summaryrefslogtreecommitdiffstats
Commit message (Collapse)AuthorAgeFilesLines
* linux: reenable symbols in backtraces in debug buildsevan@chromium.org2009-12-312-10/+18
| | | | | | | | This is easy enough to fix, and would be occasionally helpful. Review URL: http://codereview.chromium.org/524015 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@35405 0039d316-1c4b-4281-b951-d872f2087c98
* GTK: Remove OwnedWidgetGtk wrapper for the GtkMenu* in MenuGtk.estade@chromium.org2009-12-313-24/+17
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The "execute the command later" workaround, which was in place to avoid OwnedWidget-DCHECKing when selecting menu items that might cause cascading object-destruction effects, caused problems with the Browser::SetBlockCommandExecution() workaround, the reasoning for which can be found in BrowserWindowGtk and is reproduced here: // This piece of code is based on the fact that calling // gtk_window_activate_key() method against |window_| may only trigger a // browser command execution, by matching either a global accelerator // defined in above |kAcceleratorMap| or the accelerator key of a menu // item defined in chrome/browser/gtk/standard_menus.cc. // // Here we need to retrieve the command id (if any) associated to the // keyboard event. Instead of looking up the command id in above // |kAcceleratorMap| table by ourselves, we block the command execution of // the |browser_| object then send the keyboard event to the |window_| by // calling gtk_window_activate_key() method, as if we are activating an // accelerator key. Then we can retrieve the command id from the // |browser_| object. // // Pros of this approach: // 1. We can handle accelerators defined not only in above // |kAcceleratorMap| table, but also those in standard_menus.cc. // 2. We don't need to care about keyboard layout problem, as // gtk_window_activate_key() method handles it for us. // // Cons: // 1. The logic is a little complicated. // 2. We should be careful not to introduce any accelerators that trigger // customized code instead of browser commands. The easiest and cleanest thing is just to remove the OwnedWidgetGtk. BUG=none TEST=interactive ui tests pass again Review URL: http://codereview.chromium.org/522028 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@35404 0039d316-1c4b-4281-b951-d872f2087c98
* Re-enable two DB layout tests.dumi@chromium.org2009-12-311-7/+0
| | | | | | | | | TEST=none BUG=30081, 30082 Review URL: http://codereview.chromium.org/515074 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@35403 0039d316-1c4b-4281-b951-d872f2087c98
* bsds: views/ and webkit/ support for FreeBSD/OpenBSDevan@chromium.org2009-12-3129-74/+85
| | | | | | | | | - add a TOOLKIT_USES_GTK #define to catch GTK and Views+GTK - lots of ifdef twiddling Review URL: http://codereview.chromium.org/519030 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@35402 0039d316-1c4b-4281-b951-d872f2087c98
* GTK: implement page action context menus.estade@chromium.org2009-12-312-35/+59
| | | | | | | | | BUG=30656 TEST=manual Review URL: http://codereview.chromium.org/515073 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@35401 0039d316-1c4b-4281-b951-d872f2087c98
* Implement browser action context menu on linux.estade@chromium.org2009-12-3111-58/+89
| | | | | | | | | BUG=30656 TEST=manual (right click on browser actions and test the actions) Review URL: http://codereview.chromium.org/521016 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@35400 0039d316-1c4b-4281-b951-d872f2087c98
* Rebaseline css/namespaces tests.dglazkov@chromium.org2009-12-3115-4/+33
| | | | | | | | | | TBR=ukai TEST=none BUG=29157 Review URL: http://codereview.chromium.org/523030 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@35399 0039d316-1c4b-4281-b951-d872f2087c98
* Remove some outdated valgrind suppressions.michaeln@chromium.org2009-12-301-86/+0
| | | | | | | | | BUG=27993 TEST=valgrind try runs Review URL: http://codereview.chromium.org/519029 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@35398 0039d316-1c4b-4281-b951-d872f2087c98
* Rolling 52667 -> 52672ajwong@chromium.org2009-12-306-2/+42
