GitLab

See http://codereview.chromium.org/10756018/


BUG=133570
TBR=cmp@chromium.org
Review URL: https://chromiumcodereview.appspot.com/10704171

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@146231 0039d316-1c4b-4281-b951-d872f2087c98

The policy setting crash has been fixed. Re-enable the tests.

TBR=mnissler@chromium.org
BUG=134560
TEST=

Review URL: https://chromiumcodereview.appspot.com/10692175

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@146230 0039d316-1c4b-4281-b951-d872f2087c98

BUG=135482
TEST=none


Review URL: https://chromiumcodereview.appspot.com/10700150

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@146229 0039d316-1c4b-4281-b951-d872f2087c98

CacheEntry is now complex enough to be a class, rather than a struct.

BUG=136625
TEST=out/Release/unit_tests --gtest_filter=GData*

Review URL: https://chromiumcodereview.appspot.com/10698157

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@146227 0039d316-1c4b-4281-b951-d872f2087c98

This CL moves the contents of ui_resources_standard.grd into ui_resources.grd and theme_resources_standard.grd into theme_resources.grd. This also eliminates the following Grit outputs:

ui_resources.pak  ==>  duplicated in each scaled-resource pak
ui_resources_standard.h  ==>  merged into ui_resources.h
ui_resources_standard_map.h  ==>  renamed to ui_resources_map.h
ui_resources_standard_map.cc  ==>  renamed to ui_resources_map.cc

theme_resources.rc  ==>  merged into theme_resources_standard.rc
theme_resources.pak  ==>  duplicated in each scaled-resource pak
theme_resources_standard.h  ==>  merged into theme_resources.h
theme_resources_standard_map.h  ==>  merged into theme_resources_map.h
theme_resources_standard_map.cc  ==>  merged into theme_resources_map.cc

The image duplication is intentional; we may revisit this later.

Currently the standard-size images are included twice in the Windows build, once as an rc and once as a pak. I'll fix this in a followup CL.


TBR=ben@chromium.org,brettw@chromium.org,darin@chromium.org,stevenjb@chromium.org
BUG=133570
TEST=Trybots, local build

Review URL: https://chromiumcodereview.appspot.com/10756018

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@146224 0039d316-1c4b-4281-b951-d872f2087c98

BUG=123739
TEST=try


Review URL: https://chromiumcodereview.appspot.com/10695150

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@146222 0039d316-1c4b-4281-b951-d872f2087c98

BUG=None
TEST=None


Review URL: https://chromiumcodereview.appspot.com/10752006

TBR=shishir@chromium.org
Review URL: https://chromiumcodereview.appspot.com/10704169

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@146221 0039d316-1c4b-4281-b951-d872f2087c98

Some of the tests, such as layout tests, need to run on a seperate thread.
Adding an extra to the intent to achieve this.
When invoking the test, users can use 'adb shell am start -e RunInSubThread -n testName'

BUG=
TEST=


Review URL: https://chromiumcodereview.appspot.com/10696172

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@146220 0039d316-1c4b-4281-b951-d872f2087c98

This is a partial port of WebAudio's SincResampler from WebKit formatted and
culled for use by Chrome Media.

We can't directly use the one in WebKit as it's layed under a ton of abstraction
and is tightly coupled with WebKit objects.

Test generates a swept sine wave and calculates the RMS error for common sample
rates (via UMA stats).

MultiChannelResampler and AudioRenderMixer changes to support resampling will
come in later CLs.

The 1000 ft view is that MultiChannelResampler will implement SincResampler::
AudioSourceProvider and AudioRendererMixer will implement a new
MultiChannelResampler::MultiChannelAudioSourceProvider interface.

When resampling is necessary AudioRenderMixer will feed itself into a
MultiChannelResampler instance which will poll data as necessary and feed it
channel by channel into a set of SincResamplers (one for each channel).

We want to resample post-mixing since resampling is a much more expensive
operation.

Original for reference:
http://git.chromium.org/gitweb/?p=external/Webkit.git&a=blob&f=Source/WebCore/platform/audio/SincResampler.cpp

Visual plot of 44100 to 48000 for reference; red line is the resampled signal
and the green line is the reference signal. Ideally only the pure signal (green)
should be seen.  Any bit of the resampled signal showing (red) is where the
resampling algorithm is incorrect: http://i.imgur.com/1vsaI.png

BUG=133637
TEST=New unittests.

Review URL: https://chromiumcodereview.appspot.com/10702050

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@146219 0039d316-1c4b-4281-b951-d872f2087c98

This preference appears to be unused and the code for Hixie76 has been removed
from WebKit.

Review URL: https://chromiumcodereview.appspot.com/10689165

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@146218 0039d316-1c4b-4281-b951-d872f2087c98

(otherwise only half (the common half...:( ) of the sid string would get encoded...)

this change is super trivial and only fixes the resulting suffix

R=grt@chromium.org
BUG=133810
TEST=Make sure Chrome gets the proper suffix


Review URL: https://chromiumcodereview.appspot.com/10703147

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@146217 0039d316-1c4b-4281-b951-d872f2087c98

to base\win\metro.

