doc: Mention memory allocation in the fuzz testing section
[libav.git] / doc / developer.texi
1 \input texinfo @c -*- texinfo -*-
2
3 @settitle Developer Documentation
4 @titlepage
5 @center @titlefont{Developer Documentation}
6 @end titlepage
7
8 @top
9
10 @contents
11
12 @chapter Developers Guide
13
14 @section API
15 @itemize @bullet
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.
18
19 @item libavformat is the library containing the file format handling (mux and
20 demux code for several formats). Look at @file{avplay.c} to use it in a
21 player. See @file{libavformat/output-example.c} to use it to generate
22 audio or video streams.
23
24 @end itemize
25
26 @section Integrating libav in your program
27
28 Shared libraries should be used whenever is possible in order to reduce
29 the effort distributors have to pour to support programs and to ensure
30 only the public API is used.
31
32 You can use Libav in your commercial program, but you must abide to the
33 license, LGPL or GPL depending on the specific features used, please refer
34 to @uref{http://libav.org/legal.html, our legal page} for a quick checklist and to
35 the following links for the exact text of each license:
36 @uref{http://git.libav.org/?p=libav.git;a=blob;f=COPYING.GPLv2, GPL version 2},
37 @uref{http://git.libav.org/?p=libav.git;a=blob;f=COPYING.GPLv3, GPL version 3},
38 @uref{http://git.libav.org/?p=libav.git;a=blob;f=COPYING.LGPLv2.1, LGPL version 2.1},
39 @uref{http://git.libav.org/?p=libav.git;a=blob;f=COPYING.LGPLv3, LGPL version 3}.
40 Any modification to the source code can be suggested for inclusion.
41 The best way to proceed is to send your patches to the
42 @uref{https://lists.libav.org/mailman/listinfo/libav-devel, libav-devel}
43 mailing list.
44
45 @anchor{Coding Rules}
46 @section Coding Rules
47
48 @subsection Code formatting conventions
49 The code is written in K&R C style. That means the following:
50 @itemize @bullet
51 @item
52 The control statements are formatted by putting space between the statement
53 and parenthesis in the following way:
54 @example
55 for (i = 0; i < filter->input_count; i++) @{
56 @end example
57 @item
58 The case statement is always located at the same level as the switch itself:
59 @example
60 switch (link->init_state) @{
61 case AVLINK_INIT:
62 continue;
63 case AVLINK_STARTINIT:
64 av_log(filter, AV_LOG_INFO, "circular filter chain detected");
65 return 0;
66 @end example
67 @item
68 Braces in function declarations are written on the new line:
69 @example
70 const char *avfilter_configuration(void)
71 @{
72 return LIBAV_CONFIGURATION;
73 @}
74 @end example
75 @item
76 Do not check for NULL values by comparison, @samp{if (p)} and
77 @samp{if (!p)} are correct; @samp{if (p == NULL)} and @samp{if (p != NULL)}
78 are not.
79 @item
80 In case of a single-statement if, no curly braces are required:
81 @example
82 if (!pic || !picref)
83 goto fail;
84 @end example
85 @item
86 Do not put spaces immediately inside parentheses. @samp{if (ret)} is
87 a valid style; @samp{if ( ret )} is not.
88 @end itemize
89
90 There are the following guidelines regarding the indentation in files:
91 @itemize @bullet
92 @item
93 Indent size is 4.
94 @item
95 The TAB character is forbidden outside of Makefiles as is any
96 form of trailing whitespace. Commits containing either will be
97 rejected by the git repository.
98 @item
99 You should try to limit your code lines to 80 characters; however, do so if
100 and only if this improves readability.
101 @end itemize
102 The presentation is one inspired by 'indent -i4 -kr -nut'.
103
104 The main priority in Libav is simplicity and small code size in order to
105 minimize the bug count.
106
107 @subsection Comments
108 Use the JavaDoc/Doxygen format (see examples below) so that code documentation
109 can be generated automatically. All nontrivial functions should have a comment
110 above them explaining what the function does, even if it is just one sentence.
