General Documentation


Table of Contents


General Documentation

1. external libraries

FFmpeg can be hooked up with a number of external libraries to add support for more formats. None of them are used by default, their use has to be explicitly requested by passing the appropriate flags to `./configure'.

1.1 AMR

AMR comes in two different flavors, wideband and narrowband. FFmpeg can make use of the AMR wideband (floating-point mode) and the AMR narrowband (floating-point mode) reference decoders and encoders. Go to http://www.penguin.cz/~utx/amr and follow the instructions for installing the libraries. Then pass --enable-libamr-nb and/or --enable-libamr-wb to configure to enable the libraries. Note that libamr is copyrighted without any sort of license grant. This means that you can use it if you legally obtained it but you are not allowed to redistribute it in any way. Any FFmpeg binaries with libamr support you create are non-free and unredistributable!

2. Supported File Formats and Codecs

You can use the -formats option to have an exhaustive list.

2.1 File Formats

FFmpeg supports the following file formats through the libavformat library:
Name Encoding Decoding Comments
4xm X 4X Technologies format, used in some games.
ADTS AAC audio X X
American Laser Games MM X Multimedia format used in games like Mad Dog McCree.
ASF X X
AVI X X
AVM2 (Flash 9) X X Only embedded audio is decoded.
AVS X Multimedia format used by the Creature Shock game.
Bethsoft VID X Used in some games from Bethesda Softworks.
BFI X Brute Force & Ignorance, used in Flash Traffic: City of Angels.
C93 X Used in the game Cyberia from Interplay.
CIN X Multimedia format used by Delphine Software games.
Creative VOC X X Created for the Sound Blaster Pro.
CRYO APC X Audio format used in some games by CRYO Interactive Entertainment.
DV X X
DXA X This format is used in the non-Windows version of the Feeble Files game and different game cutscenes repacked for use with ScummVM.
Electronic Arts Multimedia X Used in various EA games; files have extensions like WVE and UV2.
FLIC X .fli/.flc files
FLV X X Macromedia Flash video files
GXF X X General eXchange Format SMPTE 360M, used by Thomson Grass Valley playout servers.
id Cinematic X Used in Quake II.
id RoQ X X Used in Quake III, Jedi Knight 2, other computer games.
IFF X Interchange File Format
Interplay MVE X Format used in various Interplay computer games.
LMLM4 X Used by Linux Media Labs MPEG-4 PCI boards
Matroska X X
MAXIS EA XA X Used in Sim City 3000; file extension .xa.
Monkey's Audio X
Motion Pixels MVI X
MOV/QuickTime X X
MPEG audio X X
MPEG-1 systems X X muxed audio and video
MPEG-2 PS X X also known as VOB file
MPEG-2 TS X also known as DVB Transport Stream
MPEG-4 X X MPEG-4 is a variant of QuickTime.
MSN TCP webcam X Used by MSN Messenger webcam streams.
MXF X X Material eXchange Format SMPTE 377M, used by D-Cinema, broadcast industry.
Nullsoft Video X
NUT X X NUT Open Container Format
OMA X Audio format used in Sony Sonic Stage and Sony Vegas.
PlayStation STR X
PVA X Used by TechnoTrend DVB PCI boards.
raw AC-3 X X
raw CRI ADX audio X X
raw MJPEG X X
raw MPEG video X X
raw MPEG-4 video X X
raw PCM 8/16/32 bits, 32/64-bit floating point, mu-law/A-law X X
raw Shorten audio X
RealMedia X X
RL2 X Audio and video format used in some games by Entertainment Software Partners.
Sega FILM/CPK X Used in many Sega Saturn console games.
SEQ X Tiertex .seq files used in the DOS CD-ROM version of the game Flashback.
Sierra Online X .sol files used in Sierra Online games.
Sierra VMD X Used in Sierra CD-ROM games.
SIFF X Audio and video format used in some games by Beam Software.
Smacker X Multimedia format used by many games.
SUN AU format X X
THP X Used on the Nintendo GameCube.
WAV X X
WC3 Movie X Multimedia format used in Origin's Wing Commander III computer game.
Westwood Studios VQA/AUD X Multimedia formats used in Westwood Studios games.

X means that encoding (resp. decoding) is supported.

