[draft] flush visual state prototype

flush visual state prototype

BUG=

[mkosiba (inactive), 2014-12-09 18:28:09]

mkosiba@chromium.org changed reviewers:
+ boliu@chromium.org, nduca@chromium.org

[mkosiba (inactive), 2014-12-09 18:35:59]

here is how I'd envision the hookup for the API to look like.

https://codereview.chromium.org/240163005/ implements RenderWidget::QueueMessage
which is where the fun is.

[mnaganov (inactive), 2014-12-10 10:39:54]

mnaganov@chromium.org changed reviewers:
+ mnaganov@chromium.org

[mnaganov (inactive), 2014-12-10 10:39:54]

https://codereview.chromium.org/784343002/diff/1/content/browser/frame_host/r...
File content/browser/frame_host/render_frame_host_impl.cc (right):

https://codereview.chromium.org/784343002/diff/1/content/browser/frame_host/r...
content/browser/frame_host/render_frame_host_impl.cc:179: for(const auto& iter:
flush_visual_state_callbacks_) {
Considering the recent issue we had with Java Bridge, when people block the UI
thread until they receive a signal via an injected object (which means that
something on their page is "ready"). Do you think people will want to do the
same with VisualStateListener?

It looks like it's hard to detect such a deadlock from our code. Will we only
rely on the docs explaining how to use it correctly?

[mkosiba (inactive), 2014-12-10 15:25:53]

mkosiba@chromium.org changed reviewers:
+ igsolla@chromium.org

[mkosiba (inactive), 2014-12-10 15:27:18]

mkosiba@chromium.org changed reviewers:
+ vmpstr@chromium.org

[Ignacio Solla, 2015-01-12 13:32:47]

On 2014/12/10 15:27:19, mkosiba wrote:


Taking over this change here: https://codereview.chromium.org/831903004/