1
0
mirror of https://github.com/mpv-player/mpv synced 2025-01-01 12:22:22 +00:00
Commit Graph

106 Commits

Author SHA1 Message Date
diego
e0bc115b4d cosmetics: Rename struct vf_instance_s --> vf_instance.
git-svn-id: svn://svn.mplayerhq.hu/mplayer/trunk@30684 b3059339-0415-0410-9bf9-f77b7e298cf2
2010-02-21 15:48:03 +00:00
diego
28cf6ea2be Rename open() vf initialization function to vf_open().
This avoids clashes with fcntl.h under certain circumstances.


git-svn-id: svn://svn.mplayerhq.hu/mplayer/trunk@30680 b3059339-0415-0410-9bf9-f77b7e298cf2
2010-02-21 13:40:49 +00:00
diego
23db422c3d Add missing license headers to all files in the libmpcodecs directory.
git-svn-id: svn://svn.mplayerhq.hu/mplayer/trunk@30463 b3059339-0415-0410-9bf9-f77b7e298cf2
2010-01-30 16:57:40 +00:00
reimar
cbcae4856c Make the scale filter prefer yuv conversions that do not need chroma scaling.
Also always try keeping input and output format the same first.


git-svn-id: svn://svn.mplayerhq.hu/mplayer/trunk@30409 b3059339-0415-0410-9bf9-f77b7e298cf2
2010-01-23 19:00:09 +00:00
reimar
86f1430a59 Remove the query_format_cache, it is extra complexity that makes improvements
difficult while the speed advantage in normal use cases is not measurable.


git-svn-id: svn://svn.mplayerhq.hu/mplayer/trunk@30408 b3059339-0415-0410-9bf9-f77b7e298cf2
2010-01-23 18:37:47 +00:00
reimar
f1714582b2 Simplify loop condition
git-svn-id: svn://svn.mplayerhq.hu/mplayer/trunk@30407 b3059339-0415-0410-9bf9-f77b7e298cf2
2010-01-23 18:28:47 +00:00
reimar
ef395a2f08 Replace deprecated sws_scale_ordered usages by sws_scale (which does the same).
git-svn-id: svn://svn.mplayerhq.hu/mplayer/trunk@30192 b3059339-0415-0410-9bf9-f77b7e298cf2
2010-01-03 17:12:12 +00:00
reimar
0eb6b9aa95 One more fix for -vf scale with alpha.
git-svn-id: svn://svn.mplayerhq.hu/mplayer/trunk@30159 b3059339-0415-0410-9bf9-f77b7e298cf2
2010-01-01 01:25:35 +00:00
reimar
c0cae7b327 Also pass alpha plane to swscale, avoids crashes when converting between
YUV with alpha and some other format supporting alpha.


git-svn-id: svn://svn.mplayerhq.hu/mplayer/trunk@30158 b3059339-0415-0410-9bf9-f77b7e298cf2
2010-01-01 01:06:34 +00:00
reimar
8afc25d710 Add support for YUV format with alpha and fix the codecs.conf entry for vp6a
to use it.
Fixes playback of samples in http://samples.mplayerhq.hu/FLV/flash_with_alpha/


git-svn-id: svn://svn.mplayerhq.hu/mplayer/trunk@30157 b3059339-0415-0410-9bf9-f77b7e298cf2
2009-12-31 23:45:07 +00:00
reimar
0ff930da82 Try to put the list of output formats for vf_scale in a more sensible
order.
While the 16-bit yuv formats should not be preferred over the 8-bit ones,
it seems reasonable to probe them directly after the equivalent 8-bit formats.


git-svn-id: svn://svn.mplayerhq.hu/mplayer/trunk@30155 b3059339-0415-0410-9bf9-f77b7e298cf2
2009-12-31 23:25:21 +00:00
reimar
a326622304 Add support for 16-bit per component YUV formats.
git-svn-id: svn://svn.mplayerhq.hu/mplayer/trunk@30152 b3059339-0415-0410-9bf9-f77b7e298cf2
2009-12-31 22:53:25 +00:00
reimar
22c0b2ed57 Support 440P in scale filter.
git-svn-id: svn://svn.mplayerhq.hu/mplayer/trunk@30142 b3059339-0415-0410-9bf9-f77b7e298cf2
2009-12-30 12:06:09 +00:00
reimar
4198e9d45b Remove unreachable parsing code: for filters that use the option parser,
the args argument to open will always be NULL and vf->priv will always be
!= NULL.