| | | | | | | | | | | | Rebaselining LayoutTests/svg/custom/use-on-symbol-inside-pattern-expected.svg. TBR=dglazkov BUG=none TEST=green bots. Review URL: http://codereview.chromium.org/515070 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@35397 0039d316-1c4b-4281-b951-d872f2087c98
* Be sure to create and populate the meta table within the transaction that ↵michaeln@chromium.org2009-12-301-6/+8
| | | | | | | | | | | creates the rest of the schema. BUG=none TEST=existing tests apply Review URL: http://codereview.chromium.org/518023 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@35396 0039d316-1c4b-4281-b951-d872f2087c98
* Disabling the URLRequestTestHTTP.GetTest_NoCache test for ChromeFrame as ↵ananta@chromium.org2009-12-301-0/+4
| | | | | | | | | | | | this has been consistently failing on the chrome frame builder. TBR=tommi Review URL: http://codereview.chromium.org/515071 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@35395 0039d316-1c4b-4281-b951-d872f2087c98
* Add NSTemporaryDirectory() to the list of ignored to catch all possible ↵thomasvl@chromium.org2009-12-301-6/+19
| | | | | | | | | | values. Also handle all the forms of the paths with links resolved. TEST=page cycles, browser_tests all get back to being fast. normal profiles still exclude the files listed in the first bug. BUG=25959,30002 Review URL: http://codereview.chromium.org/523026 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@35394 0039d316-1c4b-4281-b951-d872f2087c98
* Use https for GAIA URL.cevans@chromium.org2009-12-301-1/+1
| | | | | | | | | BUG=NONE TEST=NONE Review URL: http://codereview.chromium.org/522022 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@35393 0039d316-1c4b-4281-b951-d872f2087c98
* Make the 0 value of the enum match 'fail' to satisfy DCHECKs.evan@chromium.org2009-12-301-4/+4
| | | | git-svn-id: svn://svn.chromium.org/chrome/trunk/src@35392 0039d316-1c4b-4281-b951-d872f2087c98
* Roll webkit 52633 -> 52667.ajwong@chromium.org2009-12-3085-441/+446
| | | | | | | | | | | Rebaseline a bunch of SVG test broken to due webkit r52647. BUG=25977,31028,29737,31350 TEST=mostly green layout test try runs. Review URL: http://codereview.chromium.org/515067 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@35391 0039d316-1c4b-4281-b951-d872f2087c98
* linux: use XShmPutImage when SHM Pixmap APIs unavailableevan@chromium.org2009-12-304-58/+86
| | | | | | | | | | | | | | SHM Pixmaps were frequently unavailable (which makes some sense: you're asking for a shared memory handle into a graphics-card-side memory buffer). But we can still do an SHM copy from pixels->pixmap as long as we only go in write only direction, via the XShmPutImage API. BUG=31301 Review URL: http://codereview.chromium.org/522023 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@35390 0039d316-1c4b-4281-b951-d872f2087c98
* Make back forward menu model a MenuModel.estade@chromium.org2009-12-3017-568/+401
| | | | | | | | | | | On Linux, implement just enough MenuModel support in GtkMenu. On Linux + windows, remove wrapper classes. On Mac, just change a few function calls around. BUG=none TEST=unit test + manually clicking around on it in debug mode Review URL: http://codereview.chromium.org/501168 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@35389 0039d316-1c4b-4281-b951-d872f2087c98
* Re-enable the ExtensionApiTest.Storage test.dumi@chromium.org2009-12-301-2/+1
| | | | | | | | | | BUG=27216 TEST=none Review URL: http://codereview.chromium.org/523021 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@35388 0039d316-1c4b-4281-b951-d872f2087c98
* Defense-in-depth: initialize some variables.cevans@chromium.org2009-12-302-2/+5
| | | | | | | | | BUG=NONE TEST=NONE Review URL: http://codereview.chromium.org/517025 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@35387 0039d316-1c4b-4281-b951-d872f2087c98
* Don't defend against Initialize being called twice, that should not happen. michaeln@chromium.org2009-12-301-7/+6
| | | | | | | | | | TEST=none BUG=26754 TBR=mark Review URL: http://codereview.chromium.org/524013 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@35386 0039d316-1c4b-4281-b951-d872f2087c98
* Minor cleanup, removed unused URLRequestJob::GetMoreData() method.michaeln@chromium.org2009-12-307-31/+0
