Fixes for surfaces bindings exposed by wm_flow app

Fixes for surfaces bindings exposed by wm_flow app

DisplayManager wasn't handling multiply nested things correctly and the
client library wasn't handling uploading to multiple views from the same
application. This fixes those.  There's still some z-ordering issues that
show up pretty visibly. This hacks up texture alpha so you can see
partially through views to try to debug.

BUG=

Committed: https://crrev.com/048f8474044145b282f61d780817002df9137b78
Cr-Commit-Position: refs/heads/master@{#294866}

[jamesr, 2014-09-13 01:10:39]

jamesr@chromium.org changed reviewers:
+ ben@chromium.org

[jamesr, 2014-09-13 01:10:40]

This doesn't totally fix things but makes wm_flow a bit more usable.  The
z-order seems inconsistent between different apps (I can show you monday) so I'm
not sure exactly what order everything is supposed to draw in.  Maybe some apps
are embedding things backwards?

[jamesr, 2014-09-13 01:13:55]

https://codereview.chromium.org/555953007/diff/1/mojo/services/public/cpp/vie...
File mojo/services/public/cpp/view_manager/lib/bitmap_uploader.cc (right):

https://codereview.chromium.org/555953007/diff/1/mojo/services/public/cpp/vie...
mojo/services/public/cpp/view_manager/lib/bitmap_uploader.cc:157:
texture_state->vertex_opacity.push_back(0.25f * i);
don't want to land this since it makes everything very hard to see, but with it
you can see the embedded apps in wm_flow are there they're just being covered up
by something.  However the window decorations appear to be painting in the
correct order (at least they paint in an order consistent with hit testing).  So
I'm a little bit flummoxed.

[Ben Goodger (Google), 2014-09-13 01:22:22]

https://codereview.chromium.org/555953007/diff/1/mojo/services/public/cpp/vie...
File mojo/services/public/cpp/view_manager/lib/view.cc (right):

https://codereview.chromium.org/555953007/diff/1/mojo/services/public/cpp/vie...
mojo/services/public/cpp/view_manager/lib/view.cc:308: if (manager_) {
nit: no braces

https://codereview.chromium.org/555953007/diff/1/mojo/services/public/cpp/vie...
File mojo/services/public/cpp/view_manager/lib/view_manager_client_impl.cc
(right):

https://codereview.chromium.org/555953007/diff/1/mojo/services/public/cpp/vie...
mojo/services/public/cpp/view_manager/lib/view_manager_client_impl.cc:143:
delete it->second;
rather than track this here, do you think it makes sense now to move the bitmap
uploader instance to the view class?

[jamesr, 2014-09-13 04:12:57]

Oh, I probably do need to iterate backwards (since the quad list goes that
way) and do a preorder traversal to get things nesting correctly.

To unsubscribe from this group and stop receiving emails from it, send an email
to chromium-reviews+unsubscribe@chromium.org.

[Ben Goodger (Google), 2014-09-15 17:36:51]

lgtm

https://codereview.chromium.org/555953007/diff/40001/mojo/services/public/cpp...
File mojo/services/public/cpp/view_manager/lib/view.cc (right):

https://codereview.chromium.org/555953007/diff/40001/mojo/services/public/cpp...
mojo/services/public/cpp/view_manager/lib/view.cc:403: void
View::LocalCreateBitmapUploader() {
don't need the prefix "Local" ... that convention was only for methods that
apply state locally adjacent to a corresponding push to the service.

[jamesr, 2014-09-15 17:43:34]

The CQ bit was checked by jamesr@chromium.org

[commit-bot: I haz the power, 2014-09-15 18:02:11]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patchset/555953007/60001

[commit-bot: I haz the power, 2014-09-15 19:28:40]

Committed patchset #4 (id:60001) as 9cf469ee282cc9a8a9792fc12bf7eb3ab3e942e8

[commit-bot: I haz the power, 2014-09-15 19:30:54]

Patchset 4 (id:??) landed as
https://crrev.com/048f8474044145b282f61d780817002df9137b78
Cr-Commit-Position: refs/heads/master@{#294866}