git-svn-id: svn://svn.mplayerhq.hu/mplayer/trunk@29945 b3059339-0415-0410-9bf9-f77b7e298cf2
2009-11-21 19:50:33 +00:00
reimar
6fa77ca503 Allow vf_scale to output to RGB48, though still prefer the 24 and 32 bit formats.
git-svn-id: svn://svn.mplayerhq.hu/mplayer/trunk@29943 b3059339-0415-0410-9bf9-f77b7e298cf2
2009-11-21 18:58:03 +00:00
cehoyos
b92666b582 Add IMGFMT_RGB48LE and IMGFMT_RGB48BE.
git-svn-id: svn://svn.mplayerhq.hu/mplayer/trunk@29486 b3059339-0415-0410-9bf9-f77b7e298cf2
2009-08-11 17:51:11 +00:00
diego
e5751ced4f Replace WORDS_BIGENDIAN by HAVE_BIGENDIAN in all internal code.
git-svn-id: svn://svn.mplayerhq.hu/mplayer/trunk@29443 b3059339-0415-0410-9bf9-f77b7e298cf2
2009-07-26 19:53:00 +00:00
diego
6e9cbdc104 whitespace cosmetics: Remove all trailing whitespace.
git-svn-id: svn://svn.mplayerhq.hu/mplayer/trunk@29305 b3059339-0415-0410-9bf9-f77b7e298cf2
2009-05-13 02:58:57 +00:00
reimar
49e2373fe7 Consistently use MP_MAX_PLANES as size for plane pointer/stride arrays in libmpcodecs.
This might avoid some issues since sws_scale in some cases assumes these
have at least 4 valid entries.


git-svn-id: svn://svn.mplayerhq.hu/mplayer/trunk@29103 b3059339-0415-0410-9bf9-f77b7e298cf2
2009-03-29 14:46:59 +00:00
gpoirier
0f224d32dc revert #ifdef WORDS_BIGENDIAN => #if WORDS_BIGENDIAN changes from r28331
git-svn-id: svn://svn.mplayerhq.hu/mplayer/trunk@28334 b3059339-0415-0410-9bf9-f77b7e298cf2
2009-01-16 13:19:29 +00:00
gpoirier
5538dcf03b fix compilation on non x86 machines (PPC here)
git-svn-id: svn://svn.mplayerhq.hu/mplayer/trunk@28331 b3059339-0415-0410-9bf9-f77b7e298cf2
2009-01-16 10:07:14 +00:00
reimar
93d598c7b2 Lots and lots of #ifdef ARCH_... -> #if ARCH_...
and #ifdef HAVE_MMX etc -> #if HAVE_MMX.
There might be still more that need to be fixed.


git-svn-id: svn://svn.mplayerhq.hu/mplayer/trunk@28325 b3059339-0415-0410-9bf9-f77b7e298cf2
2009-01-16 09:21:21 +00:00
diego
6b52a2e974 Change all occurrences of asm and __asm to __asm__, same as was done for FFmpeg.
Neither variant is valid C99 syntax, but __asm__ is the most portable variant.


git-svn-id: svn://svn.mplayerhq.hu/mplayer/trunk@27788 b3059339-0415-0410-9bf9-f77b7e298cf2
2008-10-16 18:59:27 +00:00
michael
5f5225d0b8 Factorize "int i".
git-svn-id: svn://svn.mplayerhq.hu/mplayer/trunk@26735 b3059339-0415-0410-9bf9-f77b7e298cf2
2008-05-11 12:37:04 +00:00
diego
2f690cb652 Consistently #include mpbswap.h instead of bswap.h everywhere.
git-svn-id: svn://svn.mplayerhq.hu/mplayer/trunk@26574 b3059339-0415-0410-9bf9-f77b7e298cf2
2008-04-28 14:43:30 +00:00
reimar
b95c80ca86 Make all vf_info_t structs const
git-svn-id: svn://svn.mplayerhq.hu/mplayer/trunk@25249 b3059339-0415-0410-9bf9-f77b7e298cf2
2007-12-02 14:57:15 +00:00
diego
2c74f01a5f warning fix:
vf_scale.c:359:92: warning: trigraph ??) ignored, use -trigraphs to enable