| | | | | | | | | BUG=none TEST=none Review URL: http://codereview.chromium.org/515063 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@35385 0039d316-1c4b-4281-b951-d872f2087c98
* Added pause button to DownloadItemMac via IB and IBAction to receive the event.thakis@chromium.org2009-12-304-39/+767
| | | | | | | | | | BUG=23069 TEST=Download large file, click pause in download item's menu to pause/unpause. Review URL: http://codereview.chromium.org/524009 Patch from sig11@reprehensible.net. git-svn-id: svn://svn.chromium.org/chrome/trunk/src@35384 0039d316-1c4b-4281-b951-d872f2087c98
* Apply style guideline for variable names for browser action container.mhm@chromium.org2009-12-301-2/+2
| | | | | | | | | | | Change arrowLocation to arrow_location, to follow Chromium code standards. BUG=494008 TEST=None Review URL: http://codereview.chromium.org/522027 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@35383 0039d316-1c4b-4281-b951-d872f2087c98
* [Mac] address perf regressions on botsthomasvl@chromium.org2009-12-301-0/+13
| | | | | | | | | | This isn't the prettiest, but it should get the perf back where we want it. If the file we're asked to exclude is in /var or /private, ignore it. The problem on the bots is the TimeMachine prefs plist is growing unbounded as we keep adding 4 files each run for temporary profiles. The load/modify/write gets slower with each addition. This quick/dirty check to avoid that. Since the original bug for the backup of this data is still open, I figure landing this is the right thing to do until that is fully addressed. BUG=25959,30002 TEST=page cycles, browser_tests all get back to being fast. normal profiles still exclude the files listed in the first bug. Review URL: http://codereview.chromium.org/523024 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@35382 0039d316-1c4b-4281-b951-d872f2087c98
* Revert enabled DevToolsSanityTest.TestConsoleLog test.apavlov@chromium.org2009-12-301-1/+1
| | | | | | | | | | Fails on Linux TBR=pfeldman@chromium.org Review URL: http://codereview.chromium.org/519028 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@35381 0039d316-1c4b-4281-b951-d872f2087c98
* Enable DevToolsSanityTest.TestConsoleLog test (seems flaky)apavlov@chromium.org2009-12-301-1/+1
| | | | | | | | | | TBR=pfeldman@chromium.org BUG=none TEST=DevToolsSanityTest.TestConsoleLog Review URL: http://codereview.chromium.org/521015 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@35380 0039d316-1c4b-4281-b951-d872f2087c98
* If we receive a redirect response, we should copy the http referer field ↵japhet@chromium.org2009-12-301-0/+9
| | | | | | | | | | | from the old request. BUG=7357 TEST=none Review URL: http://codereview.chromium.org/470010 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@35379 0039d316-1c4b-4281-b951-d872f2087c98
* Fix a race in WorkerThreadTicker test.phajdan.jr@chromium.org2009-12-301-3/+3
| | | | | | | | | TEST=none BUG=30582 Review URL: http://codereview.chromium.org/506073 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@35378 0039d316-1c4b-4281-b951-d872f2087c98
* Replaces the % character with \x when generating Windows shortcuts via jschuh@chromium.org2009-12-303-11/+19
| | | | | | | | | | | | File->"Create application shortcuts." The \x is converted back to % in handling the --app switch. BUG=23693 TEST=None Review URL: http://codereview.chromium.org/515028 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@35377 0039d316-1c4b-4281-b951-d872f2087c98
* [Linux/Mac] clean up the shared memory file on disk used by stats in tests.thomasvl@chromium.org2009-12-301-7/+33
| | | | | | | | BUG=none TEST=files shouldn't be left around. Review URL: http://codereview.chromium.org/518019 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@35376 0039d316-1c4b-4281-b951-d872f2087c98
* [Mac] Fixup the new tab button imagesthomasvl@chromium.org2009-12-302-705/+43
| | | | | | | | | | | | - Don't set the images in the xib since cocoa can't find them in the other bundle during tests (and maybe not during normal app runtime) - Make sure the images are setup during view start. - Create constants for the image names since some are used in more then one place. BUG=30342 TEST=no longer get the error messages on test runs Review URL: http://codereview.chromium.org/515061 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@35375 0039d316-1c4b-4281-b951-d872f2087c98
