Chromium Code Reviews
chromiumcodereview-hr@appspot.gserviceaccount.com (chromiumcodereview-hr) | Please choose your nickname with Settings | Help | Chromium Project | Gerrit Changes | Sign out
(398)

Issue 13116: Adding media filters interface definitions. (Closed)

Created:
12 years ago by scherkus (not reviewing)
Modified:
9 years, 7 months ago
CC:
chromium-reviews_googlegroups.com
Visibility:
Public.

Description

Adding media filters interface definitions.The scheduler (forward declared and mentioned in comments) will arrive in a later patch. The buffers are already checked in under media/base/buffers.h Committed: http://src.chromium.org/viewvc/chrome?view=rev&revision=6393

Patch Set 1 #

Patch Set 2 : '' #

Patch Set 3 : '' #

Unified diffs Side-by-side diffs Delta from patch set Stats (+185 lines, -0 lines) Patch
A media/base/filters.h View 1 2 1 chunk +180 lines, -0 lines 0 comments Download
M media/base/media.cc View 1 chunk +1 line, -0 lines 0 comments Download
M media/build/media.vcproj View 1 chunk +4 lines, -0 lines 0 comments Download

Messages

Total messages: 3 (0 generated)
scherkus (not reviewing)
Checking in the media filter interface definitions. We've built some very successful prototypes based on ...
12 years ago (2008-12-04 02:22:20 UTC) #1
Peter Kasting
I'm clueless about this kind of thing but it LGTM. http://codereview.chromium.org/13116/diff/8/204 File media/base/filters.h (right): http://codereview.chromium.org/13116/diff/8/204#newcode82 ...
12 years ago (2008-12-04 19:01:39 UTC) #2
scherkus (not reviewing)
12 years ago (2008-12-04 19:38:02 UTC) #3
Updated the definition of DataSourceInterface::Read to return size_t and an
error constant.

http://codereview.chromium.org/13116/diff/8/204
File media/base/filters.h (right):

http://codereview.chromium.org/13116/diff/8/204#newcode82
Line 82: virtual int Read(char* data, size_t size) = 0;
On 2008/12/04 19:01:39, pkasting wrote:
> Nit: Looks like this should return a size_t?  (If you're uncomfortable with
> returning -1 directly on failure, you could define some sort of error constant
> as -1 [the way the STL does with npos]).

Good suggestion.  I'll use the -1 representation of size_t, which is the maximum
possible value of size_t.

Powered by Google App Engine
This is Rietveld 408576698