GitLab

This is a resubmit of https://codereview.chromium.org/12041059

BUG=175308, 170921

Review URL: https://codereview.chromium.org/12249003

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@182187 0039d316-1c4b-4281-b951-d872f2087c98

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@182181 0039d316-1c4b-4281-b951-d872f2087c98

The launcher install is now done by the webstore, so this code is no longer
needed.

BUG=None


Review URL: https://chromiumcodereview.appspot.com/12208146

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@182179 0039d316-1c4b-4281-b951-d872f2087c98

Introducing chrome::GetTotalBrowserCount() in browser_finder.cc which counts browsers across all desktops and profiles (this is the desired count for most BrowserList::empty() and BrowserList::size() consumers).

Otherwise calling a specific BrowserListImpl's size() when appropriate.

BUG=129187


Review URL: https://chromiumcodereview.appspot.com/12212120

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@182178 0039d316-1c4b-4281-b951-d872f2087c98

Renamed  --is-asan-mac-builder to --is-asan-mac-builder-hackfix per Nico's suggestion.
Added more comments.

BUG=170629, 174873
TBR=thakis
Review URL: https://codereview.chromium.org/12255003

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@182175 0039d316-1c4b-4281-b951-d872f2087c98

This causes a PRECONDITION error when two clients try to delete duplicated files.
The conflict is ignorable on this case, so we can just drop the etag matching. 

(notry, since most of try jobs are passed, and failed try looks not related to the patch)

BUG=175682
NOTRY=true

Review URL: https://chromiumcodereview.appspot.com/12209115

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@182174 0039d316-1c4b-4281-b951-d872f2087c98

It turns out that r181918, which pinned the Clang revision on Mac ASan
bots worked a little too well, pinning the version on *all* Mac bots.

This reverts the Clang roll. We will then fix update.sh to get the Mac
ASan pinning correct, and then re-land the roll.

BUG=174873, 170629

Review URL: https://codereview.chromium.org/12256002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@182173 0039d316-1c4b-4281-b951-d872f2087c98

The shortcut code on mac can only handle v2 apps, so we keep the menu item
hidden there.

BUG=175691
R=estade@chromium.org


Review URL: https://chromiumcodereview.appspot.com/12223120

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@182172 0039d316-1c4b-4281-b951-d872f2087c98

Android browser compositor does not have access directly to the
LayerTreeHost, only the compositor. This CL allows to set the
device scale factor from the compositor interface.

BUG=175589


Review URL: https://chromiumcodereview.appspot.com/12220110

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@182171 0039d316-1c4b-4281-b951-d872f2087c98

Clean up some manifest unit tests.

BUG=159265
TBR=sky@chromium.org

Review URL: https://chromiumcodereview.appspot.com/12042096

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@182170 0039d316-1c4b-4281-b951-d872f2087c98

BUG=166797
TEST=run_py_tests.py testSwitchToFrame


Review URL: https://chromiumcodereview.appspot.com/12217149

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@182169 0039d316-1c4b-4281-b951-d872f2087c98

This CL fixes some typos, broken links and other small issues in extension docs. 

Thanks for Sunny Jerry (ratsunny@gmail.com). Some of these mistakes were discovered by him and listed at https://code.google.com/p/chrome-documentation-chs/source/browse/documentation/error_list.txt .

BUG=173630,168952,119535,167355
TEST=none

Review URL: https://chromiumcodereview.appspot.com/12223068

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@182168 0039d316-1c4b-4281-b951-d872f2087c98

In logging.h, the NOTIMPLEMENTED_POLICY macro defines the behavior of the NOTIMPLEMENTED() macro.  For case 5 (log only once per call-site), code that uses stream output operators would fail to compile.  For example, in ui/aura/root_window_host_linux.cc:747:

  NOTIMPLEMENTED() << "Unsupported bits-per-pixel " << image->bits_per_pixel;

The solution is to add EAT_STREAM_PARAMETERS to the end of the definition in base/logging.h.  This will log call sites as "not implemented" but will not log the additional custom messages.

TEST=Compiled chrome with compiler define set to NOTIMPLEMENTED_POLICY=5 and ran to confirm desired NOTIMPLEMENTED() behaviors.


