malloc padding to avoid reading past the malloc()ed area.
[libav.git] / doc / ffmpeg-doc.texi
CommitLineData
9181577c
FB
1\input texinfo @c -*- texinfo -*-
2
3@settitle FFmpeg Documentation
4@titlepage
5@sp 7
6@center @titlefont{FFmpeg Documentation}
7@sp 3
8@end titlepage
9
10
11@chapter Introduction
12
13FFmpeg is a very fast video and audio converter. It can also grab from
14a live audio/video source.
15
16The command line interface is designed to be intuitive, in the sense
4c5f7207
DB
17that FFmpeg tries to figure out all parameters that can possibly be
18derived automatically. You usually only have to specify the target
19bitrate you want.
9181577c
FB
20
21FFmpeg can also convert from any sample rate to any other, and resize
22video on the fly with a high quality polyphase filter.
23
24@chapter Quick Start
25
e99c4e10 26@c man begin EXAMPLES
9181577c
FB
27@section Video and Audio grabbing
28
e99c4e10
FB
29FFmpeg can use a video4linux compatible video source and any Open Sound
30System audio source:
31
9181577c 32@example
e99c4e10 33ffmpeg /tmp/out.mpg
9181577c
FB
34@end example
35
e99c4e10 36Note that you must activate the right video source and channel before
4c5f7207
DB
37launching FFmpeg with any TV viewer such as xawtv
38(@url{http://bytesex.org/xawtv/}) by Gerd Knorr. You also
39have to set the audio recording levels correctly with a
e99c4e10 40standard mixer.
9181577c 41
6bf40f39 42@section Video and Audio file format conversion
9181577c 43
4c5f7207 44* FFmpeg can use any supported file format and protocol as input:
9181577c
FB
45
46Examples:
47
4c5f7207 48* You can use YUV files as input:
9181577c
FB
49
50@example
e99c4e10 51ffmpeg -i /tmp/test%d.Y /tmp/out.mpg
9181577c
FB
52@end example
53
e99c4e10 54It will use the files:
9181577c 55@example
e99c4e10
FB
56/tmp/test0.Y, /tmp/test0.U, /tmp/test0.V,
57/tmp/test1.Y, /tmp/test1.U, /tmp/test1.V, etc...
9181577c
FB
58@end example
59
e99c4e10
FB
60The Y files use twice the resolution of the U and V files. They are
61raw files, without header. They can be generated by all decent video
62decoders. You must specify the size of the image with the @option{-s} option
4c5f7207 63if FFmpeg cannot guess it.
9181577c 64
4c5f7207 65* You can input from a raw YUV420P file:
9181577c
FB
66
67@example
e99c4e10 68ffmpeg -i /tmp/test.yuv /tmp/out.avi
9181577c
FB
69@end example
70
4c5f7207
DB
71test.yuv is a file containing raw YUV planar data. Each frame is composed
72of the Y plane followed by the U and V planes at half vertical and
e99c4e10 73horizontal resolution.
9181577c 74
4c5f7207 75* You can output to a raw YUV420P file:
9181577c
FB
76
77@example
f926cbe7 78ffmpeg -i mydivx.avi hugefile.yuv
9181577c
FB
79@end example
80
81* You can set several input files and output files:
82
83@example
e99c4e10 84ffmpeg -i /tmp/a.wav -s 640x480 -i /tmp/a.yuv /tmp/a.mpg
9181577c
FB
85@end example
86
4c5f7207
DB
87Converts the audio file a.wav and the raw YUV video file a.yuv
88to MPEG file a.mpg.
9181577c 89
6bf40f39 90* You can also do audio and video conversions at the same time:
9181577c
FB
91
92@example
e99c4e10 93ffmpeg -i /tmp/a.wav -ar 22050 /tmp/a.mp2
9181577c
FB
94@end example
95
4c5f7207 96Converts a.wav to MPEG audio at 22050Hz sample rate.
9181577c
FB
97
98* You can encode to several formats at the same time and define a
e99c4e10 99mapping from input stream to output streams:
9181577c
FB
100
101@example
e99c4e10 102ffmpeg -i /tmp/a.wav -ab 64 /tmp/a.mp2 -ab 128 /tmp/b.mp2 -map 0:0 -map 0:0
9181577c
FB
103@end example
104
4c5f7207
DB
105Converts a.wav to a.mp2 at 64 kbits and to b.mp2 at 128 kbits. '-map
106file:index' specifies which input stream is used for each output
e99c4e10 107stream, in the order of the definition of output streams.
9181577c
FB
108
109* You can transcode decrypted VOBs
110
111@example
e99c4e10 112ffmpeg -i snatch_1.vob -f avi -vcodec mpeg4 -b 800 -g 300 -bf 2 -acodec mp3 -ab 128 snatch.avi
9181577c
FB
113@end example
114
4c5f7207
DB
115This is a typical DVD ripping example; the input is a VOB file, the
116output an AVI file with MPEG-4 video and MP3 audio. Note that in this
117command we use B-frames so the MPEG-4 stream is DivX5 compatible, and
118GOP size is 300 which means one intra frame every 10 seconds for 29.97fps
119input video. Furthermore, the audio stream is MP3-encoded so you need
120to enable LAME support by passing @code{--enable-mp3lame} to configure.
121The mapping is particularly useful for DVD transcoding
e99c4e10 122to get the desired audio language.
9181577c 123
4c5f7207 124NOTE: To see the supported input formats, use @code{ffmpeg -formats}.
e99c4e10 125@c man end
9181577c
FB
126
127@chapter Invocation
128
129@section Syntax
130
e99c4e10 131The generic syntax is:
9181577c
FB
132
133@example
e99c4e10 134@c man begin SYNOPSIS
28f88dc8 135ffmpeg [[infile options][@option{-i} @var{infile}]]... @{[outfile options] @var{outfile}@}...
e99c4e10 136@c man end
9181577c 137@end example
e99c4e10
FB
138@c man begin DESCRIPTION
139If no input file is given, audio/video grabbing is done.
9181577c 140
e99c4e10
FB
141As a general rule, options are applied to the next specified
142file. For example, if you give the @option{-b 64} option, it sets the video
4c5f7207 143bitrate of the next file. The format option may be needed for raw input
e99c4e10 144files.
9181577c 145
4c5f7207
DB
146By default, FFmpeg tries to convert as losslessly as possible: It
147uses the same audio and video parameters for the outputs as the one
e99c4e10
FB
148specified for the inputs.
149@c man end
9181577c 150
e99c4e10 151@c man begin OPTIONS
9181577c
FB
152@section Main options
153
e99c4e10 154@table @option
9181577c 155@item -L
4c5f7207 156Show license.
18bff752 157
9181577c 158@item -h
4c5f7207 159Show help.
18bff752 160
e99c4e10 161@item -formats
4c5f7207 162Show available formats, codecs, protocols, ...
18bff752 163
9181577c 164@item -f fmt
4c5f7207 165Force format.
18bff752 166
9181577c 167@item -i filename
4c5f7207 168input filename
9181577c
FB
169
170@item -y
4c5f7207 171Overwrite output files.
9181577c
FB
172
173@item -t duration
4c5f7207
DB
174Set the recording time in seconds.
175@code{hh:mm:ss[.xxx]} syntax is also supported.
9181577c 176
e83a84ac 177@item -ss position
4c5f7207
DB
178Seek to given time position in seconds.
179@code{hh:mm:ss[.xxx]} syntax is also supported.
e83a84ac 180
9181577c 181@item -title string
4c5f7207 182Set the title.
9181577c
FB
183
184@item -author string
4c5f7207 185Set the author.
