Event dispatch to OOPIFs doesn't work with mus |
||||||||||
Issue descriptionIt would appear some key code does nothing in RenderWidgetHostViewAura that is needed for event dispatch. In particular CreateDelegatedFrameHostClient() early outs for mus. And similarly RenderWidgetHostViewChildFrame does not register the frame sink id in SetFrameSinkId(). The latter is an easy fix, but not enough. I'm not familiar enough with this code to know how it needs to be wired up for mus. If someone can give me an overview, I can attempt it. At one point this was working, but lack of test coverage mean it no longer does:( This also relied on the browser seeing the whole server window hierarchy at one point. That is no longer wired up for current mus implementation. I was in the process of implementing it for this, when I realized event dispatch doesn't work at all. That bug is 866628. I'm not sure it's required anymore though. Will figure that out once I understand how all this should work. ⛆ |
|
|
,
Sep 6
,
Sep 7
,
Sep 21
,
Sep 24
,
Sep 28
The following revision refers to this bug: https://chromium.googlesource.com/chromium/src.git/+/d28ee327131cee5c2f0b2d0420f73ca34d22db4c commit d28ee327131cee5c2f0b2d0420f73ca34d22db4c Author: Scott Violet <sky@chromium.org> Date: Fri Sep 28 03:02:14 2018 aura: makes WindowPortMus use HitTestDataProviderDrawQuad This works better for hit testing with mash. BUG=879791 TEST=none Change-Id: Ie87e72933f80a77955f301da2195934109d7f569 Reviewed-on: https://chromium-review.googlesource.com/1250123 Reviewed-by: Sadrul Chowdhury <sadrul@chromium.org> Commit-Queue: Scott Violet <sky@chromium.org> Cr-Commit-Position: refs/heads/master@{#594968} [modify] https://crrev.com/d28ee327131cee5c2f0b2d0420f73ca34d22db4c/ui/aura/mus/window_port_mus.cc
,
Oct 1
The following revision refers to this bug: https://chromium.googlesource.com/chromium/src.git/+/8cb05f56ee91da91a3cb08945ddbbc89d1270ee4 commit 8cb05f56ee91da91a3cb08945ddbbc89d1270ee4 Author: Scott Violet <sky@chromium.org> Date: Mon Oct 01 19:46:35 2018 content: gets RWHVA::GetRootFrameSinkId working with single-process mash This adds HostFrameSinkManager::FindRootFrameSinkId() to find the root frame sink id. This is necessary as window->GetHost()->compositor()->frame_sink_id() is not the root frame sink in mash. BUG=879791 TEST=none Cq-Include-Trybots: luci.chromium.try:android_optional_gpu_tests_rel Change-Id: I5a1520b2ba144cb231d254059810ff153add3a49 Reviewed-on: https://chromium-review.googlesource.com/1249152 Commit-Queue: Scott Violet <sky@chromium.org> Reviewed-by: kylechar <kylechar@chromium.org> Reviewed-by: Sadrul Chowdhury <sadrul@chromium.org> Cr-Commit-Position: refs/heads/master@{#595512} [modify] https://crrev.com/8cb05f56ee91da91a3cb08945ddbbc89d1270ee4/components/viz/host/host_frame_sink_manager.cc [modify] https://crrev.com/8cb05f56ee91da91a3cb08945ddbbc89d1270ee4/components/viz/host/host_frame_sink_manager.h [modify] https://crrev.com/8cb05f56ee91da91a3cb08945ddbbc89d1270ee4/components/viz/host/host_frame_sink_manager_unittest.cc [modify] https://crrev.com/8cb05f56ee91da91a3cb08945ddbbc89d1270ee4/content/browser/renderer_host/render_widget_host_view_aura.cc
,
Oct 1
This is now working for single-process-mash. I'm going to retarget to multi-process-mash as it still doesn't work there.
,
Oct 31
,
Yesterday
(35 hours ago)
The following revision refers to this bug: https://chromium.googlesource.com/chromium/src.git/+/b4f6394dcd59d8697e154c2e2c74eb85284f2e4b commit b4f6394dcd59d8697e154c2e2c74eb85284f2e4b Author: Mike Wasserman <msw@chromium.org> Date: Mon Jan 21 22:38:35 2019 mash: content: changes RenderWidgetHostInputEventRouter hit test querying Rebase of sky@'s < http://crrev.com/c/1239454 >. Also, make FindTargetForLocationStartingFromImpl virtual for testing. Original description: ===================================================================== Specifically this converts from starting the hit testing from the root to hit testing from the frame-sink of the RenderWidgetHost. As the event is already received by the RenderWidgetHost hit testing can start from there, rather than the root. This matters in the mash case because of hot hit testing works. ===================================================================== TODO: After this change, we may be able to revert http://crrev.com/c/1249152 (sky@ suggested this should be possible, but it breaks SPM iframe targeting afaict) Bug: 879791, 920931 Test: SingleProcessMash web content (w/iframe) hit testing works with a11y magnifier. Change-Id: Ie6c53ecfdba9444c62a9b3e3a7f573f0ec27ce6b Reviewed-on: https://chromium-review.googlesource.com/c/1411650 Reviewed-by: Sadrul Chowdhury <sadrul@chromium.org> Reviewed-by: Ria Jiang <riajiang@chromium.org> Commit-Queue: Michael Wasserman <msw@chromium.org> Cr-Commit-Position: refs/heads/master@{#624687} [modify] https://crrev.com/b4f6394dcd59d8697e154c2e2c74eb85284f2e4b/components/viz/host/hit_test/hit_test_query.cc [modify] https://crrev.com/b4f6394dcd59d8697e154c2e2c74eb85284f2e4b/components/viz/host/hit_test/hit_test_query.h [modify] https://crrev.com/b4f6394dcd59d8697e154c2e2c74eb85284f2e4b/components/viz/host/hit_test/hit_test_query_unittest.cc [modify] https://crrev.com/b4f6394dcd59d8697e154c2e2c74eb85284f2e4b/content/browser/renderer_host/render_widget_host_input_event_router.cc [modify] https://crrev.com/b4f6394dcd59d8697e154c2e2c74eb85284f2e4b/content/browser/renderer_host/render_widget_host_input_event_router_unittest.cc |
|||||||
►
Sign in to add a comment |
||||||||||
Comment 1 by dtapu...@chromium.org
, Sep 4