* Updating trunk VERSION from 285.0 to 286.0chrome-bot@google.com2009-12-301-1/+1
| | | | git-svn-id: svn://svn.chromium.org/chrome/trunk/src@35374 0039d316-1c4b-4281-b951-d872f2087c98
* Fix linux/mac treejar@chromium.org2009-12-301-0/+2
| | | | | | | | | non-debug builds had unused static function tbr=mbelshe Review URL: http://codereview.chromium.org/523023 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@35373 0039d316-1c4b-4281-b951-d872f2087c98
* Provide a quick and dirty way to reset about:objects datajar@chromium.org2009-12-304-45/+289
| | | | | | | | | | | | | | | | | | | | | To make it easier to use the about:object profiling facility, I put in a quick/dirty way to reset all profile stats to 0 (as if there were no births, deaths, etc.). This code is only activated under debug builds (or if a developer inists in a private build). These stats don't impact semantics of the browser, so the hackish approach to clearing the data counts can't instigate a crash, and it makes it much easier to look at changes in the stats. While changing the code, I also added a lot of comments, and did a few minor cleanups items. I also officially added about:tasks as a replacement for about:objects, as this is really how the service is used in Chrome. r=mbelshe Review URL: http://codereview.chromium.org/100297 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@35372 0039d316-1c4b-4281-b951-d872f2087c98
* Fix up rowbytes vs. width desynchronization, and fix failure to initialize ↵cevans@chromium.org2009-12-301-8/+6
| | | | | | | | | | | | | entire bitmap memory. To fix up the rowbytes value properly, we simply don't send it via IPC any more, and recalulate it from width and depth in the trusted code. It's a cheap calculation. Also one bonus fix: don't use an unintialized IconInfo if deserialization fails. BUG=31307 TEST=Manual; ran with breakpoints on the failure paths. Review URL: http://codereview.chromium.org/517023 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@35371 0039d316-1c4b-4281-b951-d872f2087c98
* Dropping unneeded set of vsprops.bradnelson@google.com2009-12-304-38/+4
| | | | | | | | | | | | | | | These vsprops used to be used to setup the branding #defines. This is now handled in common.gypi. This is an incremental step toward dropping vsprops entirely for release. BUG=None TEST=None TBR=maruel Review URL: http://codereview.chromium.org/523020 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@35367 0039d316-1c4b-4281-b951-d872f2087c98
* UITest class should derive from PlatformTest instead of testing::Testnirnimesh@chromium.org2009-12-302-3/+3
| | | | | | | | | | This is so that ui tests using UITest have a valid NSAutoreleasePool, which PlatformTest provides, but testing::Test does not. Review URL: http://codereview.chromium.org/519026 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@35366 0039d316-1c4b-4281-b951-d872f2087c98
* Disable interactive ui test DevToolsSanityTest.TestConsoleLog which went red ↵michaeln@chromium.org2009-12-301-1/+1
| | | | | | | | | | | | with a webkit roll. TEST=none BUG=31302 TBR=pam Review URL: http://codereview.chromium.org/518024 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@35365 0039d316-1c4b-4281-b951-d872f2087c98
* Use the standard DeleteTask rather than rolling my own.ajwong@chromium.org2009-12-301-13/+1
| | | | | | | | | TBR=scherkus BUG=none TEST=none git-svn-id: svn://svn.chromium.org/chrome/trunk/src@35364 0039d316-1c4b-4281-b951-d872f2087c98
* Removed reference to Windows-only debug variable. This makes it soakalin@chromium.org2009-12-302-4/+4
| | | | | | | | | | | that libjingle logging happens on non-Windows platforms. BUG=none TEST=manual Review URL: http://codereview.chromium.org/523012 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@35363 0039d316-1c4b-4281-b951-d872f2087c98
* Add a range of sanity checks to the clipboard object handling. In ↵cevans@chromium.org2009-12-301-4/+23
