1 \input texinfo @c -*- texinfo -*-
3 @settitle Developer Documentation
5 @center @titlefont{Developer Documentation}
12 @chapter Developers Guide
16 @item libavcodec is the library containing the codecs (both encoding and
17 decoding). Look at @file{libavcodec/apiexample.c} to see how to use it.
19 @item libavformat is the library containing the file format handling (mux and
20 demux code for several formats). Look at @file{ffplay.c} to use it in a
21 player. See @file{libavformat/output-example.c} to use it to generate
22 audio or video streams.
26 @section Integrating libavcodec or libavformat in your program
28 You can integrate all the source code of the libraries to link them
29 statically to avoid any version problem. All you need is to provide a
30 'config.mak' and a 'config.h' in the parent directory. See the defines
31 generated by ./configure to understand what is needed.
33 You can use libavcodec or libavformat in your commercial program, but
34 @emph{any patch you make must be published}. The best way to proceed is
35 to send your patches to the FFmpeg mailing list.
40 FFmpeg is programmed in the ISO C90 language with a few additional
41 features from ISO C99, namely:
44 the @samp{inline} keyword;
48 designated struct initializers (@samp{struct s x = @{ .i = 17 @};})
50 compound literals (@samp{x = (struct s) @{ 17, 23 @};})
53 These features are supported by all compilers we care about, so we will not
54 accept patches to remove their use unless they absolutely do not impair
55 clarity and performance.
57 All code must compile with GCC 3.3. Currently, FFmpeg also
58 compiles with several other compilers, such as the Compaq ccc compiler
59 or Sun Studio 9, and we would like to keep it that way unless it would
60 be exceedingly involved. To ensure compatibility, please do not use any
61 additional C99 features or GCC extensions. Especially watch out for:
64 mixing statements and declarations;
66 @samp{long long} (use @samp{int64_t} instead);
68 @samp{__attribute__} not protected by @samp{#ifdef __GNUC__} or similar;
70 GCC statement expressions (@samp{(x = (@{ int y = 4; y; @})}).
74 The presentation is one inspired by 'indent -i4 -kr -nut'.
75 The TAB character is forbidden outside of Makefiles as is any
76 form of trailing whitespace. Commits containing either will be
77 rejected by the git repository.
79 The main priority in FFmpeg is simplicity and small code size in order to
80 minimize the bug count.
82 Comments: Use the JavaDoc/Doxygen
83 format (see examples below) so that code documentation
84 can be generated automatically. All nontrivial functions should have a comment
85 above them explaining what the function does, even if it is just one sentence.
86 All structures and their member variables should be documented, too.
99 typedef struct Foobar@{
100 int var1; /**< var1 description */
101 int var2; ///< var2 description
102 /** var3 description */
110 * @@param my_parameter description of my_parameter
111 * @@return return value description
113 int myfunc(int my_parameter)
117 fprintf and printf are forbidden in libavformat and libavcodec,
118 please use av_log() instead.
120 Casts should be used only when necessary. Unneeded parentheses
121 should also be avoided if they don't make the code easier to understand.
123 @section Development Policy
127 Contributions should be licensed under the LGPL 2.1, including an
128 "or any later version" clause, or the MIT license. GPL 2 including
129 an "or any later version" clause is also acceptable, but LGPL is
132 You must not commit code which breaks FFmpeg! (Meaning unfinished but
133 enabled code which breaks compilation or compiles but does not work or
134 breaks the regression tests)
135 You can commit unfinished stuff (for testing etc), but it must be disabled
136 (#ifdef etc) by default so it does not interfere with other developers'
139 You do not have to over-test things. If it works for you, and you think it
140 should work for others, then commit. If your code has problems
141 (portability, triggers compiler bugs, unusual environment etc) they will be
142 reported and eventually fixed.
144 Do not commit unrelated changes together, split them into self-contained
145 pieces. Also do not forget that if part B depends on part A, but A does not
146 depend on B, then A can and should be committed first and separate from B.
147 Keeping changes well split into self-contained parts makes reviewing and
148 understanding them on the commit log mailing list easier. This also helps
149 in case of debugging later on.
150 Also if you have doubts about splitting or not splitting, do not hesitate to
151 ask/discuss it on the developer mailing list.
153 Do not change behavior of the programs (renaming options etc) or public
154 API or ABI without first discussing it on the ffmpeg-devel mailing list.
155 Do not remove functionality from the code. Just improve!
157 Note: Redundant code can be removed.
159 Do not commit changes to the build system (Makefiles, configure script)
160 which change behavior, defaults etc, without asking first. The same
161 applies to compiler warning fixes, trivial looking fixes and to code
162 maintained by other developers. We usually have a reason for doing things
163 the way we do. Send your changes as patches to the ffmpeg-devel mailing
164 list, and if the code maintainers say OK, you may commit. This does not
165 apply to files you wrote and/or maintain.