BUG=133560
R=cpu
Review URL: https://chromiumcodereview.appspot.com/10693133

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@146214 0039d316-1c4b-4281-b951-d872f2087c98

BUG=136629
TEST=no
TBR=aa@chromium.org

Review URL: https://chromiumcodereview.appspot.com/10690144

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@146213 0039d316-1c4b-4281-b951-d872f2087c98

Rename/re-comment tests and added new death tests for WeakPtr.

Review URL: https://chromiumcodereview.appspot.com/10690080

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@146212 0039d316-1c4b-4281-b951-d872f2087c98

R=jar@chromium.org
BUG=108171


Review URL: https://chromiumcodereview.appspot.com/10694131

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@146211 0039d316-1c4b-4281-b951-d872f2087c98

These tests are failing since the MS 2008 - 2010 switch.  Turning them off to make the tree green until someone takes the bug and figures out how to fix it.

BUG=132133
TEST=gpu win release bots
TBR=nduca
Review URL: https://chromiumcodereview.appspot.com/10540111

TBR=zmo@google.com
Review URL: https://chromiumcodereview.appspot.com/10704161

TBR=jbates@chromium.org
Review URL: https://chromiumcodereview.appspot.com/10695160

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@146210 0039d316-1c4b-4281-b951-d872f2087c98

Revert 146192 - Revert 146164 - Always query for the latest RenderViewHost from WebContents before sending the NOtBlacklistedMessage, since RenderViewHost can be changed if users navigate to a different site. We have seen crashes (highly possibly) because of this (http://code.google.com/p/chromium/issues/detail?id=134191)


BUG=134191
TEST=PasswordFormManagerTest

Review URL: https://chromiumcodereview.appspot.com/10704147

TBR=zysxqn@google.com
Review URL: https://chromiumcodereview.appspot.com/10696177

TBR=michaeln@google.com
Review URL: https://chromiumcodereview.appspot.com/10701145

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@146206 0039d316-1c4b-4281-b951-d872f2087c98

The Inner class was being leaked because it was marked DeleteOnUIThread, but the last reference is released after the UI message loop goes away. So I removed that trait.

BUG=133449
TEST=no

Review URL: https://chromiumcodereview.appspot.com/10695130

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@146205 0039d316-1c4b-4281-b951-d872f2087c98

Now UpdateCacheWithSubDirectoryType() is trivial enough (it used to be bigger)
to be gone.

BUG=136625
TEST=out/Release/unit_tests --gtest_filter=GData*

Review URL: https://chromiumcodereview.appspot.com/10692171

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@146203 0039d316-1c4b-4281-b951-d872f2087c98

BUG=127222
TEST=manual test.


Review URL: https://chromiumcodereview.appspot.com/10444049

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@146202 0039d316-1c4b-4281-b951-d872f2087c98

Modify ppapi/generators/idl_parser.py to
0: not require a file-level comment (but generate the same parse tree structure),
1: actually support ext_attrs (modifiers) for dictionaries, and
2: support [ext_attr=(symbols|values)].

Modify json_schema_compiler to
0: use "base::Value" and any_helper.ANY_CLASS instead of Value and Any in order to support ArrayBuffers named |value| or |any|,
1: actually test that namespaces and dictionaries are sorted correctly,
2: fix HGenerator._FieldDependencyOrder(),
3: support [inline_doc] on dictionaries and enums,
4: support descriptions on enums,
5: support documentation_permissions_required,
6: support [legalValues=(values...)].

Review URL: https://chromiumcodereview.appspot.com/10639020

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@146201 0039d316-1c4b-4281-b951-d872f2087c98

If an account with hosts registered is accessed on a machine on which those hosts are cached, then the host list should not display the "no hosts" icon while the host list is being refreshed. This CL fixes a bug related to the asynchronous nature of getting the local host id for the purpose of filtering it from the host list, which caused the cached host list never to be displayed.

Note that if the account has hosts registered, but is accessed from a computer with no cache, then the "no hosts" icon will still flash briefly (although it will generally be hidden behind the infographic).

BUG=135288
TEST=Manual

Review URL: https://chromiumcodereview.appspot.com/10736020

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@146200 0039d316-1c4b-4281-b951-d872f2087c98

When this is enabled for windows, we stop getting crash reports.
See https://code.google.com/p/chromium/issues/detail?id=135480

BUG=135480
TEST=


Review URL: https://chromiumcodereview.appspot.com/10736039

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@146199 0039d316-1c4b-4281-b951-d872f2087c98

BUG=99330
TEST=Under opaque theme, resize window really small. The minimum width shouldn't cause tabstrip elements to be cutoff.