9181577c
FB
186
187@item -copyright string
4c5f7207 188Set the copyright.
9181577c
FB
189
190@item -comment string
4c5f7207 191Set the comment.
9181577c 192
99db6420 193@item -target type
4c5f7207
DB
194Specify target file type ("vcd", "svcd", "dvd", "dv", "pal-vcd",
195"ntsc-svcd", ... ). All the format options (bitrate, codecs,
196buffer sizes) are then set automatically. You can just type:
99db6420
FB
197
198@example
199ffmpeg -i myfile.avi -target vcd /tmp/vcd.mpg
200@end example
201
4c5f7207
DB
202Nevertheless you can specify additional options as long as you know
203they do not conflict with the standard, as in:
791d8d1d
MB
204
205@example
206ffmpeg -i myfile.avi -target vcd -bf 2 /tmp/vcd.mpg
207@end example
208
99db6420 209@item -hq
4c5f7207 210Activate high quality settings.
18bff752 211
401fbdde 212@item -itsoffset offset
4c5f7207
DB
213Set the input time offset in seconds.
214@code{[-]hh:mm:ss[.xxx]} syntax is also supported.
215This option affects all the input files that follow it.
216The offset is added to the timestamps of the input files.
217Specifying a positive offset means that the corresponding
218streams are delayed by 'offset' seconds.
401fbdde 219
9181577c
FB
220@end table
221
222@section Video Options
223
e99c4e10 224@table @option
9181577c 225@item -b bitrate
4c5f7207 226Set the video bitrate in kbit/s (default = 200 kb/s).
18bff752 227@item -r fps
4c5f7207 228Set frame rate (default = 25).
18bff752 229@item -s size
4c5f7207
DB
230Set frame size. The format is @samp{wxh} (default = 160x128).
231The following abbreviations are recognized:
5ee03c86 232@table @samp
18bff752
FB
233@item sqcif
234128x96
235@item qcif
236176x144
237@item cif
238352x288
239@item 4cif
240704x576
241@end table
242
243@item -aspect aspect
4c5f7207 244Set aspect ratio (4:3, 16:9 or 1.3333, 1.7777).
18bff752 245@item -croptop size
4c5f7207 246Set top crop band size (in pixels).
18bff752 247@item -cropbottom size
4c5f7207 248Set bottom crop band size (in pixels).
18bff752 249@item -cropleft size
4c5f7207 250Set left crop band size (in pixels).
18bff752 251@item -cropright size
4c5f7207 252Set right crop band size (in pixels).
1ff93ffc 253@item -padtop size
4c5f7207 254Set top pad band size (in pixels).
1ff93ffc 255@item -padbottom size
4c5f7207 256Set bottom pad band size (in pixels).
1ff93ffc 257@item -padleft size
4c5f7207 258Set left pad band size (in pixels).
1ff93ffc 259@item -padright size
4c5f7207 260Set right pad band size (in pixels).
5b79a73e 261@item -padcolor (hex color)
4c5f7207
DB
262Set color of padded bands. The value for padcolor is expressed
263as a six digit hexadecimal number where the first two digits
264represent red, the middle two digits green and last two digits
265blue (default = 000000 (black)).
9181577c 266@item -vn
4c5f7207 267Disable video recording.
9181577c 268@item -bt tolerance
4c5f7207 269Set video bitrate tolerance (in kbit/s).
18bff752 270@item -maxrate bitrate
4c5f7207 271Set max video bitrate tolerance (in kbit/s).
18bff752 272@item -minrate bitrate
4c5f7207 273Set min video bitrate tolerance (in kbit/s).
18bff752 274@item -bufsize size
4c5f7207 275Set rate control buffer size (in kbit).
18bff752 276@item -vcodec codec
4c5f7207 277Force video codec to @var{codec}. Use the @code{copy} special value to
5ee03c86 278tell that the raw codec data must be copied as is.
18bff752 279@item -sameq
4c5f7207 280Use same video quality as source (implies VBR).
9181577c
FB
281
282@item -pass n
4c5f7207
DB
283Select the pass number (1 or 2). It is useful to do two pass
284encoding. The statistics of the video are recorded in the first
285pass and the video is generated at the exact requested bitrate
286in the second pass.
9181577c
FB
287
288@item -passlogfile file
4c5f7207 289Set two pass logfile name to @var{file}.
9181577c
FB
290
291@end table
292
18bff752 293@section Advanced Video Options
9181577c 294
e99c4e10 295@table @option
9181577c 296@item -g gop_size
4c5f7207 297Set the group of pictures size.
9181577c 298@item -intra
4c5f7207 299Use only intra frames.
9181577c 300@item -qscale q
4c5f7207 301Use fixed video quantiser scale (VBR).
9181577c 302@item -qmin q
4c5f7207 303minimum video quantiser scale (VBR)
9181577c 304@item -qmax q
4c5f7207 305maximum video quantiser scale (VBR)
9181577c 306@item -qdiff q
4c5f7207 307maximum difference between the quantiser scales (VBR)
9181577c
FB
308@item -qblur blur
309video quantiser scale blur (VBR)
310@item -qcomp compression
311video quantiser scale compression (VBR)
18bff752
FB
312
313@item -rc_init_cplx complexity
4c5f7207 314initial complexity for single pass encoding
18bff752 315@item -b_qfactor factor
4c5f7207 316qp factor between P- and B-frames
18bff752 317@item -i_qfactor factor
4c5f7207 318qp factor between P- and I-frames
18bff752 319@item -b_qoffset offset
4c5f7207 320qp offset between P- and B-frames
18bff752 321@item -i_qoffset offset
4c5f7207 322qp offset between P- and I-frames
18bff752 323@item -rc_eq equation
4c5f7207
DB
324Set rate control equation (@pxref{FFmpeg formula
325evaluator}) (default = @code{tex^qComp}).
5ee03c86
FB
326@item -rc_override override
327rate control override for specific intervals
18bff752 328@item -me method
4c5f7207
DB
329Set motion estimation method to @var{method}.
330Available methods are (from lowest to best quality):
18bff752
FB
331@table @samp
332@item zero
5ee03c86 333Try just the (0, 0) vector.
18bff752
FB
334@item phods
335@item log
336@item x1
337@item epzs
338(default method)
339@item full
340exhaustive search (slow and marginally better than epzs)
341@end table
342
343@item -dct_algo algo
4c5f7207 344Set DCT algorithm to @var{algo}. Available values are:
18bff752
FB
345@table @samp
346@item 0
347FF_DCT_AUTO (default)
348@item 1
349FF_DCT_FASTINT
350@item 2
351FF_DCT_INT
352@item 3
353FF_DCT_MMX
354@item 4
355FF_DCT_MLIB
356@item 5
357FF_DCT_ALTIVEC
358@end table
359
360@item -idct_algo algo
4c5f7207 361Set IDCT algorithm to @var{algo}. Available values are:
18bff752
FB
362@table @samp
363@item 0
364FF_IDCT_AUTO (default)
365@item 1
366FF_IDCT_INT
367@item 2
368FF_IDCT_SIMPLE
369@item 3
370FF_IDCT_SIMPLEMMX
371@item 4
372FF_IDCT_LIBMPEG2MMX
373@item 5
374FF_IDCT_PS2
375@item 6
376FF_IDCT_MLIB
377@item 7
378FF_IDCT_ARM
379@item 8
380FF_IDCT_ALTIVEC
381@item 9
382FF_IDCT_SH4
383@item 10
384FF_IDCT_SIMPLEARM
385@end table
386
387@item -er n
4c5f7207 388Set error resilience to @var{n}.