2.2 Image Formats

FFmpeg can read and write images for each frame of a video sequence. The following image formats are supported:
Name Encoding Decoding Comments
.Y.U.V X X one raw file per component
animated GIF X X Only uncompressed GIFs are generated.
JPEG X X Progressive JPEG is not supported.
PAM X X PAM is a PNM extension with alpha support.
PCX X PC Paintbrush
PGM, PPM X X
PGMYUV X X PGM with U and V components in YUV 4:2:0
PNG X X 2/4 bpp not supported yet
PTX X V.Flash PTX format
RAS X Sun Rasterfile
SGI X X SGI RGB image format
Targa X Targa (.TGA) image format
TIFF X X YUV, JPEG and some extension is not supported yet.

X means that encoding (resp. decoding) is supported.

2.3 Video Codecs

Name Encoding Decoding Comments
4X Video X Used in certain computer games.
American Laser Games Video X Used in games like Mad Dog McCree.
AMV X Used in Chinese MP3 players.
Apple Animation X X fourcc: 'rle '
Apple Graphics X fourcc: 'smc '
Apple MJPEG-B X
Apple QuickDraw X fourcc: qdrw
Apple Video X fourcc: rpza
Asus v1 X X fourcc: ASV1
Asus v2 X X fourcc: ASV2
ATI VCR1 X fourcc: VCR1
ATI VCR2 X fourcc: VCR2
Autodesk RLE X fourcc: AASC
AVID DNxHD X X aka SMPTE VC3
AVS video X Video encoding used by the Creature Shock game.
Bethsoft VID X Used in some games from Bethesda Softworks.
C93 video X Codec used in Cyberia game.
CamStudio X fourcc: CSCD
Cin video X Codec used in Delphine Software games.
Cinepak X
Cirrus Logic AccuPak X fourcc: CLJR
Creative YUV X fourcc: CYUV
Dirac E E supported through external libdirac/libschroedinger libraries
Duck TrueMotion v1 X fourcc: DUCK
Duck TrueMotion v2 X fourcc: TM20
DV X X
DXA Video X Codec originally used in Feeble Files game.
Electronic Arts CMV X Used in NHL 95 game.
Electronic Arts TGV X
Electronic Arts TGQ X
FFmpeg Video 1 X X experimental lossless codec (fourcc: FFV1)
Flash Screen Video X X fourcc: FSV1
FLIC video X
FLV X X Sorenson H.263 used in Flash
Fraps FPS1 X
H.261 X X
H.263(+) X X also known as RealVideo 1.0
H.264 E X encoding supported through external library libx264
HuffYUV X X
IBM Ultimotion X fourcc: ULTI
id Cinematic video X Used in Quake II.
id RoQ X X Used in Quake III, Jedi Knight 2, other computer games.
Intel Indeo 3 X
Interplay Video X Used in Interplay .MVE files.
JPEG-LS X X fourcc: MJLS, lossless and near-lossless is supported.
KMVC X Codec used in Worms games.
LOCO X
lossless MJPEG X X
Microsoft RLE X
Microsoft Video-1 X
Mimic X Used in MSN Messenger Webcam streams.
Miro VideoXL X fourcc: VIXL
MJPEG X X
Motion Pixels Video X
MPEG-1 X X
MPEG-2 X X
MPEG-4 X X
MSMPEG4 V1 X X
MSMPEG4 V2 X X
MSMPEG4 V3 X X
MSZH X Part of LCL
On2 VP3 X still experimental
On2 VP5 X fourcc: VP50
On2 VP6 X fourcc: VP60,VP61,VP62
planar RGB X fourcc: 8BPS
QPEG X fourccs: QPEG, Q1.0, Q1.1
RealVideo 1.0 X X
RealVideo 2.0 X X
RealVideo 3.0 X still far from ideal
RealVideo 4.0 X
Renderware TXD X Texture dictionaries used by the Renderware Engine.
RTjpeg X Video encoding used in NuppelVideo files.
Smacker video X Video encoding used in Smacker.
Snow X X experimental wavelet codec (fourcc: SNOW)
Sony PlayStation MDEC X
Sorenson Video 1 X X fourcc: SVQ1
Sorenson Video 3 X fourcc: SVQ3
Sunplus MJPEG X fourcc: SP5X
TechSmith Camtasia X fourcc: TSCC
Theora E X encoding supported through external library libtheora
THP X Used on the Nintendo GameCube.
Tiertex Seq video X Codec used in DOS CD-ROM FlashBack game.
VC-1 X
VMD Video X Used in Sierra VMD files.
VMware Video X Codec used in videos captured by VMware.
Westwood VQA X
Winnov WNV1 X
WMV7 X X
WMV8 X X
WMV9 X not completely working
Xan/WC3 X Used in Wing Commander III .MVE files.
ZLIB X X part of LCL, encoder experimental
ZMBV X X Encoder works only in PAL8.

