- 17 Sep, 2013 26 commits
-
-
avi@chromium.org authored
BUG=54748 TEST=load a password-protected PDF on Windows, see fancy new UI Review URL: https://chromiumcodereview.appspot.com/23478036 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@223509 0039d316-1c4b-4281-b951-d872f2087c98
-
huangs@chromium.org authored
We wish to make MostLikely URL recommendations use local thumbnail. To increase the likelihood of a hit, this CL enables a thumbnail request to perform URL prefix match. For example, if thumbnail is stored for ==> http://www.chromium.org/Home but not ==> http://www.chromium.org/ then previously the request ==> chrome://thumb/http://www.chromium.org/ would fail. This CL aims to create a fallback, by adding "thumb2" that matches prefix, i.e.: ==> chrome://thumb2/http://www.chromium.org/ would match "http://www.chromium.org/Home" and display the corresponding thumbnail. Details: - We consider "URL prefix" match, which is not same as "string prefix" match. Specifics: --- Need identical protocol/scheme, host, and port (GURL is smart about this). --- Query strings are ignored. --- For path, we require entire path components to match. E.g., "base/test" is a prefix of "base/test/sub", but is NOT a prefix of "base/testing" or "best/test-case". - If multiple matches exist, the first URL-lexicographical match is taken. This allows the "nearest" children to be matched, but favors siblings that are alphabetically closer. For example, "base" prefers "base/test" over "base/test/sub", but also prefers "base/deep/sub/dir/" over "base/test" (since "deep" < "test"). - To implement the above match, a new comparator is used in the thumbnail cache. - Adding new test TopSitesCacheTest, which also tests existing code. - Adding url_utils.cc in chrome\browser\history, along with unit tests. - Adding the "chrome://thumb2" path, which causes most of the 1-2-line changes in files. BUG=284634 TEST=TopSitesCacheTest.* Review URL: https://chromiumcodereview.appspot.com/23477033 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@223508 0039d316-1c4b-4281-b951-d872f2087c98
-
jennyz@chromium.org authored
Fix the bluetooth audio device id changing on the fly issue and the active device inconsistent issue caused by multiple NodesChanged signal received for the same node change. BUG=290485 Review URL: https://chromiumcodereview.appspot.com/23620055 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@223507 0039d316-1c4b-4281-b951-d872f2087c98
-
bryaneyler@google.com authored
BUG=245025 R=eroman@chromium.org,ellyjones@chromium.org Review URL: https://chromiumcodereview.appspot.com/23569007 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@223506 0039d316-1c4b-4281-b951-d872f2087c98
-
asvitkine@chromium.org authored
Found by the Mac OS X 'leaks' tool; fixed on a plane. BUG=292905 Review URL: https://chromiumcodereview.appspot.com/23503064 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@223505 0039d316-1c4b-4281-b951-d872f2087c98
-
dshi@chromium.org authored
In run_test.Main, there are two places that logging level may be changed: 1. BrowserFinderOptions, when calling parser.parse_args, in which logging level is set based on verbosity's value. 2. run_test.Main, in which logging level is set to WARN if verbosity is 0. Such change on logging level causes problem to caller like autotest, which relies on logging level to log details during the test, and output messages from different logging level to different log file, e.g., client.DEBUG, client.INFO etc. This CL added some fix to: 1. Cache the logging level at the beginning of the Main method. 2. Always restore logging level at the end of the Main method. BUG=chromium:284763 TEST=verified in local dut with autotest: test_that --fast -b lumpy 172.22.75.225 telemetry_UnitTests Review URL: https://chromiumcodereview.appspot.com/23458027 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@223504 0039d316-1c4b-4281-b951-d872f2087c98
-
asvitkine@chromium.org authored
Found by the Mac OS X 'leaks' tool; fixed on a plane. BUG=292905 Review URL: https://chromiumcodereview.appspot.com/24188002 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@223503 0039d316-1c4b-4281-b951-d872f2087c98
-
thestig@chromium.org authored
Heapcheck: Remove suppressions for SafeBrowsingDatabaseManagerTest.CheckCorrespondsListType because the bug has been fixed. BUG=289509 NOTRY=true TBR=noamsml@chromium.org Review URL: https://chromiumcodereview.appspot.com/23950009 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@223502 0039d316-1c4b-4281-b951-d872f2087c98
-
shadi@chromium.org authored
BUG=291379 Review URL: https://chromiumcodereview.appspot.com/24159003 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@223501 0039d316-1c4b-4281-b951-d872f2087c98
-
danakj@chromium.org authored
This moves the release callback out of the TextureMailbox so that we can use it for other resource types than just texture mailboxes. While doing this, we make a SingleReleaseCallback class that is held in a scoped_ptr. This class DCHECKs that the callback is run before it is destroyed, and ensures clear ownership semantics as you must Pass() the callback around. No change in behaviour, covered by existing tests. R=piman BUG=263069 Review URL: https://chromiumcodereview.appspot.com/23648014 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@223500 0039d316-1c4b-4281-b951-d872f2087c98
