dany/libwebp
1
0
Fork 0
mirror of https://github.com/webmproject/libwebp.git synced 2024-11-20 12:28:26 +01:00
Code Issues Actions Packages Projects Releases Wiki Activity
1,693 Commits 34 Branches 64 Tags 14 MiB
b524e3369c
Commit Graph

1693 Commits

This Branch
This Branch
All Branches
Author SHA1 Message Date
skal
7e2d65950f Merge "remove the PACK() bit-packing tricks" 2013-09-04 23:55:41 -07:00
skal
c13fecf908 remove the PACK() bit-packing tricks 
was too smart for its own good :)
This is more ARM-friendly, since it removes a mult.

Change-Id: If146034c8efa2e71e3eaaf1230cb553884a42ebb
 2013-09-05 08:53:36 +02:00
Pascal Massimino
2fd091c9ae Merge "use NULL for lf_stats_ testing, not bool" 2013-09-04 07:18:42 -07:00
Pascal Massimino
b11c9d6284 dwebp: use default dct_method 
No significant speed diff was observed.
Avoids idct mismatch artifacts.

Change-Id: I9ca64d165d313f23b4aa231e7a97a20213ff86c7
 2013-09-03 15:48:41 -07:00
James Zern
4bb8465f8c Merge "(de)mux.h: wrap pseudo-code in /* */" 2013-09-03 15:47:42 -07:00
skal
cfb56b1707 make -pass option work with token buffers 
-pass 2 can be useful sometimes. More passes usually don't help more.
This change is a step toward being able to re-code the whole picture
with varying parameter (when token buffer is used).

Change-Id: Ia2538e2069a53c080e2ad248c18a1e04623a9304
 2013-09-03 23:37:42 +02:00
James Zern
5416aab479 (de)mux.h: wrap pseudo-code in /* */ 
easier to copy / read as a block

Change-Id: Ia22689a13afd8ea2325dcdf432e35fc802d8177a
 2013-09-03 14:31:02 -07:00
Pascal Massimino
35dba337a3 use NULL for lf_stats_ testing, not bool 
Change-Id: I14c6341bce1c745b7f2d1b790f2d4f8441d01be6
 2013-09-03 08:45:59 -07:00
skal
733a7faae4 enc->Iterator memory cleanup 
* move yuv_in_/out_* scratch buffers to iterator
* add y_top_/uv_top_ shortcuts in iterator

That's ~3k of stack size instead of heap.
But it allows having several iterators work in parallel.

Change-Id: I6a437c0f2ef1e5d398c1d6a2fd4974fa0869f0c1
 2013-08-31 23:38:11 +02:00
Urvang Joshi
e81fac86dd Add support for "no blend" in webpmux binary 
Change-Id: I6d07b120626317df73f1a6f026931c5b9485a340
 2013-08-26 18:04:52 -07:00
Urvang Joshi
3b80bc4859 gif2webp: Separate out each step into a method 
We now have ReadFrame(), OptimizeFrame(), EncodeFrame() and
DisposeFrame()
methods.

Change-Id: I522834bad18dd6a7a3ddac7c00dfd829c48248f8
 2013-08-22 11:14:51 -07:00
skal
bef7e9ccd1 Add doc precision about demux object keeping pointers to data. 
Change-Id: I3d2139f975eedcce36606e586e0cbd6fa7d207e6
 2013-08-21 11:09:37 -07:00
James Zern
61405a143d dwebp: enable stdout output with WIC 
Change-Id: Ieb73a414784480945bba6cb2687468517e24e755
 2013-08-20 19:59:30 -07:00
pascal massimino
6eabb88637 Merge "Animated WebP: add "do no blend" option to spec" 2013-08-20 19:14:17 -07:00
skal
be20decb5c fix compilation for BITS 24 
in_bits is const. Trying to apply bswap on it, one gets the error message:
error: read-only variable 'in_bits' used as 'asm' output

