libwebp: upstream animation cherry-picks

libwebp: upstream animation cherry-picks

WebPIterator now also denotes if the frame has alpha.
Support for "Do not blend" in demux library.

These two improve the decoding performance of animated WebP images.

BUG=

Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=218714

[fbarchard, 2013-08-21 02:00:13]

https://codereview.chromium.org/22802020/diff/1/third_party/libwebp/demux/dem...
File third_party/libwebp/demux/demux.c (right):

https://codereview.chromium.org/22802020/diff/1/third_party/libwebp/demux/dem...
third_party/libwebp/demux/demux.c:259: frame->has_alpha_ |= features.has_alpha;
if you dont check the pointer, to enable alpha, theres a risk that the user may
pass a null by accident?  May want to check the pointer as well.

https://codereview.chromium.org/22802020/diff/1/third_party/libwebp/webp/mux.h
File third_party/libwebp/webp/mux.h (right):

https://codereview.chromium.org/22802020/diff/1/third_party/libwebp/webp/mux....
third_party/libwebp/webp/mux.h:202: WebPMuxAnimBlend   blend_method;    // Blend
operation for the frame.
Is WebPMuxAnimBlend   blend_method; a full int?

[urvang (Google), 2013-08-21 02:16:51]

https://codereview.chromium.org/22802020/diff/1/third_party/libwebp/demux/dem...
File third_party/libwebp/demux/demux.c (right):

https://codereview.chromium.org/22802020/diff/1/third_party/libwebp/demux/dem...
third_party/libwebp/demux/demux.c:259: frame->has_alpha_ |= features.has_alpha;
On 2013/08/21 02:00:13, fbarchard wrote:
> if you dont check the pointer, to enable alpha, theres a risk that the user
may
> pass a null by accident?  May want to check the pointer as well.

Nope. StoreFrame() is a static method, so 'frame' is always non-null here.

https://codereview.chromium.org/22802020/diff/1/third_party/libwebp/webp/mux.h
File third_party/libwebp/webp/mux.h (right):

https://codereview.chromium.org/22802020/diff/1/third_party/libwebp/webp/mux....
third_party/libwebp/webp/mux.h:202: WebPMuxAnimBlend   blend_method;    // Blend
operation for the frame.
On 2013/08/21 02:00:13, fbarchard wrote:
> Is WebPMuxAnimBlend   blend_method; a full int?

I guess the question comes from the fact that we decremented padding size by 1.

IIRC, the size of WebPMuxAnimBlend is not guaranteed to be that of an int (as 1
byte can hold it too), but most implementations would make the size to be that
of an int.

[fbarchard, 2013-08-21 02:18:52]

lgtm.

On 2013/08/21 02:16:51, urvang wrote:
>
https://codereview.chromium.org/22802020/diff/1/third_party/libwebp/demux/dem...
> File third_party/libwebp/demux/demux.c (right):
> 
>
https://codereview.chromium.org/22802020/diff/1/third_party/libwebp/demux/dem...
> third_party/libwebp/demux/demux.c:259: frame->has_alpha_ |=
features.has_alpha;
> On 2013/08/21 02:00:13, fbarchard wrote:
> > if you dont check the pointer, to enable alpha, theres a risk that the user
> may
> > pass a null by accident?  May want to check the pointer as well.
> 
> Nope. StoreFrame() is a static method, so 'frame' is always non-null here.
> 
> https://codereview.chromium.org/22802020/diff/1/third_party/libwebp/webp/mux.h
> File third_party/libwebp/webp/mux.h (right):
> 
>
https://codereview.chromium.org/22802020/diff/1/third_party/libwebp/webp/mux....
> third_party/libwebp/webp/mux.h:202: WebPMuxAnimBlend   blend_method;    //
Blend
> operation for the frame.
> On 2013/08/21 02:00:13, fbarchard wrote:
> > Is WebPMuxAnimBlend   blend_method; a full int?
> 
> I guess the question comes from the fact that we decremented padding size by
1.
> 
> IIRC, the size of WebPMuxAnimBlend is not guaranteed to be that of an int (as
1
> byte can hold it too), but most implementations would make the size to be that
> of an int.

Right.  I looked it up and its an enum.  By default these are size of int.

[commit-bot: I haz the power, 2013-08-21 02:20:33]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/urvang@google.com/22802020/1

[commit-bot: I haz the power, 2013-08-21 05:00:52]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/urvang@google.com/22802020/1

[commit-bot: I haz the power, 2013-08-21 12:45:20]

Change committed as 218714