GTTF: No more FLAKY_ .

GTTF: No more FLAKY_ .

Please note that FLAKY_ tests have been ignored anyway. When tests started
crashing, people just flipped that to DISABLED_ . Why not go straight to
DISABLED_ then, so that we avoid wasting time on stupid test prefix games?

With DISABLED_ it is clear to everyone that there is no coverage from given
test. FLAKY_ creates an illusion of coverage, while in fact the test is still
ignored.

If a FLAKY_ test fails and nobody notices, does it still make a sound? ;-)

Finally, note that gtest has a --gtest_also_run_disabled_tests if you need
to run tests manually.

TBR=jam

BUG=none

Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=174472

[Paweł Hajdan Jr., 2012-12-21 22:50:02]

Please review:

John: entire CL
Chris: FYI,
https://codereview.chromium.org/11664013/diff/1/media/ffmpeg/ffmpeg_regressio...
needs some love I guess. My suggestion is create a copy of the test without
ASSERTs/EXPECTs and undisable the tests - that may accomplish the goal.

[jam, 2012-12-21 23:12:21]

lgtm

https://codereview.chromium.org/11664013/diff/1/base/test/test_suite.h
File base/test/test_suite.h (left):

https://codereview.chromium.org/11664013/diff/1/base/test/test_suite.h#oldcode26
base/test/test_suite.h:26: typedef bool (*TestMatch)(const testing::TestInfo&);
this isn't used anymore, remove

https://codereview.chromium.org/11664013/diff/1/chrome/common/time_format_uni...
File chrome/common/time_format_unittest.cc (right):

https://codereview.chromium.org/11664013/diff/1/chrome/common/time_format_uni...
chrome/common/time_format_unittest.cc:52: TEST(TimeFormat, RelativeDate) {
seems like this should be DISABLED