X means that encoding (resp. decoding) is supported. E means that support is provided through an external library.

2.4 Audio Codecs

Name Encoding Decoding Comments
4X IMA ADPCM X
8SVX audio X
AAC E X encoding supported through external library libfaac
AC-3 IX IX
AMR-NB E E supported through external library libamrnb
AMR-WB E E supported through external library libamrwb
AMV IMA ADPCM X Used in AMV files
Apple lossless audio X X QuickTime fourcc 'alac'
Apple MACE 3 X
Apple MACE 6 X
ATRAC 3 X
CD-ROM XA ADPCM X
Cin audio X Codec used in Delphine Software International games.
Creative ADPCM X 16 -> 4, 8 -> 4, 8 -> 3, 8 -> 2
CRI ADX ADPCM X X Used in Sega Dreamcast games.
DSP Group TrueSpeech X
DTS Coherent Audio X
Duck DK3 IMA ADPCM X Used in some Sega Saturn console games.
Duck DK4 IMA ADPCM X Used in some Sega Saturn console games.
DV audio X
Electronic Arts ADPCM X Used in various EA titles.
Enhanced AC-3 X
FLAC lossless audio IX X
G.726 ADPCM X X
GSM E E supported through external library libgsm
GSM_MS E E supported through external library libgsm
id RoQ DPCM X X Used in Quake III, Jedi Knight 2, other computer games.
Intel Music Coder X
Interplay MVE DPCM X Used in various Interplay computer games.
MAXIS EA ADPCM X Used in Sim City 3000.
Microsoft ADPCM X X
MLP/TrueHD X Used in DVD-Audio and Blu-Ray discs.
Monkey's Audio X Only versions 3.97-3.99 are supported.
MPEG audio layer 3 E IX encoding supported through external library LAME
MPEG audio layer 2 IX IX
MS IMA ADPCM X X
Musepack X SV7 and SV8 are supported.
Nellymoser ASAO X X
QCELP / PureVoice X
Qdesign QDM2 X There are still some distortions.
QT IMA ADPCM X X
RA144 X Real 14400 bit/s codec
RA288 X Real 28800 bit/s codec
RADnet IX IX Real low bitrate AC-3 codec
Real COOK X All versions except 5.1 are supported.
Shorten X
Sierra Online DPCM X Used in Sierra Online game audio files.
Smacker audio X
SMJPEG IMA ADPCM X Used in certain Loki game ports.
Sonic X X experimental codec
Sonic lossless X X experimental codec
Speex E supported through external library libspeex
THP ADPCM X Used on the Nintendo GameCube.
True Audio (TTA) X
Vorbis X X
WavPack X
Westwood Studios IMA ADPCM X Used in Westwood Studios games like Command and Conquer.
WMA v1/v2 X X
Xan DPCM X Used in Origin's Wing Commander IV AVI files.

X means that encoding (resp. decoding) is supported. E means that support is provided through an external library. I means that an integer-only version is available, too (ensures high performance on systems without hardware floating point support).

2.5 Subtitle Formats

Name Muxing Demuxing Encoding Decoding Rendering
ASS/SSA X X
DVB X X X X X
DVD X X X X X

X means that the feature is supported.

3. Platform Specific information

3.1 BSD

BSD make will not build FFmpeg, you need to install and use GNU Make (`gmake').

3.2 Windows

To get help and instructions for building FFmpeg under Windows, check out the FFmpeg Windows Help Forum at http://ffmpeg.arrozcru.org/.