Change-Id: I0bef494b822c83d8ea87b1938b0e486d94de4742
 2013-08-20 18:55:00 -07:00
pascal massimino
e58cc13706 Merge "dwebp: s/unsigned char/uint8_t/" 2013-08-20 15:03:46 -07:00
James Zern
72501d4309 dwebp: s/unsigned char/uint8_t/ 
Change-Id: I579e9b0a4d03998ca1dd8683d83d30d852551e45
 2013-08-20 14:57:35 -07:00
pascal massimino
2c9633e8e0 Merge "gif2webp: Insert independent frames at regular intervals." 2013-08-20 13:59:50 -07:00
Urvang Joshi
f0d6a14b64 gif2webp: Insert independent frames at regular intervals. 
We use the 'do not blend' option for creating independent frames.
We also mark the already independent frames as 'do not blend'.

This bounds the maximum number of frames that need to be decoded to
decode a given frame, thus leading to a much better decoding performance.

Change-Id: I7cef98af2b53751ec36993fd2bd54f7f4c4aad2b
 2013-08-20 11:47:48 -07:00
James Zern
b25a6fbfdc yuv.h: fix indent 
Change-Id: I0c0bd5f7f71bc44e10134bd4f788769ec25cec1f
 2013-08-19 18:06:15 -07:00
pascal massimino
ede3602e5d Merge "cosmetics: fix indent" 2013-08-19 17:56:21 -07:00
James Zern
3a65122a53 dwebp: fix stdout related output 
logic was incorrect for the error output

Change-Id: I89e665f8065490e12e1bf3d3008087d42daee16b
 2013-08-19 17:52:33 -07:00
James Zern
388a7249c9 cosmetics: fix indent 
Change-Id: Iad0fce79886bed0d61ddf2510ce133a5355ebc1f
 2013-08-19 17:51:04 -07:00
James Zern
4c7322c86f Merge "dsp: msvc compatibility" 2013-08-19 17:42:16 -07:00
skal
d50c7e3275 Merge "5-7% faster SSE2 versions of YUV->RGB conversion functions" 2013-08-19 17:27:44 -07:00
skal
b8ab784745 Merge "simplify upsampler calls: only allow 'bottom' to be NULL" 2013-08-19 17:27:36 -07:00
skal
df6cebfa9e 5-7% faster SSE2 versions of YUV->RGB conversion functions 
The C-version gets ~7-8% slower in order to match the SSE2
output exactly. The old (now off-by-1) code is kept under
the WEBP_YUV_USE_TABLE flag for reference.

(note that calc rounding precision is slightly better ~= +0.02dB)

on ARM-neon, we somehow recover the ~4% speed that was lost by mimicking
the initial C-version (see https://gerrit.chromium.org/gerrit/#/c/41610)

Change-Id: Ia4363c5ed9b4c9edff5d932b002e57bb7814bf6f
 2013-08-19 17:05:58 -07:00
skal
ad6ac32d7c simplify upsampler calls: only allow 'bottom' to be NULL 
If 'top' was meant to be NULL, then bottom and top can be
swapped. Logic is simpler.

+ fix compilation in non-FANCY_UPSAMPLING mode

Change-Id: I7c62bbb59454017f072c0945d1ff2d24d89286ff
 2013-08-19 16:47:51 -07:00
skal
a5e8afafcb output to stdout if file name is "-" 
Doesn't work with WIC

+ redirect some info messages from stdout to stderr
+ fix the error reporting upon output-writing error

Change-Id: I92b8bd7a15e656a3f3cdfbf56299f024e39453f8
 2013-08-19 13:30:27 -07:00
James Zern
f358450feb dsp: msvc compatibility 
intrin.h is available after VS2003

