dany/libwebp
1
0
Fork 0
mirror of https://github.com/webmproject/libwebp.git synced 2025-07-14 21:09:55 +02:00
Code Issues Actions Packages Projects Releases Wiki Activity

Regression fix for alpha channels using color cache:

Browse Source 
Considering the fact that insert to/lookup from the color cache is always 32
bit, use DecodeImageData() variant in that case.

Conflicts:
	src/dec/vp8l.c

Change-Id: I6c665a6cfbd9bd10651c1e82fa54e687cbd54a2b
(cherry picked from commit a37eff47d6)
This commit is contained in:
Urvang Joshi
2013-06-17 03:45:09 -07:00
committed by James Zern
parent 2e377b53b0
commit 7288950b88
1 changed files with 5 additions and 4 deletions
Download Patch File Download Diff File Expand all files Collapse all files

9
src/dec/vp8l.c
View File

@ -1142,11 +1142,12 @@ int VP8LDecodeAlphaImageStream(int width, int height, const uint8_t* const data,
dec->action_ = READ_HDR;
if (!DecodeImageStream(width, height, 1, dec, NULL)) goto Err;
// Special case: if alpha data contains only the color indexing transform
// (a frequent case), we will use DecodeAlphaData() method that only needs
// allocation of 1 byte per pixel (alpha channel).
// Special case: if alpha data uses only the color indexing transform and
// doesn't use color cache (a frequent case), we will use DecodeAlphaData()
// method that only needs allocation of 1 byte per pixel (alpha channel).
if (dec->next_transform_ == 1 &&
dec->transforms_[0].type_ == COLOR_INDEXING_TRANSFORM) {
dec->transforms_[0].type_ == COLOR_INDEXING_TRANSFORM &&
dec->hdr_.color_cache_size_ == 0) {
bytes_per_pixel = sizeof(uint8_t);
}