18bff752
FB
389@table @samp
390@item 1
1471c6c2 391FF_ER_CAREFUL (default)
18bff752 392@item 2
5ee03c86 393FF_ER_COMPLIANT
18bff752
FB
394@item 3
395FF_ER_AGGRESSIVE
396@item 4
397FF_ER_VERY_AGGRESSIVE
398@end table
399
5ee03c86 400@item -ec bit_mask
4c5f7207 401Set error concealment to @var{bit_mask}. @var{bit_mask} is a bit mask of
5ee03c86 402the following values:
18bff752
FB
403@table @samp
404@item 1
4c5f7207 405FF_EC_GUESS_MVS (default = enabled)
18bff752 406@item 2
4c5f7207 407FF_EC_DEBLOCK (default = enabled)
18bff752
FB
408@end table
409
410@item -bf frames
4c5f7207 411Use 'frames' B-frames (supported for MPEG-1, MPEG-2 and MPEG-4).
18bff752
FB
412@item -mbd mode
413macroblock decision
414@table @samp
415@item 0
4c5f7207 416FF_MB_DECISION_SIMPLE: Use mb_cmp (cannot change it yet in FFmpeg).
18bff752 417@item 1
4c5f7207 418FF_MB_DECISION_BITS: Choose the one which needs the fewest bits.
18bff752 419@item 2
019c8838 420FF_MB_DECISION_RD: rate distortion
18bff752
FB
421@end table
422
423@item -4mv
4c5f7207 424Use four motion vector by macroblock (MPEG-4 only).
18bff752 425@item -part
4c5f7207 426Use data partitioning (MPEG-4 only).
18bff752 427@item -bug param
4c5f7207 428Work around encoder bugs that are not auto-detected.
18bff752 429@item -strict strictness
4c5f7207 430How strictly to follow the standards.
5ee03c86 431@item -aic
4c5f7207 432Enable Advanced intra coding (h263+).
5ee03c86 433@item -umv
4c5f7207 434Enable Unlimited Motion Vector (h263+)
18bff752
FB
435
436@item -deinterlace
4c5f7207 437Deinterlace pictures.
99db6420 438@item -interlace
4c5f7207
DB
439Force interlacing support in encoder (MPEG-2 and MPEG-4 only).
440Use this option if your input file is interlaced and you want
441to keep the interlaced format for minimum losses.
442The alternative is to deinterlace the input stream with
443@option{-deinterlace}, but deinterlacing introduces losses.
18bff752 444@item -psnr
4c5f7207 445Calculate PSNR of compressed frames.
18bff752 446@item -vstats
4c5f7207 447Dump video coding statistics to @file{vstats_HHMMSS.log}.
18bff752 448@item -vhook module
4c5f7207 449Insert video processing @var{module}. @var{module} contains the module
18bff752 450name and its parameters separated by spaces.
18bff752
FB
451@end table
452
453@section Audio Options
454
455@table @option
18bff752 456@item -ar freq
4c5f7207 457Set the audio sampling frequency (default = 44100 Hz).
18bff752 458@item -ab bitrate
4c5f7207 459Set the audio bitrate in kbit/s (default = 64).
18bff752 460@item -ac channels
4c5f7207 461Set the number of audio channels (default = 1).
5ee03c86 462@item -an
4c5f7207 463Disable audio recording.
5ee03c86 464@item -acodec codec
4c5f7207
DB
465Force audio codec to @var{codec}. Use the @code{copy} special value to
466specify that the raw codec data must be copied as is.
18bff752
FB
467@end table
468
469@section Audio/Video grab options
470
471@table @option
472@item -vd device
4c5f7207 473sEt video grab device (e.g. @file{/dev/video0}).
18bff752 474@item -vc channel
4c5f7207 475Set video grab channel (DV1394 only).
18bff752 476@item -tvstd standard
4c5f7207 477Set television standard (NTSC, PAL (SECAM)).
18bff752 478@item -dv1394
4c5f7207 479Set DV1394 grab.
18bff752 480@item -ad device
4c5f7207 481Set audio device (e.g. @file{/dev/dsp}).
18bff752
FB
482@end table
483
484@section Advanced options
485
486@table @option
487@item -map file:stream
4c5f7207 488Set input stream mapping.
18bff752 489@item -debug
4c5f7207 490Print specific debug info.
9181577c 491@item -benchmark
4c5f7207 492Add timings for benchmarking.
9181577c 493@item -hex
4c5f7207 494Dump each input packet.
18bff752 495@item -bitexact
4c5f7207 496Only use bit exact algorithms (for codec testing).
18bff752 497@item -ps size
4c5f7207 498Set packet size in bits.
5ee03c86 499@item -re
4c5f7207 500Read input at native frame rate. Mainly used to simulate a grab device.
5ee03c86 501@item -loop
4c5f7207
DB
502Loop over the input stream. Currently it works only for image
503streams. This option is used for automatic FFserver testing.
8108551a 504@item -loop_output number_of_times
019c8838 505Repeatedly loop output for formats that support looping such as animated GIF
4c5f7207 506(0 will loop the output infinitely).
9181577c 507@end table
18bff752
FB
508
509@node FFmpeg formula evaluator
510@section FFmpeg formula evaluator
511
512When evaluating a rate control string, FFmpeg uses an internal formula
513evaluator.
514
515The following binary operators are available: @code{+}, @code{-},
516@code{*}, @code{/}, @code{^}.
517
518The following unary operators are available: @code{+}, @code{-},
519@code{(...)}.
520
521The following functions are available:
522@table @var
523@item sinh(x)
524@item cosh(x)
525@item tanh(x)
526@item sin(x)
527@item cos(x)
528@item tan(x)
529@item exp(x)
530@item log(x)
531@item squish(x)
532@item gauss(x)
533@item abs(x)
534@item max(x, y)
535@item min(x, y)
536@item gt(x, y)
537@item lt(x, y)
538@item eq(x, y)
539@item bits2qp(bits)
540@item qp2bits(qp)
541@end table
542
543The following constants are available:
544@table @var
545@item PI
546@item E
547@item iTex
548@item pTex
549@item tex
550@item mv
551@item fCode
552@item iCount
553@item mcVar
554@item var
555@item isI
556@item isP
557@item isB
558@item avgQP
559@item qComp
560@item avgIITex
561@item avgPITex
562@item avgPPTex
563@item avgBPTex
564@item avgTex
565@end table
566
e99c4e10
FB
567@c man end
568
569@ignore
570
571@setfilename ffmpeg
572@settitle FFmpeg video converter
573
574@c man begin SEEALSO
019c8838 575ffserver(1), ffplay(1) and the HTML documentation of @file{ffmpeg}.
e99c4e10
FB
576@c man end
577
578@c man begin AUTHOR
579Fabrice Bellard
580@c man end
581
582@end ignore
9181577c
FB
583
584@section Protocols
585
4c5f7207
DB
586The filename can be @file{-} to read from standard input or to write
587to standard output.
9181577c 588
4c5f7207 589FFmpeg also handles many protocols specified with an URL syntax.
9181577c 590
4c5f7207 591Use 'ffmpeg -formats' to see a list of the supported protocols.
9181577c 592
e99c4e10 593The protocol @code{http:} is currently used only to communicate with
4c5f7207 594FFserver (see the FFserver documentation). When FFmpeg will be a
e99c4e10 595video player it will also be used for streaming :-)
9181577c
FB
596
597@chapter Tips
598
599@itemize
4c5f7207
DB
600@item For streaming at very low bitrate application, use a low frame rate
601and a small GOP size. This is especially true for RealVideo where
e99c4e10
FB
602the Linux player does not seem to be very fast, so it can miss
603frames. An example is:
9181577c
FB
604
605@example
e99c4e10 606ffmpeg -g 3 -r 3 -t 10 -b 50 -s qcif -f rv10 /tmp/b.rm
9181577c
FB
607@end example
608
609@item The parameter 'q' which is displayed while encoding is the current
4c5f7207
DB
610quantizer. The value 1 indicates that a very good quality could
611be achieved. The value 31 indicates the worst quality. If q=31 appears
e99c4e10 612too often, it means that the encoder cannot compress enough to meet
4c5f7207 613your bitrate. You must either increase the bitrate, decrease the
e99c4e10 614frame rate or decrease the frame size.