167 We refuse source indentation and other cosmetic changes if they are mixed
168 with functional changes, such commits will be rejected and removed. Every
169 developer has his own indentation style, you should not change it. Of course
170 if you (re)write something, you can use your own style, even though we would
171 prefer if the indentation throughout FFmpeg was consistent (Many projects
172 force a given indentation style - we do not.). If you really need to make
173 indentation changes (try to avoid this), separate them strictly from real
176 NOTE: If you had to put if()@{ .. @} over a large (> 5 lines) chunk of code,
177 then either do NOT change the indentation of the inner part within (do not
178 move it to the right)! or do so in a separate commit
180 Always fill out the commit log message. Describe in a few lines what you
181 changed and why. You can refer to mailing list postings if you fix a
182 particular bug. Comments such as "fixed!" or "Changed it." are unacceptable.
184 area changed: Short 1 line description
186 details describing what and why and giving references.
188 Make sure the author of the commit is set correctly. (see git commit --author)
189 If you apply a patch, send an
190 answer to ffmpeg-devel (or wherever you got the patch from) saying that
191 you applied the patch.
193 When applying patches that have been discussed (at length) on the mailing
194 list, reference the thread in the log message.
196 Do NOT commit to code actively maintained by others without permission.
197 Send a patch to ffmpeg-devel instead. If no one answers within a reasonable
198 timeframe (12h for build failures and security fixes, 3 days small changes,
199 1 week for big patches) then commit your patch if you think it is OK.
200 Also note, the maintainer can simply ask for more time to review!
202 Subscribe to the ffmpeg-cvslog mailing list. The diffs of all commits
203 are sent there and reviewed by all the other developers. Bugs and possible
204 improvements or general questions regarding commits are discussed there. We
205 expect you to react if problems with your code are uncovered.
207 Update the documentation if you change behavior or add features. If you are
208 unsure how best to do this, send a patch to ffmpeg-devel, the documentation
209 maintainer(s) will review and commit your stuff.
211 Try to keep important discussions and requests (also) on the public
212 developer mailing list, so that all developers can benefit from them.
214 Never write to unallocated memory, never write over the end of arrays,
215 always check values read from some untrusted source before using them
216 as array index or other risky things.
218 Remember to check if you need to bump versions for the specific libav
219 parts (libavutil, libavcodec, libavformat) you are changing. You need
220 to change the version integer.
221 Incrementing the first component means no backward compatibility to
222 previous versions (e.g. removal of a function from the public API).
223 Incrementing the second component means backward compatible change
224 (e.g. addition of a function to the public API or extension of an
225 existing data structure).
226 Incrementing the third component means a noteworthy binary compatible
227 change (e.g. encoder bug fix that matters for the decoder).
229 Compiler warnings indicate potential bugs or code with bad style. If a type of
230 warning always points to correct and clean code, that warning should
231 be disabled, not the code changed.
232 Thus the remaining warnings can either be bugs or correct code.
233 If it is a bug, the bug has to be fixed. If it is not, the code should
234 be changed to not generate a warning unless that causes a slowdown
235 or obfuscates the code.
237 If you add a new file, give it a proper license header. Do not copy and
238 paste it from a random place, use an existing file as template.
241 We think our rules are not too hard. If you have comments, contact us.
243 Note, these rules are mostly borrowed from the MPlayer project.
245 @section Submitting patches
247 First, read the (@pxref{Coding Rules}) above if you did not yet.
249 When you submit your patch, please use @code{git format-patch} or
250 @code{git send-email}. We cannot read other diffs :-)
252 Also please do not submit a patch which contains several unrelated changes.
253 Split it into separate, self-contained pieces. This does not mean splitting
254 file by file. Instead, make the patch as small as possible while still
255 keeping it as a logical unit that contains an individual change, even
256 if it spans multiple files. This makes reviewing your patches much easier
257 for us and greatly increases your chances of getting your patch applied.
259 Use the patcheck tool of FFmpeg to check your patch.
260 The tool is located in the tools directory.
262 Run the regression tests before submitting a patch so that you can
263 verify that there are no big problems.
265 Patches should be posted as base64 encoded attachments (or any other
266 encoding which ensures that the patch will not be trashed during
267 transmission) to the ffmpeg-devel mailing list, see
268 @url{http://lists.ffmpeg.org/mailman/listinfo/ffmpeg-devel}
270 It also helps quite a bit if you tell us what the patch does (for example
271 'replaces lrint by lrintf'), and why (for example '*BSD isn't C99 compliant
274 Also please if you send several patches, send each patch as a separate mail,
275 do not attach several unrelated patches to the same mail.
277 Your patch will be reviewed on the mailing list. You will likely be asked
278 to make some changes and are expected to send in an improved version that
279 incorporates the requests from the review. This process may go through
280 several iterations. Once your patch is deemed good enough, some developer
281 will pick it up and commit it to the official FFmpeg tree.
283 Give us a few days to react. But if some time passes without reaction,
284 send a reminder by email. Your patch should eventually be dealt with.
287 @section New codecs or formats checklist
291 Did you use av_cold for codec initialization and close functions?
293 Did you add a long_name under NULL_IF_CONFIG_SMALL to the AVCodec or
294 AVInputFormat/AVOutputFormat struct?
296 Did you bump the minor version number (and reset the micro version
297 number) in @file{avcodec.h} or @file{avformat.h}?