3.2.1 Native Windows compilation

FFmpeg can be built to run natively on Windows using the MinGW tools. Install the latest versions of MSYS and MinGW from http://www.mingw.org/. You can find detailed installation instructions in the download section and the FAQ. FFmpeg does not build out-of-the-box with the packages the automated MinGW installer provides. It also requires coreutils to be installed and many other packages updated to the latest version. The minimum version for some packages are listed below:

You will also need to pass -fno-common to the compiler to work around a GCC bug (see http://gcc.gnu.org/bugzilla/show_bug.cgi?id=37216). Within the MSYS shell, configure and make with:

./configure --enable-memalign-hack --extra-cflags=-fno-common
make
make install

This will install `ffmpeg.exe' along with many other development files to `/usr/local'. You may specify another install path using the --prefix option in `configure'. Notes:

3.2.2 Microsoft Visual C++ compatibility

As stated in the FAQ, FFmpeg will not compile under MSVC++. However, if you want to use the libav* libraries in your own applications, you can still compile those applications using MSVC++. But the libav* libraries you link to must be built with MinGW. However, you will not be able to debug inside the libav* libraries, since MSVC++ does not recognize the debug symbols generated by GCC. We strongly recommend you to move over from MSVC++ to MinGW tools. This description of how to use the FFmpeg libraries with MSVC++ is based on Microsoft Visual C++ 2005 Express Edition. If you have a different version, you might have to modify the procedures slightly.

3.2.2.1 Using static libraries

Assuming you have just built and installed FFmpeg in `/usr/local'.

  1. Create a new console application ("File / New / Project") and then select "Win32 Console Application". On the appropriate page of the Application Wizard, uncheck the "Precompiled headers" option.
  2. Write the source code for your application, or, for testing, just copy the code from an existing sample application into the source file that MSVC++ has already created for you. For example, you can copy `output_example.c' from the FFmpeg distribution.
  3. Open the "Project / Properties" dialog box. In the "Configuration" combo box, select "All Configurations" so that the changes you make will affect both debug and release builds. In the tree view on the left hand side, select "C/C++ / General", then edit the "Additional Include Directories" setting to contain the path where the FFmpeg includes were installed (i.e. `c:\msys\1.0\local\include'). Do not add MinGW's include directory here, or the include files will conflict with MSVC's.
  4. Still in the "Project / Properties" dialog box, select "Linker / General" from the tree view and edit the "Additional Library Directories" setting to contain the `lib' directory where FFmpeg was installed (i.e. `c:\msys\1.0\local\lib'), the directory where MinGW libs are installed (i.e. `c:\mingw\lib'), and the directory where MinGW's GCC libs are installed (i.e. `C:\mingw\lib\gcc\mingw32\4.2.1-sjlj'). Then select "Linker / Input" from the tree view, and add the files `libavformat.a', `libavcodec.a', `libavutil.a', `libmingwex.a', `libgcc.a', and any other libraries you used (i.e. `libz.a') to the end of "Additional Dependencies".
  5. Now, select "C/C++ / Code Generation" from the tree view. Select "Debug" in the "Configuration" combo box. Make sure that "Runtime Library" is set to "Multi-threaded Debug DLL". Then, select "Release" in the "Configuration" combo box and make sure that "Runtime Library" is set to "Multi-threaded DLL".
  6. Click "OK" to close the "Project / Properties" dialog box.
  7. MSVC++ lacks some C99 header files that are fundamental for FFmpeg. Get msinttypes from http://code.google.com/p/msinttypes/downloads/list and install it in MSVC++'s include directory (i.e. `C:\Program Files\Microsoft Visual Studio 8\VC\include').
  8. MSVC++ also does not understand the inline keyword used by FFmpeg, so you must add this line before #includeing libav*:
    #define inline _inline
    
  9. Build your application, everything should work.

3.2.2.2 Using shared libraries

This is how to create DLL and LIB files that are compatible with MSVC++:

  1. Add a call to `vcvars32.bat' (which sets up the environment variables for the Visual C++ tools) as the first line of `msys.bat'. The standard location for `vcvars32.bat' is `C:\Program Files\Microsoft Visual Studio 8\VC\bin\vcvars32.bat', and the standard location for `msys.bat' is `C:\msys\1.0\msys.bat'. If this corresponds to your setup, add the following line as the first line of `msys.bat':
    call "C:\Program Files\Microsoft Visual Studio 8\VC\bin\vcvars32.bat"
    