9181577c
FB
615
616@item If your computer is not fast enough, you can speed up the
e99c4e10
FB
617compression at the expense of the compression ratio. You can use
618'-me zero' to speed up motion estimation, and '-intra' to disable
4c5f7207 619motion estimation completely (you have only I-frames, which means it
e99c4e10 620is about as good as JPEG compression).
9181577c 621
4c5f7207 622@item To have very low audio bitrates, reduce the sampling frequency
019c8838 623(down to 22050 kHz for MPEG audio, 22050 or 11025 for AC3).
9181577c
FB
624
625@item To have a constant quality (but a variable bitrate), use the option
e99c4e10
FB
626'-qscale n' when 'n' is between 1 (excellent quality) and 31 (worst
627quality).
9181577c
FB
628
629@item When converting video files, you can use the '-sameq' option which
4c5f7207
DB
630uses the same quality factor in the encoder as in the decoder.
631It allows almost lossless encoding.
9181577c
FB
632
633@end itemize
634
635@chapter Supported File Formats and Codecs
636
637You can use the @code{-formats} option to have an exhaustive list.
638
639@section File Formats
640
6bf40f39 641FFmpeg supports the following file formats through the @code{libavformat}
0699d2fe 642library:
9181577c 643
3fc9e78f 644@multitable @columnfractions .4 .1 .1 .4
9181577c
FB
645@item Supported File Format @tab Encoding @tab Decoding @tab Comments
646@item MPEG audio @tab X @tab X
4c5f7207 647@item MPEG-1 systems @tab X @tab X
9181577c 648@tab muxed audio and video
4c5f7207 649@item MPEG-2 PS @tab X @tab X
9181577c 650@tab also known as @code{VOB} file
4c5f7207 651@item MPEG-2 TS @tab @tab X
9181577c
FB
652@tab also known as DVB Transport Stream
653@item ASF@tab X @tab X
654@item AVI@tab X @tab X
655@item WAV@tab X @tab X
656@item Macromedia Flash@tab X @tab X
4c5f7207 657@tab Only embedded audio is decoded.
2eb5d024
FB
658@item FLV @tab X @tab X
659@tab Macromedia Flash video files
9181577c 660@item Real Audio and Video @tab X @tab X
9181577c
FB
661@item Raw AC3 @tab X @tab X
662@item Raw MJPEG @tab X @tab X
663@item Raw MPEG video @tab X @tab X
664@item Raw PCM8/16 bits, mulaw/Alaw@tab X @tab X
42cad81a 665@item Raw CRI ADX audio @tab X @tab X
13dfd2b9 666@item Raw Shorten audio @tab @tab X
9181577c 667@item SUN AU format @tab X @tab X
7fe4c823 668@item NUT @tab X @tab X @tab NUT Open Container Format
019c8838 669@item QuickTime @tab X @tab X
4c5f7207
DB
670@item MPEG-4 @tab X @tab X
671@tab MPEG-4 is a variant of QuickTime.
0699d2fe 672@item Raw MPEG4 video @tab X @tab X
4fa1c4fa 673@item DV @tab X @tab X
701b603d 674@item 4xm @tab @tab X
4c5f7207 675@tab 4X Technologies format, used in some games.
4d3b1f8d
MM
676@item Playstation STR @tab @tab X
677@item Id RoQ @tab @tab X
4c5f7207 678@tab Used in Quake III, Jedi Knight 2, other computer games.
4d3b1f8d 679@item Interplay MVE @tab @tab X
4c5f7207 680@tab Format used in various Interplay computer games.
b17e9c99 681@item WC3 Movie @tab @tab X
4c5f7207 682@tab Multimedia format used in Origin's Wing Commander III computer game.
2fdf638b 683@item Sega FILM/CPK @tab @tab X
4c5f7207 684@tab Used in many Sega Saturn console games.
2fdf638b 685@item Westwood Studios VQA/AUD @tab @tab X
4c5f7207 686@tab Multimedia formats used in Westwood Studios games.
4120a53a 687@item Id Cinematic (.cin) @tab @tab X
4c5f7207 688@tab Used in Quake II.
42cad81a
MM
689@item FLIC format @tab @tab X
690@tab .fli/.flc files
da00f30e 691@item Sierra VMD @tab @tab X
4c5f7207 692@tab Used in Sierra CD-ROM games.
d08d7142 693@item Sierra Online @tab @tab X
4c5f7207 694@tab .sol files used in Sierra Online games.
38088adf 695@item Matroska @tab @tab X
ad81a9fe 696@item Electronic Arts Multimedia @tab @tab X
4c5f7207 697@tab Used in various EA games; files have extensions like WVE and UV2.
353147ed
MM
698@item Nullsoft Video (NSV) format @tab @tab X
699@end multitable
9181577c 700
4c5f7207 701@code{X} means that encoding (resp. decoding) is supported.
9181577c 702
0699d2fe
FB
703@section Image Formats
704
705FFmpeg can read and write images for each frame of a video sequence. The
706following image formats are supported:
707
3fc9e78f 708@multitable @columnfractions .4 .1 .1 .4
0699d2fe
FB
709@item Supported Image Format @tab Encoding @tab Decoding @tab Comments
710@item PGM, PPM @tab X @tab X
4c5f7207 711@item PAM @tab X @tab X @tab PAM is a PNM extension with alpha support.
2eb5d024 712@item PGMYUV @tab X @tab X @tab PGM with U and V components in YUV 4:2:0
4c5f7207
DB
713@item JPEG @tab X @tab X @tab Progressive JPEG is not supported.
714@item .Y.U.V @tab X @tab X @tab one raw file per component
715@item animated GIF @tab X @tab X @tab Only uncompressed GIFs are generated.
716@item PNG @tab X @tab X @tab 2 bit and 4 bit/pixel not supported yet.
6a91ec51 717@item SGI @tab X @tab X @tab SGI RGB image format
0699d2fe
FB
718@end multitable
719
4c5f7207 720@code{X} means that encoding (resp. decoding) is supported.