patch from the FreeImage project

Change-Id: I58a18a0db00e247f871d05e3ba99772704f0e079
 2013-08-16 20:46:16 -07:00
Pascal Massimino
43a7c8ebee Merge "cosmetics" 2013-08-15 11:24:30 -07:00
James Zern
4c5f19c148 Merge "bit_reader.h: cosmetics" 2013-08-15 11:11:03 -07:00
Pascal Massimino
f72fab7045 cosmetics 
Change-Id: Ib6fa2a7e1db8c5c48607c7097520ffca34a6cb66
 2013-08-15 04:16:37 -07:00
Pascal Massimino
14dd5e781a fix const-ness 
Change-Id: I09c85fdb90b4380c8fc03c8b5652bb9cf29b2021
 2013-08-15 00:21:15 -07:00
pascal massimino
b20aec4949 Merge "Support for 'do not blend' option in vwebp" 2013-08-14 20:01:38 -07:00
Urvang Joshi
dcf652223a Support for 'do not blend' option in vwebp 
Change-Id: I563e192fa727816d11d8647c559ec407205ef40b
 2013-08-14 19:51:46 -07:00
Urvang Joshi
d5bad03328 Animated WebP: add "do no blend" option to spec 
Marking certain frames as "do not blend" helps avoiding alpha-blending
at decode/render time.

It also helps inserting I-frames (frames which can be independently
decoded) into the animation.

Change-Id: Iaa222805db88d2f1c81104ce9d882e7c7ff8cfdb
 2013-08-14 11:50:15 -07:00
pascal massimino
a2f5f73de3 Merge "Support for "Do not blend" in mux and demux libraries" 2013-08-14 09:27:58 -07:00
Vikas Arora
e081f2f359 Pack code & extra_bits to Struct (VP8LPrefixCode). 
Also created variant VP8LPrefixEncodeBits that returns the
code & extra_bits only.
There's no impact on compression density and compression speed.

Change-Id: I2cafdd3438ac9270cd72ad9d57b383cdddfdfa4c
 2013-08-12 11:56:42 -07:00
Urvang Joshi
6284854bd5 Support for "Do not blend" in mux and demux libraries 
Change-Id: I9566a8e2d059fe1ebd9ca99c7e13311bf3f8f281
 2013-08-12 11:49:00 -07:00
Pascal Massimino
f486aaa9f8 Merge "slightly faster ParseIntraMode" 2013-08-09 02:17:16 -07:00
Pascal Massimino
d17186328c slightly faster ParseIntraMode 
+ cosmetics

Change-Id: Icb906d5f84f025ef9e04b71a73801a22cc990ee5
 2013-08-09 02:15:21 -07:00
James Zern
3ceca8ad31 bit_reader.h: cosmetics 
- use const where applicable
- drop unnecessary string.h include

Change-Id: I560eef84fe17d3925768f6817c02ea79604c4379
 2013-08-06 14:25:26 -07:00
Vikas Arora
69257f70df Create LUT for PrefixEncode. 
This speeds up lossless compression by 5%.
Change-Id: Ifd114b1d9850dc3aac74593809e7d48529d35e3d
 2013-08-05 10:20:18 -07:00
Pascal Massimino
988b70844e add WebPWorkerExecute() for convenient bypass 
This is mainly for re-using the worker structs without using the
thread.

Change-Id: I8e1be29e53874ef425b15c192fb68036b4c0a359
 2013-08-02 12:20:15 -07:00
skal
06e24987e7 Merge "VP8EncIterator clean-up" 2013-08-02 00:40:02 -07:00
skal
de4d4ad598 VP8EncIterator clean-up 
- remove unused fields from iterator
- introduce VP8IteratorSetRow() too
- rename 'done_' to 'countdown_'
- bring y_left_/u_left_/v_left_ from VP8Encoder

Change-Id: Idc1c15743157936e4cbb7002ebb5cc3c90e7f92a
 2013-08-01 23:05:54 -07:00
pascal massimino
7bbe95293f Merge "cosmetics: thread.c: drop a redundant comment" 2013-07-31 23:56:06 -07:00
James Zern
da41148560 cosmetics: thread.c: drop a redundant comment 
+ fix #if/#else/#endif comment

Change-Id: I76bfd3b123ed181897ed0feba721d5c1a3a2b0d7
 2013-07-31 22:55:51 -07:00
James Zern
feb4b6e6b3 thread.h: #ifdef when checking WEBP_USE_THREAD 
prevents a warning with -Wundef and is consistent with thread.c

Change-Id: I60fa337c3b9daeea7302e86802402cb943cdb262
 2013-07-31 22:50:57 -07:00