diff options
| author | darin@chromium.org <darin@chromium.org@0039d316-1c4b-4281-b951-d872f2087c98> | 2010-12-03 21:11:20 +0000 |
|---|---|---|
| committer | darin@chromium.org <darin@chromium.org@0039d316-1c4b-4281-b951-d872f2087c98> | 2010-12-03 21:11:20 +0000 |
| commit | 2d8099cf56475b6989911ea8e573cbeb26724f3d (patch) | |
| tree | 05b33ac2edbaf2bddddf7f75f904b3c9949ccab7 /chrome_frame/chrome_frame_activex_base.h | |
| parent | 9df79fddac6dc70909ac16207c9e63f2660ca5d1 (diff) | |
| download | chromium_src-2d8099cf56475b6989911ea8e573cbeb26724f3d.zip chromium_src-2d8099cf56475b6989911ea8e573cbeb26724f3d.tar.gz chromium_src-2d8099cf56475b6989911ea8e573cbeb26724f3d.tar.bz2 | |
Allow data URLs to trigger downloads, as they do in Firefox.
Unfortunately, many of our tests rely on being able to use WebURLLoader to load
data: URLs when there is no ResourceLoaderBridge implementation. In those tests,
we would crash if we try to use the ResourceLoaderBridge. To workaround that,
and because it probably a good optimization anyways, I decided to check if the
data URL has a supported MIME type, and if it does, then I let it load directly
as before.
Since data URLs may be very large, I modified DataURL::Parse to skip parsing the
'data' section of the URL if the corresponding out param is null.
R=tony
BUG=38546
TEST=none (I would like to add a download test, but they are all disabled or flaky.)
Review URL: http://codereview.chromium.org/5542001
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@68212 0039d316-1c4b-4281-b951-d872f2087c98
Diffstat (limited to 'chrome_frame/chrome_frame_activex_base.h')
0 files changed, 0 insertions, 0 deletions