    Alternatively, you may start the `Visual Studio 2005 Command Prompt', and run `c:\msys\1.0\msys.bat' from there.
  2. Within the MSYS shell, run lib.exe. If you get a help message from `Microsoft (R) Library Manager', this means your environment variables are set up correctly, the `Microsoft (R) Library Manager' is on the path and will be used by FFmpeg to create MSVC++-compatible import libraries.
  3. Build FFmpeg with
    ./configure --enable-shared --enable-memalign-hack
    make
    make install
    
    Your install path (`/usr/local/' by default) should now have the necessary DLL and LIB files under the `bin' directory.

To use those files with MSVC++, do the same as you would do with the static libraries, as described above. But in Step 4, you should only need to add the directory where the LIB files are installed (i.e. `c:\msys\usr\local\bin'). This is not a typo, the LIB files are installed in the `bin' directory. And instead of adding `libxx.a' files, you should add `avcodec.lib', `avformat.lib', and `avutil.lib'. There should be no need for `libmingwex.a', `libgcc.a', and `wsock32.lib', nor any other external library statically linked into the DLLs. The `bin' directory contains a bunch of DLL files, but the ones that are actually used to run your application are the ones with a major version number in their filenames (i.e. `avcodec-51.dll').

3.2.3 Cross compilation for Windows with Linux

You must use the MinGW cross compilation tools available at http://www.mingw.org/. Then configure FFmpeg with the following options:

./configure --target-os=mingw32 --cross-prefix=i386-mingw32msvc-

(you can change the cross-prefix according to the prefix chosen for the MinGW tools). Then you can easily test FFmpeg with Wine (http://www.winehq.com/).

3.2.4 Compilation under Cygwin

The main issue with the 1.5.x Cygwin versions is that newlib, its C library, does not contain llrint(). You need to upgrade to the unstable 1.7.x versions, or leverage the implementation in MinGW (as explained below). Just install your Cygwin with all the "Base" packages, plus the following "Devel" ones:

binutils, gcc-core, make, subversion, mingw-runtime, diffutils

The experimental gcc4 package is still buggy, hence please use the official gcc 3.4.4 or a 4.2.x compiled from source by yourself. Install the current binutils-20080624-2 as they work fine (the old binutils-20060709-1 proved buggy on shared builds). Then create a small library that just contains llrint():

ar x /usr/lib/mingw/libmingwex.a llrint.o
ar cq /usr/local/lib/libllrint.a llrint.o

Then run

./configure --enable-static --disable-shared  --extra-ldflags='-L /usr/local/lib' --extra-libs='-l llrint'

to make a static build or

./configure --enable-shared --disable-static  --extra-ldflags='-L /usr/local/lib' --extra-libs='-l llrint'

to build shared libraries. If you want to build FFmpeg with additional libraries, download Cygwin "Devel" packages for Ogg and Vorbis from any Cygwin packages repository:

libogg-devel, libvorbis-devel

These library packages are only available from Cygwin Ports (http://sourceware.org/cygwinports/) :

yasm, libSDL-devel, libdirac-devel, libfaac-devel, libfaad-devel, libgsm-devel,
libmp3lame-devel, libschroedinger1.0-devel, speex-devel, libtheora-devel,
libxvidcore-devel

The recommendation for libnut and x264 is to build them from source by yourself, as they evolve too quickly for Cygwin Ports to be up to date.

3.2.5 Crosscompilation for Windows under Cygwin

With Cygwin you can create Windows binaries that do not need the cygwin1.dll. Just install your Cygwin as explained before, plus these additional "Devel" packages:

gcc-mingw-core, mingw-runtime, mingw-zlib

and add some special flags to your configure invocation. For a static build run

./configure --target-os=mingw32 --enable-memalign-hack --enable-static --disable-shared --extra-cflags=-mno-cygwin --extra-libs=-mno-cygwin

and for a build with shared libraries

./configure --target-os=mingw32 --enable-memalign-hack --enable-shared --disable-static --extra-cflags=-mno-cygwin --extra-libs=-mno-cygwin

3.3 BeOS

BeOS support is broken in mysterious ways.