0699d2fe 721
9181577c
FB
722@section Video Codecs
723
3fc9e78f 724@multitable @columnfractions .4 .1 .1 .4
9181577c 725@item Supported Codec @tab Encoding @tab Decoding @tab Comments
4c5f7207
DB
726@item MPEG-1 video @tab X @tab X
727@item MPEG-2 video @tab X @tab X
728@item MPEG-4 @tab X @tab X @tab also known as DivX4/5
9181577c
FB
729@item MSMPEG4 V1 @tab X @tab X
730@item MSMPEG4 V2 @tab X @tab X
4c5f7207 731@item MSMPEG4 V3 @tab X @tab X @tab also known as DivX3
9181577c 732@item WMV7 @tab X @tab X
4c5f7207 733@item WMV8 @tab X @tab X @tab not completely working
b06b45c4 734@item H.261 @tab X @tab X
4c5f7207 735@item H.263(+) @tab X @tab X @tab also known as RealVideo 1.0
b06b45c4 736@item H.264 @tab @tab X
9181577c 737@item MJPEG @tab X @tab X
4c5f7207 738@item lossless MJPEG @tab X @tab X
d6896c49
AB
739@item Apple MJPEG-B @tab @tab X
740@item Sunplus MJPEG @tab @tab X @tab fourcc: SP5X
4fa1c4fa 741@item DV @tab X @tab X
4c5f7207
DB
742@item HuffYUV @tab X @tab X
743@item FFmpeg Video 1 @tab X @tab X @tab experimental lossless codec (fourcc: FFV1)
744@item FFmpeg Snow @tab X @tab X @tab experimental wavelet codec (fourcc: SNOW)
701b603d 745@item Asus v1 @tab X @tab X @tab fourcc: ASV1
4d3b1f8d 746@item Asus v2 @tab X @tab X @tab fourcc: ASV2
701b603d 747@item Creative YUV @tab @tab X @tab fourcc: CYUV
d2bfadc0 748@item Sorenson Video 1 @tab X @tab X @tab fourcc: SVQ1
701b603d
MM
749@item Sorenson Video 3 @tab @tab X @tab fourcc: SVQ3
750@item On2 VP3 @tab @tab X @tab still experimental
d6896c49 751@item Theora @tab @tab X @tab still experimental
cdb2c1ca 752@item Intel Indeo 3 @tab @tab X
9303d49c 753@item FLV @tab X @tab X @tab Sorenson H.263 used in Flash
4d3b1f8d 754@item ATI VCR1 @tab @tab X @tab fourcc: VCR1
7fe4c823 755@item ATI VCR2 @tab @tab X @tab fourcc: VCR2
4d3b1f8d 756@item Cirrus Logic AccuPak @tab @tab X @tab fourcc: CLJR
4c5f7207 757@item 4X Video @tab @tab X @tab Used in certain computer games.
4d3b1f8d 758@item Sony Playstation MDEC @tab @tab X
4c5f7207
DB
759@item Id RoQ @tab @tab X @tab Used in Quake III, Jedi Knight 2, other computer games.
760@item Xan/WC3 @tab @tab X @tab Used in Wing Commander III .MVE files.
761@item Interplay Video @tab @tab X @tab Used in Interplay .MVE files.
070ed1bc 762@item Apple Animation @tab @tab X @tab fourcc: 'rle '
42cad81a 763@item Apple Graphics @tab @tab X @tab fourcc: 'smc '
2fdf638b 764@item Apple Video @tab @tab X @tab fourcc: rpza
d08d7142 765@item Apple QuickDraw @tab @tab X @tab fourcc: qdrw
2fdf638b
MM
766@item Cinepak @tab @tab X
767@item Microsoft RLE @tab @tab X
768@item Microsoft Video-1 @tab @tab X
4120a53a 769@item Westwood VQA @tab @tab X
4c5f7207 770@item Id Cinematic Video @tab @tab X @tab Used in Quake II.
1dc1ed99 771@item Planar RGB @tab @tab X @tab fourcc: 8BPS
42cad81a 772@item FLIC video @tab @tab X
9a4117d5 773@item Duck TrueMotion v1 @tab @tab X @tab fourcc: DUCK
dce76c20 774@item Duck TrueMotion v2 @tab @tab X @tab fourcc: TM20
4c5f7207 775@item VMD Video @tab @tab X @tab Used in Sierra VMD files.
a273bbfb
RT
776@item MSZH @tab @tab X @tab Part of LCL
777@item ZLIB @tab X @tab X @tab Part of LCL, encoder experimental
9d53d58e 778@item TechSmith Camtasia @tab @tab X @tab fourcc: TSCC
d0a0bbd2 779@item IBM Ultimotion @tab @tab X @tab fourcc: ULTI
ab711b3c 780@item Miro VideoXL @tab @tab X @tab fourcc: VIXL
acfd8f0f 781@item QPEG @tab @tab X @tab fourccs: QPEG, Q1.0, Q1.1
9c7fb608 782@item LOCO @tab @tab X @tab
a8a15e9d 783@item Winnov WNV1 @tab @tab X @tab
589f8220 784@item Autodesk Animator Studio Codec @tab @tab X @tab fourcc: AASC
b81f8949 785@item Fraps FPS1 @tab @tab X @tab
9181577c
FB
786@end multitable
787
4c5f7207 788@code{X} means that encoding (resp. decoding) is supported.
9181577c 789
4c5f7207
DB
790See @url{http://www.mplayerhq.hu/~michael/codec-features.html} to
791get a precise comparison of the FFmpeg MPEG-4 codec compared to
792other implementations.
60837265 793
9181577c
FB
794@section Audio Codecs
795
796@multitable @columnfractions .4 .1 .1 .1 .7
797@item Supported Codec @tab Encoding @tab Decoding @tab Comments
798@item MPEG audio layer 2 @tab IX @tab IX
799@item MPEG audio layer 1/3 @tab IX @tab IX
4c5f7207 800@tab MP3 encoding is supported through the external library LAME.
37776c3b 801@item AC3 @tab IX @tab IX
4c5f7207 802@tab liba52 is used internally for decoding.
34d7008d 803@item Vorbis @tab X @tab X
4c5f7207 804@tab Supported through the external library libvorbis.
4745b5bf 805@item WMA V1/V2 @tab @tab X
f6fa7a6c 806@item AAC @tab X @tab X
4c5f7207 807@tab Supported through the external library libfaac/libfaad.
d4c3c5a6 808@item Microsoft ADPCM @tab X @tab X
4d3b1f8d
MM
809@item MS IMA ADPCM @tab X @tab X
810@item QT IMA ADPCM @tab @tab X
811@item 4X IMA ADPCM @tab @tab X
d4e437df 812@item G.726 ADPCM @tab X @tab X
b17e9c99 813@item Duck DK3 IMA ADPCM @tab @tab X
4c5f7207 814@tab Used in some Sega Saturn console games.
b17e9c99 815@item Duck DK4 IMA ADPCM @tab @tab X
4c5f7207 816@tab Used in some Sega Saturn console games.
2fdf638b 817@item Westwood Studios IMA ADPCM @tab @tab X
4c5f7207 818@tab Used in Westwood Studios games like Command and Conquer.
7d8379f2 819@item SMJPEG IMA ADPCM @tab @tab X
4c5f7207 820@tab Used in certain Loki game ports.
42cad81a
MM
821@item CD-ROM XA ADPCM @tab @tab X
822@item CRI ADX ADPCM @tab X @tab X
4c5f7207 823@tab Used in Sega Dreamcast games.
7d8379f2 824@item Electronic Arts ADPCM @tab @tab X
4c5f7207 825@tab Used in various EA titles.
b3bfb299 826@item Creative ADPCM @tab @tab X
2eb5d024
FB
827@item RA144 @tab @tab X
828@tab Real 14400 bit/s codec
829@item RA288 @tab @tab X
830@tab Real 28800 bit/s codec
37776c3b 831@item RADnet @tab X @tab IX
4c5f7207 832@tab Real low bitrate AC3 codec, liba52 is used for decoding.
2eb5d024 833@item AMR-NB @tab X @tab X
4c5f7207 834@tab Supported through an external library.
d663a1fd 835@item AMR-WB @tab X @tab X
4c5f7207 836@tab Supported through an external library.
2eb5d024 837@item DV audio @tab @tab X
4d3b1f8d 838@item Id RoQ DPCM @tab @tab X
4c5f7207 839@tab Used in Quake III, Jedi Knight 2, other computer games.
