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.
39 There are 3 ways by which code gets into ffmpeg.
41 @item Submiting Patches to the main developer mailing list
42 see @ref{Submitting patches} for details.
43 @item Directly commiting changes to the main tree.
44 @item Commiting changes to a git clone, for example on github.com or
45 gitorious.org. And asking us to merge these changes.
48 Whichever way, changes should be reviewed by the maintainer of the code
49 before they are commited. And they should follow the @ref{Coding Rules}.
50 The developer making the commit and the author are responsible for their changes
51 and should try to fix issues their commit causes.
56 FFmpeg is programmed in the ISO C90 language with a few additional
57 features from ISO C99, namely:
60 the @samp{inline} keyword;
64 designated struct initializers (@samp{struct s x = @{ .i = 17 @};})
66 compound literals (@samp{x = (struct s) @{ 17, 23 @};})
69 These features are supported by all compilers we care about, so we will not
70 accept patches to remove their use unless they absolutely do not impair
71 clarity and performance.
73 All code must compile with recent versions of GCC and a number of other
74 currently supported compilers. To ensure compatibility, please do not use
75 additional C99 features or GCC extensions. Especially watch out for:
78 mixing statements and declarations;
80 @samp{long long} (use @samp{int64_t} instead);
82 @samp{__attribute__} not protected by @samp{#ifdef __GNUC__} or similar;
84 GCC statement expressions (@samp{(x = (@{ int y = 4; y; @})}).
88 The presentation is one inspired by 'indent -i4 -kr -nut'.
89 The TAB character is forbidden outside of Makefiles as is any
90 form of trailing whitespace. Commits containing either will be
91 rejected by the git repository.
93 The main priority in FFmpeg is simplicity and small code size in order to
94 minimize the bug count.
96 Comments: Use the JavaDoc/Doxygen
97 format (see examples below) so that code documentation
98 can be generated automatically. All nontrivial functions should have a comment
99 above them explaining what the function does, even if it is just one sentence.
100 All structures and their member variables should be documented, too.
113 typedef struct Foobar@{
114 int var1; /**< var1 description */
115 int var2; ///< var2 description
116 /** var3 description */
124 * @@param my_parameter description of my_parameter
125 * @@return return value description
127 int myfunc(int my_parameter)
131 fprintf and printf are forbidden in libavformat and libavcodec,
132 please use av_log() instead.
134 Casts should be used only when necessary. Unneeded parentheses
135 should also be avoided if they don't make the code easier to understand.
137 @section Development Policy
141 Contributions should be licensed under the LGPL 2.1, including an
142 "or any later version" clause, or the MIT license. GPL 2 including
143 an "or any later version" clause is also acceptable, but LGPL is
146 You must not commit code which breaks FFmpeg! (Meaning unfinished but
147 enabled code which breaks compilation or compiles but does not work or
148 breaks the regression tests)
149 You can commit unfinished stuff (for testing etc), but it must be disabled
150 (#ifdef etc) by default so it does not interfere with other developers'
153 You do not have to over-test things. If it works for you, and you think it
154 should work for others, then commit. If your code has problems
155 (portability, triggers compiler bugs, unusual environment etc) they will be
156 reported and eventually fixed.
158 Do not commit unrelated changes together, split them into self-contained
159 pieces. Also do not forget that if part B depends on part A, but A does not
160 depend on B, then A can and should be committed first and separate from B.
161 Keeping changes well split into self-contained parts makes reviewing and
162 understanding them on the commit log mailing list easier. This also helps
163 in case of debugging later on.
164 Also if you have doubts about splitting or not splitting, do not hesitate to
165 ask/discuss it on the developer mailing list.
167 Do not change behavior of the programs (renaming options etc) or public
168 API or ABI without first discussing it on the ffmpeg-devel mailing list.
169 Do not remove functionality from the code. Just improve!
171 Note: Redundant code can be removed.
173 Do not commit changes to the build system (Makefiles, configure script)
174 which change behavior, defaults etc, without asking first. The same
175 applies to compiler warning fixes, trivial looking fixes and to code
176 maintained by other developers. We usually have a reason for doing things
177 the way we do. Send your changes as patches to the ffmpeg-devel mailing
178 list, and if the code maintainers say OK, you may commit. This does not
179 apply to files you wrote and/or maintain.
181 We refuse source indentation and other cosmetic changes if they are mixed
182 with functional changes, such commits will be rejected and removed. Every
183 developer has his own indentation style, you should not change it. Of course
184 if you (re)write something, you can use your own style, even though we would
185 prefer if the indentation throughout FFmpeg was consistent (Many projects
186 force a given indentation style - we do not.). If you really need to make
187 indentation changes (try to avoid this), separate them strictly from real
190 NOTE: If you had to put if()@{ .. @} over a large (> 5 lines) chunk of code,
191 then either do NOT change the indentation of the inner part within (do not
192 move it to the right)! or do so in a separate commit
194 Always fill out the commit log message. Describe in a few lines what you
195 changed and why. You can refer to mailing list postings if you fix a
196 particular bug. Comments such as "fixed!" or "Changed it." are unacceptable.