git-svn-id: svn://svn.mplayerhq.hu/mplayer/trunk@24269 b3059339-0415-0410-9bf9-f77b7e298cf2
2007-08-28 17:06:00 +00:00
reimar
6b7e82c9eb 100l, fix compile error in r23498 due to bad copy in paste
git-svn-id: svn://svn.mplayerhq.hu/mplayer/trunk@23499 b3059339-0415-0410-9bf9-f77b7e298cf2
2007-06-07 20:01:05 +00:00
reimar
34af64d86b RGB8/BGR8 IMGFMTs are paletted in case of swscale input.
Actually they are also for output, but swscale does not support PIX_FMT_PAL8 output.


git-svn-id: svn://svn.mplayerhq.hu/mplayer/trunk@23498 b3059339-0415-0410-9bf9-f77b7e298cf2
2007-06-07 19:11:25 +00:00
reimar
13fa313b55 Remove useless fastmemcpy.h include for filters that do not even do any memcpy.
git-svn-id: svn://svn.mplayerhq.hu/mplayer/trunk@23391 b3059339-0415-0410-9bf9-f77b7e298cf2
2007-05-27 20:36:04 +00:00
michael
28f35b52f6 swap palette on big endian
git-svn-id: svn://svn.mplayerhq.hu/mplayer/trunk@22246 b3059339-0415-0410-9bf9-f77b7e298cf2
2007-02-17 12:58:35 +00:00
michael
8fd50571b0 palette init code for IMGFMT_RGB4 and IMGFMT_RG4B
git-svn-id: svn://svn.mplayerhq.hu/mplayer/trunk@22245 b3059339-0415-0410-9bf9-f77b7e298cf2
2007-02-17 11:40:01 +00:00
michael
6b328af8eb add IMGFMT_RGB8 palette init code what moron actually added init code just for half of the formats?!
git-svn-id: svn://svn.mplayerhq.hu/mplayer/trunk@22243 b3059339-0415-0410-9bf9-f77b7e298cf2
2007-02-17 11:34:25 +00:00
michael
29aefa57b4 another ridiculous palette bug which has nothing to do with the new palete support in sws
git-svn-id: svn://svn.mplayerhq.hu/mplayer/trunk@22242 b3059339-0415-0410-9bf9-f77b7e298cf2
2007-02-17 11:04:13 +00:00
michael
40bbdc2358 PAL8 ...
git-svn-id: svn://svn.mplayerhq.hu/mplayer/trunk@22237 b3059339-0415-0410-9bf9-f77b7e298cf2
2007-02-16 21:17:41 +00:00
reimar
6d32d91eca Mark m_struct_t defaults as const
git-svn-id: svn://svn.mplayerhq.hu/mplayer/trunk@22045 b3059339-0415-0410-9bf9-f77b7e298cf2
2007-01-28 16:48:01 +00:00
lucabe
1f88518638 Use PIX_FMT_* instead of IMGFMT_* when calling sws_getContext()
git-svn-id: svn://svn.mplayerhq.hu/mplayer/trunk@19876 b3059339-0415-0410-9bf9-f77b7e298cf2
2006-09-17 15:02:13 +00:00
reimar
fa94660a60 A value of 2 is valid for noup, too
git-svn-id: svn://svn.mplayerhq.hu/mplayer/trunk@19526 b3059339-0415-0410-9bf9-f77b7e298cf2
2006-08-24 17:21:03 +00:00
michael
14ddfdddf8 vertical scaler with accurate rounding, some people on doom9 can see +-1 errors
the +-1 issue is limited to >2tap vertical filters, so bilinear upscale was unaffected
the new code is sometime faster sometimes slower but the difference is significant (~20%) so its optional and enabled with arnd=1


