summaryrefslogtreecommitdiffstats
Commit message (Collapse)AuthorAgeFilesLines
* Roll WebKit to r66934 on behalf of zmo@chromium.orgvangelis@chromium.org2010-09-081-1/+1
| | | | git-svn-id: svn://svn.chromium.org/chrome/trunk/src@58780 0039d316-1c4b-4281-b951-d872f2087c98
* Break out of nested loops in AutoFillManager::GetAutoFillSuggestionsisherman@chromium.org2010-09-081-1/+1
| | | | | | | | | BUG=none TEST=none Review URL: http://codereview.chromium.org/3355012 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@58779 0039d316-1c4b-4281-b951-d872f2087c98
* New authorization framework for sync. johnnyg@chromium.org2010-09-0881-2035/+1314
| | | | | | | | | | | | | | | | | | To quote chron's original patch (http://codereview.chromium.org/3148036/show) <blockquote> This patch removes: authenticator.cc, auth_watcher.cc removes calls to user_settings.cc, removes an authenticate PB request to the server, and moves token storage into the Chrome TokenService. This patch introduces the SigninManager, which is an interim solution for user management prior to moving the system into chrome. Other changes include removing the dependency on the sync backend to be running while the sync wizard is intially displayed. This means that the backend can be brought up in response to credentials becoming available. The backend now is always provided credentials on startup. If an auth error occurs, it propogates it up via a notification. Some event handlers were removed and streamlined for more straightforward sync system startup. </blockquote> BUG=51001, 50293, 35158 TEST=Unit tests && Start up sync, log in, log out, run with expired credentials, run with new gaia credentials, run with gaia credentials updated while system is syncing. Try logging in with incorrect username. Trigger CAPTCHA. Try logging out and in repeatedly. Check about:sync works. Try going offline and back online again. Expire gaia credentials and try renewing it with the UI dialog. Review URL: http://codereview.chromium.org/3305003 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@58778 0039d316-1c4b-4281-b951-d872f2087c98
* gtk: Use CHROMEGTK_CALLBACK in PrintDialogGtk and move a thunk implementationjhawkins@chromium.org2010-09-082-11/+11
| | | | | | | | | | | to the source file. BUG=none TEST=none Review URL: http://codereview.chromium.org/3306012 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@58777 0039d316-1c4b-4281-b951-d872f2087c98
* Implement ISimpleDOM COM interface in BrowserAccessibility. Also updatectguil@chromium.org2010-09-083-71/+543
| | | | | | | | | | | | | | a few IAccessible2 functions with some of the new attributes that are available now. BUG=48185 TEST=None Committed: http://src.chromium.org/viewvc/chrome?view=rev&revision=58491 Review URL: http://codereview.chromium.org/3219009 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@58776 0039d316-1c4b-4281-b951-d872f2087c98
* Fix Firefox data import for Mac.mirandac@chromium.org2010-09-072-2/+8
| | | | | | | | | | | | | r58258 broke Firefox import on the Mac by trying to use a value from g_browser_process, which is not accessible in Mac out-of-process import. Because this value isn't used or needed for the Mac build, don't even create it in the Mac version. BUG=54265 TEST=Firefox import works on Mac again. Review URL: http://codereview.chromium.org/3341016 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@58775 0039d316-1c4b-4281-b951-d872f2087c98
* Add locking around store_birthday to ensure atomic overwrites and avoid racetim@chromium.org2010-09-072-3/+14
| | | | | | | | | | | | | | conditions. BUG=50217 TEST=Remove suppressions in chrome/test/data/valgrind/sync_unit_tests.gtest-tsan.txt, then run sync_unit_tests with tsan to confirm no race conditions involving store_birthday. Original patch by: zea@chromium.org Original review: http://codereview.chromium.org/3333011/show Review URL: http://codereview.chromium.org/3340008 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@58774 0039d316-1c4b-4281-b951-d872f2087c98
* Delayed event support in JingleThread.sergeyu@chromium.org2010-09-072-23/+86
| | | | | | | | | BUG=54347 TEST=unittests Review URL: http://codereview.chromium.org/3333014 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@58773 0039d316-1c4b-4281-b951-d872f2087c98
* Add a VideoDecodeContext that provides resources for a VideoDecodeEnginehclam@chromium.org2010-09-076-0/+208
| | | | | | | | | | | Also define a Gles2VideoDecodeContext to be used in the chrome renderer. BUG=53714 TEST=none Review URL: http://codereview.chromium.org/3233003 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@58772 0039d316-1c4b-4281-b951-d872f2087c98
* Changes the max width for bookmark menus back to 400 and leaves themsky@chromium.org2010-09-074-9/+19
