summaryrefslogtreecommitdiff
path: root/tests/DumpRenderTree2
AgeCommit message (Collapse)Author
2013-08-25Remove WebViewClassic specific test code Jonathan Dixon
Bug: 10427705 These tested internals of WebViewClassic and will no longer build when that is removed Change-Id: I42af538dcb0343e9eff0b3fa85d7ff39f19c7c61
2012-05-29Fix style for Geolocation mock mechanism Steve Block
See I6d88d5dce5c2148812b191a5b452718bf0854aeb Change-Id: I1179319f066664393aa5f7324df89f8057a9e922
2012-05-22Switch Geolocation DRT methods to control client-based mock Steve Block
Note that we plumb DRT Geolocation mock calls via WebView. This is required because the WebCore client-based mock is not a static but is tied to the Page. See WebKit change Ib74a3c05991593e75c3138415d4ac0bf0c9aefa9. Bug: 6511338 Change-Id: I6d88d5dce5c2148812b191a5b452718bf0854aeb
2012-04-23Seperate interface and implementation of 4 WebView classes Jonathan Dixon
GeolocationPermissionsClassic CookieManagerClassic WebIconDatabaseClassic WebStorageClassic Also creats a WebViewFactory top level class - this remains hidden for now, as it's currently only used implicitly by the other public WebView classes to create the provider instances. Bug: 5626244 Change-Id: Id0ca1c16d8058f31a86414bbc0e8a55db4b907ba
2012-04-12Minor clean-up in DeviceOrientation and DeviceMotion Steve Block
No functional change. See corresponding external/webkit change https://android-git.corp.google.com/g/#change,125700 Change-Id: I2693328cb058820587ac43dd3121818959efd2d0
2012-03-02Refactor WebView to be a thin proxy class Jonathan Dixon
Splits interface and implementation; all client calls are forwarded to an abstract WebViewProvider interface, and the existing implementation moved into the WebViewClassic implementor of this interface. Originally taken from a snapshot from the development branch, by: git diff HEAD 9a4c328a54cc05e5 | git apply - but then rebased to keep up to date with master Interdepends on webkit and Browser changes: https://android-git.corp.google.com/g/158979 https://android-git.corp.google.com/g/167911 Change-Id: I91403f32654ff308934e95c832d17b292a7d9b2e
2011-05-24In DRT2, correctly handle the case of zero tests Steve Block
Bug: 3291382 Change-Id: I8cde1d8d78a57407c585b02ca8ba3f3f893a609b
2011-05-24Update DRT2 to reflect WebKit code movement Steve Block
See http://trac.webkit.org/changeset/74301 Change-Id: Ibe495cdc1f50f65db5663dbb63af9b9c2db050d0
2011-05-24Add extra error handling to DRT2 launcher script Steve Block
Bug: 3291220 Change-Id: I6f8e57458cd4334a998eb256af303ffcd9c50d64
2011-01-24A few small cleanning-ups. Huahui Wu
1. Remove mDeferMultiTouch in WebView, it was for testing only but we don't need it now since we always pass MultiTouch to WebKit. 2. Remove the use of mDeferMultiTouch in DRT tests. 3. Correct the index of getX(Y) for the second touch point in the debug string of MotionEvent. Change-Id: Ib63cfc5935af1a169ed26b2b138f74908492bc18
2011-01-21b/3369602 Using Math.round() to convert float coordinates to integers. Huahui Wu
Change-Id: If4da01046c909667ce57a849569ae6b1acefb939
2011-01-13Improve error handling (NPE in DRT2) Philippe Marti
Bug: 3284126 Change-Id: I7f8c6259f2932d2b1a10a2f9612866786d425da9
2010-11-23pass layout test list using external file Guang Zhu
Currently layout test list is passed between two processes in intent, which does not work for large lists. This change store the list in external storage folder (per app based, actual location determined by framework) instead. Change-Id: Ib175f766159f5e4f831d6b9ace9f1c4caef7298d
2010-11-11Fix EventSender mouse events to correctly convert from screen coordinates Steve Block
This is required for LayoutTest fast/dom/Node/mutation-blur.html Change-Id: Ib82f897ab61ac3a8e8eacee8ec08f5242c43c736
2010-11-11Refactor EventSenderImpl.MousePoint Steve Block
This factors out a new Point class and a new createViewPointFromContentCoordinates() method. This will allow Point and the conversion method to be re-used with mouse events. Change-Id: I2d689d0ffa0ad7ba01d0e90f78c812604ab96234
2010-11-09Include .php tests in DumpRenderTree2. Iain Merrick
I've manually checked the existing LayoutTests for false positives. There are non-test .php files, but they're all in "resources" subdirectories, so they should be caught by isTestDir(). Added an extra @warning to isTestFile() just in case. Bug: 3164771 Change-Id: Ib282acf46f6e839439ce521c4b2cb5820fcd263c
