diff options
| author | jcivelli@chromium.org <jcivelli@chromium.org@0039d316-1c4b-4281-b951-d872f2087c98> | 2010-06-26 01:32:22 +0000 |
|---|---|---|
| committer | jcivelli@chromium.org <jcivelli@chromium.org@0039d316-1c4b-4281-b951-d872f2087c98> | 2010-06-26 01:32:22 +0000 |
| commit | 56ba1c2f48dab0e3069108692352339ae99b11e0 (patch) | |
| tree | 8d3a7d4ddbb1ffc26d8d95b22e41106e1be05a53 /chrome/common/worker_messages.h | |
| parent | 4416c5a0892d7230203654abe4952794da270308 (diff) | |
| download | chromium_src-56ba1c2f48dab0e3069108692352339ae99b11e0.zip chromium_src-56ba1c2f48dab0e3069108692352339ae99b11e0.tar.gz chromium_src-56ba1c2f48dab0e3069108692352339ae99b11e0.tar.bz2 | |
TranslateInfoBarDelegate2 were leaked.
(Relanding, the changed unit-test was flaky.
Turns out comparing pointers in that context to see that an object changed is a bad idea as a new object might be at the exact same address as an old object.)
This revealed that the NormalTranslate unit-test was incorrect (it was not testing that changing the source or target language from the after translate infobar causes a translation). This also revealed that returning reference string in the delegate is not a brilliant idea as when we switch infobars, the new infobar might be using a string from the old infobar that might have been deleted.
BUG=None
TEST=Linux heap check bot should be green.
Review URL: http://codereview.chromium.org/2878005
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@50920 0039d316-1c4b-4281-b951-d872f2087c98
Diffstat (limited to 'chrome/common/worker_messages.h')
0 files changed, 0 insertions, 0 deletions