| | | | | | | | | | | | at 800 for the wrench menu. Apparently folks don't like wide bookmark menus. BUG=50653 TEST=see bug Review URL: http://codereview.chromium.org/3317011 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@58771 0039d316-1c4b-4281-b951-d872f2087c98
* Mac: Fix window shadow when compositor is active.thakis@chromium.org2010-09-071-3/+52
| | | | | | | | | BUG=53382 TEST=Enable compositor. Window shadow should look normal, even after resizing the window. The shadow of other windows (prefs, bubbles, etc) should look like it did before. Review URL: http://codereview.chromium.org/3360014 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@58770 0039d316-1c4b-4281-b951-d872f2087c98
* Revert 58768 - Adding the sync-notification-host command line parameter to ↵rsimha@chromium.org2010-09-0724-200/+141
| | | | | | | | | | | | | | | | | | | the integration tests. The sync integration tests currently access the network to communicate with the notification server. We need to move to a model where all server accesses are stubbed out by local servers. This is the first step towards implementing a local notification server. BUG=53933,53931 TEST=sync_integration_tests Review URL: http://codereview.chromium.org/3326007 TBR=rsimha@chromium.org Review URL: http://codereview.chromium.org/3318016 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@58769 0039d316-1c4b-4281-b951-d872f2087c98
* Adding the sync-notification-host command line parameter to the integration ↵rsimha@chromium.org2010-09-0724-141/+200
| | | | | | | | | | | | | | | | tests. The sync integration tests currently access the network to communicate with the notification server. We need to move to a model where all server accesses are stubbed out by local servers. This is the first step towards implementing a local notification server. BUG=53933,53931 TEST=sync_integration_tests Review URL: http://codereview.chromium.org/3326007 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@58768 0039d316-1c4b-4281-b951-d872f2087c98
* In OnEnableSpdy(bool), use "npn,force-alt-protocol" so we could use ↵lzheng@chromium.org2010-09-071-1/+1
| | | | | | | | | | alternate protocol to test spdy number for http. TEST=none BUG=none Review URL: http://codereview.chromium.org/3315008 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@58766 0039d316-1c4b-4281-b951-d872f2087c98
* Adds a dependency on chrome.gyp:crash_service from ↵slightlyoff@chromium.org2010-09-071-0/+3
| | | | | | | | | | | chrome_frame.gyp:chrome_frame_tests since the tests refuse to start without the crash service built. TEST=clobber, build chrome_frame_tests, see if they run BUG=None Review URL: http://codereview.chromium.org/3295004 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@58765 0039d316-1c4b-4281-b951-d872f2087c98
* Adding a sample for the extensions idle API.kurrik@chromium.org2010-09-0710-2/+237
| | | | | | | | | | | | Rendered the samples directory for the new sample. News examples order changed in samples.(json|html) due to previous CL's sorting fix. BUG=None TEST=None Review URL: http://codereview.chromium.org/3315010 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@58764 0039d316-1c4b-4281-b951-d872f2087c98
* Clean up benchmarks and pop warning message on error triggered by redirect.lzheng@chromium.org2010-09-071-1/+6
| | | | | | | | | | | | Clean up benchmarks[] between tests. Otherwise, if there is a bad url, it will stuck there forever till we result the extension. Pop out a warning if the url is redirected and thus could not find an entry in benchmarks[]. This should have only to the parent frames since in script.js, we don't post event for subframes result (window.parent != window). TEST=none BUG=none Review URL: http://codereview.chromium.org/3318015 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@58763 0039d316-1c4b-4281-b951-d872f2087c98
* Add "?" to query strings if original url has no query string.lzheng@chromium.org2010-09-073-11/+26
| | | | | | | | BUG=none TEST=protocol_manager_unittest.cc Review URL: http://codereview.chromium.org/3244005 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@58762 0039d316-1c4b-4281-b951-d872f2087c98
* Add mouse/keyboard event support to Chromoting client (Pepper and X11).garykac@chromium.org2010-09-077-6/+49
| | | | | | | | | BUG=50247 TEST=none Review URL: http://codereview.chromium.org/3341005 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@58761 0039d316-1c4b-4281-b951-d872f2087c98
* Change the display of chrome://net-internals/#data.eroman@chromium.org2010-09-074-33/+121
| | | | | | | | | The help text is no longer inline, and it shows the number of events that have been captured so far. BUG=53168 Review URL: http://codereview.chromium.org/3351013 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@58760 0039d316-1c4b-4281-b951-d872f2087c98
