Create a ResourceMessageFilter to filter resource related IPCs. This gets ri...

Create a ResourceMessageFilter to filter resource related IPCs.  This gets rid of the awkward ResourceDispatcherHost::Receiver interface and allows a bunch of cleanup.  I've also generalized the filtering done in WorkerProcessHost and moved it to ChildProcessHost (since it's now used to add the ResourceMessageFilter).

Committed: http://src.chromium.org/viewvc/chrome?view=rev&revision=69335

[brettw, 2010-12-15 21:31:10]

LGTM, this is much nicer.

http://codereview.chromium.org/5874002/diff/58002/chrome/browser/plugin_proce...
File chrome/browser/plugin_process_host.cc (right):

http://codereview.chromium.org/5874002/diff/58002/chrome/browser/plugin_proce...
chrome/browser/plugin_process_host.cc:72: request_id);
Indent 2 more spaces.

http://codereview.chromium.org/5874002/diff/58002/chrome/browser/renderer_hos...
File chrome/browser/renderer_host/browser_render_process_host.cc (right):

http://codereview.chromium.org/5874002/diff/58002/chrome/browser/renderer_hos...
chrome/browser/renderer_host/browser_render_process_host.cc:224: namespace {
Usually we'd put anonymous namespace first. In this case, it seems like most of
the stuff previously in this file can be in this anonymous namespace. Can you
mess around with that?

http://codereview.chromium.org/5874002/diff/58002/chrome/browser/renderer_hos...
chrome/browser/renderer_host/browser_render_process_host.cc:232: :
request_context_(profile->GetRequestContext()),
2 more spaces indent.

http://codereview.chromium.org/5874002/diff/58002/chrome/browser/renderer_hos...
File chrome/browser/renderer_host/resource_dispatcher_host_unittest.cc (right):

http://codereview.chromium.org/5874002/diff/58002/chrome/browser/renderer_hos...
chrome/browser/renderer_host/resource_dispatcher_host_unittest.cc:406: // the
host delegate acts as a second one so we can have some requests
Style nit: can you capitalize this?

http://codereview.chromium.org/5874002/diff/58002/chrome/browser/renderer_hos...
chrome/browser/renderer_host/resource_dispatcher_host_unittest.cc:412:
has_canceled_(false), received_after_canceled_(0) {
Style nit: can you move the last initializer to its own line?

http://codereview.chromium.org/5874002/diff/58002/chrome/browser/renderer_hos...
File chrome/browser/renderer_host/resource_message_filter.cc (right):

http://codereview.chromium.org/5874002/diff/58002/chrome/browser/renderer_hos...
chrome/browser/renderer_host/resource_message_filter.cc:32: bool
ResourceMessageFilter::OnMessageReceived(const IPC::Message& message,
Style nit: blank line above this.

http://codereview.chromium.org/5874002/diff/58002/chrome/common/child_process...
File chrome/common/child_process_host.h (right):

http://codereview.chromium.org/5874002/diff/58002/chrome/common/child_process...
chrome/common/child_process_host.h:67: void
AddFilter(IPC::ChannelProxy::MessageFilter* filter);
Can you make explicit that this takes a reference to the given filter?

[jam, 2010-12-15 21:41:04]

http://codereview.chromium.org/5874002/diff/58002/chrome/browser/plugin_proce...
File chrome/browser/plugin_process_host.cc (right):

http://codereview.chromium.org/5874002/diff/58002/chrome/browser/plugin_proce...
chrome/browser/plugin_process_host.cc:72: request_id);
On 2010/12/15 21:31:10, brettw wrote:
> Indent 2 more spaces.

Done.

http://codereview.chromium.org/5874002/diff/58002/chrome/browser/renderer_hos...
File chrome/browser/renderer_host/browser_render_process_host.cc (right):

http://codereview.chromium.org/5874002/diff/58002/chrome/browser/renderer_hos...
chrome/browser/renderer_host/browser_render_process_host.cc:224: namespace {
On 2010/12/15 21:31:10, brettw wrote:
> Usually we'd put anonymous namespace first. In this case, it seems like most
of
> the stuff previously in this file can be in this anonymous namespace. Can you
> mess around with that?

I had tried that, but those classes are forward declared in the header and used
in scoped_ptrs, so it can't be done

http://codereview.chromium.org/5874002/diff/58002/chrome/browser/renderer_hos...
chrome/browser/renderer_host/browser_render_process_host.cc:232: :
request_context_(profile->GetRequestContext()),
On 2010/12/15 21:31:10, brettw wrote:
> 2 more spaces indent.

Done.

http://codereview.chromium.org/5874002/diff/58002/chrome/browser/renderer_hos...
File chrome/browser/renderer_host/resource_dispatcher_host_unittest.cc (right):

http://codereview.chromium.org/5874002/diff/58002/chrome/browser/renderer_hos...
chrome/browser/renderer_host/resource_dispatcher_host_unittest.cc:406: // the
host delegate acts as a second one so we can have some requests
On 2010/12/15 21:31:10, brettw wrote:
> Style nit: can you capitalize this?

Done.

http://codereview.chromium.org/5874002/diff/58002/chrome/browser/renderer_hos...
chrome/browser/renderer_host/resource_dispatcher_host_unittest.cc:412:
has_canceled_(false), received_after_canceled_(0) {
On 2010/12/15 21:31:10, brettw wrote:
> Style nit: can you move the last initializer to its own line?

Done.

http://codereview.chromium.org/5874002/diff/58002/chrome/browser/renderer_hos...
File chrome/browser/renderer_host/resource_message_filter.cc (right):

http://codereview.chromium.org/5874002/diff/58002/chrome/browser/renderer_hos...
chrome/browser/renderer_host/resource_message_filter.cc:32: bool
ResourceMessageFilter::OnMessageReceived(const IPC::Message& message,
On 2010/12/15 21:31:10, brettw wrote:
> Style nit: blank line above this.

Done.

http://codereview.chromium.org/5874002/diff/58002/chrome/common/child_process...
File chrome/common/child_process_host.h (right):

http://codereview.chromium.org/5874002/diff/58002/chrome/common/child_process...
chrome/common/child_process_host.h:67: void
AddFilter(IPC::ChannelProxy::MessageFilter* filter);
On 2010/12/15 21:31:10, brettw wrote:
> Can you make explicit that this takes a reference to the given filter?

Done.