| Commit message (Collapse) | Author | Age | Files | Lines |
... | |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
This is an effort to speed up compile and link time, and also minimizing the
size of the intermediary .o files on disk. For example, just moving the
constructor/destructor from the classes in chrome/browser/pref_member.{cc,h}
netted a 368k drop in total .o file size. In aggregate, this shrinks
libbrowser.a by 10 megabytes, and a few odd megabytes on most other chrome .a files.
A lot of this was done before I started harvesting what the most included
symbols were across all of chrome's code. Most of them are in webkit, but
there's plenty in base/ that are used everywhere to keep me busy for several
patches to come.
BUG=none
TEST=none
Review URL: http://codereview.chromium.org/3012001
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@52528 0039d316-1c4b-4281-b951-d872f2087c98
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Previously committed as r52349 and r52336. Related commits: r52367, r52364
and r52343.
Rerunning trybots due to previous trybot breakage.
TEST=none
BUG=none
Review URL: http://codereview.chromium.org/2965015
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@52496 0039d316-1c4b-4281-b951-d872f2087c98
|
|
|
|
|
|
|
|
|
| |
BUG=49123
TEST=Start autocompleting something, check that things line up.
Review URL: http://codereview.chromium.org/2968012
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@52432 0039d316-1c4b-4281-b951-d872f2087c98
|
|
|
|
|
|
| |
Review URL: http://codereview.chromium.org/2982009
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@52367 0039d316-1c4b-4281-b951-d872f2087c98
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
(I accidentally committed a previous version of the patch (the fixed version
was in codereview since I had switched to my mac to address this exact compile
failure.)
BUG=none
TEST=none
TBR=evan
Review URL: http://codereview.chromium.org/2937010
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@52349 0039d316-1c4b-4281-b951-d872f2087c98
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
- Some cleanup to url_request.h
- Remove task.h from headers that don't need it.
- Remove histogram.h from message_loop.h
BUG=none
TEST=none
Review URL: http://codereview.chromium.org/2986002
TBR=erg@chromium.org
Review URL: http://codereview.chromium.org/2969006
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@52343 0039d316-1c4b-4281-b951-d872f2087c98
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
- Some cleanup to url_request.h
- Remove task.h from headers that don't need it.
- Remove histogram.h from message_loop.h
BUG=none
TEST=none
Review URL: http://codereview.chromium.org/2986002
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@52336 0039d316-1c4b-4281-b951-d872f2087c98
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Previously a right-click in omnibox or on a page action would have a
context menu _and_ the omnibox popup. Clicking a page action or star
would mess up the omnibox selection. This brings things in line with
other platforms.
Also fix to not change selection if an omnibox decoration handles the click.
BUG=40145
TEST=Clicking star with omnibox popup up should close popup.
TEST=Clicking star should not change omnibox selection.
TEST=Right-click in omnibox with popup up should close popup.
Review URL: http://codereview.chromium.org/2900006
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@52126 0039d316-1c4b-4281-b951-d872f2087c98
|
|
|
|
|
|
|
|
|
| |
BUG=43602
TEST=Good https:// sites are always green, never blue.
Review URL: http://codereview.chromium.org/2805078
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@51879 0039d316-1c4b-4281-b951-d872f2087c98
|
|
|
|
|
|
|
|
|
|
| |
InMemoryURLIndex.
BUG=19736
TEST=No change in behavior when flag is not used. Should not crash when flag is used.
Review URL: http://codereview.chromium.org/2861041
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@51555 0039d316-1c4b-4281-b951-d872f2087c98
|
|
|
|
|
|
|
|
|
|
|
|
| |
Also fix a bug where we'd keep an extension's old keyword even if it changed
on upgrade/reload.
BUG=46475
BUG=48091
Review URL: http://codereview.chromium.org/2807033
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@51540 0039d316-1c4b-4281-b951-d872f2087c98
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
use them from views and gfx.
Sorry for the new patch on this and not an update. Not sure what happened.
