GitLab

BUG=159969


Review URL: https://chromiumcodereview.appspot.com/11308355

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@171137 0039d316-1c4b-4281-b951-d872f2087c98

Bookmark Sync may try to Move an entry to the location where it already is placed.
Updating the folder's timestamp with such no-op moves unwillingly disorders
the "recently used" list of folders; it should better be avoided.

BUG=162764

Review URL: https://chromiumcodereview.appspot.com/11411346

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@171134 0039d316-1c4b-4281-b951-d872f2087c98

This is blocking https://bugs.webkit.org/show_bug.cgi?id=102726 since
QuotaStatusCode hard codes the values which prevents us from changing those
values.

BUG=None


Review URL: https://chromiumcodereview.appspot.com/11434107

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@171133 0039d316-1c4b-4281-b951-d872f2087c98

- Add MonitorDeviceChange() for PPB_AudioInput_Dev.
- Change EnumerateDevices() to use PP_ArrayOutput.
- Move device enumeration code out of the audio input implementation, so that it can be shared by video capture.
- Update the audio_input manual test.
- Add unittests for the device enumeration code.

TEST=None
BUG=137799


Review URL: https://chromiumcodereview.appspot.com/11411047

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@171132 0039d316-1c4b-4281-b951-d872f2087c98

Targets with dependencies (direct/indirect) on a java target receive the
chromium_*.jar paths in the variable input_jars_paths. Targets should
use that rather than hardcoding where they think the jar will be.

Change native_test_apk.xml to build with a list of jars in
INPUT_JARS_PATHS rather than using all jars in a specific output
directory.

Make generate_native_test.py pass the --jars argument to ant with
-DINPUT_JARS_PATHS. This supports those targets (WebKit) that are not
updated to pass the ant-arg directly.

BUG=


Review URL: https://chromiumcodereview.appspot.com/11308351

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@171131 0039d316-1c4b-4281-b951-d872f2087c98

BUG=none
TEST=none

Review URL: https://codereview.chromium.org/11418309

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@171130 0039d316-1c4b-4281-b951-d872f2087c98

BUG=162110


Review URL: https://chromiumcodereview.appspot.com/11299339

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@171129 0039d316-1c4b-4281-b951-d872f2087c98

It's not used anymore.


Review URL: https://chromiumcodereview.appspot.com/11414317

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@171128 0039d316-1c4b-4281-b951-d872f2087c98

BUG=


Review URL: https://chromiumcodereview.appspot.com/11428166

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@171127 0039d316-1c4b-4281-b951-d872f2087c98

Part of the new metrics suggested for Spellcheck.

R=rlp@chromium.org
BUG=none.


Review URL: https://chromiumcodereview.appspot.com/11308339

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@171126 0039d316-1c4b-4281-b951-d872f2087c98

BUG=157690


Review URL: https://chromiumcodereview.appspot.com/11416327

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@171125 0039d316-1c4b-4281-b951-d872f2087c98

Now that we only use the host part of the given origin, we need to call
DirectoryTitleToOrigin() before making a origin GURL.

BUG=156041
TEST=manual

Review URL: https://codereview.chromium.org/11421222

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@171124 0039d316-1c4b-4281-b951-d872f2087c98

BUG=159265


Review URL: https://chromiumcodereview.appspot.com/11411183

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@171123 0039d316-1c4b-4281-b951-d872f2087c98

NOTRY=true
BUG=none


Review URL: https://chromiumcodereview.appspot.com/11411367

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@171122 0039d316-1c4b-4281-b951-d872f2087c98

Original shortcut keys (Ctrl+Alt+F6/F7) are often the keys to switch VTs and cannot be used as the magnifier shortcuts on almost all Linux workstations.

This CL adds additional shortcut keys only for debug mode.

BUG=none
TEST=manual

Review URL: https://codereview.chromium.org/11421223

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@171121 0039d316-1c4b-4281-b951-d872f2087c98

Cleanup: remove redundant shelf code.
- replaces switch to get values for alignment with template
- eliminate alignment member variables and use shelf_layout_manager where possible.

other minor clean ups: remove unnecessary namespace, use member variable.