2010-11-09Fix DumpRenderTree2 to not attempt to read expected results over HTTPS Steve Block
This is required now that we have HTTPS tests in the triaged set of tests. Change-Id: I650518a4536ed903f637a44c0901b1e38bdc1a84
2010-11-09Fix FsUtils.readDataFromUrl() to do the network request on a background thread Steve Block
This is required now that doing network requests on the main thread throws NetworkOnMainThreadException. Also updates TestsListPreloaderThread to create the FileFilter lazily and removes a superfluous FileFilter param from the Summarizer constructor. Change-Id: If15fddac934cd8be611f6693b25837657abbddc3
2010-10-18Fix DumpRenderTree2 to correctly handle crashed tests Steve Block
Change-Id: I894987d5b8f24ea31893aefe3cb6450c2736aa90
2010-10-18Fix DumpRenderTree2 script to allow only one path to be specified Steve Block
Change-Id: I4a85ff69554a9c76af0b2deea747b8414660fee2
2010-10-18Improve error handling in DumpRenderTree2 scripts Steve Block
Change-Id: Idba4acc22dd037a001ea5c05cbd6a29292f6de2c
2010-10-14Implement ↵ Steve Block
LayoutTestController.overridePreference('WebKitUsesPageCachePreferenceKey') This is required for tests that test PageCache behaviour, eg - fast/dom/DeviceMotion/no-page-cache.html - fast/dom/DeviceOrientation/no-page-cache.html - fast/dom/Geolocation/no-page-cache.html Change-Id: Ie5862f6c86718142c949be81a85358ca6fac628a
2010-09-29Fix some bugs in DumpRenderTree2 when forming HTML diffs Steve Block
Change-Id: Id98c1b7dde1961c2dab4214a36ad43916baf8011
2010-09-29Use the empty string when expected text results are missing in DumpRenderTree2 Steve Block
Change-Id: I204583ecc9c6a8017349124d5813c693c7f00dc9
2010-09-29Fix a bug with the expected result path in DumpRenderTree2 Steve Block
If we fail to read from any of the possible expected result paths, we should not record an expected result path. Change-Id: Ica96b791c14a7755154be4ec9480be4d13f47a19
2010-09-29Correctly implement EventSender mouse methods in DumpRenderTree2 Steve Block
This is taken from a similar change to DumpRenderTree in https://android-git.corp.google.com/g/62449 Change-Id: I01418100ace1cf1ccfbb4fd28d72b99b54e0c140
2010-09-29Trim empty lines when comparing text results in DumpRenderTree2 Steve Block
Change-Id: I94c9de8e6131f6b60456700e9d5a3b27c8c9d1c3
2010-09-28add multiple device support to layout test driver script Guang Zhu
pass on the serial number onto adb command so that it don't get confused when multiple devices are connected Change-Id: I0b16f9da941d77425bcc5f782393f8800e035871
2010-09-28Reduce the AppCache maximum size to a more reasonable value in DumpRenderTree2 Steve Block
Without this change, any use of AppCache fails after the test http/tests/appcache/max-size.html has been run. The test in question passes, but seems to leave the database in a corrupted state, causing all future database transactions to fail. The only fix is to remove the database file. This behaviour is only observed with DumpRenderTree2, not DumpRenderTree, so I don't think this is a WebView problem. I'm not sure what the root cause is, but I think it make sense to submit this fix now in order to get the LayoutTests stable with DumpRenderTree2. Change-Id: I763c09406e66839676b839e3cdeba3b9427c1cb9
2010-09-28Merge changes Ic62f60d9,I04a01634 Steve Block
* changes: Provide a useful message in DumpRenderTree2 GUI when the host server is not running Prevent DumpRenderTree2 from crashing when the host server is not running
2010-09-28Provide a useful message in DumpRenderTree2 GUI when the host server is not ↵ Steve Block
running Bug: 3010758 Change-Id: Ic62f60d96055e7ad1e475215ca384306cdfa859c
2010-09-28Prevent DumpRenderTree2 from crashing when the host server is not running Steve Block
Bug: 3010758 Change-Id: I04a01634544d223d225b264827d3cf18be74e1e0
2010-09-28Update run_layout_tests.py to give a useful error if DumpRenderTree2 is not ↵ Steve Block
found on the device Bug: 3006981 Change-Id: Ie789eb726bc051305371d4958c16ceca980ea1d1
2010-09-26Bug 2372180: pass multi-touch events from browser to webkit. Huahui Wu
Since the UI uses multi-touch for zooming, the event is passed to webkit only when the webpage doesn't allow zooming. 1. Updated the data structure TouchEventData in WebViewCore.java to take multiple points and used it everywhere. 2. Added a passMultiTouchToWebkit() function to do what its name means. 3. Added the multi-point touch support to dumprendertree. Change-Id: Iafc1aa7b21a587ad26efd2f124b9a66316297ab8