-
thestig@chromium.org authored
BUG=291236 NOTRY=true TBR=yangguo@chromium.org Review URL: https://chromiumcodereview.appspot.com/23620056 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@223499 0039d316-1c4b-4281-b951-d872f2087c98
-
asvitkine@chromium.org authored
Fixed at altitude 12000 meters. BUG=227391 TEST=See bug. Review URL: https://chromiumcodereview.appspot.com/23591052 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@223498 0039d316-1c4b-4281-b951-d872f2087c98
-
sergeyu@chromium.org authored
TBR=sergeyu@chromium.org BUG=281781 Review URL: https://codereview.chromium.org/23514062 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@223497 0039d316-1c4b-4281-b951-d872f2087c98
-
thestig@chromium.org authored
- Move a WontFix leak to the intentional section, and update the suppression to match the only current instance of the leak. - Widen a suppression for an inspector memory error. - Add new suppression for a WebCore::Document::decoder NULL deref. BUG=73299,288801,293088 NOTRY=true TBR=thakis@chromium.org Review URL: https://chromiumcodereview.appspot.com/23902038 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@223496 0039d316-1c4b-4281-b951-d872f2087c98
-
rkc@chromium.org authored
The feedback manifest doesn't contain the correct icons to show on the app shelf. Add them. R=xiyuan@chromium.org BUG=291381 Review URL: https://chromiumcodereview.appspot.com/23866010 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@223495 0039d316-1c4b-4281-b951-d872f2087c98
-
sergeyu@chromium.org authored
The new proxy is based on ppapi::proxy::PluginResource and ppapi::host::ResourceHost which simplifies code significantly. Also the permission check is consistent with socket APIs now. BUG=281781 Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=223482 Review URL: https://chromiumcodereview.appspot.com/23819033 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@223494 0039d316-1c4b-4281-b951-d872f2087c98
-
jsbell@chromium.org authored
When porting the transaction back-end from Blink we brought along an AVLTree implementation for the in-memory key/value store for transactions. Replace this with a std::map since it provides similar performance guarantees (red-black tree) and we like to delete code. TBR=jam@chromium.org R=alecflett@chromium.org Review URL: https://chromiumcodereview.appspot.com/17462005 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@223493 0039d316-1c4b-4281-b951-d872f2087c98
-
ananta@chromium.org authored
Currently the Tooltip widget is created in the tooltip timer which also handles setting the tooltip text on the Tooltip object. There is a timing window between this and the mouse event coming in, which ends up causing the tooltips to not show up if the underlying root changes in between. Fixes bug https://code.google.com/p/chromium/issues/detail?id=292530 BUG=292530 R=sky@chromium.org, sky Review URL: https://codereview.chromium.org/23958012 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@223492 0039d316-1c4b-4281-b951-d872f2087c98
-
dpolukhin@chromium.org authored
BUG=292963 TEST=manual TBR=dskelton@chromium.org NOTRY=true Review URL: https://chromiumcodereview.appspot.com/23591056 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@223490 0039d316-1c4b-4281-b951-d872f2087c98
-
wonsik@chromium.org authored
BUG=NONE R=qinmin@chromium.org Review URL: https://chromiumcodereview.appspot.com/24065004 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@223487 0039d316-1c4b-4281-b951-d872f2087c98
-
tfarina@chromium.org authored
Soon PerfTimer will move into base/timer/ directory as it is already used by production code. This splits the logging stuff into their owns files and also moves PerfTimeLogger into its own files. BUG=None TEST=base_unittests, ipc_perftests, etc... R=brettw@chromium.org Review URL: https://chromiumcodereview.appspot.com/23985006 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@223486 0039d316-1c4b-4281-b951-d872f2087c98
-
thestig@chromium.org authored
BUG=293024,293033,293043,293049,293058 NOTRY=true TBR=thakis@chromium.org Review URL: https://chromiumcodereview.appspot.com/23470010 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@223485 0039d316-1c4b-4281-b951-d872f2087c98
-
ikarienator@chromium.org authored
> Simplify PPB_NetworkMonitor proxy. > > The new proxy is based on ppapi::proxy::PluginResource and > ppapi::host::ResourceHost which simplifies code significantly. Also > the permission check is consistent with socket APIs now. > > BUG=281781 > > Review URL: https://chromiumcodereview.appspot.com/23819033 TBR=sergeyu@chromium.org Review URL: https://codereview.chromium.org/23463037 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@223484 0039d316-1c4b-4281-b951-d872f2087c98
-
jam@chromium.org authored
This also fixes a number of issues I saw when testing Netflix: -fix http status header not reaching plugin -send a Content-Type header if one isn't specified by the plugin for POSTs -null check info->headers, needed for data: URLs This doesn't support responding to 200s after a byte-range request. I'm gathering stats now to see how often this happens. BUG=286074 R=ananta@chromium.org, jschuh@chromium.org Review URL: https://codereview.chromium.org/23830007 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@223483 0039d316-1c4b-4281-b951-d872f2087c98