299 Did you register it in @file{allcodecs.c} or @file{allformats.c}?
301 Did you add the CodecID to @file{avcodec.h}?
303 If it has a fourcc, did you add it to @file{libavformat/riff.c},
304 even if it is only a decoder?
306 Did you add a rule to compile the appropriate files in the Makefile?
307 Remember to do this even if you're just adding a format to a file that is
308 already being compiled by some other rule, like a raw demuxer.
310 Did you add an entry to the table of supported formats or codecs in
311 @file{doc/general.texi}?
313 Did you add an entry in the Changelog?
315 If it depends on a parser or a library, did you add that dependency in
318 Did you @code{git add} the appropriate files before committing?
320 Did you make sure it compiles standalone, i.e. with
321 @code{configure --disable-everything --enable-decoder=foo}
322 (or @code{--enable-demuxer} or whatever your component is)?
326 @section patch submission checklist
330 Does 'make fate' pass with the patch applied?
332 Was the patch generated with git format-patch or send-email?
334 Did you sign off your patch? (git commit -s)
335 See @url{http://kerneltrap.org/files/Jeremy/DCO.txt} for the meaning
338 Did you provide a clear git commit log message?
340 Is the patch against latest FFmpeg git master branch?
342 Are you subscribed to ffmpeg-dev?
343 (the list is subscribers only due to spam)
345 Have you checked that the changes are minimal, so that the same cannot be
346 achieved with a smaller patch and/or simpler final code?
348 If the change is to speed critical code, did you benchmark it?
350 If you did any benchmarks, did you provide them in the mail?
352 Have you checked that the patch does not introduce buffer overflows or
353 other security issues?
355 Did you test your decoder or demuxer against damaged data? If no, see
356 tools/trasher and the noise bitstream filter. Your decoder or demuxer
357 should not crash or end in a (near) infinite loop when fed damaged data.
359 Does the patch not mix functional and cosmetic changes?
361 Did you add tabs or trailing whitespace to the code? Both are forbidden.
363 Is the patch attached to the email you send?
365 Is the mime type of the patch correct? It should be text/x-diff or
366 text/x-patch or at least text/plain and not application/octet-stream.
368 If the patch fixes a bug, did you provide a verbose analysis of the bug?
370 If the patch fixes a bug, did you provide enough information, including
371 a sample, so the bug can be reproduced and the fix can be verified?
372 Note please do not attach samples >100k to mails but rather provide a
373 URL, you can upload to ftp://upload.ffmpeg.org
375 Did you provide a verbose summary about what the patch does change?
377 Did you provide a verbose explanation why it changes things like it does?
379 Did you provide a verbose summary of the user visible advantages and
380 disadvantages if the patch is applied?
382 Did you provide an example so we can verify the new feature added by the
385 If you added a new file, did you insert a license header? It should be
386 taken from FFmpeg, not randomly copied and pasted from somewhere else.
388 You should maintain alphabetical order in alphabetically ordered lists as
389 long as doing so does not break API/ABI compatibility.
391 Lines with similar content should be aligned vertically when doing so
392 improves readability.
394 Consider to add a regression test for your code.
396 If you added YASM code please check that things still work with --disable-yasm
399 @section Patch review process
401 All patches posted to ffmpeg-devel will be reviewed, unless they contain a
402 clear note that the patch is not for the git master branch.
403 Reviews and comments will be posted as replies to the patch on the
404 mailing list. The patch submitter then has to take care of every comment,
405 that can be by resubmitting a changed patch or by discussion. Resubmitted
406 patches will themselves be reviewed like any other patch. If at some point
407 a patch passes review with no comments then it is approved, that can for
408 simple and small patches happen immediately while large patches will generally
409 have to be changed and reviewed many times before they are approved.
410 After a patch is approved it will be committed to the repository.
412 We will review all submitted patches, but sometimes we are quite busy so
413 especially for large patches this can take several weeks.
415 When resubmitting patches, please do not make any significant changes
416 not related to the comments received during review. Such patches will
417 be rejected. Instead, submit significant changes or new features as
420 @section Regression tests
422 Before submitting a patch (or committing to the repository), you should at least
423 test that you did not break anything.
425 The regression tests build a synthetic video stream and a synthetic
426 audio stream. These are then encoded and decoded with all codecs or
427 formats. The CRC (or MD5) of each generated file is recorded in a
428 result file. A 'diff' is launched to compare the reference results and
429 the result file. The output is checked immediately after each test
432 The regression tests then go on to test the FFserver code with a
433 limited set of streams. It is important that this step runs correctly
436 Run 'make test' to test all the codecs and formats. Commands like
437 'make regtest-mpeg2' can be used to run a single test. By default,
438 make will abort if any test fails. To run all tests regardless,
439 use make -k. To get a more verbose output, use 'make V=1 test' or
442 Run 'make fulltest' to test all the codecs, formats and FFserver.
444 [Of course, some patches may change the results of the regression tests. In
445 this case, the reference results of the regression tests shall be modified