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 Submitting Patches to the main developer mailing list
42 see @ref{Submitting patches} for details.
43 @item Directly committing changes to the main tree.
44 @item Committing 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 committed. 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 @subsection Code formatting conventions
58 There are the following guidelines regarding the indentation in files:
63 The TAB character is forbidden outside of Makefiles as is any
64 form of trailing whitespace. Commits containing either will be
65 rejected by the git repository.
67 You should try to limit your code lines to 80 characters; however, do so if
68 and only if this improves readability.
70 The presentation is one inspired by 'indent -i4 -kr -nut'.
72 The main priority in FFmpeg is simplicity and small code size in order to
73 minimize the bug count.
76 Use the JavaDoc/Doxygen format (see examples below) so that code documentation
77 can be generated automatically. All nontrivial functions should have a comment
78 above them explaining what the function does, even if it is just one sentence.
79 All structures and their member variables should be documented, too.
81 Avoid Qt-style and similar Doxygen syntax with @code{!} in it, i.e. replace
82 @code{//!} with @code{///} and similar. Also @@ syntax should be employed
83 for markup commands, i.e. use @code{@@param} and not @code{\param}.
97 typedef struct Foobar@{
98 int var1; /**< var1 description */
99 int var2; ///< var2 description
100 /** var3 description */
108 * @@param my_parameter description of my_parameter
109 * @@return return value description
111 int myfunc(int my_parameter)
115 @subsection C language features
117 FFmpeg is programmed in the ISO C90 language with a few additional
118 features from ISO C99, namely:
121 the @samp{inline} keyword;
125 designated struct initializers (@samp{struct s x = @{ .i = 17 @};})
127 compound literals (@samp{x = (struct s) @{ 17, 23 @};})
130 These features are supported by all compilers we care about, so we will not
131 accept patches to remove their use unless they absolutely do not impair
132 clarity and performance.
134 All code must compile with recent versions of GCC and a number of other
135 currently supported compilers. To ensure compatibility, please do not use
136 additional C99 features or GCC extensions. Especially watch out for:
139 mixing statements and declarations;
141 @samp{long long} (use @samp{int64_t} instead);
143 @samp{__attribute__} not protected by @samp{#ifdef __GNUC__} or similar;
145 GCC statement expressions (@samp{(x = (@{ int y = 4; y; @})}).
148 @subsection Naming conventions
149 All names are using underscores (_), not CamelCase. For example, @samp{avfilter_get_video_buffer} is
150 a valid function name and @samp{AVFilterGetVideo} is not. The exception from this are type names, like
151 for example structs and enums; they should always be in the CamelCase
154 There are following conventions for naming variables and functions:
157 For local variables no prefix is required.
159 For variables and functions declared as @code{static} no prefixes are required.
161 For variables and functions used internally by the library, @code{ff_} prefix
163 For example, @samp{ff_w64_demuxer}.
165 For variables and functions used internally across multiple libraries, use
166 @code{avpriv_}. For example, @samp{avpriv_aac_parse_header}.
168 For exported names, each library has its own prefixes. Just check the existing
169 code and name accordingly.
172 @subsection Miscellanous conventions
175 fprintf and printf are forbidden in libavformat and libavcodec,
176 please use av_log() instead.
178 Casts should be used only when necessary. Unneeded parentheses
179 should also be avoided if they don't make the code easier to understand.
182 @subsection Editor configuration
183 In order to configure Vim to follow FFmpeg formatting conventions, paste
184 the following snippet into your @file{.vimrc}:
186 " indentation rules for FFmpeg: 4 spaces, no tabs
190 " allow tabs in Makefiles
191 autocmd FileType make set noexpandtab shiftwidth=8 softtabstop=8
192 " Trailing whitespace and tabs are forbidden, so highlight them.
193 highlight ForbiddenWhitespace ctermbg=red guibg=red
194 match ForbiddenWhitespace /\s\+$\|\t/
195 " Do not highlight spaces at the end of line while typing on that line.
