Ehsan Karamad 800aec8a7e Bug 1466260 [wpt PR 11309] - Revert "Cleanup plugin element frames when the layout tree is detached", a=testonly ... Automatic update from web-platform-testsRevert "Cleanup plugin element frames when the layout tree is detached" This reverts commit 596e6b3281a8cc2c4b646553e4cde169a29a1d1d. Reason for revert: DetachLayoutTree could be called at inopportune times such as style recalc or during layout. This caused several crashes. Original change's description: > Cleanup plugin element frames when the layout tree is detached > > Unlike frames, plugin elements (<embed> and <object>) are updated > when their layout tree is rebuilt. However, due to an oversight, > <embed> or <object> elements displaying subframes would only have > the associated content view (FrameView) torn down, leaving a > dangling content frame. > > This led to some interesting side effects: > - When transitioning from a local frame to a plugin, the content > frame would remain live, with JS still happily running. > - When transitioning from a remote frame to a plugin, > - When navigating a remote frame from one URL to another, the > element would stop updating since the FrameView would be torn > down but a new one would never be created. Note that this was > not (as much of) a problem for local frames, since local frames > re-create the LocalFrameView on every navigation. > With this CL, if a plugin element has an associated content frame, > use that to clean up the state associated with the element when the > layout tree is detached. This can cause the browser context to be > re-created: this matches the behavior of Edge and Firefox. > > Note that there is still one edge case where Blink behaves oddly: > if the navigation fails and should display fallback content, the > content frame still remains attached. This will be addressed in > followups. > > Link to whatwg issue: https://github.com/whatwg/html/issues/3576 > > Bug: 776510, 781880 > Cq-Include-Trybots: master.tryserver.chromium.linux:linux_layout_tests_layout_ng > Change-Id: Id18605fbe602ea6c0076c1d579345cdcf28cc984 > Reviewed-on: https://chromium-review.googlesource.com/996314 > Commit-Queue: Ehsan Karamad <ekaramad@chromium.org> > Reviewed-by: Daniel Cheng <dcheng@chromium.org> > Reviewed-by: Alex Moshchuk <alexmos@chromium.org> > Cr-Commit-Position: refs/heads/master@{#561768} TBR=dcheng@chromium.org,creis@chromium.org,alexmos@chromium.org,ekaramad@chromium.org # Not skipping CQ checks because original CL landed > 1 day ago. Bug: 776510, 781880, 847216, 846708 Change-Id: I762631b92463352bea9bf3102d90a13b72776786 Cq-Include-Trybots: master.tryserver.chromium.linux:linux_layout_tests_layout_ng Reviewed-on: https://chromium-review.googlesource.com/1083500 Reviewed-by: Daniel Cheng <dcheng@chromium.org> Reviewed-by: Alex Moshchuk <alexmos@chromium.org> Reviewed-by: Ehsan Karamad <ekaramad@chromium.org> Commit-Queue: Ehsan Karamad <ekaramad@chromium.org> Cr-Commit-Position: refs/heads/master@{#563902} -- wpt-commits: 69a48d90cac792566f14d734d2bae4feda462829 wpt-pr: 11309		2018-06-25 21:21:04 +00:00
..
should-load.html
should-not-load.html