diff options
| author | pkasting@chromium.org <pkasting@chromium.org@0039d316-1c4b-4281-b951-d872f2087c98> | 2012-04-03 01:34:14 +0000 |
|---|---|---|
| committer | pkasting@chromium.org <pkasting@chromium.org@0039d316-1c4b-4281-b951-d872f2087c98> | 2012-04-03 01:34:14 +0000 |
| commit | 4172081a0e4cbb694293dad2d5ed4ba4f2cd26b4 (patch) | |
| tree | 77cdae20592a241afafa8f86a3f75d4be87c42bf /chrome/test/functional/memory.py | |
| parent | 313f73ea0b7fd817741ed866770dcbebd9d6a66a (diff) | |
| download | chromium_src-4172081a0e4cbb694293dad2d5ed4ba4f2cd26b4.zip chromium_src-4172081a0e4cbb694293dad2d5ed4ba4f2cd26b4.tar.gz chromium_src-4172081a0e4cbb694293dad2d5ed4ba4f2cd26b4.tar.bz2 | |
Reland r129998: When input is "" (or " " with trim_whitespace true), SplitString() should return an empty vector, not a vector of one empty string.
Brett and I discussed this for a while and felt this would be wise, whereas dropping all empty segments entirely (e.g. converting "a,,b" to a vector of two elements instead of three) was probably unwise.
This also simplifies the code some.
Fixing this also required changing the code in mime_util.cc to handle empty vectors to "are codecs valid" oracle functions (in which case we return false). I also fixed some style issues there. It also required avoiding passing the empty string in a test in extension_api_unittest.cc; Aaron assures me that this code is not expected to be defensive against such inputs, but at his suggestion I also added some CHECK()s to the API.
BUG=none
TEST=none
Review URL: https://chromiumcodereview.appspot.com/9958076
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@130285 0039d316-1c4b-4281-b951-d872f2087c98
Diffstat (limited to 'chrome/test/functional/memory.py')
0 files changed, 0 insertions, 0 deletions