Integrating back into using the external ppapi/cpp wrappers.

Integrating back into using the external ppapi/cpp wrappers.

This should bring us close to Gary's original npapi code.

BUG=none
TEST=can connect to simple host

Committed: http://src.chromium.org/viewvc/chrome?view=rev&revision=51857

[awong, 2010-07-03 01:43:45]

This hooks up the verbatim decoder, and makes the pepper plugin actually
connect. Yay!

[hclam, 2010-07-03 01:51:20]

very nice! good work!

Alpha

2010年7月2日18:43 <ajwong@chromium.org>:

> Reviewers: garykac, Alpha,
>
> Message:
> This hooks up the verbatim decoder, and makes the pepper plugin actually
> connect. Yay!
>
> Description:
> Integrating back into using the external ppapi/cpp wrappers.
>
> Please review this at http://codereview.chromium.org/2858037/show
>
> SVN Base: git://codf21.jail.google.com/chromium.git
>
> Affected files:
>  M remoting/client/chromoting_client.h
>  M remoting/client/chromoting_client.cc
>  M remoting/client/plugin/chromoting_plugin.h
>  M remoting/client/plugin/chromoting_plugin.cc
>  M remoting/client/plugin/pepper_entrypoints.cc
>  A remoting/client/plugin/pepper_util.h
>  A remoting/client/plugin/pepper_util.cc
>  M remoting/client/plugin/pepper_view.h
>  M remoting/client/plugin/pepper_view.cc
>  M remoting/host/session_manager.cc
>  M remoting/remoting.gyp
>
>
>

[awong, 2010-07-07 18:25:53]

okay...cleaned up and ready for a review.

[garykac, 2010-07-07 21:16:45]

LGTM.  Only a few nits.

http://codereview.chromium.org/2858037/diff/11001/12001
File remoting/client/chromoting_client.cc (right):

http://codereview.chromium.org/2858037/diff/11001/12001#newcode127
remoting/client/chromoting_client.cc:127: LOG(INFO) << "Client sayd DoRepaint "
<< MessageLoop::current();
says? said?  My edit-distance metric is failing me here. ^_^

http://codereview.chromium.org/2858037/diff/11001/12004
File remoting/client/plugin/chromoting_plugin.h (right):

http://codereview.chromium.org/2858037/diff/11001/12004#newcode45
remoting/client/plugin/chromoting_plugin.h:45: // TODO(ajwong): Mimetype doesn't
really make sense for us as the trigger
This comment is no longer accurate since we need the mimetype to trigger the
internal plugin.  Also, the protocol has changed to chrome://remoting.

http://codereview.chromium.org/2858037/diff/11001/12005
File remoting/client/plugin/pepper_entrypoints.cc (right):

http://codereview.chromium.org/2858037/diff/11001/12005#newcode31
remoting/client/plugin/pepper_entrypoints.cc:31: DCHECK(MessageLoop::current()
!= NULL);
Shouldn't there always be a MessageLoop()?  When will it be NULL?

http://codereview.chromium.org/2858037/diff/11001/12007
File remoting/client/plugin/pepper_util.h (right):

http://codereview.chromium.org/2858037/diff/11001/12007#newcode16
remoting/client/plugin/pepper_util.h:16: // Function for adaptin a Chromium
style Task into a
nit: adapting (or adaptin' if you prefer)

http://codereview.chromium.org/2858037/diff/11001/12007#newcode18
remoting/client/plugin/pepper_util.h:18: // as |user_data|.  This funciton will
invoke Task::Run() on |user_data| when
s/funciton/function/

[awong, 2010-07-08 16:53:32]

All fixed. submitting.
On 2010/07/07 21:16:45, garykac wrote:
> LGTM.  Only a few nits.
> 
> http://codereview.chromium.org/2858037/diff/11001/12001
> File remoting/client/chromoting_client.cc (right):
> 
> http://codereview.chromium.org/2858037/diff/11001/12001#newcode127
> remoting/client/chromoting_client.cc:127: LOG(INFO) << "Client sayd DoRepaint
"
> << MessageLoop::current();
> says? said?  My edit-distance metric is failing me here. ^_^
> 
> http://codereview.chromium.org/2858037/diff/11001/12004
> File remoting/client/plugin/chromoting_plugin.h (right):
> 
> http://codereview.chromium.org/2858037/diff/11001/12004#newcode45
> remoting/client/plugin/chromoting_plugin.h:45: // TODO(ajwong): Mimetype
doesn't
> really make sense for us as the trigger
> This comment is no longer accurate since we need the mimetype to trigger the
> internal plugin.  Also, the protocol has changed to chrome://remoting.
> 
> http://codereview.chromium.org/2858037/diff/11001/12005
> File remoting/client/plugin/pepper_entrypoints.cc (right):
> 
> http://codereview.chromium.org/2858037/diff/11001/12005#newcode31
> remoting/client/plugin/pepper_entrypoints.cc:31: DCHECK(MessageLoop::current()
> != NULL);
> Shouldn't there always be a MessageLoop()?  When will it be NULL?
> 
> http://codereview.chromium.org/2858037/diff/11001/12007
> File remoting/client/plugin/pepper_util.h (right):
> 
> http://codereview.chromium.org/2858037/diff/11001/12007#newcode16
> remoting/client/plugin/pepper_util.h:16: // Function for adaptin a Chromium
> style Task into a
> nit: adapting (or adaptin' if you prefer)
> 
> http://codereview.chromium.org/2858037/diff/11001/12007#newcode18
> remoting/client/plugin/pepper_util.h:18: // as |user_data|.  This funciton
will
> invoke Task::Run() on |user_data| when
> s/funciton/function/