111 All structures and their member variables should be documented, too.
112
113 Avoid Qt-style and similar Doxygen syntax with @code{!} in it, i.e. replace
114 @code{//!} with @code{///} and similar. Also @@ syntax should be employed
115 for markup commands, i.e. use @code{@@param} and not @code{\param}.
116
117 @example
118 /**
119 * @@file
120 * MPEG codec.
121 * @@author ...
122 */
123
124 /**
125 * Summary sentence.
126 * more text ...
127 * ...
128 */
129 typedef struct Foobar@{
130 int var1; /**< var1 description */
131 int var2; ///< var2 description
132 /** var3 description */
133 int var3;
134 @} Foobar;
135
136 /**
137 * Summary sentence.
138 * more text ...
139 * ...
140 * @@param my_parameter description of my_parameter
141 * @@return return value description
142 */
143 int myfunc(int my_parameter)
144 ...
145 @end example
146
147 @subsection C language features
148
149 Libav is programmed in the ISO C90 language with a few additional
150 features from ISO C99, namely:
151 @itemize @bullet
152 @item
153 the @samp{inline} keyword;
154 @item
155 @samp{//} comments;
156 @item
157 designated struct initializers (@samp{struct s x = @{ .i = 17 @};})
158 @item
159 compound literals (@samp{x = (struct s) @{ 17, 23 @};})
160 @end itemize
161
162 These features are supported by all compilers we care about, so we will not
163 accept patches to remove their use unless they absolutely do not impair
164 clarity and performance.
165
166 All code must compile with recent versions of GCC and a number of other
167 currently supported compilers. To ensure compatibility, please do not use
168 additional C99 features or GCC extensions. Especially watch out for:
169 @itemize @bullet
170 @item
171 mixing statements and declarations;
172 @item
173 @samp{long long} (use @samp{int64_t} instead);
174 @item
175 @samp{__attribute__} not protected by @samp{#ifdef __GNUC__} or similar;
176 @item
177 GCC statement expressions (@samp{(x = (@{ int y = 4; y; @})}).
178 @end itemize
179
180 @subsection Naming conventions
181 All names are using underscores (_), not CamelCase. For example,
182 @samp{avfilter_get_video_buffer} is a valid function name and
183 @samp{AVFilterGetVideo} is not. The only exception from this are structure
184 names; they should always be in the CamelCase
185
186 There are following conventions for naming variables and functions:
187 @itemize @bullet
188 @item
189 For local variables no prefix is required.
190 @item
191 For variables and functions declared as @code{static} no prefixes are required.
192 @item
193 For variables and functions used internally by the library, @code{ff_} prefix
194 should be used.
195 For example, @samp{ff_w64_demuxer}.
196 @item
197 For variables and functions used internally across multiple libraries, use
198 @code{avpriv_}. For example, @samp{avpriv_aac_parse_header}.
199 @item
200 For exported names, each library has its own prefixes. Just check the existing
201 code and name accordingly.
202 @end itemize
203
204 @subsection Miscellaneous conventions
205 @itemize @bullet
206 @item
207 fprintf and printf are forbidden in libavformat and libavcodec,
208 please use av_log() instead.
209 @item
210 Casts should be used only when necessary. Unneeded parentheses
211 should also be avoided if they don't make the code easier to understand.
212 @end itemize
213
214 @subsection Editor configuration
215 In order to configure Vim to follow Libav formatting conventions, paste
216 the following snippet into your @file{.vimrc}:
217 @example
218 " Indentation rules for Libav: 4 spaces, no tabs.
219 set expandtab
220 set shiftwidth=4
221 set softtabstop=4
222 set cindent
223 set cinoptions=(0
224 " Allow tabs in Makefiles.
225 autocmd FileType make set noexpandtab shiftwidth=8 softtabstop=8
226 " Trailing whitespace and tabs are forbidden, so highlight them.
227 highlight ForbiddenWhitespace ctermbg=red guibg=red
228 match ForbiddenWhitespace /\s\+$\|\t/
229 " Do not highlight spaces at the end of line while typing on that line.
