GitLab

Previously, if the viewport ever moved farther than the skewport limits
in a single frame, the skewport math would break.  It would first
intersect with the max skewport (becoming empty) and then unioning with
the viewport (becoming the viewport).  By moving the union first, it
prevents this empty condition, leaving a valid skewport in the right
direction.

This addresses a case where prepainting wasn't happening when it should
have.

BUG=446645

Review URL: https://codereview.chromium.org/1021663002

Cr-Commit-Position: refs/heads/master@{#321428}

Reuse was disabled in #317499 because CrOS corruption rates were spiking and
were too close to the stable branch point to gather more history.  Re-enabling
to allow monitoring in dev/beta channels with the hope that the rise we saw on
CrOS was normal fluctuations and unrelated to this change.

BUG=460568

Review URL: https://codereview.chromium.org/1023773003

Cr-Commit-Position: refs/heads/master@{#321427}

Original CL reverted here: https://codereview.chromium.org/1013003002/

Nothing has changed from that CL above. We are relanding after Oshima
prevented marking events created from the same native event as "repeat"
in this CL https://codereview.chromium.org/1017323002/

Menus used to block detecting user activity. The fix works by making the
UserActivityDetector a PlatformEventObserver rather than an
EventHandler.
Also made sure that PlatformEventSource has an instance in
ozone_platform_test.cc

TBR=sadrul@chromium.org,spang@chromium.org,oshima@chromium.org,jamescook@chromium.org
BUG=462735
TEST=ui_base_unittests --gtest_filter=UserActivityDetectorTest.*

Review URL: https://codereview.chromium.org/1024583003

Cr-Commit-Position: refs/heads/master@{#321426}

Summary of changes available at:
https://chromium.googlesource.com/skia/+log/3ef71e3..5553bed

CQ_INCLUDE_TRYBOTS=tryserver.blink:linux_blink_rel
TBR=egdaniel@google.com

Review URL: https://codereview.chromium.org/1023633004

Cr-Commit-Position: refs/heads/master@{#321425}

- Extends Channel to PostMessageChannel to switch saml_injected.js to
  use postMessage instead of extension messaging API when used with the
  new Authenticator;
- Port Saml logic in main.js and background.js into saml_handler.js to
  be used with the new Authenticator;
- Hook up confirm/verify password flow etc with new Authenticator;

BUG=461505

Review URL: https://codereview.chromium.org/1004753004

Cr-Commit-Position: refs/heads/master@{#321424}

If on a MacOS machine, iossim and Chromium.app are built than we can
launch and use them for running tests.

BUG=468565

Review URL: https://codereview.chromium.org/1015313002

Cr-Commit-Position: refs/heads/master@{#321423}

R=inferno@chromium.org,tsepez@chromium.org
BUG=450268
NOPRESUBMIT=true

Review URL: https://codereview.chromium.org/1025483002

Cr-Commit-Position: refs/heads/master@{#321422}

The NotificationMessageFilter still receives a number of IPC messages on
the UI thread. It is possible that the IPC message gets handled after
the profile has been shut down, so make sure that this isn't the case
by verifying that the RenderProcessHost still exists.

BUG=457590

Review URL: https://codereview.chromium.org/1023633003

Cr-Commit-Position: refs/heads/master@{#321421}

BUG=468762

Review URL: https://codereview.chromium.org/1024473003

Cr-Commit-Position: refs/heads/master@{#321420}

This CL moves key injection logic setRemapKey() and injectKeyCombination() into
the plugin layer.

With this change, we no longer need to pass the default remap keys from the
ApplicationDelegate through the SessionConnector to the DesktopConnectedView
in order to setup a key remapping.

Instead, we can simply call plugin.setRemapKeys('a->b,b->c') on the delegate.

BUG=465878

Review URL: https://codereview.chromium.org/1022473004

Cr-Commit-Position: refs/heads/master@{#321419}

R=brianderson@chromium.org

Review URL: https://codereview.chromium.org/1024593002

Cr-Commit-Position: refs/heads/master@{#321418}

Summary of changes available at:
https://chromium.googlesource.com/chromium/blink/+log/5e13200..22ac63e

TBR=pilgrim@chromium.org,yurys@chromium.org

Review URL: https://codereview.chromium.org/1020893003

Cr-Commit-Position: refs/heads/master@{#321417}

This is a formatting change, simplifying the existing code. No
functional change intended.

Also simplified how the DisplayState::selected_mode is populated when
parsing the displays.

BUG=none

Review URL: https://codereview.chromium.org/1026443002

Cr-Commit-Position: refs/heads/master@{#321416}

This change alters the functionality of getObjectAttr slightly: it no
longer accepts null or Array objects as "object" values.  I made this
change because the type checking code seems designed to mirror the
semantics of JSON, where each value belongs to exactly one of the
following categories: array, boolean, number, null, object, or string.