198 area changed: Short 1 line description
200 details describing what and why and giving references.
202 Make sure the author of the commit is set correctly. (see git commit --author)
203 If you apply a patch, send an
204 answer to ffmpeg-devel (or wherever you got the patch from) saying that
205 you applied the patch.
207 When applying patches that have been discussed (at length) on the mailing
208 list, reference the thread in the log message.
210 Do NOT commit to code actively maintained by others without permission.
211 Send a patch to ffmpeg-devel instead. If no one answers within a reasonable
212 timeframe (12h for build failures and security fixes, 3 days small changes,
213 1 week for big patches) then commit your patch if you think it is OK.
214 Also note, the maintainer can simply ask for more time to review!
216 Subscribe to the ffmpeg-cvslog mailing list. The diffs of all commits
217 are sent there and reviewed by all the other developers. Bugs and possible
218 improvements or general questions regarding commits are discussed there. We
219 expect you to react if problems with your code are uncovered.
221 Update the documentation if you change behavior or add features. If you are
222 unsure how best to do this, send a patch to ffmpeg-devel, the documentation
223 maintainer(s) will review and commit your stuff.
225 Try to keep important discussions and requests (also) on the public
226 developer mailing list, so that all developers can benefit from them.
228 Never write to unallocated memory, never write over the end of arrays,
229 always check values read from some untrusted source before using them
230 as array index or other risky things.
232 Remember to check if you need to bump versions for the specific libav
233 parts (libavutil, libavcodec, libavformat) you are changing. You need
234 to change the version integer.
235 Incrementing the first component means no backward compatibility to
236 previous versions (e.g. removal of a function from the public API).
237 Incrementing the second component means backward compatible change
238 (e.g. addition of a function to the public API or extension of an
239 existing data structure).
240 Incrementing the third component means a noteworthy binary compatible
241 change (e.g. encoder bug fix that matters for the decoder).
243 Compiler warnings indicate potential bugs or code with bad style. If a type of
244 warning always points to correct and clean code, that warning should
245 be disabled, not the code changed.
246 Thus the remaining warnings can either be bugs or correct code.
247 If it is a bug, the bug has to be fixed. If it is not, the code should
248 be changed to not generate a warning unless that causes a slowdown
249 or obfuscates the code.
251 If you add a new file, give it a proper license header. Do not copy and
252 paste it from a random place, use an existing file as template.
255 We think our rules are not too hard. If you have comments, contact us.
257 Note, these rules are mostly borrowed from the MPlayer project.
259 @anchor{Submitting patches}
260 @section Submitting patches
262 First, read the @ref{Coding Rules} above if you did not yet, in particular
263 the rules regarding patch submission.
265 When you submit your patch, please use @code{git format-patch} or
266 @code{git send-email}. We cannot read other diffs :-)
268 Also please do not submit a patch which contains several unrelated changes.
269 Split it into separate, self-contained pieces. This does not mean splitting
270 file by file. Instead, make the patch as small as possible while still
271 keeping it as a logical unit that contains an individual change, even
272 if it spans multiple files. This makes reviewing your patches much easier
273 for us and greatly increases your chances of getting your patch applied.
275 Use the patcheck tool of FFmpeg to check your patch.
276 The tool is located in the tools directory.
278 Run the @ref{Regression Tests} before submitting a patch in order to verify
279 it does not cause unexpected problems.
281 Patches should be posted as base64 encoded attachments (or any other
282 encoding which ensures that the patch will not be trashed during
283 transmission) to the ffmpeg-devel mailing list, see
284 @url{http://lists.ffmpeg.org/mailman/listinfo/ffmpeg-devel}
286 It also helps quite a bit if you tell us what the patch does (for example
287 'replaces lrint by lrintf'), and why (for example '*BSD isn't C99 compliant
290 Also please if you send several patches, send each patch as a separate mail,
291 do not attach several unrelated patches to the same mail.
293 Your patch will be reviewed on the mailing list. You will likely be asked
294 to make some changes and are expected to send in an improved version that
295 incorporates the requests from the review. This process may go through
296 several iterations. Once your patch is deemed good enough, some developer
297 will pick it up and commit it to the official FFmpeg tree.
299 Give us a few days to react. But if some time passes without reaction,
300 send a reminder by email. Your patch should eventually be dealt with.
