Skip to content

GitLab

T
tangled
Switch branch/tag
  1. 05 Oct, 2017 4 commits
    • Kevin Bailey's avatar
      Revert "[omnibox] Change ACMatches from vector to list" · 0dd59492
      Kevin Bailey authored 
      This reverts commit 3aae0039.

Reason for revert: Increase in crashes on Android, surrounding 'AutocompleteController::DeleteMatch'. crbug.com/771603

Original change's description:
> [omnibox] Change ACMatches from vector to list
> 
> The bulk of heavy work done on ACMatches is std::sort() and
> std::rotate.  This process takes a noticeable amount of time, and it's
> run several times per user keystroke.  Making ACMatches a list instead
> of a vector means that elements can be moved around trivially, instead
> of being copied.  The execution time for the "long poles" goes down to
> a tenth of the previous time.
> 
> Bug: 
> Change-Id: Id663a58c05e39f7a2310d97cbccaa679fc3d1b70
> Reviewed-on: https://chromium-review.googlesource.com/667436
> Commit-Queue: Kevin Bailey <krb@chromium.org>
> Reviewed-by: Peter Kasting <pkasting@chromium.org>
> Cr-Commit-Position: refs/heads/master@{#503015}

TBR=pkasting@chromium.org,krb@chromium.org

# Not skipping CQ checks because original CL landed > 1 day ago.

Change-Id: If4b0eee9705613d57b8a743ca693fff4e4e0cc00
Reviewed-on: https://chromium-review.googlesource.com/700854
Commit-Queue: Kevin Bailey <krb@chromium.org>
Reviewed-by: default avatarKevin Bailey <krb@chromium.org>
Reviewed-by: default avatarJustin Donnelly <jdonnelly@chromium.org>
Reviewed-by: default avatarMark Pearson <mpearson@chromium.org>
Cr-Commit-Position: refs/heads/master@{#506577}
      0dd59492
    • Avi Drissman's avatar
      Make SigninViewControllerDelegate unregister itself as modal dialog delegate. · 0ff0ce07
      Avi Drissman authored 
      BUG=771142
TEST=crashes go away

Change-Id: I6bb1c0986bc16aba253129c57db08eaefc4cc83a
Reviewed-on: https://chromium-review.googlesource.com/698666Reviewed-by: default avatarMike Wittman <wittman@chromium.org>
Reviewed-by: default avatarScott Violet <sky@chromium.org>
Commit-Queue: Avi Drissman <avi@chromium.org>
Cr-Commit-Position: refs/heads/master@{#506576}
      0ff0ce07
    • Trent Apted's avatar
      Include the NSWindow on events synthesized from blink::WebInputEvent for guest views · f8a0c12d
      Trent Apted authored 
      RenderWidgetHostViewGuest started generating synthetic keyboard NSEvents
from blink::WebInputEvent in r489154. These events didn't include an
NSWindow which meant that a fallback path was triggered in
-[CommandDispatcher redispatchKeyEvent] which caused a second NSEvent to
be synthesized, to add the NSWindow back on.

This "worked" for Cocoa (NativeAppWindowCocoa), but not for views-
backed guest views, causing WebViewInteractiveTest.EditCommands/0 to
regress and WebViewInteractiveTest.TextSelection/1 to time out.

Fix those by passing GetNativeView() to the NativeWebKeyboardEvent
constructor in RenderWidgetHostViewGuest::OnHandleInputEvent(). This
also avoids triggering the fallback path in CommandDispatcher, so only
one synthetic NSEvent is created.

Bug: 770547
Cq-Include-Trybots: master.tryserver.chromium.linux:linux_site_isolation
Change-Id: I50ded74f7563f755dd6d735750437bbbafff39e2
Reviewed-on: https://chromium-review.googlesource.com/700174Reviewed-by: default avatarDave Tapuska <dtapuska@chromium.org>
Reviewed-by: default avatarAvi Drissman <avi@chromium.org>
Commit-Queue: Avi Drissman <avi@chromium.org>
Cr-Commit-Position: refs/heads/master@{#506575}
      f8a0c12d
    • Greg Thompson's avatar
      Deflake ui_controls::SendMouseMove on Windows. · 522146e8
      Greg Thompson authored 
      On Windows, SendMouseMove must transform a point in screen coordinates
into the range (0..2^16-1) for consumption by SendInput. Windows then
converts this back to screen coordinates. This change rounds up rather
than down in the initial conversion so that Windows lands on the correct
point when it seemly truncates.

Previously, we'd see off-by-one errors from time to time. While this may
seem insignificant at first blush, it's problematic when SendMouseMove
is called to move the mouse to a point where it already resides. No
events are generated by SendInput in this case, so the hook proc will
never be called. This shouldn't be a problem since SendMouseMove detects
this case and short-circuits the request. With the ob1, however,
SendMouseMove thinks it needs to move the mouse by one point, yet the
bad math results in a call to SendInput to move the mouse to its current
position. This makes tests timeout and all of us sad.

BUG=390564,639940,710436,726659,728488,764415
R=mathp@chromium.org, sky@chromium.org

Change-Id: I1daebe8281db0a799ef7266c3c1765de65580f03
Reviewed-on: https://chromium-review.googlesource.com/686898Reviewed-by: default avatarScott Violet <sky@chromium.org>
Reviewed-by: default avatarMathieu Perreault <mathp@chromium.org>
Commit-Queue: Greg Thompson <grt@chromium.org>
Cr-Commit-Position: refs/heads/master@{#506574}
      522146e8
  3. 04 Oct, 2017 36 commits