3.4 OS/2

For information about compiling FFmpeg on OS/2 see http://www.edm2.com/index.php/FFmpeg.

4. Developers Guide

4.1 API

4.2 Integrating libavcodec or libavformat in your program

You can integrate all the source code of the libraries to link them statically to avoid any version problem. All you need is to provide a 'config.mak' and a 'config.h' in the parent directory. See the defines generated by ./configure to understand what is needed. You can use libavcodec or libavformat in your commercial program, but any patch you make must be published. The best way to proceed is to send your patches to the FFmpeg mailing list.

4.3 Coding Rules

FFmpeg is programmed in the ISO C90 language with a few additional features from ISO C99, namely:

These features are supported by all compilers we care about, so we will not accept patches to remove their use unless they absolutely do not impair clarity and performance. All code must compile with GCC 2.95 and GCC 3.3. Currently, FFmpeg also compiles with several other compilers, such as the Compaq ccc compiler or Sun Studio 9, and we would like to keep it that way unless it would be exceedingly involved. To ensure compatibility, please do not use any additional C99 features or GCC extensions. Especially watch out for:

Indent size is 4. The presentation is the one specified by 'indent -i4 -kr -nut'. The TAB character is forbidden outside of Makefiles as is any form of trailing whitespace. Commits containing either will be rejected by the Subversion repository. The main priority in FFmpeg is simplicity and small code size in order to minimize the bug count. Comments: Use the JavaDoc/Doxygen format (see examples below) so that code documentation can be generated automatically. All nontrivial functions should have a comment above them explaining what the function does, even if it is just one sentence. All structures and their member variables should be documented, too.

/**
 * @file mpeg.c
 * MPEG codec.
 * @author ...
 */

/**
 * Summary sentence.
 * more text ...
 * ...
 */
typedef struct Foobar{
    int var1; /**< var1 description */
    int var2; ///< var2 description
    /** var3 description */
    int var3;
} Foobar;

/**
 * Summary sentence.
 * more text ...
 * ...
 * @param my_parameter description of my_parameter
 * @return return value description
 */
int myfunc(int my_parameter)
...

fprintf and printf are forbidden in libavformat and libavcodec, please use av_log() instead. Casts should be used only when necessary. Unneeded parentheses should also be avoided if they don't make the code easier to understand.

