diff options
| author | yusukes@chromium.org <yusukes@chromium.org@0039d316-1c4b-4281-b951-d872f2087c98> | 2010-10-14 10:41:57 +0000 |
|---|---|---|
| committer | yusukes@chromium.org <yusukes@chromium.org@0039d316-1c4b-4281-b951-d872f2087c98> | 2010-10-14 10:41:57 +0000 |
| commit | c85183ef13d045a18c97a0cf3678d4883bf60ba9 (patch) | |
| tree | 633b9d8a08ace93011e002f4f33960ade1630a9b /build/common.gypi | |
| parent | 38d85f96292ba88bac7fe28d30e67fd0814b494f (diff) | |
| download | chromium_src-c85183ef13d045a18c97a0cf3678d4883bf60ba9.zip chromium_src-c85183ef13d045a18c97a0cf3678d4883bf60ba9.tar.gz chromium_src-c85183ef13d045a18c97a0cf3678d4883bf60ba9.tar.bz2 | |
Tiny cleanup: remove all use of kFallbackInputMethodId from the Chrome tree.
I also added ResetUseStubImpl() calls to some unit tests. Without them, a test in unit_tests, namely OfflineLoadPageTest.OfflinePaeProceed, crashes by NULL pointer dereference inside the newly added keyboard_library->GetHardwareKeyboardLayoutName() call:
1. Test X calls keyboard_library->GetXXX().
chromeos::CrosLibrary::Library<chromeos::KeyboardLibrary>::library_ becomes non-NULL, but the real libcros function is not called since EnsureLoaded() returns false.
2. Test Y (e.g. proxy_config_service_impl_unittest.cc) calls SetUseStubImpl(), but forgets to call ResetUseStubImpl().
3. OfflineLoadPageTest.OfflinePaeProceed calls keyboard_library->GetHardwareKeyboardLayoutName().
This time, since 'stub mode' is still on, EnsureLoaded() returns true, and the function pointer for GetHardwareKeyboardLayoutName libcros API, which is still NULL, is used.
BUG=chromium-os:6563
TEST=ran emerge, ran try bot.
Review URL: http://codereview.chromium.org/3735001
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@62530 0039d316-1c4b-4281-b951-d872f2087c98
Diffstat (limited to 'build/common.gypi')
0 files changed, 0 insertions, 0 deletions