Review URL: https://chromiumcodereview.appspot.com/12221152

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@182167 0039d316-1c4b-4281-b951-d872f2087c98

1. Add a border bar to the top of the footer button area.
2. Change the inset size of the footer and scroll_content. The existing
   code scrolls the contents right after the button content, but a
   buffer is necessary.


Review URL: https://chromiumcodereview.appspot.com/12221126

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@182166 0039d316-1c4b-4281-b951-d872f2087c98

With the new full page NTP the detached bookmark bar was no longer visible.

The problem was that we relied on an instance of NewTabUI to control the bookmark bar visibility. The new full page NTP no longer has an instance of NewTabUI so the bookmark bar was never visible.

Fix was to move the relevant code out of NewTabUI and into BookmarkTabHelper.

BUG=175146
TEST=Ran with instant extended enabled. Verified that the detached bookmark bar was visible.
Added a new test BookmarkInstantExtendedBrowsertest.DetachedBookmarkBar. Verified that the test fails without my patch and passes with it.

Review URL: https://chromiumcodereview.appspot.com/12217127

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@182165 0039d316-1c4b-4281-b951-d872f2087c98

BUG=None
Review URL: https://codereview.chromium.org/12221167

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@182164 0039d316-1c4b-4281-b951-d872f2087c98

BUG=155269

Review URL: https://chromiumcodereview.appspot.com/12220123

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@182163 0039d316-1c4b-4281-b951-d872f2087c98

When I added the resource provider fence concept, I didn't
extend the read-fence long enough (it currently only protects
the current frame). This extends the read-lock for one frame.

BUG=172995

Review URL: https://chromiumcodereview.appspot.com/12220095

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@182162 0039d316-1c4b-4281-b951-d872f2087c98

ShutdownAndBlock function can be called multiple times safely.
So checking |shutdown_completed_| in OnDisonnected function is not necessary.

BUG=174431
TEST=Checked no crash at sing out.


Review URL: https://chromiumcodereview.appspot.com/12212133

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@182161 0039d316-1c4b-4281-b951-d872f2087c98

BUG=174924
TEST=on lumpy, launch the settings page, and confirm the setting is hidden.

R=nona@chromium.org
TBR=nkostylev@chromium.org
# TBR for just inverting the flag...

Review URL: https://chromiumcodereview.appspot.com/12207143

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@182160 0039d316-1c4b-4281-b951-d872f2087c98

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@182159 0039d316-1c4b-4281-b951-d872f2087c98

This was broken by https://chromiumcodereview.appspot.com/12096096,
which refactored the code causing a 304 response to no longer
resulting in an early return.

I'll write a test in a follow-up CL.

BUG=175795
TEST=VariationsService logging shows that the seed
does not get reset when we get a 304.

Review URL: https://chromiumcodereview.appspot.com/12251003

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@182151 0039d316-1c4b-4281-b951-d872f2087c98

After switching the HudLayer to use skia's font rendering the FontAtlas has
become obsolete. This change removes the FontAtlas and all the related
plumbing.

BUG=172284


Review URL: https://chromiumcodereview.appspot.com/12212099

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@182150 0039d316-1c4b-4281-b951-d872f2087c98

I'm refactoring these APIs in the WebKit repo but can't sanely stage changes to
the API without breaking these OVERRIDE annotations since the interface and
implementations are in separate repositories. The clang plugin knows to not
enforce OVERRIDE for APIs in the WebKit namespace so this won't break that.


BUG=


Review URL: https://chromiumcodereview.appspot.com/12250021

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@182149 0039d316-1c4b-4281-b951-d872f2087c98

This function showed up as hot in the trace and should not have been explicitly implemented as non-inline (pushing the args for the function call costs as much as the copy itself so operator = should *never* not be inlined if it's just doing a normal copy)

BUG=175855


Review URL: https://chromiumcodereview.appspot.com/12250019

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@182148 0039d316-1c4b-4281-b951-d872f2087c98

There is timing issue around initialization, if the InputContext is initialized
after InputMethodManager, Reset function will be called with NULL object proxy.
This cause a crash. If the object proxy is not ready, calling Reset is not
necessary. So just doing nothing is fine.

BUG=174248
TEST=None


Review URL: https://chromiumcodereview.appspot.com/12223084

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@182147 0039d316-1c4b-4281-b951-d872f2087c98

And put them in unnamed namespace.