2010-09-22Fix DumpRenderTree2 to clear AppCache between tests Steve Block
Change-Id: I1bafc79bbdbec0df7e4c8039488478a3bfcb3bd4
2010-09-22Fix DumpRenderTree2 to wait for the WebCore thread to become ready Steve Block
This is required to make sure that the WebCore thread's message handlers are ready to receive settings. Change-Id: I2c4df95e3b7409535b739ada957d93a7e6cd78d6
2010-09-22Fix a bug in DumpRenderTree2's run_apache2.py script Steve Block
We use a printf style approach when setting the run command in the command string to be executed. When doing so, the printf must be applied to only the relevant part of the string. This avoids problems in the case that other parts of the string contain special character sequences, such as '%s' in a custom log directive. Change-Id: I87df773a2872320386a34cb6a5041a9e0ba6f31e
2010-09-22Fix DumpRenderTree2 to make AbstractResult serializable Steve Block
This should have been doen as part of https://android-git.corp.google.com/g/#change,65283 Change-Id: I7a4681f98014ff5a7546a7346fc5523e01f62e19
2010-09-22Fix DumpRenderTree2 to handle SSL errors Steve Block
This fixes http/tests/appcache/different-https-origin-resource-main.html Change-Id: I90e151c027b2d8c55ea82104c02e4a641b933024
2010-09-22Merge "Fix DumpRenderTree2 to correctly add JavaScript interfaces to new ↵ Steve Block
windows"
2010-09-17Updated Summarizer to use SQLite DB. Maksymilian Osowski
It avoids OOM issues by dumping the memory to the DB every few hundred tests and then reading them in chunks when generating summaries, etc. Change-Id: I5f63d960d456f17e9f2189230a7ae0ca5607ce51
2010-09-17Changed Summarizer to dump generated html code to file every few hundred ↵ Maksymilian Osowski
tests to avoid OOM errors. Change-Id: Id0917d45ffab2e47d67d17d629f342add5292d4e
2010-09-17Fix DumpRenderTree2 to serve HTTP tests on http://127.0.0.1:8000 Steve Block
Currently DumpRenderTree2 serves HTTP tests on http://localhost:8080. Some of the tests hardcode 'http://127.0.0.1:8000' for resources, so we must use this scheme and port for the requests to be same-origin. This fixes LayoutTest http/tests/appcache/credential-url.html Change-Id: I2a9c402fc7b0e0b62eb4a278d67ac0e63ef96e8b
2010-09-17Merge "Updated ManagerService to call Summarizer.summarize() in a separate ↵ Maksymilian Osowski
thread."
2010-09-17Merge "Changed the behaviour of run_apache2.py restart." Maksymilian Osowski
2010-09-17Changed the behaviour of run_apache2.py restart. Maksymilian Osowski
Now it explicitly stops and starts the server and not just invokes apache2 -k restart Change-Id: I74cfa9c3388552fcc56b62c5ab614262ca540d0b
2010-09-17Updated ManagerService to call Summarizer.summarize() in a separate thread. Maksymilian Osowski
This could potentially take a long time and result in ANR. Change-Id: Ib8b246e627e841217b68c955d92c97801e93d923
2010-09-17Merge "Update DumpRenderTree2 to handle failing tests where no meaningful ↵ Steve Block
diff is available"
2010-09-17Fix DumpRenderTree2 to correctly add JavaScript interfaces to new windows Steve Block
When adding JavaScript interfaces to a WebView, they must be passed to the WebView constructor, rather than to WebView.addJavascriptInterface() if they are to be used immediately without the risk of race conditions. This fixes LayoutTest http/tests/appcache/crash-when-navigating-away-then-back.html This was fixed for DumpRenderTree in https://android-git.corp.google.com/g/23804 Change-Id: I1581d6d6cfc1b35be1ec2f2e5b5b944da2b4b935
2010-09-17Update DumpRenderTree2 to handle failing tests where no meaningful diff is ↵ Steve Block
available Currently, DumpRenderTree2 asigns the following result codes {PASS, FAIL_RESULT_DIFFERS, FAIL_NO_EXPECTED_RESULT, FAIL_TIMED_OUT, FAIL_CRASHED} This is not strictly correct, as a test may fail for any of three reasons - crashing, timing out, or the the result not being as expected. Therefore, it's possible for a test to fail even if the result is as expected. This patch updates AbstractResult to handle each of the three reasons for failure separately. We then test all three to determine if the test has passed. This allows us to correctly report whether or not the result differs from expected for a failing test. Change-Id: I7adcfe72c4dd0bd3de2e1b868d9807be6eb5bddf