git-svn-id: svn://svn.mplayerhq.hu/mplayer/trunk@19177 b3059339-0415-0410-9bf9-f77b7e298cf2
2006-07-24 10:36:06 +00:00
lucabe
cc30aae17b Move postproc ---> libswscale
git-svn-id: svn://svn.mplayerhq.hu/mplayer/trunk@18866 b3059339-0415-0410-9bf9-f77b7e298cf2
2006-06-30 12:00:31 +00:00
diego
eebf08b978 Convert all if(verbose>X) to mp_msg_test calls.
git-svn-id: svn://svn.mplayerhq.hu/mplayer/trunk@17933 b3059339-0415-0410-9bf9-f77b7e298cf2
2006-03-24 08:12:03 +00:00
michael
02b199e5e9 passing pts through the filter layer (lets see if pts or cola comes out at the end)
git-svn-id: svn://svn.mplayerhq.hu/mplayer/trunk@17907 b3059339-0415-0410-9bf9-f77b7e298cf2
2006-03-21 21:26:42 +00:00
michael
593dda77e9 10l (dont limit dimension components independantly if noup)
git-svn-id: svn://svn.mplayerhq.hu/mplayer/trunk@17675 b3059339-0415-0410-9bf9-f77b7e298cf2
2006-02-24 15:22:59 +00:00
michael
8f51734daf no upscale flag so automatic downscaling is possible in mencoder
git-svn-id: svn://svn.mplayerhq.hu/mplayer/trunk@17647 b3059339-0415-0410-9bf9-f77b7e298cf2
2006-02-18 21:12:56 +00:00
rathann
e7db4ccf1a Patch by Stefan Huehner / stefan % huehner ! org \
patch replaces '()' for the correct '(void)' in function
declarations/prototypes which have no parameters. The '()' syntax tell
thats there is a variable list of arguments, so that the compiler cannot
check this. The extra CFLAG '-Wstrict-declarations' shows those cases.

Comments about a similar patch applied to ffmpeg:

That in C++ these mean the same, but in ANSI C the semantics are
different; function() is an (obsolete) K&R C style forward declaration,
it basically means that the function can have any number and any types
of parameters, effectively completely preventing the compiler from doing
any sort of type checking. -- Erik Slagter

Defining functions with unspecified arguments is allowed but bad.
With arguments unspecified the compiler can't report an error/warning
if the function is called with incorrect arguments. -- Måns Rullgård


git-svn-id: svn://svn.mplayerhq.hu/mplayer/trunk@17567 b3059339-0415-0410-9bf9-f77b7e298cf2
2006-02-09 14:08:03 +00:00
ods15
87b66affe9 deobfuscate some very simple code...
git-svn-id: svn://svn.mplayerhq.hu/mplayer/trunk@17216 b3059339-0415-0410-9bf9-f77b7e298cf2
2005-12-17 20:00:16 +00:00
diego
36e1974be0 Unify include path handling, -I.. is in CFLAGS.
git-svn-id: svn://svn.mplayerhq.hu/mplayer/trunk@17013 b3059339-0415-0410-9bf9-f77b7e298cf2
2005-11-18 14:39:25 +00:00
alex
6ee561dbc4 10l
git-svn-id: svn://svn.mplayerhq.hu/mplayer/trunk@15538 b3059339-0415-0410-9bf9-f77b7e298cf2
2005-05-21 15:00:27 +00:00
nicodvb
0cde424897 subtracting 8 from negative w and h rounds the dimension to the closest multiple of 16
git-svn-id: svn://svn.mplayerhq.hu/mplayer/trunk@14925 b3059339-0415-0410-9bf9-f77b7e298cf2
2005-03-06 21:15:24 +00:00
syrjala
c260a1139d Improved NV12/NV21 support.
- Fixed PlanarToNV12Wrapper() and made it handle NV21.
- Added yuv2nv12XinC() to handle software scaling.
- Added NV12/NV21 handling to various places.
- Removed NV12 from vf_hue and vf_spp as they don't look like they can
  actually handle it.


git-svn-id: svn://svn.mplayerhq.hu/mplayer/trunk@14716 b3059339-0415-0410-9bf9-f77b7e298cf2
2005-02-16 23:47:00 +00:00