230 autocmd InsertEnter * match ForbiddenWhitespace /\t\|\s\+\%#\@@<!$/
231 @end example
232
233 For Emacs, add these roughly equivalent lines to your @file{.emacs.d/init.el}:
234 @example
235 (c-add-style "libav"
236 '("k&r"
237 (c-basic-offset . 4)
238 (indent-tabs-mode . nil)
239 (show-trailing-whitespace . t)
240 (c-offsets-alist
241 (statement-cont . (c-lineup-assignments +)))
242 )
243 )
244 (setq c-default-style "libav")
245 @end example
246
247 @section Development Policy
248
249 @enumerate
250 @item
251 Contributions should be licensed under the LGPL 2.1, including an
252 "or any later version" clause, or the MIT license. GPL 2 including
253 an "or any later version" clause is also acceptable, but LGPL is
254 preferred.
255 @item
256 All the patches MUST be reviewed in the mailing list before they are
257 committed.
258 @item
259 The Libav coding style should remain consistent. Changes to
260 conform will be suggested during the review or implemented on commit.
261 @item
262 Patches should be generated using @code{git format-patch} or directly sent
263 using @code{git send-email}.
264 Please make sure you give the proper credit by setting the correct author
265 in the commit.
266 @item
267 The commit message should have a short first line in the form of
268 a @samp{topic: short description} as a header, separated by a newline
269 from the body consisting of an explanation of why the change is necessary.
270 Referring to the issue on the bug tracker does not exempt to report an
271 excerpt of the bug.
272 @item
273 Work in progress patches should be sent to the mailing list with the [WIP]
274 or the [RFC] tag.
275 @item
276 Branches in public personal repos are advised as way to
277 work on issues collaboratively.
278 @item
279 You do not have to over-test things. If it works for you and you think it
280 should work for others, send it to the mailing list for review.
281 If you have doubt about portability please state it in the submission so
282 people with specific hardware could test it.
283 @item
284 Do not commit unrelated changes together, split them into self-contained
285 pieces. Also do not forget that if part B depends on part A, but A does not
286 depend on B, then A can and should be committed first and separate from B.
287 Keeping changes well split into self-contained parts makes reviewing and
288 understanding them on the commit log mailing list easier. This also helps
289 in case of debugging later on.
290 @item
291 Patches that change behavior of the programs (renaming options etc) or
292 public API or ABI should be discussed in depth and possible few days should
293 pass between discussion and commit.
294 Changes to the build system (Makefiles, configure script) which alter
295 the expected behavior should be considered in the same regard.
296 @item
297 When applying patches that have been discussed (at length) on the mailing
298 list, reference the thread in the log message.
299 @item
300 Subscribe to the
301 @uref{https://lists.libav.org/mailman/listinfo/libav-devel, libav-devel} and
302 @uref{https://lists.libav.org/mailman/listinfo/libav-commits, libav-commits}
303 mailing lists.
304 Bugs and possible improvements or general questions regarding commits
305 are discussed on libav-devel. We expect you to react if problems with
306 your code are uncovered.
307 @item
308 Update the documentation if you change behavior or add features. If you are
309 unsure how best to do this, send an [RFC] patch to libav-devel.
310 @item
311 All discussions and decisions should be reported on the public developer
312 mailing list, so that there is a reference to them.
313 Other media (e.g. IRC) should be used for coordination and immediate
314 collaboration.
315 @item
316 Never write to unallocated memory, never write over the end of arrays,
317 always check values read from some untrusted source before using them
318 as array index or other risky things. Always use valgrind to double-check.
319 @item
320 Remember to check if you need to bump versions for the specific libav
321 parts (libavutil, libavcodec, libavformat) you are changing. You need
322 to change the version integer.
323 Incrementing the first component means no backward compatibility to
324 previous versions (e.g. removal of a function from the public API).
325 Incrementing the second component means backward compatible change
326 (e.g. addition of a function to the public API or extension of an
327 existing data structure).
328 Incrementing the third component means a noteworthy binary compatible
329 change (e.g. encoder bug fix that matters for the decoder).