* Adds the following to CancelableRequestConsumer:sky@chromium.org2010-09-073-0/+72
| | | | | | | | | | | | | | | | | | GetClientDataForCurrentRequest: The particular use case I need this for is that I'm using the same CancelableRequestConsumer for multiple sources so that I don't know whic CancelableRequestProvider to pass in to GetClientData. None-the-less it seems a lot easier to use this when always get the provider and handle. GetFirstHandleForClientData: For this one I know I want to cancel a request for a given client data but don't know the associated handle. BUG=none TEST=none Review URL: http://codereview.chromium.org/3311012 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@58759 0039d316-1c4b-4281-b951-d872f2087c98
* [GTK] change workaround for weird RTL label alignmentestade@chromium.org2010-09-075-13/+53
| | | | | | | | | | | | | | | When a GtkLabel has line wrapping and an integer size request, the text block is left aligned even in RTL locales. It will wrap at the left (which is correct) and be right justified (also correct) but is flush with the left of the widget allocation instead of the right of the widget allocation. seems not to occur when the size is controled via gtk_label_set_width_chars() instead of gtk_widget_set_size_request(). (also goes away when line wrapping is turned off) TODO: file a reduction upstream. BUG=52857 TEST=launch chrome in Hebrew and look at the under the hood tab of options. Checkbox labels should look correct Review URL: http://codereview.chromium.org/3355007 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@58758 0039d316-1c4b-4281-b951-d872f2087c98
* [Chrome OS] Make Chrome call session_manager to trigger exitcmasone@google.com2010-09-072-1/+13
| | | | | | | | | | | | | On Chromium OS, instead of letting the browser exit directly, we want to allow another process to trigger and watch shutdown, because otherwise we have no way of dealing with chrome hanging on exit. BUG=chromium-os:5395 TEST=Install chrome on a device, log in and log out. Look at /var/log/session_manager to determine that it triggered the browser exit. Review URL: http://codereview.chromium.org/3112012 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@58757 0039d316-1c4b-4281-b951-d872f2087c98
* Added GPU stat for whether a GPU can report context lost.apatrick@chromium.org2010-09-0725-186/+339
| | | | | | | | | | | | | | | On Windows it checks to see if the IDirect3D9 object supports Vista and later features (meaning it won't routinely report lost contexts). On Linux and Mac we can just check if its EGL versus some other GL that doesn't report lost contexts. I routed the stats to the renderer process so webkit code can query whether lost contexts are likely. I didn't wire up breakpad yet. TEST=try BUG=52318 Review URL: http://codereview.chromium.org/3149016 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@58755 0039d316-1c4b-4281-b951-d872f2087c98
* Reset the desktop notifications settings to default.tfarina@chromium.org2010-09-073-0/+12
| | | | | | | | | | | | When the user clicks in "Reset to default" button on the Preferences/Options dialog, we should reset the desktop notifications settings to the default default option, like we already do for the other content settings. BUG=49656 TEST=see bug. Review URL: http://codereview.chromium.org/3332007 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@58754 0039d316-1c4b-4281-b951-d872f2087c98
* Simple change to the focus-stealing behaviour of CF in response to ↵twiz@google.com2010-09-071-1/+8
| | | | | | | | | | | | | WM_NOTIFYPARENT messages. The hosted CF instance may have spawned a pop-up window. We do not want that pop-up window to lose focus/or dismiss itself when interacting with the CF instance as a result of the SetFocus(...) call here. To prevent this problem, we compare the root owner windows of the focus and plugin windows, instead of just testing !IsChild. Reviewer note: The semantics of comparing root owner windows are slightly different from before. Will this modified test be safe for the purposes of the original test? BUG=None TEST=None Review URL: http://codereview.chromium.org/2063009 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@58749 0039d316-1c4b-4281-b951-d872f2087c98
* Roll webkit from r66862 to r66893.zmo@google.com2010-09-071-1/+1
| | | | git-svn-id: svn://svn.chromium.org/chrome/trunk/src@58748 0039d316-1c4b-4281-b951-d872f2087c98
* Clean up tabbed content settings pages.estade@chromium.org2010-09-071-16/+1
| | | | | | | | | | | Remove errant buttons and change a <span> to an <h4>. BUG=none TEST=manual Review URL: http://codereview.chromium.org/3298016 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@58747 0039d316-1c4b-4281-b951-d872f2087c98