Review URL: https://chromiumcodereview.appspot.com/8172010

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@146197 0039d316-1c4b-4281-b951-d872f2087c98

BUG=134017, 117261
TEST=


Review URL: https://chromiumcodereview.appspot.com/10626007

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@146196 0039d316-1c4b-4281-b951-d872f2087c98

It is still possible to focus these buttons using tab, so this does not regress accessibility.

BUG=136872
TEST=none

Review URL: https://chromiumcodereview.appspot.com/10694139

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@146195 0039d316-1c4b-4281-b951-d872f2087c98

BUG=134917
TEST=nothing yet.
TBR=dhollowa

Review URL: https://chromiumcodereview.appspot.com/10703148

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@146194 0039d316-1c4b-4281-b951-d872f2087c98

TEST=added
BUG=132616


Review URL: https://chromiumcodereview.appspot.com/10759006

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@146193 0039d316-1c4b-4281-b951-d872f2087c98

Revert 146164 - Always query for the latest RenderViewHost from WebContents before sending the NOtBlacklistedMessage, since RenderViewHost can be changed if users navigate to a different site. We have seen crashes (highly possibly) because of this (http://code.google.com/p/chromium/issues/detail?id=134191)


BUG=134191
TEST=PasswordFormManagerTest

Review URL: https://chromiumcodereview.appspot.com/10704147

TBR=zysxqn@google.com
Review URL: https://chromiumcodereview.appspot.com/10696177

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@146192 0039d316-1c4b-4281-b951-d872f2087c98

Previously this function takes a DocumentEntry* pointer obtained from UploadFileInfo.
This implicitly requires caller to keep the DocumentEntry object (or the UploadFileInfo
struct) alive until the call is complete.

Changing this argument to a scoped pointer makes ownership transfer explicit.

BUG=none
TEST=unit_tests (GData*), lumpy

Review URL: https://chromiumcodereview.appspot.com/10735059

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@146190 0039d316-1c4b-4281-b951-d872f2087c98

These result in compiler warnings with -Wsequence-point

BUG=136625
TEST=out/Release/unit_tests --gtest_filter=GData*

Review URL: https://chromiumcodereview.appspot.com/10702154

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@146189 0039d316-1c4b-4281-b951-d872f2087c98

Details in bug.

TBR=kkania@chromium.org
BUG=136875
TEST=

Review URL: https://chromiumcodereview.appspot.com/10704162

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@146188 0039d316-1c4b-4281-b951-d872f2087c98

TBR=dyu@chromium.org
BUG=136875
TEST=

Review URL: https://chromiumcodereview.appspot.com/10702153

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@146187 0039d316-1c4b-4281-b951-d872f2087c98

BUG=126093
TEST=Unit test, manual testing.


Review URL: https://chromiumcodereview.appspot.com/10636019

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@146186 0039d316-1c4b-4281-b951-d872f2087c98

These should belong to CacheEntry as non-static member functions.

BUG=136625
TEST=out/Release/unit_tests --gtest_filter=GData*

Review URL: https://chromiumcodereview.appspot.com/10692154

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@146183 0039d316-1c4b-4281-b951-d872f2087c98

Remove browser::FindTabbedBrowser call in bookmark_utils.cc. All callers, with the exception of metro-specific code, already have a navigator. By moving the browser::FindTabbedBrowser call there, it's clear that in that spot we need to use the metro-specific context.

BUG=129187
Review URL: https://chromiumcodereview.appspot.com/10692164

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@146182 0039d316-1c4b-4281-b951-d872f2087c98

BUG=NONE
TEST=asan bots are now happy

Review URL: https://chromiumcodereview.appspot.com/10736021

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@146181 0039d316-1c4b-4281-b951-d872f2087c98

The underlying issue is that the GR expects individual touch events (one touch per event) while javascript expects composite events (containing information on each touch point in a single event.)

BUG=130205
TEST=


Review URL: https://chromiumcodereview.appspot.com/10702096

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@146180 0039d316-1c4b-4281-b951-d872f2087c98

Change GDataCache::UpdateCacheWithSubDirectoryType() and
GDataCacheMetadata::UpdateCache() to take CacheEntry.

This is in preparation for removing IsCache/SetCache/ClearCache*
functions from GDataCache.

BUG=136625
TEST=out/Release/unit_tests --gtest_filter=GData*

Review URL: https://chromiumcodereview.appspot.com/10735049

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@146179 0039d316-1c4b-4281-b951-d872f2087c98

The transaction-abort test was flaky in Chromium, fixed by WK90412/r122291. The others were
disabled to allow landing WK89379/r121492 which renamed the tests.

BUG=83226
TEST=browser_tests --gtest_filter='IndexedDBLayoutTest.TransactionTests'


Review URL: https://chromiumcodereview.appspot.com/10736036

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@146178 0039d316-1c4b-4281-b951-d872f2087c98