330 @item
331 Compiler warnings indicate potential bugs or code with bad style.
332 If it is a bug, the bug has to be fixed. If it is not, the code should
333 be changed to not generate a warning unless that causes a slowdown
334 or obfuscates the code.
335 If a type of warning leads to too many false positives, that warning
336 should be disabled, not the code changed.
337 @item
338 If you add a new file, give it a proper license header. Do not copy and
339 paste it from a random place, use an existing file as template.
340 @end enumerate
341
342 We think our rules are not too hard. If you have comments, contact us.
343
344 Note, some rules were borrowed from the MPlayer project.
345
346 @section Submitting patches
347
348 First, read the @ref{Coding Rules} above if you did not yet, in particular
349 the rules regarding patch submission.
350
351 As stated already, please do not submit a patch which contains several
352 unrelated changes.
353 Split it into separate, self-contained pieces. This does not mean splitting
354 file by file. Instead, make the patch as small as possible while still
355 keeping it as a logical unit that contains an individual change, even
356 if it spans multiple files. This makes reviewing your patches much easier
357 for us and greatly increases your chances of getting your patch applied.
358
359 Use the patcheck tool of Libav to check your patch.
360 The tool is located in the tools directory.
361
362 Run the @ref{Regression Tests} before submitting a patch in order to verify
363 it does not cause unexpected problems.
364
365 It also helps quite a bit if you tell us what the patch does (for example
366 'replaces lrint by lrintf'), and why (for example '*BSD isn't C99 compliant
367 and has no lrint()'). This kind of explanation should be the body of the
368 commit message.
369
370 Also please if you send several patches, send each patch as a separate mail,
371 do not attach several unrelated patches to the same mail.
372
373 Patches should be posted to the
374 @uref{https://lists.libav.org/mailman/listinfo/libav-devel, libav-devel}
375 mailing list. Use @code{git send-email} when possible since it will properly
376 send patches without requiring extra care. If you cannot, then send patches
377 as base64-encoded attachments, so your patch is not trashed during
378 transmission.
379
380 Your patch will be reviewed on the mailing list. You will likely be asked
381 to make some changes and are expected to send in an improved version that
382 incorporates the requests from the review. This process may go through
383 several iterations. Once your patch is deemed good enough, it will be
384 committed to the official Libav tree.
385
386 Give us a few days to react. But if some time passes without reaction,
387 send a reminder by email. Your patch should eventually be dealt with.
388
389
390 @section New codecs or formats checklist
391
392 @enumerate
393 @item
394 Did you use av_cold for codec initialization and close functions?
395 @item
396 Did you add a long_name under NULL_IF_CONFIG_SMALL to the AVCodec or
397 AVInputFormat/AVOutputFormat struct?
398 @item
399 Did you bump the minor version number (and reset the micro version
400 number) in @file{libavcodec/version.h} or @file{libavformat/version.h}?
401 @item
402 Did you register it in @file{allcodecs.c} or @file{allformats.c}?
403 @item
404 Did you add the AVCodecID to @file{avcodec.h}?
405 When adding new codec IDs, also add an entry to the codec descriptor
406 list in @file{libavcodec/codec_desc.c}.
407 @item
408 If it has a FourCC, did you add it to @file{libavformat/riff.c},
409 even if it is only a decoder?
410 @item
411 Did you add a rule to compile the appropriate files in the Makefile?
412 Remember to do this even if you are just adding a format to a file that
413 is already being compiled by some other rule, like a raw demuxer.
414 @item
415 Did you add an entry to the table of supported formats or codecs in
416 @file{doc/general.texi}?
417 @item
418 Did you add an entry in the Changelog?
419 @item
420 If it depends on a parser or a library, did you add that dependency in
421 configure?
422 @item
423 Did you @code{git add} the appropriate files before committing?
424 @item
425 Did you make sure it compiles standalone, i.e. with
426 @code{configure --disable-everything --enable-decoder=foo}
427 (or @code{--enable-demuxer} or whatever your component is)?