BUG=151417
TEST=none

Review URL: https://chromiumcodereview.appspot.com/11434099

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@171120 0039d316-1c4b-4281-b951-d872f2087c98

 - Remove addOneShortEventListener + navigateTo pattern.
 - Remove if (list.parentId == newParentId) pattern.
 - Replace boolean parameter opt_updateHashNow to callback function to avoid using boolean argument.

This patch also fixes bug 163582 to set focus to folder name input box at right time.

This re-factoring reduces 30 lines of total code size.
* navigateTo(): Change to take callback to call at BookmarkList.load event.
* updateParentId(): Add comments and replace if-statement by conditional operator.
* showInFolder(): Change to check whether list.selectedItem is null or not, before using it. Rename callback function for navigateTo() from f() to selectItem().
* newFolder(): Change to set edit mode for new folder after navigateTo().

Related review:
https://codereview.chromium.org/11411162

BUG=163582

Review URL: https://codereview.chromium.org/11280208

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@171119 0039d316-1c4b-4281-b951-d872f2087c98

BUG=158086

Review URL: https://codereview.chromium.org/11316329

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@171118 0039d316-1c4b-4281-b951-d872f2087c98

Added method RenderView::FromRoutingID to expose internal lookup function.

BUG=155398


Review URL: https://chromiumcodereview.appspot.com/11316160

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@171117 0039d316-1c4b-4281-b951-d872f2087c98

Clean up some code in the GesturePoint to provide centralized
access to the scroll deltas so that smoothing of scroll start
is easier to implement.

BUG=141653


Review URL: https://chromiumcodereview.appspot.com/11415293

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@171116 0039d316-1c4b-4281-b951-d872f2087c98

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@171113 0039d316-1c4b-4281-b951-d872f2087c98

This change is in preparation for threading certain ui animations. When we do
this, the actual duration of an element will not be known ahead of time, since 
dispatching the element to the compositor thread will involve a small delay.

BUG=164206

Review URL: https://chromiumcodereview.appspot.com/11316315

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@171112 0039d316-1c4b-4281-b951-d872f2087c98

Tiles now have a contents scale that they are rastered with, so that non-1.0
contents scales work.  Additionally, PictureLayer derives from
ContentsScalingLayer so that it picks up the right content bounds and default
contents scale.

BUG=155209


Review URL: https://chromiumcodereview.appspot.com/11421210

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@171111 0039d316-1c4b-4281-b951-d872f2087c98

BUG=107201,133436
TEST=no visible change


Review URL: https://chromiumcodereview.appspot.com/11418260

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@171110 0039d316-1c4b-4281-b951-d872f2087c98

This changes the Android port of Chrome to depend on the pref
kSyncSessionsGUID to be set before the SessionModelAssociator is
initialized.

We now also force-removes any foreign sessions that are using the
current client's GUID as a session tag.

This also lead to the removal of the Android specific session utils.

BUG=157419


Review URL: https://chromiumcodereview.appspot.com/11414013

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@171109 0039d316-1c4b-4281-b951-d872f2087c98

Some device's concept of time might be off.
To prevent corrupt texture runtime measurements from corrupting
the texture upload logic, we clamp the runtimes to the range
of [1,15000] microseconds.

If the corrupt runtimes are 0, this patch will effectively cause
us to always upload all pending textures.

This is a workaround until the async texture upload feature lands.

BUG=162699


Review URL: https://chromiumcodereview.appspot.com/11434111

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@171108 0039d316-1c4b-4281-b951-d872f2087c98

R=sky
BUG=163454
Depends on: https://codereview.chromium.org/11418217/
NOTRY=true

Review URL: https://chromiumcodereview.appspot.com/11299262

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@171107 0039d316-1c4b-4281-b951-d872f2087c98

Marked IPC::ChannelProxy as non thread-safe. Added DCHECKs to verify that public methods (with Send() being the only exception) are called on the thread that created the object. 

Tests calling Send() from a wrong thread should be fixed first before similar check can be added to Send(). See http://crbug.com/163523 for details.

BUG=163091,163523

Review URL: https://chromiumcodereview.appspot.com/11308278

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@171106 0039d316-1c4b-4281-b951-d872f2087c98

Refactor ProcessSingleton to allow a Desktop process to activate Metro Chrome instead of grabbing the singleton (while making sure it simply RDVs if there already is an existing Chrome).