BUG=none
TEST=none
Review URL: http://codereview.chromium.org/2809047
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@51409 0039d316-1c4b-4281-b951-d872f2087c98
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Started out just trying to change PrefService::GetString and ::SetString. This snowballed a little bit. Had to change a bunch of url strings in search_engines/ from wstring to string (some of them may be better off as GURLs, but UTF-8 is a step in the right direction, since that's what GURL uses internally, as well as externally via its setters/getters).
TODO (later patch): things that ask for accepted languages should use std::string, not std::wstring.
BUG=none
TEST=try bots
Review URL: http://codereview.chromium.org/2854015
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@51157 0039d316-1c4b-4281-b951-d872f2087c98
|
|
|
|
|
|
|
|
|
|
|
| |
linux/views.
BUG=none
TEST=none
Review URL: http://codereview.chromium.org/2811023
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@51124 0039d316-1c4b-4281-b951-d872f2087c98
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
attempt caused some flakiness on the new tab warm startup test. The
only difference between this and the last is that I'm leaving some
suspcious code in TabStripModel::InsertTabContents and
Browser::AddTabWithURL. I'll try to isolate why those changes caused
flakiness after I land the rest.
Here's the description:
Moves Browser::AddTypes to TabStripModel. This patch is primarily
cleanup before I fix 29933, but has a couple of interesting bits
beyond the enum change:
. AddTabContents now supports adding pinned.
. Converted InsertTabContents call in extensionstabmodule to pass in
nothing (Rafael said old code was wrong).
BUG=29933
TEST=none
TBR=ben@chromium.org
Review URL: http://codereview.chromium.org/2866024
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@50981 0039d316-1c4b-4281-b951-d872f2087c98
|
|
|
|
|
|
|
|
|
|
|
|
| |
We are now doing more than DNS pre-resolution, and want
to do TCP/IP pre-warming. This change uses the updated
class names etc., and tries to use Predictor as the
common element in most names.
r=mbelshe
Review URL: http://codereview.chromium.org/2866026
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@50951 0039d316-1c4b-4281-b951-d872f2087c98
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Move these files from cocoa/ to cocoa/location_bar/:
autocomplete_text_field.h
autocomplete_text_field.mm
autocomplete_text_field_cell.h
autocomplete_text_field_cell.mm
autocomplete_text_field_cell_unittest.mm
autocomplete_text_field_editor.h
autocomplete_text_field_editor.mm
autocomplete_text_field_editor_unittest.mm
autocomplete_text_field_unittest.mm
autocomplete_text_field_unittest_helper.h
autocomplete_text_field_unittest_helper.mm
location_bar_view_mac.h
location_bar_view_mac.mm
location_bar_view_mac_unittest.mm
BUG=none
TEST=none
Review URL: http://codereview.chromium.org/2807024
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@50873 0039d316-1c4b-4281-b951-d872f2087c98
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
The DNS prefetching system has been extended to handle TCP/IP
preconnection. As a result, class names and function names will
be migrated to variations on Predictor* and Predict* to better
indicate functionality.
This change is only meant to include name changes, comment
updates, and no semantic changes.
r=mbelshe
Review URL: http://codereview.chromium.org/2813031
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@50792 0039d316-1c4b-4281-b951-d872f2087c98
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
cleanup before I fix 29933, but has a couple of interesting bits
beyond the enum change:
. AddTabContents now supports adding pinned.
. Nuked duplicate code in Browser::addTabWithURL that invoked
wasHidden on the TabContents. This code is already in TabStripModel.
. Moved code for setting visibility of tabcontents from
TabStripModel::AddTabContents to InsertTabContentsAt. Since everything
ends up in InsertTabContentsAt it should be there.
. Converted InsertTabContents call in extensionstabmodule to pass in
nothing (Rafael said old code was wrong).
BUG=29933
TEST=none
Review URL: http://codereview.chromium.org/2863021
TBR=sky@chromium.org
Review URL: http://codereview.chromium.org/2849025
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@50776 0039d316-1c4b-4281-b951-d872f2087c98
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Unlike Windows, in Linux, we do not draw the strings in autocomplete popup by ourselves. Instead, we rely on Pango to draw the strings. We only need to pass the text, the attribute, and the width to Pango, and Pango will draw the text correctly (taking care of visual runs for bidi text).
