Patch fails otherwise on non-Windows platforms.
Review URL: http://codereview.chromium.org/14478

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@7093 0039d316-1c4b-4281-b951-d872f2087c98

Created cr-mac entries for some tests we were failing. The tree is a copy of
  the webkit data (we match it), but the console output is different, so that
  matches the cr-win version.  but it's a new file since it combo that doesn't
  exist elsewhere.
Lint passes.
Review URL: http://codereview.chromium.org/14479

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@7092 0039d316-1c4b-4281-b951-d872f2087c98

set_store to ClearStore and properly remove an observer I added. And I'm
disabling a test that needs more investigation.

Makes the interactive bookmark bar view tests faster by disabling
saving. This is why BookmarkBarViewTest9 keeps timing out.

BUG=none
TEST=none

Review URL: http://codereview.chromium.org/14471

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@7091 0039d316-1c4b-4281-b951-d872f2087c98

Google Chrome Channel Changer - Minor update to add stable channel, some updates to text description.

Review URL: http://codereview.chromium.org/14455

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@7090 0039d316-1c4b-4281-b951-d872f2087c98

necessary for the KURL.h refactor since it's becoming more and more difficult
to do my hack where the old types are defined to a different name.

This also adds deep copy to GKURL.
Review URL: http://codereview.chromium.org/14161

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@7089 0039d316-1c4b-4281-b951-d872f2087c98

TBR=evan

Review URL: http://codereview.chromium.org/14166


git-svn-id: svn://svn.chromium.org/chrome/trunk/src@7088 0039d316-1c4b-4281-b951-d872f2087c98

I tested locally with gcc 4.3, but gcc 4.2 behaves differently and fails.
Review URL: http://codereview.chromium.org/14165

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@7087 0039d316-1c4b-4281-b951-d872f2087c98

(forgot to include this file)


git-svn-id: svn://svn.chromium.org/chrome/trunk/src@7086 0039d316-1c4b-4281-b951-d872f2087c98

The tests are currently marked as failing only WIN, when really they fail on LINUX as well. This rebaseline fixes them on linux. See issue 14434 for reasoning behind the rebaseline.

Review URL: http://codereview.chromium.org/14145

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@7085 0039d316-1c4b-4281-b951-d872f2087c98

Review URL: http://codereview.chromium.org/14480


git-svn-id: svn://svn.chromium.org/chrome/trunk/src@7084 0039d316-1c4b-4281-b951-d872f2087c98

Review URL: http://codereview.chromium.org/14481

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@7083 0039d316-1c4b-4281-b951-d872f2087c98

TBR=estade

Review URL: http://codereview.chromium.org/14482


git-svn-id: svn://svn.chromium.org/chrome/trunk/src@7082 0039d316-1c4b-4281-b951-d872f2087c98

when we ask it to draw a scrollbar button. It's not clear if it expects
the background to already be filled in, or if we are brearking it by
giving it WebKit metrics. Either way, it's messing up our pixeltest
baselines with undefined pixels so we paint seafoam-green under
scrollbar buttons before GTK draws.

Review URL: http://codereview.chromium.org/14477


git-svn-id: svn://svn.chromium.org/chrome/trunk/src@7081 0039d316-1c4b-4281-b951-d872f2087c98

Review URL: http://codereview.chromium.org/14164


git-svn-id: svn://svn.chromium.org/chrome/trunk/src@7080 0039d316-1c4b-4281-b951-d872f2087c98

Review URL: http://codereview.chromium.org/14466

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@7079 0039d316-1c4b-4281-b951-d872f2087c98

Review URL: http://codereview.chromium.org/14163

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@7078 0039d316-1c4b-4281-b951-d872f2087c98

Removed the use of the ConsoleMessage::operator== as it has been responsible for crashes while running the pagecycler tests. Console messages which are the same will now appear as separate messages in the inspector instead of apearing as a single message with a count indicator.
Review URL: http://codereview.chromium.org/14458

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@7077 0039d316-1c4b-4281-b951-d872f2087c98

BUG=2186
TEST=none


git-svn-id: svn://svn.chromium.org/chrome/trunk/src@7076 0039d316-1c4b-4281-b951-d872f2087c98

Review URL: http://codereview.chromium.org/14470

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@7075 0039d316-1c4b-4281-b951-d872f2087c98

The rankings module now works with any list, not just
the first one. This is part of the support needed for
new eviction algorithms.

note: Rankings::CheckList() is basically the code that
was implementing the old Rankings::SelfCheck() (moved
with this cl).

The disk cache behavior is not changing with this cl.