This change also changes the type of values thrown from strings to Error objects so that there are more likely to be useful stack traces when an error occurs.

BUG=

Review URL: https://codereview.chromium.org/1015553003

Cr-Commit-Position: refs/heads/master@{#321415}

Fix for a crasher in the browser seen while dispatching mouse enter or mouse exit messages via the root view.

Based on a number of crash dumps I looked at starting from M40 onwards, the crash occurs in the RootView::NotifyEnterExitOfDescendant function while dereferencing a NULL view parameter.

The parameter passed as the view, is the mouse_move_handler_ member which is checked for validity before calling this
function. However the disassembly in the crash dump clearly suggests that the parameter passed on the stack is NULL.

Looking at the code the only way that could happen if we end up in a nested invocation to the root view, which could
potentially happen in the context of a modal loop. I could not repro that hypothesis however.

Given that this is a browser crash and seems to be occurring frequently enough, I think this warrants a NULL check
for the mouse_move_handler_ before calling the RootView::NotifyEnterExitOfDescendant function.

I added a CHECK for the view parameter in the NotifyEnterExitOfDescendant  function in case there are additional callsites
added in the future.

BUG=467356
TEST=No test at the moment as I could not verify the hypothesis with actions in the UI.

Review URL: https://codereview.chromium.org/996103009

Cr-Commit-Position: refs/heads/master@{#321414}

Fix a mistake I made in http://crrev.com/1003853004.
Before that CL, the "More tools" submenu was incorrectly using IDC_ZOOM_MENU
command, and I fixed that. But then I incorrectly added checks for
IDC_MORE_TOOLS_MENU wherever it was checking for IDC_ZOOM_MINUS, causing this
incorrect spacing bug. This removes those checks, but we still need a check
to make sure it is enabled, because when it was incorrectly using IDC_ZOOM_MENU,
it was getting enabled by BrowserCommandController.

BUG=468644

Review URL: https://codereview.chromium.org/1016413002

Cr-Commit-Position: refs/heads/master@{#321413}

BUG=466721,431489

Review URL: https://codereview.chromium.org/1020783002

Cr-Commit-Position: refs/heads/master@{#321412}

Previously |cdm_id| is guarded by the ENABLE_BROWSER_CDMS macro, which makes a
lot of code more complicated and brings very little savings. This CL always
defines |cdm_id| to clean this up.

BUG=432998

Review URL: https://codereview.chromium.org/835683002

Cr-Commit-Position: refs/heads/master@{#321411}

Removes the unused shutdown() function from WebSchedulerImpl now that it is
no longer used by Blink.

BUG=467369

Review URL: https://codereview.chromium.org/1022443002

Cr-Commit-Position: refs/heads/master@{#321410}

The banner engagement persisted data should be per-origin. This test
ensures that is the case.

BUG=468148

Review URL: https://codereview.chromium.org/1019763002

Cr-Commit-Position: refs/heads/master@{#321409}

BUG=468474

Review URL: https://codereview.chromium.org/1019943002

Cr-Commit-Position: refs/heads/master@{#321408}

BUG=349185

Review URL: https://codereview.chromium.org/1000863003

Cr-Commit-Position: refs/heads/master@{#321407}

Re-mask all unmasked cards when the user disables autofill integration.

Delete all server cards and addresses when the user disables the sync type or stops syncing.

Fix a bug where addresses were always written to the DB on startup because profiles were being created differently from protos than they were from sqlite so wouldn't match.

Optimize autofill profile comparisons. I was offended by the number of autofill profile and card comparisons we do at startup, the small number of items in the list means std::set is operating at a worst-case. I switched to brute-force with a "give up and rewrite the data" escape valve if you have a large number of addresses.

BUG=450843

Review URL: https://codereview.chromium.org/1007003007

Cr-Commit-Position: refs/heads/master@{#321406}

BUG=463780

Review URL: https://codereview.chromium.org/1019303004

Cr-Commit-Position: refs/heads/master@{#321405}

Summary of changes available at:
https://chromium.googlesource.com/skia/+log/647211f..3ef71e3

CQ_INCLUDE_TRYBOTS=tryserver.blink:linux_blink_rel
TBR=egdaniel@google.com

Review URL: https://codereview.chromium.org/1009603004

Cr-Commit-Position: refs/heads/master@{#321404}

This deduplication becomes a particularly good win when implementing
GetThreadPriority() in https://codereview.chromium.org/1006933003/

BUG=456903,468793

Review URL: https://codereview.chromium.org/1015503002

Cr-Commit-Position: refs/heads/master@{#321403}

Revert of Allow users to add third-party VPNs from the settings page (patchset #5 id:80001 of https://codereview.chromium.org/1017443002/)