4d3b1f8d 840@item Interplay MVE DPCM @tab @tab X
4c5f7207 841@tab Used in various Interplay computer games.
b17e9c99 842@item Xan DPCM @tab @tab X
4c5f7207 843@tab Used in Origin's Wing Commander IV AVI files.
d08d7142 844@item Sierra Online DPCM @tab @tab X
4c5f7207 845@tab Used in Sierra Online game audio files.
d6896c49
AB
846@item Apple MACE 3 @tab @tab X
847@item Apple MACE 6 @tab @tab X
13dfd2b9
MM
848@item FLAC lossless audio @tab @tab X
849@item Shorten lossless audio @tab @tab X
f770ee03
MM
850@item Apple lossless audio @tab @tab X
851@tab QuickTime fourcc 'alac'
33a4d8b1 852@item FFmpeg Sonic @tab X @tab X
4c5f7207 853@tab experimental lossy/lossless codec
d9b1c197
RT
854@item Qdesign QDM2 @tab @tab X
855@tab there are still some distortions
440e7988
RT
856@item Real COOK @tab @tab X
857@tab All versions except 5.1 are supported
9181577c
FB
858@end multitable
859
4c5f7207 860@code{X} means that encoding (resp. decoding) is supported.
9181577c 861
4c5f7207
DB
862@code{I} means that an integer-only version is available, too (ensures high
863performance on systems without hardware floating point support).
9181577c 864
47d944d2
FB
865@chapter Platform Specific information
866
867@section Linux
868
4c5f7207
DB
869FFmpeg should be compiled with at least GCC 2.95.3. GCC 3.2 is the
870preferred compiler now for FFmpeg. All future optimizations will depend on
47d944d2
FB
871features only found in GCC 3.2.
872
873@section BSD
874
875@section Windows
876
b030b284
FB
877@subsection Native Windows compilation
878
879@itemize
880@item Install the current versions of MSYS and MinGW from
881@url{http://www.mingw.org/}. You can find detailed installation
882instructions in the download section and the FAQ.
883
4c5f7207 884@item If you want to test the FFplay, also download
50f52fcd
FB
885the MinGW development library of SDL 1.2.x
886(@file{SDL-devel-1.2.x-mingw32.tar.gz}) from
4c5f7207 887@url{http://www.libsdl.org}. Unpack it in a temporary directory, and
50f52fcd 888unpack the archive @file{i386-mingw32msvc.tar.gz} in the MinGW tool
988a9f9e
FB
889directory. Edit the @file{sdl-config} script so that it gives the
890correct SDL directory when invoked.
50f52fcd 891
e926e391 892@item Extract the current version of FFmpeg.
b030b284
FB
893
894@item Start the MSYS shell (file @file{msys.bat}).
895
4c5f7207
DB
896@item Change to the FFmpeg directory and follow
897 the instructions of how to compile FFmpeg (file
50f52fcd
FB
898@file{INSTALL}). Usually, launching @file{./configure} and @file{make}
899suffices. If you have problems using SDL, verify that
900@file{sdl-config} can be launched from the MSYS command line.
901
4c5f7207
DB
902@item You can install FFmpeg in @file{Program Files/FFmpeg} by typing
903@file{make install}. Don't forget to copy @file{SDL.dll} to the place
904you launch @file{ffplay} from.
50f52fcd 905
b030b284
FB
906@end itemize
907
988a9f9e
FB
908Notes:
909@itemize
50f52fcd 910
988a9f9e
FB
911@item The target @file{make wininstaller} can be used to create a
912Nullsoft based Windows installer for FFmpeg and FFplay. @file{SDL.dll}
4c5f7207 913must be copied to the FFmpeg directory in order to build the
988a9f9e
FB
914installer.
915
4c5f7207 916@item By using @code{./configure --enable-shared} when configuring FFmpeg,
988a9f9e
FB
917you can build @file{avcodec.dll} and @file{avformat.dll}. With
918@code{make install} you install the FFmpeg DLLs and the associated
919headers in @file{Program Files/FFmpeg}.
920
4c5f7207
DB
921@item Visual C++ compatibility: If you used @code{./configure --enable-shared}
922when configuring FFmpeg, FFmpeg tries to use the Microsoft Visual
988a9f9e 923C++ @code{lib} tool to build @code{avcodec.lib} and
4c5f7207 924@code{avformat.lib}. With these libraries you can link your Visual C++
be0efc0c 925code directly with the FFmpeg DLLs (see below).
988a9f9e
FB
926
927@end itemize
b030b284 928
be0efc0c
MB
929@subsection Visual C++ compatibility
930
931FFmpeg will not compile under Visual C++ -- and it has too many
932dependencies on the GCC compiler to make a port viable. However,
933if you want to use the FFmpeg libraries in your own applications,
934you can still compile those applications using Visual C++. An
935important restriction to this is that you have to use the
936dynamically linked versions of the FFmpeg libraries (i.e. the
937DLLs), and you have to make sure that Visual-C++-compatible
938import libraries are created during the FFmpeg build process.
939
940This description of how to use the FFmpeg libraries with Visual C++ is
941based on Visual C++ 2005 Express Edition Beta 2. If you have a different
942version, you might have to modify the procedures slightly.
943
944Here are the step-by-step instructions for building the FFmpeg libraries
945so they can be used with Visual C++:
946
947@enumerate
948
949@item Install Visual C++ (if you haven't done so already).
950
951@item Install MinGW and MSYS as described above.
952
953@item Add a call to @file{vcvars32.bat} (which sets up the environment
954variables for the Visual C++ tools) as the first line of
955@file{msys.bat}. The standard location for @file{vcvars32.bat} is
956@file{C:\Program Files\Microsoft Visual Studio 8\VC\bin\vcvars32.bat},
957and the standard location for @file{msys.bat} is
958@file{C:\msys\1.0\msys.bat}. If this corresponds to your setup, add the
959following line as the first line of @file{msys.bat}:
960
961@code{call "C:\Program Files\Microsoft Visual Studio 8\VC\bin\vcvars32.bat"}
962
963@item Start the MSYS shell (file @file{msys.bat}) and type @code{link.exe}.
964If you get a help message with the command line options of @code{link.exe},
965this means your environment variables are set up correctly, the
966Microsoft linker is on the path and will be used by FFmpeg to
967create Visual-C++-compatible import libraries.
968
e926e391 969@item Extract the current version of FFmpeg and change to the FFmpeg directory.
be0efc0c
MB
970
971@item Type the command
972@code{./configure --enable-shared --enable-memalign-hack} to configure and,
973if that didn't produce any errors, type @code{make} to build FFmpeg.
974
975@item The subdirectories @file{libavformat}, @file{libavcodec}, and
976@file{libavutil} should now contain the files @file{avformat.dll},
977@file{avformat.lib}, @file{avcodec.dll}, @file{avcodec.lib},
978@file{avutil.dll}, and @file{avutil.lib}, respectively. Copy the three
979DLLs to your System32 directory (typically @file{C:\Windows\System32}).
980
981@end enumerate
982
983And here is how to use these libraries with Visual C++:
984
985@enumerate
986
987@item Create a new console application ("File / New / Project") and then
988select "Win32 Console Application". On the appropriate page of the
989Application Wizard, uncheck the "Precompiled headers" option.
990
991@item Write the source code for your application, or, for testing, just
992copy the code from an existing sample application into the source file
993that Visual C++ has already created for you. (Note that your source
994filehas to have a @code{.cpp} extension; otherwise, Visual C++ won't
995compile the FFmpeg headers correctly because in C mode, it doesn't
996recognize the @code{inline} keyword.) For example, you can copy
997@file{output_example.c} from the FFmpeg distribution (but you will
998have to make minor modifications so the code will compile under
999C++, see below).