4.4 Development Policy

  1. Contributions should be licensed under the LGPL 2.1, including an "or any later version" clause, or the MIT license. GPL 2 including an "or any later version" clause is also acceptable, but LGPL is preferred.
  2. You must not commit code which breaks FFmpeg! (Meaning unfinished but enabled code which breaks compilation or compiles but does not work or breaks the regression tests) You can commit unfinished stuff (for testing etc), but it must be disabled (#ifdef etc) by default so it does not interfere with other developers' work.
  3. You do not have to over-test things. If it works for you, and you think it should work for others, then commit. If your code has problems (portability, triggers compiler bugs, unusual environment etc) they will be reported and eventually fixed.
  4. Do not commit unrelated changes together, split them into self-contained pieces. Also do not forget that if part B depends on part A, but A does not depend on B, then A can and should be committed first and separate from B. Keeping changes well split into self-contained parts makes reviewing and understanding them on the commit log mailing list easier. This also helps in case of debugging later on. Also if you have doubts about splitting or not splitting, do not hesitate to ask/discuss it on the developer mailing list.
  5. Do not change behavior of the program (renaming options etc) without first discussing it on the ffmpeg-devel mailing list. Do not remove functionality from the code. Just improve! Note: Redundant code can be removed.
  6. Do not commit changes to the build system (Makefiles, configure script) which change behavior, defaults etc, without asking first. The same applies to compiler warning fixes, trivial looking fixes and to code maintained by other developers. We usually have a reason for doing things the way we do. Send your changes as patches to the ffmpeg-devel mailing list, and if the code maintainers say OK, you may commit. This does not apply to files you wrote and/or maintain.
  7. We refuse source indentation and other cosmetic changes if they are mixed with functional changes, such commits will be rejected and removed. Every developer has his own indentation style, you should not change it. Of course if you (re)write something, you can use your own style, even though we would prefer if the indentation throughout FFmpeg was consistent (Many projects force a given indentation style - we do not.). If you really need to make indentation changes (try to avoid this), separate them strictly from real changes. NOTE: If you had to put if(){ .. } over a large (> 5 lines) chunk of code, then either do NOT change the indentation of the inner part within (do not move it to the right)! or do so in a separate commit
  8. Always fill out the commit log message. Describe in a few lines what you changed and why. You can refer to mailing list postings if you fix a particular bug. Comments such as "fixed!" or "Changed it." are unacceptable.
  9. If you apply a patch by someone else, include the name and email address in the log message. Since the ffmpeg-cvslog mailing list is publicly archived you should add some SPAM protection to the email address. Send an answer to ffmpeg-devel (or wherever you got the patch from) saying that you applied the patch.
  10. When applying patches that have been discussed (at length) on the mailing list, reference the thread in the log message.
  11. Do NOT commit to code actively maintained by others without permission. Send a patch to ffmpeg-devel instead. If no one answers within a reasonable timeframe (12h for build failures and security fixes, 3 days small changes, 1 week for big patches) then commit your patch if you think it is OK. Also note, the maintainer can simply ask for more time to review!
  12. Subscribe to the ffmpeg-cvslog mailing list. The diffs of all commits are sent there and reviewed by all the other developers. Bugs and possible improvements or general questions regarding commits are discussed there. We expect you to react if problems with your code are uncovered.
  13. Update the documentation if you change behavior or add features. If you are unsure how best to do this, send a patch to ffmpeg-devel, the documentation maintainer(s) will review and commit your stuff.
  14. Try to keep important discussions and requests (also) on the public developer mailing list, so that all developers can benefit from them.
  15. Never write to unallocated memory, never write over the end of arrays, always check values read from some untrusted source before using them as array index or other risky things.
  16. Remember to check if you need to bump versions for the specific libav parts (libavutil, libavcodec, libavformat) you are changing. You need to change the version integer. Incrementing the first component means no backward compatibility to previous versions (e.g. removal of a function from the public API). Incrementing the second component means backward compatible change (e.g. addition of a function to the public API or extension of an existing data structure). Incrementing the third component means a noteworthy binary compatible change (e.g. encoder bug fix that matters for the decoder).
  17. Compiler warnings indicate potential bugs or code with bad style. If a type of warning always points to correct and clean code, that warning should be disabled, not the code changed. Thus the remaining warnings can either be bugs or correct code. If it is a bug, the bug has to be fixed. If it is not, the code should be changed to not generate a warning unless that causes a slowdown or obfuscates the code.
  18. If you add a new file, give it a proper license header. Do not copy and paste it from a random place, use an existing file as template.

We think our rules are not too hard. If you have comments, contact us. Note, these rules are mostly borrowed from the MPlayer project.

4.5 Submitting patches

First, (see section 4.3 Coding Rules) above if you did not yet. When you submit your patch, try to send a unified diff (diff '-up' option). We cannot read other diffs :-) Also please do not submit a patch which contains several unrelated changes. Split it into separate, self-contained pieces. This does not mean splitting file by file. Instead, make the patch as small as possible while still keeping it as a logical unit that contains an individual change, even if it spans multiple files. This makes reviewing your patches much easier for us and greatly increases your chances of getting your patch applied. Run the regression tests before submitting a patch so that you can verify that there are no big problems. Patches should be posted as base64 encoded attachments (or any other encoding which ensures that the patch will not be trashed during transmission) to the ffmpeg-devel mailing list, see http://lists.mplayerhq.hu/mailman/listinfo/ffmpeg-devel It also helps quite a bit if you tell us what the patch does (for example 'replaces lrint by lrintf'), and why (for example '*BSD isn't C99 compliant and has no lrint()') Also please if you send several patches, send each patch as a separate mail, do not attach several unrelated patches to the same mail.