The only small problems are that:
when the UI is RTL, for pure English text, we need to insert LRE mark so that the ending punctuation is drawn correctly as is (not flipped), for example "....(choose location)" will not be drawn as "(....(choose location" in RTL UI.
Also, for some (not all) pure English text, when eliding, the ellipsis appears at the very left of the string in RTL UI. I do not know the reason of such inconsistency, but looks like adding LRE mark to pure English also forces the ellipsis appears at the right of the string.
BUG=43800
TEST=Omnibox should elide long URLs, mixed LTR/RTL text, etc. correctly in both LTR and RTL locales
Review URL: http://codereview.chromium.org/2824027
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@50773 0039d316-1c4b-4281-b951-d872f2087c98
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
cleanup before I fix 29933, but has a couple of interesting bits
beyond the enum change:
. AddTabContents now supports adding pinned.
. Nuked duplicate code in Browser::addTabWithURL that invoked
wasHidden on the TabContents. This code is already in TabStripModel.
. Moved code for setting visibility of tabcontents from
TabStripModel::AddTabContents to InsertTabContentsAt. Since everything
ends up in InsertTabContentsAt it should be there.
. Converted InsertTabContents call in extensionstabmodule to pass in
nothing (Rafael said old code was wrong).
BUG=29933
TEST=none
Review URL: http://codereview.chromium.org/2863021
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@50752 0039d316-1c4b-4281-b951-d872f2087c98
|
|
|
|
|
|
|
|
|
| |
BUG=none
TEST=none
Review URL: http://codereview.chromium.org/2831023
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@50743 0039d316-1c4b-4281-b951-d872f2087c98
|
|
|
|
|
|
|
|
|
|
|
|
| |
- Rename Canvas to CanvasSkia.
- Create a subclass Canvas that inherits from CanvasSkia for compatibility.
BUG=none
TEST=none
Review URL: http://codereview.chromium.org/2862025
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@50664 0039d316-1c4b-4281-b951-d872f2087c98
|
|
|
|
|
|
|
|
|
|
|
|
| |
in history is for bookmark-related stuff (the bookmarks system is still wstring-based).
The substantial change here is in RTL to make a string16 variant of the functions and in changing the WordIterator to use string16 (this cleaned up some weird utf-32 code).
TEST=none
BUG=none
Review URL: http://codereview.chromium.org/2808017
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@50597 0039d316-1c4b-4281-b951-d872f2087c98
|
|
|
|
|
|
| |
Review URL: http://codereview.chromium.org/2876002
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@50590 0039d316-1c4b-4281-b951-d872f2087c98
|
|
|
|
|
|
|
|
|
|
|
| |
omnibox.
BUG=44736
TEST=see bug
Review URL: http://codereview.chromium.org/2809030
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@50544 0039d316-1c4b-4281-b951-d872f2087c98
|
|
|
|
|
|
|
|
|
|
|
|
| |
- Don't enter "keyword" mode unless the extension is enabled and allowed in
incognito.
- Show extension suggest results for incognito windows.
BUG=46933
Review URL: http://codereview.chromium.org/2810020
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@50498 0039d316-1c4b-4281-b951-d872f2087c98
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
left-clicking).
Eat middle-mouse clicks on the omnibox to avoid tripping a CRichEditCtrl bug.
Original patch by Jared Wein (see http://codereview.chromium.org/2241003 ), r=me.
BUG=6873, 30134
TEST=Middle-click the omnibox, then try to click it (or any other UI element), and make sure the click has an effect. Select a portion of the URL and right-click on the omnibox. Notice that the selection does not change. Clear the selection and right-click on the omnibox and notice that the URL is automatically selected.
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@50420 0039d316-1c4b-4281-b951-d872f2087c98
|
|
|
|
|
|
|
|
|
|
|
|
| |
- Moved the const to autocomplete_edit_view_win.cc
- Deleted the file.
BUG=None
TEST=trybots
Review URL: http://codereview.chromium.org/2864013
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@50417 0039d316-1c4b-4281-b951-d872f2087c98
|
|
|
|
|
|
|
|
|
|
|
|
| |
- Remove some of the boilerplate strings.