1000
1001@item Open the "Project / Properties" dialog box. In the "Configuration"
1002combo box, select "All Configurations" so that the changes you make will
1003affect both debug and release builds. In the tree view on the left hand
1004side, select "C/C++ / General", then edit the "Additional Include
1005Directories" setting to contain the complete paths to the
1006@file{libavformat}, @file{libavcodec}, and @file{libavutil}
1007subdirectories of your FFmpeg directory. Note that the directories have
1008to be separated using semicolons. Now select "Linker / General" from the
1009tree view and edit the "Additional Library Directories" setting to
1010contain the same three directories.
1011
1012@item Still in the "Project / Properties" dialog box, select "Linker / Input"
1013from the tree view, then add the files @file{avformat.lib},
1014@file{avcodec.lib}, and @file{avutil.lib} to the end of the "Additional
1015Dependencies". Note that the names of the libraries have to be separated
1016using spaces.
1017
765c3440 1018@item Now, select "C/C++ / Code Generation" from the tree view. Select
be0efc0c
MB
1019"Debug" in the "Configuration" combo box. Make sure that "Runtime
1020Library" is set to "Multi-threaded Debug DLL". Then, select "Release" in
1021the "Configuration" combo box and make sure that "Runtime Library" is
1022set to "Multi-threaded DLL".
1023
1024@item Click "OK" to close the "Project / Properties" dialog box and build
1025the application. Hopefully, it should compile and run cleanly. If you
1026used @file{output_example.c} as your sample application, you will get a
1027few compiler errors, but they are easy to fix. The first type of error
1028occurs because Visual C++ doesn't allow an @code{int} to be converted to
1029an @code{enum} without a cast. To solve the problem, insert the required
1030casts (this error occurs once for a @code{CodecID} and once for a
1031@code{CodecType}). The second type of error occurs because C++ requires
1032the return value of @code{malloc} to be cast to the exact type of the
1033pointer it is being assigned to. Visual C++ will complain that, for
1034example, @code{(void *)} is being assigned to @code{(uint8_t *)} without
1035an explicit cast. So insert an explicit cast in these places to silence
1036the compiler. The third type of error occurs because the @code{snprintf}
1037library function is called @code{_snprintf} under Visual C++. So just
1038add an underscore to fix the problem. With these changes,
1039@file{output_example.c} should compile under Visual C++, and the
1040resulting executable should produce valid video files.
1041
1042@end enumerate
1043
b030b284
FB
1044@subsection Cross compilation for Windows with Linux
1045
1046You must use the MinGW cross compilation tools available at
1047@url{http://www.mingw.org/}.
1048
4c5f7207 1049Then configure FFmpeg with the following options:
b030b284
FB
1050@example
1051./configure --enable-mingw32 --cross-prefix=i386-mingw32msvc-
1052@end example
019c8838 1053(you can change the cross-prefix according to the prefix chosen for the
b030b284
FB
1054MinGW tools).
1055
4c5f7207 1056Then you can easily test FFmpeg with Wine
b030b284
FB
1057(@url{http://www.winehq.com/}).
1058
019c8838 1059@section Mac OS X
47d944d2
FB
1060
1061@section BeOS
1062
1063The configure script should guess the configuration itself.
1064Networking support is currently not finished.
1065errno issues fixed by Andrew Bachmann.
1066
1067Old stuff:
1068
7dd611c9 1069François Revol - revol at free dot fr - April 2002
47d944d2
FB
1070
1071The configure script should guess the configuration itself,
4c5f7207 1072however I still didn't test building on the net_server version of BeOS.
47d944d2 1073
4c5f7207 1074FFserver is broken (needs poll() implementation).
47d944d2 1075
4c5f7207
DB
1076There are still issues with errno codes, which are negative in BeOS, and
1077that FFmpeg negates when returning. This ends up turning errors into
47d944d2
FB
1078valid results, then crashes.
1079(To be fixed)
1080
34d7008d 1081@chapter Developers Guide
9181577c
FB
1082
1083@section API
1084@itemize
1085@item libavcodec is the library containing the codecs (both encoding and
4c5f7207 1086decoding). Look at @file{libavcodec/apiexample.c} to see how to use it.
9181577c 1087
4c5f7207
DB
1088@item libavformat is the library containing the file format handling (mux and
1089demux code for several formats). Look at @file{ffplay.c} to use it in a
a93b9dba
FB
1090player. See @file{output_example.c} to use it to generate audio or video
1091streams.
1092
9181577c
FB
1093@end itemize
1094
1095@section Integrating libavcodec or libavformat in your program
1096
1097You can integrate all the source code of the libraries to link them
1098statically to avoid any version problem. All you need is to provide a
1099'config.mak' and a 'config.h' in the parent directory. See the defines
1100generated by ./configure to understand what is needed.
1101
1102You can use libavcodec or libavformat in your commercial program, but
1103@emph{any patch you make must be published}. The best way to proceed is
4c5f7207 1104to send your patches to the FFmpeg mailing list.
9181577c 1105
80d1c272 1106@node Coding Rules
9181577c
FB
1107@section Coding Rules
1108
4c5f7207 1109FFmpeg is programmed in the ISO C90 language with a few additional
cb6e87c1
FH
1110features from ISO C99, namely:
1111@itemize @bullet
1112@item
1113the @samp{inline} keyword;
1114@item
1115@samp{//} comments;
1116@item
1117designated struct initializers (@samp{struct s x = @{ .i = 17 @};})
1118@item
114afa9c 1119compound literals (@samp{x = (struct s) @{ 17, 23 @};})
cb6e87c1
FH
1120@end itemize
1121
1122These features are supported by all compilers we care about, so we won't
1123accept patches to remove their use unless they absolutely don't impair
1124clarity and performance.
1125
4c5f7207 1126All code must compile with GCC 2.95 and GCC 3.3. Currently, FFmpeg also
cb6e87c1
FH
1127compiles with several other compilers, such as the Compaq ccc compiler
1128or Sun Studio 9, and we would like to keep it that way unless it would
1129be exceedingly involved. To ensure compatibility, please don't use any
4c5f7207 1130additional C99 features or GCC extensions. Especially watch out for:
cb6e87c1
FH
1131@itemize @bullet
1132@item
1133mixing statements and declarations;
1134@item
1135@samp{long long} (use @samp{int64_t} instead);
1136@item
1137@samp{__attribute__} not protected by @samp{#ifdef __GNUC__} or similar;
1138@item
019c8838 1139GCC statement expressions (@samp{(x = (@{ int y = 4; y; @})}).
cb6e87c1 1140@end itemize
9181577c 1141
cb6e87c1 1142Indent size is 4. The TAB character should not be used.
9181577c
FB
1143The presentation is the one specified by 'indent -i4 -kr'.
1144
4c5f7207 1145Main priority in FFmpeg is simplicity and small code size (=less
9181577c
FB
1146bugs).
1147
4c5f7207
DB
1148Comments: Use the JavaDoc/Doxygen
1149format (see examples below) so that code documentation
1150can be generated automatically. All nontrivial functions should have a comment
1151above them explaining what the function does, even if it's just one sentence.
1152All structures and their member variables should be documented, too.
f4888b83
MN
1153@example
1154/**
2185824b
MR
1155 * @@file mpeg.c
1156 * MPEG codec.
1157 * @@author ...
1158 */
f4888b83
MN
1159
1160/**
2185824b
MR
1161 * Summary sentence.
1162 * more text ...
1163 * ...
1164 */
114afa9c 1165typedef struct Foobar@{
2185824b
MR
1166 int var1; /**< var1 description */
1167 int var2; ///< var2 description
1168 /** var3 description */
1169 int var3;
114afa9c 1170@} Foobar;
f4888b83
MN
1171
1172/**
2185824b
MR
1173 * Summary sentence.