| | | | | | | | | | | particular, fix an integer overflow that might be an issue on Linux. Other changes are to avoid OOB reads and calling front() or [] on an empty or insufficiently sized vector. BUG=31928 TEST=NONE Review URL: http://codereview.chromium.org/522024 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@35362 0039d316-1c4b-4281-b951-d872f2087c98
* WebKit update r52554->r52633.pam@chromium.org2009-12-302-10/+27
| | | | | | | | | TBR=dimich BUG=none TEST=layout tests Review URL: http://codereview.chromium.org/515052 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@35361 0039d316-1c4b-4281-b951-d872f2087c98
* Don't shoehorn a size_t into an int -- especially if the int is used in an ↵cevans@chromium.org2009-12-301-2/+2
| | | | | | | | | | | subsequent allocation. BUG=NONE TEST=NONE Review URL: http://codereview.chromium.org/518022 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@35360 0039d316-1c4b-4281-b951-d872f2087c98
* linux: fix X SharedMemory API types and commentsevan@chromium.org2009-12-303-6/+7
| | | | | | | | Some functions actually return a ShmSeg, not an XID. Review URL: http://codereview.chromium.org/519024 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@35359 0039d316-1c4b-4281-b951-d872f2087c98
* Expect appcache/remove-cache layout test to fail on OSX for now.michaeln@chromium.org2009-12-301-0/+4
| | | | | | | | | | BUG=31300 TEST=none Review URL: http://codereview.chromium.org/521014 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@35358 0039d316-1c4b-4281-b951-d872f2087c98
* Use Xvfb when running the tests.jrg@chromium.org2009-12-291-0/+63
| | | | | | | | | | | | | | Normal buildbot-run tests that need an X display connection (e.g. ui_tests) are executed by RunPythonCommandInBuildDir() from scripts/slave/slave_utils.py which takes care of this. However, we are currently running coverage_posix.py from within a build command which itself is not run a s a "python command in build dir". Thus, no Xvfb. TODO(jrg): transition coverage generation to a distinct buildbot step. Review URL: http://codereview.chromium.org/524007 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@35357 0039d316-1c4b-4281-b951-d872f2087c98
* Revert 35339 - Change WDS to use the DB thread rather than its own thread.michaeln@google.com2009-12-294-34/+48
| | | | | | | | | | | | | This cleanup was requested by brettw and was started to make it easier for the sync service to post tasks to the WDS thread (now the DB thread). This simplifies the WDS a bit since it no longer has to manage its own thread, and can assume that the DB thread is running throughout its lifetime. One change in behavior that is significant is that previous to this change, the WDS worker thread would always be joined when Shutdown() was called from Profile::~Profile(). Now the Shutdown() method schedules a task that can extend the lifetime of the WDS past the lifetime of the Profile instance. Review URL: http://codereview.chromium.org/524003 TBR=skrul@chromium.org Review URL: http://codereview.chromium.org/521013 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@35356 0039d316-1c4b-4281-b951-d872f2087c98
* Adding the DB performance tests.dumi@chromium.org2009-12-2918-6/+728
| | | | | | | | | | BUG=none TEST=none Review URL: http://codereview.chromium.org/465141 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@35355 0039d316-1c4b-4281-b951-d872f2087c98
* AppCacheDatabase and SQL based AppCacheStorageImpl. michaeln@chromium.org2009-12-2925-49/+3637
| | | | | | | | | | | | | | | Still nothing is being written to disk with this CL, in-memory SQLite and DiskCaches are being utilized. Responses are not yet being removed from the DiskCasche when the should be. Once that's done (in the next CL), we'll start saving things on disk. BUG=none TEST=appcache_database_unittest.cc, appcache_storage_impl_unittest.cc Review URL: http://codereview.chromium.org/518020 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@35354 0039d316-1c4b-4281-b951-d872f2087c98
* Continue migration from query_helpers.* to sqlite_utils.*zork@google.com2009-12-294-153/+158
| | | | | | | | BUG=none TEST=Run sync_unit_tests Review URL: http://codereview.chromium.org/509023 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@35353 0039d316-1c4b-4281-b951-d872f2087c98
generated by cgit v1.1 at 2025-01-31 08:48:58 +0000