196 autocmd InsertEnter * match ForbiddenWhitespace /\t\|\s\+\%#\@@<!$/
199 For Emacs, add these roughly equivalent lines to your @file{.emacs.d/init.el}:
201 (setq c-default-style "k&r")
202 (setq-default c-basic-offset 4)
203 (setq-default indent-tabs-mode nil)
204 (setq-default show-trailing-whitespace t)
207 @section Development Policy
211 Contributions should be licensed under the LGPL 2.1, including an
212 "or any later version" clause, or the MIT license. GPL 2 including
213 an "or any later version" clause is also acceptable, but LGPL is
216 You must not commit code which breaks FFmpeg! (Meaning unfinished but
217 enabled code which breaks compilation or compiles but does not work or
218 breaks the regression tests)
219 You can commit unfinished stuff (for testing etc), but it must be disabled
220 (#ifdef etc) by default so it does not interfere with other developers'
223 You do not have to over-test things. If it works for you, and you think it
224 should work for others, then commit. If your code has problems
225 (portability, triggers compiler bugs, unusual environment etc) they will be
226 reported and eventually fixed.
228 Do not commit unrelated changes together, split them into self-contained
229 pieces. Also do not forget that if part B depends on part A, but A does not
230 depend on B, then A can and should be committed first and separate from B.
231 Keeping changes well split into self-contained parts makes reviewing and
232 understanding them on the commit log mailing list easier. This also helps
233 in case of debugging later on.
234 Also if you have doubts about splitting or not splitting, do not hesitate to
235 ask/discuss it on the developer mailing list.
237 Do not change behavior of the programs (renaming options etc) or public
238 API or ABI without first discussing it on the ffmpeg-devel mailing list.
239 Do not remove functionality from the code. Just improve!
241 Note: Redundant code can be removed.
243 Do not commit changes to the build system (Makefiles, configure script)
244 which change behavior, defaults etc, without asking first. The same
245 applies to compiler warning fixes, trivial looking fixes and to code
246 maintained by other developers. We usually have a reason for doing things
247 the way we do. Send your changes as patches to the ffmpeg-devel mailing
248 list, and if the code maintainers say OK, you may commit. This does not
249 apply to files you wrote and/or maintain.
251 We refuse source indentation and other cosmetic changes if they are mixed
252 with functional changes, such commits will be rejected and removed. Every
253 developer has his own indentation style, you should not change it. Of course
254 if you (re)write something, you can use your own style, even though we would
255 prefer if the indentation throughout FFmpeg was consistent (Many projects
256 force a given indentation style - we do not.). If you really need to make
257 indentation changes (try to avoid this), separate them strictly from real
260 NOTE: If you had to put if()@{ .. @} over a large (> 5 lines) chunk of code,
261 then either do NOT change the indentation of the inner part within (do not
262 move it to the right)! or do so in a separate commit
264 Always fill out the commit log message. Describe in a few lines what you
265 changed and why. You can refer to mailing list postings if you fix a
266 particular bug. Comments such as "fixed!" or "Changed it." are unacceptable.
268 area changed: Short 1 line description
270 details describing what and why and giving references.
272 Make sure the author of the commit is set correctly. (see git commit --author)
273 If you apply a patch, send an
274 answer to ffmpeg-devel (or wherever you got the patch from) saying that
275 you applied the patch.
277 When applying patches that have been discussed (at length) on the mailing
278 list, reference the thread in the log message.
280 Do NOT commit to code actively maintained by others without permission.
281 Send a patch to ffmpeg-devel instead. If no one answers within a reasonable
282 timeframe (12h for build failures and security fixes, 3 days small changes,
283 1 week for big patches) then commit your patch if you think it is OK.
284 Also note, the maintainer can simply ask for more time to review!
286 Subscribe to the ffmpeg-cvslog mailing list. The diffs of all commits
287 are sent there and reviewed by all the other developers. Bugs and possible
288 improvements or general questions regarding commits are discussed there. We
289 expect you to react if problems with your code are uncovered.
291 Update the documentation if you change behavior or add features. If you are
292 unsure how best to do this, send a patch to ffmpeg-devel, the documentation
293 maintainer(s) will review and commit your stuff.