* This CL contains the following changes:sanjeevr@chromium.org2010-09-078-63/+186
| | | | | | | | | | | | | Added "location" and "drivername" tags for Windows printers. Saved a hash of all tags with the printer. Checked for this during printer update and reuploaded new tags if tags have changed. All proxy generated tags have a special __cp__ prefix. BUG=None. TEST=Test with changed printer tags. The proxy should make an update request with the new tags. Review URL: http://codereview.chromium.org/3342015 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@58746 0039d316-1c4b-4281-b951-d872f2087c98
* Fixes bookmark menus to deal with multiple menus sharing the samesky@chromium.org2010-09-072-48/+123
| | | | | | | | | | | mnemonic. BUG=49728 TEST=see bug Review URL: http://codereview.chromium.org/3325015 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@58745 0039d316-1c4b-4281-b951-d872f2087c98
* [Chrome OS] Add the owner to the whitelist on first logincmasone@google.com2010-09-076-20/+72
| | | | | | | | | BUG=None TEST=unit tests, run on the device and then check the whitelist for the owner. Review URL: http://codereview.chromium.org/3340012 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@58744 0039d316-1c4b-4281-b951-d872f2087c98
* Add libEGL.pdb and libGLESv2.pdb to the list of copied symbol filesjamesr@chromium.org2010-09-071-1/+3
| | | | | | | | | | | These DLLs are generated from ANGLE. The pdb's are needed to debug archived continuous builds. BUG=none TEST=Check chrome-win32-syms.zip on the builders Review URL: http://codereview.chromium.org/3310017 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@58743 0039d316-1c4b-4281-b951-d872f2087c98
* Changes the mnemonic for 'Tools' from 't' to 'l'. It was overlappingsky@chromium.org2010-09-071-2/+2
| | | | | | | | | | | with the 't' in 'new tab'. BUG=49728 TEST=make sure that alt-f 'l' works. Review URL: http://codereview.chromium.org/3340016 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@58742 0039d316-1c4b-4281-b951-d872f2087c98
* Make parallel test launcher print output as it's written by the child.phajdan.jr@chromium.org2010-09-071-8/+61
| | | | | | | | | | | Add protection against hang when the child leaks a process. BUG=54098 TEST=none Review URL: http://codereview.chromium.org/3360008 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@58741 0039d316-1c4b-4281-b951-d872f2087c98
* [Mac] Snap fullscreen window to screen after exiting spaces rohitrao@chromium.org2010-09-071-0/+9
| | | | | | | | | | | | | | | | | | | | Users can arbitrarily move fullscreen windows. If the fullscreen window changes screen then the fullscreen controller will snap it to the screen. If the fullscreen window doesn't change screen then the window will remain where it is which looks weird. This patch fixes the issue by always snapping the window's frame to the screen any time it has moved. An alternative fix would be to override a new 10.6 API: -[NSWindow isMovable] and return NO. This would allow users to move windows between spaces but not the position inside a space. I think detecting window move is better because: - it works on 10.5 also - users with multiple monitors can uses space to move the window between monitors Patch by Sailesh Agrawal <fnepal@gmail.com>. BUG=41693 TEST=Verified that the full screen window snaps to the screen when moved in Spaces. Verified that changing the screen resolution, mirroring, etc... all correct cause the full screen window to snap to the screen. Review URL: http://codereview.chromium.org/3343009 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@58740 0039d316-1c4b-4281-b951-d872f2087c98
* dom-ui options: Correctly set enabled state of commit button when showing ↵csilv@chromium.org2010-09-071-4/+9
| | | | | | | | | | the dialog. BUG=none TEST=Open 'Clear Browsing Data' dialog when no options are checked, commit button should be disabled. Review URL: http://codereview.chromium.org/3352006 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@58739 0039d316-1c4b-4281-b951-d872f2087c98
* dom-ui options: Improve how proxy preferences are observed.csilv@chromium.org2010-09-071-2/+1
| | | | | | | | BUG=none TEST=none Review URL: http://codereview.chromium.org/3366009 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@58738 0039d316-1c4b-4281-b951-d872f2087c98
* Makes --disable-hang-monitor apply to navigations and closing tabs.creis@google.com2010-09-072-1/+8
| | | | | | | | | BUG=none TEST=Pass --disable-hang-monitor and close a tab with a slow onunload event. Review URL: http://codereview.chromium.org/3327011 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@58737 0039d316-1c4b-4281-b951-d872f2087c98
* Reland r58230.willchan@chromium.org2010-09-076-52/+111