- Allow the description field to be fully customizable and stylable.
BUG=46478
BUG=46479
Review URL: http://codereview.chromium.org/2849015
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@50411 0039d316-1c4b-4281-b951-d872f2087c98
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
canonicalization). Nearly every caller was already doing this.
This in turn allows us to do better fixup/canonicalization of view-source: URLs. We now convert "view-source:google.com" into "view-source:http://google.com/".
With a few changes scattered through the omnibox code, this also means we can do HTTP-stripping on view-source: URLs, and support the user typing in things like the case above. This also fixes some weirdness where if you tried to type something starting with "view-source:", the What You Typed match in the dropdown would show only a scheme, or a scheme plus "http:", in some cases.
BUG=46612
TEST="view-source:google.com" should work.
Review URL: http://codereview.chromium.org/2817011
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@50290 0039d316-1c4b-4281-b951-d872f2087c98
|
|
|
|
|
|
|
|
|
|
| |
navigate you to an error page.
BUG=46480
Review URL: http://codereview.chromium.org/2819006
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@50178 0039d316-1c4b-4281-b951-d872f2087c98
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Round off the omnibox corners like adjacent toolbar buttons. Round
off the popup corners reflection-like. Put back the slight gap
between them.
BUG=45762
TEST=Above stuff. Findbar field should not be rounded.
Review URL: http://codereview.chromium.org/2863007
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@50169 0039d316-1c4b-4281-b951-d872f2087c98
|
|
|
|
|
|
|
| |
TBR=thakis@chromium.org
Review URL: http://codereview.chromium.org/2835007
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@50000 0039d316-1c4b-4281-b951-d872f2087c98
|
|
|
|
| |
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@49984 0039d316-1c4b-4281-b951-d872f2087c98
|
|
|
|
|
|
|
|
|
|
|
|
| |
As discussed on irc, we're getting rid of const NSObjects, and we're making all properties nonatomic.
const: All of cocoa takes nonconst NSObjects, and passing a const NSString to a function that takes NSString is a const violation. gcc doesn't complain about this, but clang intentionally does. constness is usually done via immutable base classes and mutable subclasses in cocoa anyway, so getting rid of const isn't that bad. Also, we don't really have a choice.
nonatomic: Some of our classes have custom setters that are not @synchronized. If the @property for that is not non-atomic, then the interface claims that the method is synchronized but the implementation actually isn't. That's a bug. gcc happens not to complain about this, but clang does. Since we shouldn't need atomic properties anywhere, the simple rule is now to just make all properties nonatomic.
Review URL: http://codereview.chromium.org/2769014
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@49808 0039d316-1c4b-4281-b951-d872f2087c98
|
|
|
|
|
|
|
|
|
|
|
| |
silly because it would capture it in UTF16, then convert to wide, send it to
the browser, then convert it to UTF-8 for FTS.
TEST=none
BUG=none
Review URL: http://codereview.chromium.org/2714012
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@49594 0039d316-1c4b-4281-b951-d872f2087c98
|
|
|
|
|
|
|
|
|
|
|
| |
directory.
BUG=None
TEST=trybots
Review URL: http://codereview.chromium.org/2657003
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@49024 0039d316-1c4b-4281-b951-d872f2087c98
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Snow Leopard deprecated -[NSImage setFlipped:] for good reasons. This
implements a method similar to the new
-[NSImage drawInRect:*:respectFlipped:] method, which allows images to
be drawn so that they always look right visually regardless of the
-isFlipped status of the view being drawn.
BUG=38943
TEST=Autocomplete, download, and reload button look right.
Review URL: http://codereview.chromium.org/2587003
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@48954 0039d316-1c4b-4281-b951-d872f2087c98
|
|
|
|
|
|
|
|
|
|
|
|
| |
especially helps when the scheme has also been stripped, as it makes the hostname look less unbalanced. We're careful to avoid stripping the slash when doing so would confuse the omnibox.
This also moves to more aggressive stripping and/or unescaping in several places. In general, it seems like we should be as aggressive as is feasible.
BUG=43587
TEST=Visit google.com. There should be no slash in the address bar.