295 Try to keep important discussions and requests (also) on the public
296 developer mailing list, so that all developers can benefit from them.
298 Never write to unallocated memory, never write over the end of arrays,
299 always check values read from some untrusted source before using them
300 as array index or other risky things.
302 Remember to check if you need to bump versions for the specific libav*
303 parts (libavutil, libavcodec, libavformat) you are changing. You need
304 to change the version integer.
305 Incrementing the first component means no backward compatibility to
306 previous versions (e.g. removal of a function from the public API).
307 Incrementing the second component means backward compatible change
308 (e.g. addition of a function to the public API or extension of an
309 existing data structure).
310 Incrementing the third component means a noteworthy binary compatible
311 change (e.g. encoder bug fix that matters for the decoder).
313 Compiler warnings indicate potential bugs or code with bad style. If a type of
314 warning always points to correct and clean code, that warning should
315 be disabled, not the code changed.
316 Thus the remaining warnings can either be bugs or correct code.
317 If it is a bug, the bug has to be fixed. If it is not, the code should
318 be changed to not generate a warning unless that causes a slowdown
319 or obfuscates the code.
321 If you add a new file, give it a proper license header. Do not copy and
322 paste it from a random place, use an existing file as template.
325 We think our rules are not too hard. If you have comments, contact us.
327 Note, these rules are mostly borrowed from the MPlayer project.
329 @anchor{Submitting patches}
330 @section Submitting patches
332 First, read the @ref{Coding Rules} above if you did not yet, in particular
333 the rules regarding patch submission.
335 When you submit your patch, please use @code{git format-patch} or
336 @code{git send-email}. We cannot read other diffs :-)
338 Also please do not submit a patch which contains several unrelated changes.
339 Split it into separate, self-contained pieces. This does not mean splitting
340 file by file. Instead, make the patch as small as possible while still
341 keeping it as a logical unit that contains an individual change, even
342 if it spans multiple files. This makes reviewing your patches much easier
343 for us and greatly increases your chances of getting your patch applied.
345 Use the patcheck tool of FFmpeg to check your patch.
346 The tool is located in the tools directory.
348 Run the @ref{Regression tests} before submitting a patch in order to verify
349 it does not cause unexpected problems.
351 Patches should be posted as base64 encoded attachments (or any other
352 encoding which ensures that the patch will not be trashed during
353 transmission) to the ffmpeg-devel mailing list, see
354 @url{http://lists.ffmpeg.org/mailman/listinfo/ffmpeg-devel}
356 It also helps quite a bit if you tell us what the patch does (for example
357 'replaces lrint by lrintf'), and why (for example '*BSD isn't C99 compliant
360 Also please if you send several patches, send each patch as a separate mail,
361 do not attach several unrelated patches to the same mail.
363 Your patch will be reviewed on the mailing list. You will likely be asked
364 to make some changes and are expected to send in an improved version that
365 incorporates the requests from the review. This process may go through
366 several iterations. Once your patch is deemed good enough, some developer
367 will pick it up and commit it to the official FFmpeg tree.
369 Give us a few days to react. But if some time passes without reaction,
370 send a reminder by email. Your patch should eventually be dealt with.
373 @section New codecs or formats checklist
377 Did you use av_cold for codec initialization and close functions?
379 Did you add a long_name under NULL_IF_CONFIG_SMALL to the AVCodec or
380 AVInputFormat/AVOutputFormat struct?
382 Did you bump the minor version number (and reset the micro version
383 number) in @file{libavcodec/version.h} or @file{libavformat/version.h}?
385 Did you register it in @file{allcodecs.c} or @file{allformats.c}?
387 Did you add the CodecID to @file{avcodec.h}?
389 If it has a fourCC, did you add it to @file{libavformat/riff.c},
390 even if it is only a decoder?
392 Did you add a rule to compile the appropriate files in the Makefile?
393 Remember to do this even if you're just adding a format to a file that is
394 already being compiled by some other rule, like a raw demuxer.
396 Did you add an entry to the table of supported formats or codecs in
397 @file{doc/general.texi}?