| | | | | | | | | | | Simple 2-line fix. BUG=53138 TEST=Make sure ethernet is plugged in. Run Chrome. Unplug ethernet. Does not crash. Review URL: http://codereview.chromium.org/3371001 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@58736 0039d316-1c4b-4281-b951-d872f2087c98
* Don't call freeaddrinfo(NULL) to avoid crash on FreeBSD.vandebo@chromium.org2010-09-071-2/+4
| | | | | | | | | | | | Apparently freeaddrinfo(NULL) is undefined (or at least FreeBSD reads the appropriate RFCs that way) and FreeBSD crashes in this case. BUG=FreeBSD crash TEST=no crash after change on FreeBSD. Review URL: http://codereview.chromium.org/3360012 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@58735 0039d316-1c4b-4281-b951-d872f2087c98
* Have the Chrome Frame helper DLL's pdb not stomp on the Chrome Frame helper ↵robertshield@chromium.org2010-09-071-2/+2
| | | | | | | | | | | EXE's pdb at build time. BUG=53127 TEST=build both chrome_frame_helper and chrome_frame_helper_dll and get two pdbs. Review URL: http://codereview.chromium.org/3308012 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@58734 0039d316-1c4b-4281-b951-d872f2087c98
* Disabled Create Account link with a flag.avayvod@chromium.org2010-09-073-8/+23
| | | | | | | | | | BUG=chromium-os:5819 TEST=Check that Create Account link is not present on New User pod or login screen. Review URL: http://codereview.chromium.org/3315013 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@58733 0039d316-1c4b-4281-b951-d872f2087c98
* This CL changes our per-domain limits to be per-effective domain.rdsmith@chromium.org2010-09-077-243/+753
| | | | | | | | | | | | | | This matches FireFox behavior and is for denial of service protection when we go to FireFox expiry behavior (keeping cookies around indefinitely if they don't have a max-age and are being used); see issue (8850) for details. BUG=8850 TEST=net_unittests/net_perftests on Linux with CookieMonsterTest.* filter Review URL: http://codereview.chromium.org/3122013 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@58732 0039d316-1c4b-4281-b951-d872f2087c98
* GTTF: Make WaitForTabCountToBecome automation call not Sleep.phajdan.jr@chromium.org2010-09-0715-38/+129
| | | | | | | | | | | | Sleeping is an unreliable method to wait for things. Instead, we set up an observer. BUG=none TEST=none Review URL: http://codereview.chromium.org/3300011 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@58731 0039d316-1c4b-4281-b951-d872f2087c98
* Back out r58711. It appears to have caused video/media-related ui_testsmark@chromium.org2010-09-0710-36/+15
| | | | | | | | | | | | | | | | failures on Windows: MediaTest.VideoBearTheora MediaTest.VideoBearSilentTheora MediaTest.VideoBearWebm MediaTest.VideoBearSilentWebm UILayoutTest.MediaUILayoutTest BUG=54469 TEST=tree Review URL: http://codereview.chromium.org/3333019 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@58730 0039d316-1c4b-4281-b951-d872f2087c98
* Roll back webkit to r66862 due to a compiling error.zmo@google.com2010-09-071-1/+1
| | | | git-svn-id: svn://svn.chromium.org/chrome/trunk/src@58729 0039d316-1c4b-4281-b951-d872f2087c98
* Fix the expected number of crashes in a metrics test.nirnimesh@chromium.org2010-09-071-3/+4
| | | | | | | | | BUG= TEST= Review URL: http://codereview.chromium.org/3295015 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@58728 0039d316-1c4b-4281-b951-d872f2087c98
* Update bug reference.avi@chromium.org2010-09-071-2/+2
| | | | | | | | | | BUG=54711 TEST=no change; comment alteration. Review URL: http://codereview.chromium.org/3306010 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@58727 0039d316-1c4b-4281-b951-d872f2087c98
* Revert "net: add prober results to False Start blacklist."agl@chromium.org2010-09-077-1132/+29
| | | | | | | Broke ARM build. Need to figure out how to make gyp build the helper binary for the host target, not the build target. git-svn-id: svn://svn.chromium.org/chrome/trunk/src@58726 0039d316-1c4b-4281-b951-d872f2087c98
* Change ICU_UTIL_DATA_IMPL on Linux to match the reality.jshin@chromium.org2010-09-071-4/+2
| | | | | | | | | | | | | | | | | We've bundled the ICU data file for 'ages' on Linux, but ICU_UTIL_DATA_IMPL on Linux is still set to ICU_UTIL_DATA_FILE. Apparently, this hasn't been noticed because u_setDataDirectory and udata_setFileAccess just set the global variable pointing to the data file location. This is not totally free because u_setDataDirectory does some locking/unlocking. BUG=NONE TEST=Builds go through on Linux and binaries (chrome, test, etc) run just fine. Review URL: http://codereview.chromium.org/3329012 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@58725 0039d316-1c4b-4281-b951-d872f2087c98
generated by cgit v1.1 at 2025-02-08 17:37:52 +0000