TEST=browser_tests
TBR=ben@chromium.org


Review URL: https://chromiumcodereview.appspot.com/12220139

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@182146 0039d316-1c4b-4281-b951-d872f2087c98

turns out calling SetOverlayIcon() runs a nested message loop, which
can result in all sorts of problems (and crashes). Since we don't care
about the result I'm moving the call to a worker thread.

BUG=174768
TEST=none
R=ben@chromium.org


Review URL: https://chromiumcodereview.appspot.com/12212053

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@182145 0039d316-1c4b-4281-b951-d872f2087c98

This is preparation to replace |upload_url| of InitiateUpload methods in
DriveServiceInterface by resource id.

BUG=148632
TEST=Ran unit_tests


Review URL: https://chromiumcodereview.appspot.com/12226073

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@182144 0039d316-1c4b-4281-b951-d872f2087c98

This splits the network changed event into two separate events
and changes the events so that they report only the identifiers for
the networks, and not their properties.

Also fixed a bad bug in startDisconnect (was calling connect instead
of disconnect).

Modified the apitest so that we now run each test as a separate
browser test so that the networking stubs are reset between
tests.

BUG=chromium:168713


Review URL: https://chromiumcodereview.appspot.com/12220113

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@182143 0039d316-1c4b-4281-b951-d872f2087c98

happen with some code paths.

BUG=175600
TEST=none
R=sadrul@chromium.org


Review URL: https://chromiumcodereview.appspot.com/12212144

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@182142 0039d316-1c4b-4281-b951-d872f2087c98

BUG=174836

Review URL: https://chromiumcodereview.appspot.com/12212083

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@182141 0039d316-1c4b-4281-b951-d872f2087c98

I would also like to rename it to TouchEditingController since it is not just for selection. But that makes enough changes to the file to render git incapable of recognizing it as a move instead of a new file. So I will do the renaming in a separate CL.

BUG=115237


Review URL: https://chromiumcodereview.appspot.com/12224133

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@182139 0039d316-1c4b-4281-b951-d872f2087c98

It looks up the current preference through DesktopNotificationService
and reports failures when it's stopped.

BUG=161094


Review URL: https://chromiumcodereview.appspot.com/12221127

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@182138 0039d316-1c4b-4281-b951-d872f2087c98

Added a unit-test to BookmarkBarViewTest. It opens a context
menu. The menu is dismissed by clicking the mouse outside of
it. The underlying view gets the mouse event after the
context menu has exited.

See also: 
https://chromiumcodereview.appspot.com/11761027

BUG=171891
TEST=manual, Run: 
out/Debug/interactive_ui_tests --gtest_filter='BookmarkBarViewTest20.*'

Review URL: https://chromiumcodereview.appspot.com/12217158

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@182137 0039d316-1c4b-4281-b951-d872f2087c98

Surface caching is currently broken if any layers in the surface do not
draw themselves fully within the bounds of the surface for any reason
other than being occluded. This includes:
1) Being at non-ideal scale.
2) Being checkerboarded.
3) Missing a texture resource.
4) Having visibleContentRect clipped by something outside the surface.

This disables the feature entirely.

R=jamesr,vangelis
BUG=170713
NOTRY=true

Review URL: https://chromiumcodereview.appspot.com/12210077

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@182136 0039d316-1c4b-4281-b951-d872f2087c98

BUG=152854

Review URL: https://codereview.chromium.org/12250024

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@182135 0039d316-1c4b-4281-b951-d872f2087c98

BUG=175805
TEST=Added to browser_tests OomPriorityManagerTest. Also, manual test (see bug)

Review URL: https://codereview.chromium.org/12221159

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@182134 0039d316-1c4b-4281-b951-d872f2087c98

The problem was caused by auto focusing to the first focusable control when switching pages. To show an overlay, a owning page is shown, very often to the same current one, what is causing setting focus to the first focusable item, and scrolling to the top just before showing an overlay. This patch addresses this issue by not changing focus if the current focused control is on the active page/overlay.

TEST=Follow the bug's repro steps.
BUG=174608

Review URL: https://codereview.chromium.org/12226110

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@182133 0039d316-1c4b-4281-b951-d872f2087c98

BUG=175605
TEST=visual


Review URL: https://chromiumcodereview.appspot.com/12217146

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@182131 0039d316-1c4b-4281-b951-d872f2087c98