Reason for revert:
Invalid JS caught by Closure bots:
http://build.chromium.org/p/chromium.fyi/builders/Closure%20Compilation%20Linux/builds/18417/steps/compile/logs/stdio

Original issue's description:
> Allow users to add third-party VPNs from the settings page
>
> This CL adds "Add $provider..." entries to the "Private network" and
> "Add connection" drop-down menus on the Chrome OS settings page. The
> contents of these menus updates dynamically as VPN providers are
> added/removed.
>
> BUG=460428
> TEST=Manual
>
> Committed: https://crrev.com/4f6e1a2ded3a56e7c2d1360b94990b025a309ecb
> Cr-Commit-Position: refs/heads/master@{#321395}

TBR=michaelpg@chromium.org,stevenjb@chromium.org,isherman@chromium.org,bartfab@chromium.org
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
BUG=460428

Review URL: https://codereview.chromium.org/1026463003

Cr-Commit-Position: refs/heads/master@{#321402}

Summary of changes available at:
https://chromium.googlesource.com/chromium/blink/+log/2960bec..5e13200

TBR=pilgrim@chromium.org,yurys@chromium.org

Review URL: https://codereview.chromium.org/1016383002

Cr-Commit-Position: refs/heads/master@{#321401}

Frecon requires BSD style file locks since the lock needs to be held
past forking. So we also need to change the Ozone side since Posix and
BSD locks don't mix.

BUG=463665

Review URL: https://codereview.chromium.org/1007993007

Cr-Commit-Position: refs/heads/master@{#321400}

This adds five methods to the NotificationDatabase class to read or delete
multiple notifications, which share two actual implementations.

They will be used as follows:
- Read all data: browsing data overviews.
- Read all data for an origin: for origin storage overviews, quota
  client integration (storageQuota.queryInfo()).
- Read all data for a SWR id: to get all notifications associated
  with a given SWR (SWR.getNotifications()).
- Delete all data for an origin: clearing browsing data for a
  given origin.
- Delete all data for a SWR id: to remove notifications when a service
  worker registration gets dropped.

Actual integration with these features will be done in separate patches,
this patch just introduces the primitives in the database.

Design document:
  http://goo.gl/TciXVp

BUG=447628

Review URL: https://codereview.chromium.org/1019073002

Cr-Commit-Position: refs/heads/master@{#321399}

This causes racy sadness.

BUG=235188

Review URL: https://codereview.chromium.org/1020803002

Cr-Commit-Position: refs/heads/master@{#321398}

BUG=460581

Review URL: https://codereview.chromium.org/1003883004

Cr-Commit-Position: refs/heads/master@{#321397}

https://codereview.chromium.org/1019863002

R=tomfinegan@chromium.org

Review URL: https://codereview.chromium.org/1024523002

Cr-Commit-Position: refs/heads/master@{#321396}

This CL adds "Add $provider..." entries to the "Private network" and
"Add connection" drop-down menus on the Chrome OS settings page. The
contents of these menus updates dynamically as VPN providers are
added/removed.

BUG=460428
TEST=Manual

Review URL: https://codereview.chromium.org/1017443002

Cr-Commit-Position: refs/heads/master@{#321395}

Revert of https://codereview.chromium.org/1006003005/ and https://codereview.chromium.org/1001153002/.

We've already collected enough data to triage this bug.

BUG=464633

Review URL: https://codereview.chromium.org/1016183002

Cr-Commit-Position: refs/heads/master@{#321394}

All the major work has already been done. This CL just finishes flushing out
the documentation on tab.py and web_contents.py so that consumers know exactly
which exceptions can be thrown.

This CL contains one small behavioral change: InspectorBackend.IsInspectable()
catches DevToolsClientConnectionError and returns False. This is useful since
IsInspectable() gets called in the exception handling code, and no other
consumers of IsInspectable expect the method to throw an exception.

BUG=460625

Review URL: https://codereview.chromium.org/1007903004

Cr-Commit-Position: refs/heads/master@{#321393}

BUG=369661
TEST=tests should stay passing

Review URL: https://codereview.chromium.org/1010243002

Cr-Commit-Position: refs/heads/master@{#321392}

BUG=392620

Review URL: https://codereview.chromium.org/1006813005

Cr-Commit-Position: refs/heads/master@{#321391}

BUG=454815

Review URL: https://codereview.chromium.org/978913004

Cr-Commit-Position: refs/heads/master@{#321390}

BUG=462584
R=rvargas@chromium.org

Review URL: https://codereview.chromium.org/1017243002

Cr-Commit-Position: refs/heads/master@{#321389}