-
sergeyu@chromium.org authored
The new proxy is based on ppapi::proxy::PluginResource and ppapi::host::ResourceHost which simplifies code significantly. Also the permission check is consistent with socket APIs now. BUG=281781 Review URL: https://chromiumcodereview.appspot.com/23819033 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@223482 0039d316-1c4b-4281-b951-d872f2087c98
-
ccameron@chromium.org authored
This was previously disabled when running unit tests, but it looks like there are additional problems with WebViews, and potentially with switching in and out of compositing mode. TBR=zmo BUG=292655 Review URL: https://codereview.chromium.org/23757059 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@223480 0039d316-1c4b-4281-b951-d872f2087c98
-
- 16 Sep, 2013 14 commits
-
-
girard@chromium.org authored
At scale factors above 125%, we are over-scaling up the dpi calculation. BUG=292607 Review URL: https://chromiumcodereview.appspot.com/23851034 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@223478 0039d316-1c4b-4281-b951-d872f2087c98
-
ernstm@chromium.org authored
We still need the 'webkit' category in the trace until the reference build is updated to a revision that uses 'webkit.console' for the timeline markers. R=tonyg@chromium.org BUG=282712 Review URL: https://chromiumcodereview.appspot.com/23757054 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@223477 0039d316-1c4b-4281-b951-d872f2087c98
-
thestig@chromium.org authored
BUG=292999 NOTRY=true TBR=glider@chromium.org Review URL: https://chromiumcodereview.appspot.com/23868033 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@223476 0039d316-1c4b-4281-b951-d872f2087c98
-
frankf@chromium.org authored
PageLoadingTest.testShouldBeAbleToNavigateForwardsInTheBrowserHistory TBR=craigdh@chromium.org BUG= Review URL: https://codereview.chromium.org/23619060 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@223475 0039d316-1c4b-4281-b951-d872f2087c98
-
noamsml@chromium.org authored
In debug builds, LogInterfaces cannot run in the IO or the UI threads due to an AssertIOAllowed() call (seems the IO thread cannot run actual IO either, see https://code.google.com/p/chromium/codesearch#chromium/src/base/threading/thread_restrictions.h ) NOTRY=true Review URL: https://chromiumcodereview.appspot.com/24181004 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@223473 0039d316-1c4b-4281-b951-d872f2087c98
-
armansito@chromium.org authored
Added a utility function to NetworkState to query if a network requires activation. All UI code should use this instead of comparing the Cellular.ActivationState property themselves, as this utility function handles the "unknown" case as well. BUG=289470 R=stevenjb@chromium.org Review URL: https://codereview.chromium.org/23532065 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@223472 0039d316-1c4b-4281-b951-d872f2087c98
-
asvitkine@chromium.org authored
Found by the Mac OS X 'leaks' tool; fixed on a plane. BUG=292905 Review URL: https://chromiumcodereview.appspot.com/23606034 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@223471 0039d316-1c4b-4281-b951-d872f2087c98
-
brettw@chromium.org authored
BUG=288991 R=koz@chromium.org Review URL: https://codereview.chromium.org/23970005 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@223470 0039d316-1c4b-4281-b951-d872f2087c98
-
brettw@chromium.org authored
I was working on the Windows build and having problems finding why a file was loaded that shouldn't have been. This adds the source of the invocation to the verbose logging so you can find this information. I removed an assert which caused a debug assertion for random punctuation and isn't harmful (a real error will be reported). I moved the location-to-string conversion code into the location class so I can share it for my new logging command, and moved a bunch of other logging functions to the new .cc file. BUG=288991 R=scottmg@chromium.org Review URL: https://codereview.chromium.org/23532076 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@223469 0039d316-1c4b-4281-b951-d872f2087c98
-
tbarzic@chromium.org authored
Also, remove some uneeded dependencies from LoginState tests. BUG=NONE Review URL: https://chromiumcodereview.appspot.com/23444061 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@223468 0039d316-1c4b-4281-b951-d872f2087c98
-
chrome-admin@google.com authored
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@223466 0039d316-1c4b-4281-b951-d872f2087c98
-
eseidel@chromium.org authored
http://build.chromium.org/f/chromium/perf/dashboard/ui/changelog_blink.html?url=/trunk&range=157831:157846&mode=html TBR= BUG= Review URL: https://chromiumcodereview.appspot.com/24184002 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@223465 0039d316-1c4b-4281-b951-d872f2087c98
-
scheib@chromium.org authored
Review URL: https://chromiumcodereview.appspot.com/23757050 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@223464 0039d316-1c4b-4281-b951-d872f2087c98
-
isherman@chromium.org authored
BUG=267983 TEST=compiles R=estade@chromium.org Review URL: https://chromiumcodereview.appspot.com/23629046 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@223463 0039d316-1c4b-4281-b951-d872f2087c98
-