399 Did you add an entry in the Changelog?
401 If it depends on a parser or a library, did you add that dependency in
404 Did you @code{git add} the appropriate files before committing?
406 Did you make sure it compiles standalone, i.e. with
407 @code{configure --disable-everything --enable-decoder=foo}
408 (or @code{--enable-demuxer} or whatever your component is)?
412 @section patch submission checklist
416 Does @code{make fate} pass with the patch applied?
418 Was the patch generated with git format-patch or send-email?
420 Did you sign off your patch? (git commit -s)
421 See @url{http://kerneltrap.org/files/Jeremy/DCO.txt} for the meaning
424 Did you provide a clear git commit log message?
426 Is the patch against latest FFmpeg git master branch?
428 Are you subscribed to ffmpeg-devel?
429 (the list is subscribers only due to spam)
431 Have you checked that the changes are minimal, so that the same cannot be
432 achieved with a smaller patch and/or simpler final code?
434 If the change is to speed critical code, did you benchmark it?
436 If you did any benchmarks, did you provide them in the mail?
438 Have you checked that the patch does not introduce buffer overflows or
439 other security issues?
441 Did you test your decoder or demuxer against damaged data? If no, see
442 tools/trasher and the noise bitstream filter. Your decoder or demuxer
443 should not crash or end in a (near) infinite loop when fed damaged data.
445 Does the patch not mix functional and cosmetic changes?
447 Did you add tabs or trailing whitespace to the code? Both are forbidden.
449 Is the patch attached to the email you send?
451 Is the mime type of the patch correct? It should be text/x-diff or
452 text/x-patch or at least text/plain and not application/octet-stream.
454 If the patch fixes a bug, did you provide a verbose analysis of the bug?
456 If the patch fixes a bug, did you provide enough information, including
457 a sample, so the bug can be reproduced and the fix can be verified?
458 Note please do not attach samples >100k to mails but rather provide a
459 URL, you can upload to ftp://upload.ffmpeg.org
461 Did you provide a verbose summary about what the patch does change?
463 Did you provide a verbose explanation why it changes things like it does?
465 Did you provide a verbose summary of the user visible advantages and
466 disadvantages if the patch is applied?
468 Did you provide an example so we can verify the new feature added by the
471 If you added a new file, did you insert a license header? It should be
472 taken from FFmpeg, not randomly copied and pasted from somewhere else.
474 You should maintain alphabetical order in alphabetically ordered lists as
475 long as doing so does not break API/ABI compatibility.
477 Lines with similar content should be aligned vertically when doing so
478 improves readability.
480 Consider to add a regression test for your code.
482 If you added YASM code please check that things still work with --disable-yasm
485 @section Patch review process
487 All patches posted to ffmpeg-devel will be reviewed, unless they contain a
488 clear note that the patch is not for the git master branch.
489 Reviews and comments will be posted as replies to the patch on the
490 mailing list. The patch submitter then has to take care of every comment,
491 that can be by resubmitting a changed patch or by discussion. Resubmitted
492 patches will themselves be reviewed like any other patch. If at some point
493 a patch passes review with no comments then it is approved, that can for
494 simple and small patches happen immediately while large patches will generally
495 have to be changed and reviewed many times before they are approved.
496 After a patch is approved it will be committed to the repository.
498 We will review all submitted patches, but sometimes we are quite busy so
499 especially for large patches this can take several weeks.
501 If you feel that the review process is too slow and you are willing to try to
502 take over maintainership of the area of code you change then just clone
503 git master and maintain the area of code there. We will merge each area from
504 where its best maintained.
506 When resubmitting patches, please do not make any significant changes
507 not related to the comments received during review. Such patches will
508 be rejected. Instead, submit significant changes or new features as
511 @anchor{Regression tests}
512 @section Regression tests
514 Before submitting a patch (or committing to the repository), you should at least
515 test that you did not break anything.
517 Running 'make fate' accomplishes this, please see @url{fate.html} for details.
519 [Of course, some patches may change the results of the regression tests. In
520 this case, the reference results of the regression tests shall be modified