Use a Windows event to let the Metro process notify the Desktop process once it grabbed the singleton (no more Sleep()).

The idea is the following:

0) Any process always tries to rendez-vous before anything else.
1) A Metro process can always try to grab the singleton.
2) A Desktop process might wait if:
  a) The metro activation event exists
  b) The metro activation event doesn't exist, but ShouldActivate() returns true
    - In which case it creates the metro activation event and activates Metro Chrome (which will notify this process via the event once it grabbed the singleton; at which point this process should rendez-vous with it).

All of this is done while holding a global mutex (only releasing it while waiting and re-acquiring it before continuing when notified).

The process singleton acquiring logic has also been moved from the constructor to Create().

BUG=155585, 159869
TEST=Process singleton works as before (and better in tricky corner cases -- see http://crbug.com/159869) on Win8.

Review URL: https://chromiumcodereview.appspot.com/11099053

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@171105 0039d316-1c4b-4281-b951-d872f2087c98

BUG=164296

Review URL: https://codereview.chromium.org/11446003

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@171104 0039d316-1c4b-4281-b951-d872f2087c98

> Adding transmission times for every QUIC packet received in the AckFrame.
> Making the scheduler a strict mock for quic connection test.
> 
> Merge internal change: 39116450
> Merge internal change: 38802607
> 
> 
> Review URL: https://chromiumcodereview.appspot.com/11416155

TBR=rch@chromium.org
Review URL: https://codereview.chromium.org/11439003

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@171103 0039d316-1c4b-4281-b951-d872f2087c98

Adding support for password Autofill popups for Android.

BUG=163619


Review URL: https://chromiumcodereview.appspot.com/11299287

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@171102 0039d316-1c4b-4281-b951-d872f2087c98

A better fix for the problem this revert was addressing has been checked in in r171097.

> Disable the settings app by default.
> 
> BUG=164064
> TBR=tapted, benwells
> 
> Review URL: https://codereview.chromium.org/11428167

TBR=erikwright@chromium.org
Review URL: https://codereview.chromium.org/11441002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@171101 0039d316-1c4b-4281-b951-d872f2087c98

Allows the host side to register a ResourceHost as pending and send an ID to the plugin, to be connected to a PluginResource at a future time.

BUG=


Review URL: https://chromiumcodereview.appspot.com/11414147

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@171099 0039d316-1c4b-4281-b951-d872f2087c98

BUG=164064

Review URL: https://codereview.chromium.org/11414314
Patch from Trent Apted <tapted@chromium.org>.

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@171098 0039d316-1c4b-4281-b951-d872f2087c98

(1) Added a recursive boolean param to FilePathWatcher::Watch() function to watch for sub directory tree changes. Fixed all the calling sites.
(2) Added support to watch sub trees on Windows.
(3) Added FilePathWatcherTest.RecursiveWatch browser test.

BUG=144491
TEST=none


Review URL: https://chromiumcodereview.appspot.com/11415066

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@171097 0039d316-1c4b-4281-b951-d872f2087c98

Making the scheduler a strict mock for quic connection test.

Merge internal change: 39116450
Merge internal change: 38802607


Review URL: https://chromiumcodereview.appspot.com/11416155

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@171096 0039d316-1c4b-4281-b951-d872f2087c98

BUG=130663


Review URL: https://chromiumcodereview.appspot.com/11418259

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@171095 0039d316-1c4b-4281-b951-d872f2087c98

This changes it so that we no longer throw an unrecoverable error if we get an
auth error while bringing up the backend (which can happen if we fail to
download control types). We now let just let the backend continuing retrying
(in backoff mode), until the user resolves the auth error, at which point
we'll perform our pending sync cycle, finishing backend initialization.

BUG=163491


Review URL: https://chromiumcodereview.appspot.com/11428125

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@171094 0039d316-1c4b-4281-b951-d872f2087c98

It looks like there might be a correlation between our features and the memory regression. This will hopefully tell us which feature is causing it.

BUG=163697
TEST=None
Review URL: https://codereview.chromium.org/11299344

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@171093 0039d316-1c4b-4281-b951-d872f2087c98