1174 * more text ...
1175 * ...
1176 * @@param my_parameter description of my_parameter
1177 * @@return return value description
1178 */
f4888b83
MN
1179int myfunc(int my_parameter)
1180...
1181@end example
9181577c 1182
36d9b4e8
MN
1183fprintf and printf are forbidden in libavformat and libavcodec,
1184please use av_log() instead.
1185
7a57b90c
MN
1186@node CVS Policy
1187@section CVS Policy
1188
1189@enumerate
1190@item
1191 You must not commit code which breaks FFmpeg! (Meaning unfinished but
4c5f7207 1192 enabled code which breaks compilation or compiles but does not work or
7a57b90c
MN
1193 breaks the regression tests)
1194 You can commit unfinished stuff (for testing etc), but it must be disabled
1195 (#ifdef etc) by default so it does not interfere with other developers'
1196 work.
1197@item
1198 You don't have to over-test things. If it works for you, and you think it
4c5f7207
DB
1199 should work for others, then commit. If your code has problems
1200 (portability, triggers compiler bugs, unusual environment etc) they will be
7a57b90c
MN
1201 reported and eventually fixed.
1202@item
1203 Do not commit unrelated changes together, split them into self-contained
1204 pieces.
1205@item
1206 Do not change behavior of the program (renaming options etc) without
019c8838 1207 first discussing it on the ffmpeg-devel mailing list. Do not remove
7a57b90c
MN
1208 functionality from the code. Just improve!
1209
4c5f7207 1210 Note: Redundant code can be removed.
7a57b90c
MN
1211@item
1212 Do not commit changes to the build system (Makefiles, configure script)
019c8838 1213 which change behavior, defaults etc, without asking first. The same
7a57b90c
MN
1214 applies to compiler warning fixes, trivial looking fixes and to code
1215 maintained by other developers. We usually have a reason for doing things
019c8838 1216 the way we do. Send your changes as patches to the ffmpeg-devel mailing
7a57b90c
MN
1217 list, and if the code maintainers say OK, you may commit. This does not
1218 apply to files you wrote and/or maintain.
1219@item
1220 We refuse source indentation and other cosmetic changes if they are mixed
1221 with functional changes, such commits will be rejected and removed. Every
1222 developer has his own indentation style, you should not change it. Of course
1223 if you (re)write something, you can use your own style, even though we would
4c5f7207
DB
1224 prefer if the indentation throughout FFmpeg was consistent (Many projects
1225 force a given indentation style - we don't.). If you really need to make
7a57b90c
MN
1226 indentation changes (try to avoid this), separate them strictly from real
1227 changes.
1228
114afa9c 1229 NOTE: If you had to put if()@{ .. @} over a large (> 5 lines) chunk of code,
7a57b90c 1230 then either do NOT change the indentation of the inner part within (don't
019c8838 1231 move it to the right)! or do so in a separate commit
7a57b90c
MN
1232@item
1233 Always fill out the commit log message. Describe in a few lines what you
1234 changed and why. You can refer to mailing list postings if you fix a
1235 particular bug. Comments such as "fixed!" or "Changed it." are unacceptable.
1236@item
1237 If you apply a patch by someone else, include the name and email address in
1238 the CVS log message. Since the ffmpeg-cvslog mailing list is publicly
019c8838
DB
1239 archived you should add some SPAM protection to the email address. Send an
1240 answer to ffmpeg-devel (or wherever you got the patch from) saying that
7a57b90c
MN
1241 you applied the patch.
1242@item
1243 Do NOT commit to code actively maintained by others without permission. Send
019c8838 1244 a patch to ffmpeg-devel instead.
7a57b90c
MN
1245@item
1246 Subscribe to the ffmpeg-cvslog mailing list. The diffs of all CVS commits
1247 are sent there and reviewed by all the other developers. Bugs and possible
1248 improvements or general questions regarding commits are discussed there. We
1249 expect you to react if problems with your code are uncovered.
1250@item
1251 Update the documentation if you change behavior or add features. If you are
019c8838 1252 unsure how best to do this, send a patch to ffmpeg-devel, the documentation
7a57b90c
MN
1253 maintainer(s) will review and commit your stuff.
1254@item
1255 Revert a commit ONLY in case of a big blunder like committing something not
1256 intended to be committed or committing a wrong file, the wrong version of a
019c8838 1257 patch, CVS policy violation or broken code and you are going to recommit the
7a57b90c
MN
1258 right thing immediately.
1259
1260 Never revert changes made a long time ago or buggy code. Fix it in the
1261 normal way instead.
52f4aea0 1262@item
4c5f7207
DB
1263 Never write to unallocated memory, never write over the end of arrays,
1264 always check values read from some untrusted source before using them
1265 as array index or other risky things.
114afa9c 1266@end enumerate
7a57b90c
MN
1267
1268We think our rules are not too hard. If you have comments, contact us.
1269
1270Note, these rules are mostly borrowed from the MPlayer project.
1271
1272@subsection Renaming/moving files or content of files
019c8838 1273 You CANNOT do that. Post a request for such a change to the mailing list
7a57b90c
MN
1274 Do NOT remove & readd a file - it will kill the changelog!!!!
1275
9181577c
FB
1276@section Submitting patches
1277
80d1c272
MN
1278First, (@pxref{Coding Rules}) above if you didn't yet.
1279
6a6810b1 1280When you submit your patch, try to send a unified diff (diff '-up'
9181577c
FB
1281option). I cannot read other diffs :-)
1282
6e5c1877
MN
1283Also please do not submit patches which contain several unrelated changes.
1284Split them into individual self-contained patches; this makes reviewing
7885603a
MN
1285them much easier.
1286
9181577c 1287Run the regression tests before submitting a patch so that you can
6bf40f39 1288verify that there are no big problems.
9181577c 1289
9181577c 1290Patches should be posted as base64 encoded attachments (or any other
4c5f7207 1291encoding which ensures that the patch won't be trashed during
019c8838 1292transmission) to the ffmpeg-devel mailing list, see
67c311fa 1293@url{http://www1.mplayerhq.hu/mailman/listinfo/ffmpeg-devel}
9181577c 1294
54fcf429 1295It also helps quite a bit if you tell us what the patch does (for example
4c5f7207 1296'replaces lrint by lrintf'), and why (for example '*BSD isn't C99 compliant
54fcf429
MN
1297and has no lrint()')
1298
4c5f7207
DB
1299We reply to all submitted patches and either apply or reject with some
1300explanation why, but sometimes we are quite busy so it can take a week or two.
cf7f2b16 1301
9181577c
FB
1302@section Regression tests
1303
4c5f7207 1304Before submitting a patch (or committing to CVS), you should at least
9181577c
FB
1305test that you did not break anything.
1306
4c5f7207
DB
1307The regression tests build a synthetic video stream and a synthetic
1308audio stream. These are then encoded and decoded with all codecs or
9181577c 1309formats. The CRC (or MD5) of each generated file is recorded in a
4c5f7207 1310result file. A 'diff' is launched to compare the reference results and
9181577c
FB
1311the result file.
1312
4c5f7207 1313The regression tests then go on to test the FFserver code with a
6bf40f39
PG
1314limited set of streams. It is important that this step runs correctly
1315as well.
1316
36d9b4e8 1317Run 'make test' to test all the codecs and formats.
9181577c 1318
4c5f7207 1319Run 'make fulltest' to test all the codecs, formats and FFserver.
9181577c 1320
4c5f7207
DB
1321[Of course, some patches may change the results of the regression tests. In
1322this case, the reference results of the regression tests shall be modified
9181577c
FB
1323accordingly].
1324
1325@bye