rip out the fake testRunner and result getting / formatting code from InProcessBrowserLayoutTest
- make it use WebKitTestController to run the tests
- remove all the code for copying resources around (since we don't need to patch the tests, we can just read the original)

BUG=none
TEST=layout_browsertests passes


Committed: http://src.chromium.org/viewvc/chrome?view=rev&revision=158803

[jochen (gone - plz use gerrit), 2012-09-18 13:17:51]

plz review

[jochen (gone - plz use gerrit), 2012-09-18 18:48:26]

+dgrogan, jsbell

Looks like we need to fix the expectations for
intversion-delete-in-upgradeneeded.html and basics-shared-workers.html

I need to investigate tomorrow why intversion-close-in-oncomplete.html is timing
out

[jochen (gone - plz use gerrit), 2012-09-18 19:18:34]

Joshua rebased the test in webkit r128920

thanks

[jam, 2012-09-18 20:04:08]

https://codereview.chromium.org/10941011/diff/3001/content/browser/worker_hos...
File content/browser/worker_host/test/worker_browsertest.cc (right):

https://codereview.chromium.org/10941011/diff/3001/content/browser/worker_hos...
content/browser/worker_host/test/worker_browsertest.cc:199: //
"abort-exception-assert.html",
how did this work before?

[jochen (gone - plz use gerrit), 2012-09-18 20:06:14]

On 2012/09/18 20:04:08, John Abd-El-Malek wrote:
>
https://codereview.chromium.org/10941011/diff/3001/content/browser/worker_hos...
> File content/browser/worker_host/test/worker_browsertest.cc (right):
> 
>
https://codereview.chromium.org/10941011/diff/3001/content/browser/worker_hos...
> content/browser/worker_host/test/worker_browsertest.cc:199: //
> "abort-exception-assert.html",
> how did this work before?

Before, the workerThreadCount method just returned 0.

I could also just return zero, but it would just hide a bug in this case.

[jam, 2012-09-18 20:29:53]

lgtm

https://codereview.chromium.org/10941011/diff/3001/content/browser/worker_hos...
File content/browser/worker_host/test/worker_browsertest.cc (right):

https://codereview.chromium.org/10941011/diff/3001/content/browser/worker_hos...
content/browser/worker_host/test/worker_browsertest.cc:199: //
"abort-exception-assert.html",
On 2012/09/18 20:04:08, John Abd-El-Malek wrote:
> how did this work before?

ah, ok, lgtm

[jsbell, 2012-09-18 21:13:18]

lgtm

[commit-bot: I haz the power, 2012-09-21 21:22:01]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/jochen@chromium.org/10941011/3001

[commit-bot: I haz the power, 2012-09-22 00:46:40]

Retried try job too often for step(s) content_browsertests

[dgrogan, 2012-09-25 18:04:16]

On 2012/09/22 00:46:40, I haz the power (commit-bot) wrote:
> Retried try job too often for step(s) content_browsertests

Looks like this is just a line-endings issue where the expected text for
MessagePortTest.Tests has \r\n.

[dgrogan, 2012-09-25 18:04:31]

https://chromiumcodereview.appspot.com/10941011/diff/3001/content/test/layout...
File content/test/layout_browsertest.cc (left):

https://chromiumcodereview.appspot.com/10941011/diff/3001/content/test/layout...
content/test/layout_browsertest.cc:196:
ReplaceSubstringsAfterOffset(&actual_text, 0, "\r", "");
Looks like we now need this on the expected_text

[jochen (gone - plz use gerrit), 2012-09-25 18:06:22]

On 2012/09/25 18:04:31, dgrogan wrote:
>
https://chromiumcodereview.appspot.com/10941011/diff/3001/content/test/layout...
> File content/test/layout_browsertest.cc (left):
> 
>
https://chromiumcodereview.appspot.com/10941011/diff/3001/content/test/layout...
> content/test/layout_browsertest.cc:196:
> ReplaceSubstringsAfterOffset(&actual_text, 0, "\r", "");
> Looks like we now need this on the expected_text

I tried to reproduce this locally and couldn't.

Can you point me at which expected.txt file has \r\n? the baselines shouldn't
contain \r ever

[dgrogan, 2012-09-25 18:29:46]

> Can you point me at which expected.txt file has \r\n? the baselines shouldn't
> contain \r ever

hexdump LayoutTests/fast/events/message-channel-gc-expected.txt doesn't show any
\r\n so I bet it's something with the svn eol property. I'm going to pull down
an svn checkout to check.

[dgrogan, 2012-09-25 18:53:03]

> hexdump LayoutTests/fast/events/message-channel-gc-expected.txt doesn't show
any
> \r\n so I bet it's something with the svn eol property. I'm going to pull down
> an svn checkout to check.

Yeah, that's what's going on.
$ svn propget svn:eol-style message*txt
message-channel-gc-2-expected.txt - native
message-channel-gc-3-expected.txt - native
message-channel-gc-expected.txt - native
message-channel-listener-circular-ownership-expected.txt - native
message-port-constructor-for-deleted-document-expected.txt - native
message-port-context-destroyed-expected.txt - LF
message-port-deleted-document-expected.txt - native
message-port-deleted-frame-expected.txt - native
message-port-expected.txt - native
message-port-inactive-document-expected.txt - native

[dgrogan, 2012-09-25 19:32:13]

The properties should be fixed in wk 129539 and 129542

[commit-bot: I haz the power, 2012-09-26 07:26:30]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/jochen@chromium.org/10941011/3001

[commit-bot: I haz the power, 2012-09-26 13:14:21]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/jochen@chromium.org/10941011/3001

[commit-bot: I haz the power, 2012-09-26 15:20:09]

Change committed as 158803