Review URL: http://codereview.chromium.org/2389002
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@48691 0039d316-1c4b-4281-b951-d872f2087c98
|
|
|
|
|
|
|
|
|
|
|
| |
I've run into a couple of times this week when I needed one of these, for two different types besides bool. Time to fix the TODO.
TEST=trybots FTW, and built locally.
BUG=none
Review URL: http://codereview.chromium.org/2394001
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@48644 0039d316-1c4b-4281-b951-d872f2087c98
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
features to work on, but it's in a usable state.
When an extension is installed that specifies an omnibox keyword in its manifest, we add that keyword to the user's list of Search Engines. The user can then edit this keyword later.
I'm leveraging most of the original search engine keyword code. An extension keyword has a special URL that identifies it as an extension keyword. There is some special case code to treat these keywords slightly differently throughout
the omnibox code.
BUG=38884
Review URL: http://codereview.chromium.org/2078021
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@48503 0039d316-1c4b-4281-b951-d872f2087c98
|
|
|
|
|
|
|
|
|
| |
BUG=45271
TEST=none
Review URL: http://codereview.chromium.org/2343001
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@48499 0039d316-1c4b-4281-b951-d872f2087c98
|
|
|
|
|
|
|
|
|
|
| |
members on the AutocompleteEdit.
BUG=none
TEST=none
Review URL: http://codereview.chromium.org/2182002
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@48287 0039d316-1c4b-4281-b951-d872f2087c98
|
|
|
|
|
|
|
|
|
|
|
|
| |
The caller can use the much shorter "gfx::Rect(size())", since gfx::Rect
has a constructor that just takes a Size.
BUG=none
TEST=none
Review URL: http://codereview.chromium.org/2204001
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@48283 0039d316-1c4b-4281-b951-d872f2087c98
|
|
|
|
|
|
|
|
|
|
| |
AutocompleteEditViewWin::HandleKeystroke(). Restore keyboard layout after DefWindowProc().
BUG=14049
TEST=open LTR and RTL chrome, type in pure Hebrew, pure English and bidi characters inside omnibox, press HOME/END/shift-HOME/shift-END, keyboard layout should not be changed.
Review URL: http://codereview.chromium.org/2143006
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@48190 0039d316-1c4b-4281-b951-d872f2087c98
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
The new work will take a very different approach (using a component extension,
similar to the bookmark manager).
This basically reverts the following commits:
- http://src.chromium.org/viewvc/chrome?view=rev&revision=20595
(http://codereview.chromium.org/155067)
- http://src.chromium.org/viewvc/chrome?view=rev&revision=19918
(http://codereview.chromium.org/155051)
- http://src.chromium.org/viewvc/chrome?view=rev&revision=19906
(http://codereview.chromium.org/150207)
BUG=173
TEST=everything still builds
Review URL: http://codereview.chromium.org/2156003
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@48162 0039d316-1c4b-4281-b951-d872f2087c98
|
|
|
|
|
|
|
|
|
| |
BUG=44232
TEST=cut all the contents from the omnibox. The url (along with the protocol) should be on both PRIMARY and CLIPBOARD selections. The omnibox should now lay empty.
Review URL: http://codereview.chromium.org/2108014
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@47847 0039d316-1c4b-4281-b951-d872f2087c98
|
|
|
|
|
|
|
|
|
|
| |
location bar width, unless eliding to that would result in a width of less than 150 px.
BUG=42856
TEST=Visit https://www.barbican.org.uk/eticketing/index.asp and make the window smaller. The EV bubble should shrink, eliding in middle, until it hits a minimum size.
Review URL: http://codereview.chromium.org/2084012
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@47819 0039d316-1c4b-4281-b951-d872f2087c98
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
The TODO on AutocompleteEditViewMac about the popup needing the field
for placement and font got half done. The placement part just seemed
too intrusive to be worthwhile, but the font part seemed reasonable to
get rid of.
The others are all either not sensible anymore, or were addressed
without being removed, or have been in there long enough that if
nobody has complained by now, their subject must not matter.
BUG=none
TEST=none
Review URL: http://codereview.chromium.org/2127009
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@47732 0039d316-1c4b-4281-b951-d872f2087c98
|