Move ModuleInitializer::init out of initializeWithoutV8

Move ModuleInitializer::init out of initializeWithoutV8

It looks like that the call sites of initializeWithoutV8 doesn't need to
initialize the HTML parser thread. We can factor out ModuleInitializer.init()
from initializeWithoutV8 to initialize. With this CL, we can get rid of the dependency
from initializeWithoutV8 to core/ and modules/. I'll move initializeWithoutV8/shutdownWithoutV8
from web/ to platform/ in the next CL.

BUG=

Committed: https://crrev.com/728a85b9f9fc4e640cf4829a760a0cbc6373147e
Cr-Commit-Position: refs/heads/master@{#379785}

[haraken, 2016-03-08 04:47:49]

Description was changed from

==========
Move ModuleInitializer::init out of initializeWithoutV8

BUG=
==========

to

==========
Move ModuleInitializer::init out of initializeWithoutV8

It looks like that the call sites of initializeWithoutV8 doesn't need to
initialize the HTML parser thread. We can factor out ModuleInitializer.init()
from initializeWithoutV8 to initialize. With this CL, we can get rid of the
dependency
from initializeWithoutV8 to core/ and modules/. I'll move
initializeWithoutV8/shutdownWithoutV8
from web/ to platform/ in the next CL.

BUG=
==========

[haraken, 2016-03-08 04:47:49]

haraken@chromium.org changed reviewers:
+ kinuko@chromium.org, yhirano@chromium.org

[haraken, 2016-03-08 04:48:40]

PTAL

Sorry, I was wrong -- it looks like the call sites of initializeWithoutV8
doesn't need a HTML parser thread.

[haraken, 2016-03-08 04:48:52]

haraken@chromium.org changed reviewers:
+ sigbjornf@opera.com

[sof, 2016-03-08 06:31:27]

lgtm (but no web/ owner)

I guess that raises the question if detachMainThread() can move out also.

[haraken, 2016-03-08 06:36:11]

> I guess that raises the question if detachMainThread() can move out also.

Maybe or may not be. But here I'm more interested in breaking the dependency
from initializeWithoutV8 to core/ and modules/ so that we can move the function
into platform/ (then I'll rename it to initializePlatformAndWTF).

initializePlatformAndWTF will be used by Mojo tests:
https://codereview.chromium.org/1751563002/

[yhirano, 2016-03-08 07:05:24]

lgtm

https://codereview.chromium.org/1775563004/diff/20001/third_party/WebKit/Sour...
File third_party/WebKit/Source/web/WebKit.cpp (right):

https://codereview.chromium.org/1775563004/diff/20001/third_party/WebKit/Sour...
third_party/WebKit/Source/web/WebKit.cpp:107:
setIndexedDBClientCreateFunction(IndexedDBClientImpl::create);
[optional] Can we move this function call to ModulesInitializer?

[haraken, 2016-03-08 07:17:32]

>
https://codereview.chromium.org/1775563004/diff/20001/third_party/WebKit/Sour...
> File third_party/WebKit/Source/web/WebKit.cpp (right):
> 
>
https://codereview.chromium.org/1775563004/diff/20001/third_party/WebKit/Sour...
> third_party/WebKit/Source/web/WebKit.cpp:107:
> setIndexedDBClientCreateFunction(IndexedDBClientImpl::create);
> [optional] Can we move this function call to ModulesInitializer?

This is not that trivial, since IndexedDBClientImpl::create is in web/. I'll
consider in a follow-up CL.

[haraken, 2016-03-08 07:17:43]

The CQ bit was checked by haraken@chromium.org

[commit-bot: I haz the power, 2016-03-08 07:17:57]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1775563004/20001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1775563004/20001

[commit-bot: I haz the power, 2016-03-08 08:22:21]

Description was changed from

==========
Move ModuleInitializer::init out of initializeWithoutV8

It looks like that the call sites of initializeWithoutV8 doesn't need to
initialize the HTML parser thread. We can factor out ModuleInitializer.init()
from initializeWithoutV8 to initialize. With this CL, we can get rid of the
dependency
from initializeWithoutV8 to core/ and modules/. I'll move
initializeWithoutV8/shutdownWithoutV8
from web/ to platform/ in the next CL.

BUG=
==========

to

==========
Move ModuleInitializer::init out of initializeWithoutV8

It looks like that the call sites of initializeWithoutV8 doesn't need to
initialize the HTML parser thread. We can factor out ModuleInitializer.init()
from initializeWithoutV8 to initialize. With this CL, we can get rid of the
dependency
from initializeWithoutV8 to core/ and modules/. I'll move
initializeWithoutV8/shutdownWithoutV8
from web/ to platform/ in the next CL.

BUG=
==========

[commit-bot: I haz the power, 2016-03-08 08:22:22]

Committed patchset #2 (id:20001)

[commit-bot: I haz the power, 2016-03-08 08:23:46]

Description was changed from

==========
Move ModuleInitializer::init out of initializeWithoutV8

It looks like that the call sites of initializeWithoutV8 doesn't need to
initialize the HTML parser thread. We can factor out ModuleInitializer.init()
from initializeWithoutV8 to initialize. With this CL, we can get rid of the
dependency
from initializeWithoutV8 to core/ and modules/. I'll move
initializeWithoutV8/shutdownWithoutV8
from web/ to platform/ in the next CL.

BUG=
==========

to

==========
Move ModuleInitializer::init out of initializeWithoutV8

It looks like that the call sites of initializeWithoutV8 doesn't need to
initialize the HTML parser thread. We can factor out ModuleInitializer.init()
from initializeWithoutV8 to initialize. With this CL, we can get rid of the
dependency
from initializeWithoutV8 to core/ and modules/. I'll move
initializeWithoutV8/shutdownWithoutV8
from web/ to platform/ in the next CL.

BUG=

Committed: https://crrev.com/728a85b9f9fc4e640cf4829a760a0cbc6373147e
Cr-Commit-Position: refs/heads/master@{#379785}
==========

[commit-bot: I haz the power, 2016-03-08 08:23:47]

Patchset 2 (id:??) landed as
https://crrev.com/728a85b9f9fc4e640cf4829a760a0cbc6373147e
Cr-Commit-Position: refs/heads/master@{#379785}