Skip to content

GitLab

T
tangled
Switch branch/tag
  1. 18 Jun, 2018 34 commits
  3. 17 Jun, 2018 6 commits
    • Darren Shen's avatar
      [VK] Remove unnecessary observers of OnVirtualKeyboardStateChanged. · 6017edbb
      Darren Shen authored 
      Some classes listen to ShellObserver::OnVirtualKeyboardStateChanged in
order to add/remove KeyboardController observer when the keyboard is
enabled/disabled. However, with the simplified KeyboardController
lifetimes, we can add/remove observers whenever.

So we just add an observer in the constructor and remove it in the
destructor.

After this change, there will be no more observers of
OnVirtualKeyboardStateChanged.

Bug: 731537
Change-Id: I7274484229ff5f64ac973806e565f178a271094f
Reviewed-on: https://chromium-review.googlesource.com/1101598Reviewed-by: default avatarJames Cook <jamescook@chromium.org>
Commit-Queue: Darren Shen <shend@chromium.org>
Cr-Commit-Position: refs/heads/master@{#567918}
      6017edbb
    • Dirk Pranke's avatar
      Mark two more layout tests as failing on 10.13 debug. · e4973655
      Dirk Pranke authored 
      TBR=robertma@chromium.org
BUG=853360
NOTRY=true

Change-Id: Ic5dac923094175223964c999b74f41f82273f63f
Reviewed-on: https://chromium-review.googlesource.com/1103839
Commit-Queue: Dirk Pranke <dpranke@chromium.org>
Reviewed-by: default avatarDirk Pranke <dpranke@chromium.org>
Cr-Commit-Position: refs/heads/master@{#567917}
      e4973655
    • Dirk Pranke's avatar
      Disable browser_tests on the mac. · 57c11a94
      Dirk Pranke authored 
      Something in one of the browser_tests suites seems to be killing
machines across all Mac versions. This CL disables browser_tests,
viz_browser_tests, and surface_sync_browser_tests across all Mac
versions until we can better figure out which tests are safe and
which aren't.

Ideally we'd split some of the bots into a different pool and
just run the browser_tests in that pool so that they don't take
out the rest of the fleet, and I'll work on that next.

TBR=jbudorick@chromium.org
BUG=828051
NOTRY=true

Change-Id: I133fb9e4a3db85fce2231d0e3f906ceecece122d
Reviewed-on: https://chromium-review.googlesource.com/1103793
Commit-Queue: Dirk Pranke <dpranke@chromium.org>
Reviewed-by: default avatarDirk Pranke <dpranke@chromium.org>
Cr-Commit-Position: refs/heads/master@{#567916}
      57c11a94
    • Dirk Pranke's avatar
      Suppress more layout test failures on the Mac builders. · f3e44fd1
      Dirk Pranke authored 
      As part of cleaning up tryserver.blink and chromium.webkit I need
to green up all of the Mac builders. I think most of these failures
are either flakes or need rebaselines, but I'll suppress them first
and then work through them.

TBR=robertma@chromium.org
NOTRY=true
BUG=853360

Change-Id: Ib15695a2c7e6f0e077fe5de08b2802713e51d6d3
Reviewed-on: https://chromium-review.googlesource.com/1103714
Commit-Queue: Dirk Pranke <dpranke@chromium.org>
Reviewed-by: default avatarDirk Pranke <dpranke@chromium.org>
Cr-Commit-Position: refs/heads/master@{#567915}
      f3e44fd1
    • Abhishek Arya's avatar
      Revert "Floats and out-of-flow objects may not be adjacent to anonymous blocks." · e9d1ed95
      Abhishek Arya authored 
      This reverts commit 2394d0ac.

Reason for revert: Caused ClusterFuzz regressions - 853522, 853537, 853538, 853540.

Original change's description:
> Floats and out-of-flow objects may not be adjacent to anonymous blocks.
> 
> Floats and out-of-flow objects need to be true layout siblings of the
> inlines, or rendering will be wrong. This means that such objects should
> never be siblings of anonymous blocks, but rather inside them. This
> already works correctly for initial layout tree building, and also for
> many DOM manipulations. However, code was missing to satisfy this
> requirement if we removed a regular block that was a sibling of an
> anonymous block and either a float or out-of-flow positioned object.
> 
> This even caused a crash triggered by ruby code, which ended up mixing
> inline and block children within the same container. That is not
> allowed. This happened in the MoveAllChildrenIncludingFloatsTo() call
> inside LayoutRubyBase::MoveBlockChildren(). Added a DCHECK to
> MoveAllChildrenIncludingFloatsTo() (which could fail prior to this fix);
> When moving children from one container to another, either both or none
> of the containers must have inline children.
> 
> Bug: 852640
> Change-Id: I51d3de12c73ddd07d6b4c1aa55221b4f92359ca7
> Reviewed-on: https://chromium-review.googlesource.com/1102690
> Reviewed-by: Emil A Eklund <eae@chromium.org>
> Commit-Queue: Morten Stenshorne <mstensho@chromium.org>
> Cr-Commit-Position: refs/heads/master@{#567882}

TBR=eae@chromium.org,mstensho@chromium.org

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

Bug: 852640
Change-Id: I0af60d9eac770f21b6a0fbeccf88bb43d5efd6fb
Reviewed-on: https://chromium-review.googlesource.com/1103503Reviewed-by: default avatarAbhishek Arya <inferno@chromium.org>
Commit-Queue: Abhishek Arya <inferno@chromium.org>
Cr-Commit-Position: refs/heads/master@{#567914}
      e9d1ed95
    • Aleksei Loshkarev's avatar
      Adds metadata_fate to DataTypeController::Stop · 1333bf97
      Aleksei Loshkarev authored 
      Now ModelTypeController:Stop uses preferences to determine if data type
will be disable completely.
This CL introduce adds metadata_fate parameter which will be used
as signal for clearing metadata.

This is second CL in sequence.

Bug: 823721
Change-Id: Idfe4a70864e2f05aba9fc1ede3b0ccc2a108287a
Reviewed-on: https://chromium-review.googlesource.com/1093318
Commit-Queue: Aleksei Loshkarev <lixan@yandex-team.ru>
Reviewed-by: default avatarMikel Astiz <mastiz@chromium.org>
Reviewed-by: default avatarVasilii Sukhanov <vasilii@chromium.org>
Cr-Commit-Position: refs/heads/master@{#567913}
      1333bf97