303 @section New codecs or formats checklist
307 Did you use av_cold for codec initialization and close functions?
309 Did you add a long_name under NULL_IF_CONFIG_SMALL to the AVCodec or
310 AVInputFormat/AVOutputFormat struct?
312 Did you bump the minor version number (and reset the micro version
313 number) in @file{libavcodec/version.h} or @file{libavformat/version.h}?
315 Did you register it in @file{allcodecs.c} or @file{allformats.c}?
317 Did you add the CodecID to @file{avcodec.h}?
319 If it has a fourcc, did you add it to @file{libavformat/riff.c},
320 even if it is only a decoder?
322 Did you add a rule to compile the appropriate files in the Makefile?
323 Remember to do this even if you're just adding a format to a file that is
324 already being compiled by some other rule, like a raw demuxer.
326 Did you add an entry to the table of supported formats or codecs in
327 @file{doc/general.texi}?
329 Did you add an entry in the Changelog?
331 If it depends on a parser or a library, did you add that dependency in
334 Did you @code{git add} the appropriate files before committing?
336 Did you make sure it compiles standalone, i.e. with
337 @code{configure --disable-everything --enable-decoder=foo}
338 (or @code{--enable-demuxer} or whatever your component is)?
342 @section patch submission checklist
346 Does @code{make fate} pass with the patch applied?
348 Was the patch generated with git format-patch or send-email?
350 Did you sign off your patch? (git commit -s)
351 See @url{http://kerneltrap.org/files/Jeremy/DCO.txt} for the meaning
354 Did you provide a clear git commit log message?
356 Is the patch against latest FFmpeg git master branch?
358 Are you subscribed to ffmpeg-devel?
359 (the list is subscribers only due to spam)
361 Have you checked that the changes are minimal, so that the same cannot be
362 achieved with a smaller patch and/or simpler final code?
364 If the change is to speed critical code, did you benchmark it?
366 If you did any benchmarks, did you provide them in the mail?
368 Have you checked that the patch does not introduce buffer overflows or
369 other security issues?
371 Did you test your decoder or demuxer against damaged data? If no, see
372 tools/trasher and the noise bitstream filter. Your decoder or demuxer
373 should not crash or end in a (near) infinite loop when fed damaged data.
375 Does the patch not mix functional and cosmetic changes?
377 Did you add tabs or trailing whitespace to the code? Both are forbidden.
379 Is the patch attached to the email you send?
381 Is the mime type of the patch correct? It should be text/x-diff or
382 text/x-patch or at least text/plain and not application/octet-stream.
384 If the patch fixes a bug, did you provide a verbose analysis of the bug?
386 If the patch fixes a bug, did you provide enough information, including
387 a sample, so the bug can be reproduced and the fix can be verified?
388 Note please do not attach samples >100k to mails but rather provide a
389 URL, you can upload to ftp://upload.ffmpeg.org
391 Did you provide a verbose summary about what the patch does change?
393 Did you provide a verbose explanation why it changes things like it does?
395 Did you provide a verbose summary of the user visible advantages and
396 disadvantages if the patch is applied?
398 Did you provide an example so we can verify the new feature added by the
401 If you added a new file, did you insert a license header? It should be
402 taken from FFmpeg, not randomly copied and pasted from somewhere else.
404 You should maintain alphabetical order in alphabetically ordered lists as
405 long as doing so does not break API/ABI compatibility.
407 Lines with similar content should be aligned vertically when doing so
408 improves readability.
410 Consider to add a regression test for your code.
412 If you added YASM code please check that things still work with --disable-yasm
415 @section Patch review process
417 All patches posted to ffmpeg-devel will be reviewed, unless they contain a
418 clear note that the patch is not for the git master branch.
419 Reviews and comments will be posted as replies to the patch on the
420 mailing list. The patch submitter then has to take care of every comment,
421 that can be by resubmitting a changed patch or by discussion. Resubmitted
422 patches will themselves be reviewed like any other patch. If at some point
423 a patch passes review with no comments then it is approved, that can for
424 simple and small patches happen immediately while large patches will generally
425 have to be changed and reviewed many times before they are approved.
426 After a patch is approved it will be committed to the repository.
428 We will review all submitted patches, but sometimes we are quite busy so
429 especially for large patches this can take several weeks.
431 If you feel that the review process is too slow and you are willing to try to
432 take over maintainership of the area of code you change then just clone
433 git master and maintain the area of code there. We will merge each area from
434 where its best maintained.
436 When resubmitting patches, please do not make any significant changes
437 not related to the comments received during review. Such patches will
438 be rejected. Instead, submit significant changes or new features as
441 @section Regression tests
443 Before submitting a patch (or committing to the repository), you should at least
444 test that you did not break anything.
446 Running 'make fate' accomplishes this, please see @file{doc/fate.txt} for details.
448 [Of course, some patches may change the results of the regression tests. In
449 this case, the reference results of the regression tests shall be modified