428 @end enumerate
429
430
431 @section patch submission checklist
432
433 @enumerate
434 @item
435 Does @code{make check} pass with the patch applied?
436 @item
437 Is the patch against latest Libav git master branch?
438 @item
439 Are you subscribed to the
440 @uref{https://lists.libav.org/mailman/listinfo/libav-devel, libav-devel}
441 mailing list? (Only list subscribers are allowed to post.)
442 @item
443 Have you checked that the changes are minimal, so that the same cannot be
444 achieved with a smaller patch and/or simpler final code?
445 @item
446 If the change is to speed critical code, did you benchmark it?
447 @item
448 If you did any benchmarks, did you provide them in the mail?
449 @item
450 Have you checked that the patch does not introduce buffer overflows or
451 other security issues?
452 @item
453 Did you test your decoder or demuxer against damaged data? If no, see
454 tools/trasher, the noise bitstream filter, and
455 @uref{http://caca.zoy.org/wiki/zzuf, zzuf}. Your decoder or demuxer
456 should not crash, end in a (near) infinite loop, or allocate ridiculous
457 amounts of memory when fed damaged data.
458 @item
459 Does the patch not mix functional and cosmetic changes?
460 @item
461 Did you add tabs or trailing whitespace to the code? Both are forbidden.
462 @item
463 Is the patch attached to the email you send?
464 @item
465 Is the mime type of the patch correct? It should be text/x-diff or
466 text/x-patch or at least text/plain and not application/octet-stream.
467 @item
468 If the patch fixes a bug, did you provide a verbose analysis of the bug?
469 @item
470 If the patch fixes a bug, did you provide enough information, including
471 a sample, so the bug can be reproduced and the fix can be verified?
472 Note please do not attach samples >100k to mails but rather provide a
473 URL, you can upload to ftp://upload.libav.org
474 @item
475 Did you provide a verbose summary about what the patch does change?
476 @item
477 Did you provide a verbose explanation why it changes things like it does?
478 @item
479 Did you provide a verbose summary of the user visible advantages and
480 disadvantages if the patch is applied?
481 @item
482 Did you provide an example so we can verify the new feature added by the
483 patch easily?
484 @item
485 If you added a new file, did you insert a license header? It should be
486 taken from Libav, not randomly copied and pasted from somewhere else.
487 @item
488 You should maintain alphabetical order in alphabetically ordered lists as
489 long as doing so does not break API/ABI compatibility.
490 @item
491 Lines with similar content should be aligned vertically when doing so
492 improves readability.
493 @item
494 Make sure you check the return values of function and return appropriate
495 error codes. Especially memory allocation functions like @code{malloc()}
496 are notoriously left unchecked, which is a serious problem.
497 @end enumerate
498
499 @section Patch review process
500
501 All patches posted to the
502 @uref{https://lists.libav.org/mailman/listinfo/libav-devel, libav-devel}
503 mailing list will be reviewed, unless they contain a
504 clear note that the patch is not for the git master branch.
505 Reviews and comments will be posted as replies to the patch on the
506 mailing list. The patch submitter then has to take care of every comment,
507 that can be by resubmitting a changed patch or by discussion. Resubmitted
508 patches will themselves be reviewed like any other patch. If at some point
509 a patch passes review with no comments then it is approved, that can for
510 simple and small patches happen immediately while large patches will generally
511 have to be changed and reviewed many times before they are approved.
512 After a patch is approved it will be committed to the repository.
513
514 We will review all submitted patches, but sometimes we are quite busy so
515 especially for large patches this can take several weeks.
516
517 When resubmitting patches, if their size grew or during the review different
518 issues arisen please split the patch so each issue has a specific patch.
519
520 @anchor{Regression Tests}
521 @section Regression Tests
522
523 Before submitting a patch (or committing to the repository), you should at
524 least make sure that it does not break anything.
525
526 If the code changed has already a test present in FATE you should run it,
527 otherwise it is advised to add it.
528
529 Improvements to codec or demuxer might change the FATE results. Make sure
530 to commit the update reference with the change and to explain in the comment
531 why the expected result changed.
532
533 Please refer to @url{fate.html}.
534
535 @bye