4.6 New codecs or formats checklist

  1. Did you use av_cold for codec initialization and close functions?
  2. Did you add a long_name under NULL_IF_CONFIG_SMALL to the AVCodec or AVInputFormat/AVOutputFormat struct?
  3. Did you bump the minor version number in `avcodec.h' or `avformat.h'?
  4. Did you register it in `allcodecs.c' or `allformats.c'?
  5. Did you add the CodecID to `avcodec.h'?
  6. If it has a fourcc, did you add it to `libavformat/riff.c', even if it is only a decoder?
  7. Did you add a rule to compile the appropriate files in the Makefile? Remember to do this even if you're just adding a format to a file that is already being compiled by some other rule, like a raw demuxer.
  8. Did you add an entry to the table of supported formats or codecs in the documentation?
  9. Did you add an entry in the Changelog?
  10. If it depends on a parser or a library, did you add that dependency in configure?
  11. Did you "svn add" the appropriate files before commiting?

4.7 patch submission checklist

  1. Do the regression tests pass with the patch applied?
  2. Does make checkheaders pass with the patch applied?
  3. Is the patch a unified diff?
  4. Is the patch against latest FFmpeg SVN?
  5. Are you subscribed to ffmpeg-dev? (the list is subscribers only due to spam)
  6. Have you checked that the changes are minimal, so that the same cannot be achieved with a smaller patch and/or simpler final code?
  7. If the change is to speed critical code, did you benchmark it?
  8. If you did any benchmarks, did you provide them in the mail?
  9. Have you checked that the patch does not introduce buffer overflows or other security issues?
  10. Did you test your decoder or demuxer against damaged data? If no, see tools/trasher and the noise bitstream filter. Your decoder or demuxer should not crash or end in a (near) infinite loop when fed damaged data.
  11. Is the patch created from the root of the source tree, so it can be applied with patch -p0?
  12. Does the patch not mix functional and cosmetic changes?
  13. Did you add tabs or trailing whitespace to the code? Both are forbidden.
  14. Is the patch attached to the email you send?
  15. Is the mime type of the patch correct? It should be text/x-diff or text/x-patch or at least text/plain and not application/octet-stream.
  16. If the patch fixes a bug, did you provide a verbose analysis of the bug?
  17. If the patch fixes a bug, did you provide enough information, including a sample, so the bug can be reproduced and the fix can be verified? Note please do not attach samples >100k to mails but rather provide a URL, you can upload to ftp://upload.ffmpeg.org
  18. Did you provide a verbose summary about what the patch does change?
  19. Did you provide a verbose explanation why it changes things like it does?
  20. Did you provide a verbose summary of the user visible advantages and disadvantages if the patch is applied?
  21. Did you provide an example so we can verify the new feature added by the patch easily?
  22. If you added a new file, did you insert a license header? It should be taken from FFmpeg, not randomly copied and pasted from somewhere else.
  23. You should maintain alphabetical order in alphabetically ordered lists as long as doing so does not break API/ABI compatibility.
  24. Lines with similar content should be aligned vertically when doing so improves readability.
  25. Did you provide a suggestion for a clear commit log message?

4.8 Patch review process

All patches posted to ffmpeg-devel will be reviewed, unless they contain a clear note that the patch is not for SVN. Reviews and comments will be posted as replies to the patch on the mailing list. The patch submitter then has to take care of every comment, that can be by resubmitting a changed patch or by discussion. Resubmitted patches will themselves be reviewed like any other patch. If at some point a patch passes review with no comments then it is approved, that can for simple and small patches happen immediately while large patches will generally have to be changed and reviewed many times before they are approved. After a patch is approved it will be committed to the repository. We will review all submitted patches, but sometimes we are quite busy so especially for large patches this can take several weeks. When resubmitting patches, please do not make any significant changes not related to the comments received during review. Such patches will be rejected. Instead, submit significant changes or new features as separate patches.

4.9 Regression tests

Before submitting a patch (or committing to the repository), you should at least test that you did not break anything. The regression tests build a synthetic video stream and a synthetic audio stream. These are then encoded and decoded with all codecs or formats. The CRC (or MD5) of each generated file is recorded in a result file. A 'diff' is launched to compare the reference results and the result file. The regression tests then go on to test the FFserver code with a limited set of streams. It is important that this step runs correctly as well. Run 'make test' to test all the codecs and formats. Run 'make fulltest' to test all the codecs, formats and FFserver. [Of course, some patches may change the results of the regression tests. In this case, the reference results of the regression tests shall be modified accordingly].


This document was generated on 13 January 2009 using texi2html 1.56k.