First pass at implementing the MessageLoop interface. This includes a simple

First pass at implementing the MessageLoop interface. This includes a simple
example and a helper class. The current example just asserts due to thread
checks we have in there now, but this should provide a good starting point.

BUG=
TEST=

Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=119198

[brettw, 2012-01-06 23:48:11]

This depends on the API in:
http://codereview.chromium.org/8965082/

[bbudge, 2012-01-10 19:40:20]

http://codereview.chromium.org/9097006/diff/1/ppapi/examples/threading/thread...
File ppapi/examples/threading/threading.cc (right):

http://codereview.chromium.org/9097006/diff/1/ppapi/examples/threading/thread...
ppapi/examples/threading/threading.cc:1: // Copyright (c) 2011 The Chromium
Authors. All rights reserved.
Year

http://codereview.chromium.org/9097006/diff/1/ppapi/examples/threading/thread...
ppapi/examples/threading/threading.cc:5: #include
"ppapi/c/dev/ppb_console_dev.h"
Wow, this is a lot of includes. Is that the standard boilerplate for our
examples?

http://codereview.chromium.org/9097006/diff/1/ppapi/examples/threading/thread...
ppapi/examples/threading/threading.cc:32: thread_(this) {
I think Visual Studio will complain about using 'this' in initializers.

http://codereview.chromium.org/9097006/diff/1/ppapi/utility/threading/simple_...
File ppapi/utility/threading/simple_thread.h (right):

http://codereview.chromium.org/9097006/diff/1/ppapi/utility/threading/simple_...
ppapi/utility/threading/simple_thread.h:19: // PPAPI message loop on that
thread.
Comment doesn't seem right in the Windows case.

http://codereview.chromium.org/9097006/diff/1/ppapi/utility/threading/simple_...
ppapi/utility/threading/simple_thread.h:22: #ifdef WIN32
Is WIN32 our define for Windows in general?

http://codereview.chromium.org/9097006/diff/1/ppapi/utility/threading/simple_...
ppapi/utility/threading/simple_thread.h:37: // it. In some cases you will want
control over the message
period (.)
Also, you should probably document that 'func' can be NULL and why you would
want to do that. Alternatively, ASSERT if NULL is passed.

[brettw, 2012-01-18 17:53:09]

New snap up

http://codereview.chromium.org/9097006/diff/1/ppapi/examples/threading/thread...
File ppapi/examples/threading/threading.cc (right):

http://codereview.chromium.org/9097006/diff/1/ppapi/examples/threading/thread...
ppapi/examples/threading/threading.cc:5: #include
"ppapi/c/dev/ppb_console_dev.h"
Sorry, copy-n-pasted.

http://codereview.chromium.org/9097006/diff/1/ppapi/utility/threading/simple_...
File ppapi/utility/threading/simple_thread.h (right):

http://codereview.chromium.org/9097006/diff/1/ppapi/utility/threading/simple_...
ppapi/utility/threading/simple_thread.h:22: #ifdef WIN32
On 2012/01/10 19:40:20, bbudge1 wrote:
> Is WIN32 our define for Windows in general?

I think for the wrappers this is the best thing. For Chrome code there is OS_WIN
we should use instead, but this needs to be compilable by end-users.

[bbudge, 2012-01-23 18:15:12]

http://codereview.chromium.org/9097006/diff/6001/ppapi/proxy/ppb_message_loop...
File ppapi/proxy/ppb_message_loop_proxy.cc (right):

http://codereview.chromium.org/9097006/diff/6001/ppapi/proxy/ppb_message_loop...
ppapi/proxy/ppb_message_loop_proxy.cc:43: int64_t delay_ms);
OVERRIDE?

http://codereview.chromium.org/9097006/diff/6001/ppapi/proxy/ppb_message_loop...
ppapi/proxy/ppb_message_loop_proxy.cc:151: return PP_ERROR_BADARGUMENT;
Need to check destroyed_ here.

[brettw, 2012-01-24 05:02:17]

PTAL

[bbudge, 2012-01-25 00:35:26]

A nit and question, otherwise LGTM.

http://codereview.chromium.org/9097006/diff/15001/ppapi/examples/threading/th...
File ppapi/examples/threading/threading.cc (right):

http://codereview.chromium.org/9097006/diff/15001/ppapi/examples/threading/th...
ppapi/examples/threading/threading.cc:17:
RequestInputEvents(PP_INPUTEVENT_CLASS_MOUSE);
Would it work if this was moved to Init()? It seems out of place here.

http://codereview.chromium.org/9097006/diff/15001/ppapi/proxy/ppb_message_loo...
File ppapi/proxy/ppb_message_loop_proxy.cc (right):

http://codereview.chromium.org/9097006/diff/15001/ppapi/proxy/ppb_message_loo...
ppapi/proxy/ppb_message_loop_proxy.cc:173: // Note that the message loop must be
destroyed on the thread is was created.
is -> it, created -> created on