Review URL: http://codereview.chromium.org/14141

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@7074 0039d316-1c4b-4281-b951-d872f2087c98

The eviction age analysis can be simplified (and it will be closer to what
we really want to know) if we don't add the age of every entry, and instead
use only the first evicted entry per run.

Review URL: http://codereview.chromium.org/14448

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@7073 0039d316-1c4b-4281-b951-d872f2087c98

Review URL: http://codereview.chromium.org/14459

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@7072 0039d316-1c4b-4281-b951-d872f2087c98

On special request from someone trying to purge python2.2 from code indexed
internally at Google.


git-svn-id: svn://svn.chromium.org/chrome/trunk/src@7071 0039d316-1c4b-4281-b951-d872f2087c98

Review URL: http://codereview.chromium.org/14468

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@7070 0039d316-1c4b-4281-b951-d872f2087c98

 - It is a 'planned' crash in a CHECK(false)
 - Now is a bug because the user can trigger it by alt-click in a  <a href=data:xxxxx >

The codepath in question skips all the checks we do in the downloadmanager and on the UI command handlers.

BUG=5112

Review URL: http://codereview.chromium.org/12968

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@7069 0039d316-1c4b-4281-b951-d872f2087c98

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@7068 0039d316-1c4b-4281-b951-d872f2087c98

This CL fixes issue 4376: Press "Tab" key to navigate - Text focus is showing at wrong place in Hebrew and Arabic.
when drawing focus rectangle for CheckBox,the left point of the rectangle should be mirrored for RTL text.

http://crbug.com/4376

TEST:
1. Launch RTL Chome (for ex: set language as Hebrew).
2. Go to > Customize and Control Google Chrome > Options > Under the Hood or any other Tab
3. Press tab key to navigate thought text options
4. The focus is properly set on the text. (without this fix, the focus is not properly set on the text).

 
Review URL: http://codereview.chromium.org/14452

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@7067 0039d316-1c4b-4281-b951-d872f2087c98

(The mac version of this file was modified in webkit r38631.)

Review URL: http://codereview.chromium.org/14451


git-svn-id: svn://svn.chromium.org/chrome/trunk/src@7066 0039d316-1c4b-4281-b951-d872f2087c98

Annotate why tests fail and delete their expected results.  For two canvas tests, they only differ by font metrics, so add a new baseline for them.

Review URL: http://codereview.chromium.org/14144

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@7065 0039d316-1c4b-4281-b951-d872f2087c98

Fix Justin Haygood's last name (sorry!).  Patch by Mohamed Mansour, r=me; see http://codereview.chromium.org/14148

BUG=5558

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@7064 0039d316-1c4b-4281-b951-d872f2087c98

using layered windows incorrectly. According to the docs if you invoke
SetLayeredWindowAttributes, then you shouldn't invoke
UpdateLayeredWindow (unless you remove then add the WS_EX_LAYERED
style bit). My approach here is to disable the backing buffer for the
info bubble. I had problems toggling the WS_EX_LAYERED at the end up
the animation. The text field would not paint when I did this...

BUG=5351
TEST=bookmark a page. When the bubble appears press shift-tab and make
  sure you see a focus rect around the remove link.

Review URL: http://codereview.chromium.org/14131

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@7063 0039d316-1c4b-4281-b951-d872f2087c98

saving. This is why BookmarkBarViewTest9 keeps timing out.

BUG=none
TEST=none

Review URL: http://codereview.chromium.org/14449

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@7062 0039d316-1c4b-4281-b951-d872f2087c98

Review URL: http://codereview.chromium.org/14462

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@7061 0039d316-1c4b-4281-b951-d872f2087c98

Review URL: http://codereview.chromium.org/14159

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@7060 0039d316-1c4b-4281-b951-d872f2087c98

Very frustrating.  We seem to be seeing subtle variations in the stacks across runs.  No clue what's going on here.

TBR=brettw

Review URL: http://codereview.chromium.org/14463

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@7059 0039d316-1c4b-4281-b951-d872f2087c98

Lint passes.
Review URL: http://codereview.chromium.org/14157

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@7058 0039d316-1c4b-4281-b951-d872f2087c98

Lint passes.
Review URL: http://codereview.chromium.org/14156

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@7057 0039d316-1c4b-4281-b951-d872f2087c98

Review URL: http://codereview.chromium.org/14154

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@7056 0039d316-1c4b-4281-b951-d872f2087c98

Review URL: http://codereview.chromium.org/14153

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@7055 0039d316-1c4b-4281-b951-d872f2087c98

TBR=ager@chromium.org
Review URL: http://codereview.chromium.org/14152

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@7054 0039d316-1c4b-4281-b951-d872f2087c98