August 2013 - ffmpeg-devel - lists.ffmpeg.org

handle carriage returns in cleanws()
by 俞晓磊 18 Oct '13

18 Oct '13

When configuring ffmpeg in a cygwin shell for a non-cygwin toolchain, the generated cc_ident contains a carriage return, which is propagated into config.h. The cause seems to be that cygwin treats carriage return as a normal character. I can workaround this by adding 's/\r//' to cleanws' sed command in a patching step. I don't know whether cygwin + non-cygwin toolchain is supported by ffmpeg, but it would be nice if cleanws removes '\r' by default.

2 1

[PATCH] Store the real file location of a specified URL
by Sidney N'Guessan 18 Oct '13

18 Oct '13

Hi, To follow : http://ffmpeg.org/pipermail/ffmpeg-devel/2013-July/145549.html Patch updated in attachment. Fix for me : http://ffmpeg.org/trac/ffmpeg/ticket/2710

2 1

[PATCH]Add little-endian G726 decoder and use it for Sun AU files
by Carl Eugen Hoyos 18 Oct '13

18 Oct '13

Hi! Attached patch fixes ticket #1955 as suggested by Roman. Please comment, Carl Eugen

3 28

[RFC] doc: rewrite developer documentation
by Thilo Borgmann 17 Oct '13

17 Oct '13

Hi, I started with including some text about pull requests... I ended with a rewrite of almost the whole page, removing (IMHO) outdated info, removing redundant info, rewording many things. Most important for me is that this rewrite reorders things that seemed to have become too confusing in the old page. This is an RFC and not yet completely finished. If people tend to like the new design better than the old page, I will try to finish it and round it up for review. Please give me feedback for the overall redesign - if any detail information is missing in your eyes, this can be added later. -Thilo The following changes since commit faf7c356554d54be0ddb5f989791407e23753549: lavf/tee: add support for bitstream filtering (2013-08-16 17:58:58 +0200) are available in the git repository at: git@github.com:thiloborgmann/FFmpeg.git doc_developer for you to fetch changes up to 6f93ecdbf119640b2b0be34fd2a576c92c34d287: doc: rewrite developer documentation (2013-08-17 02:42:00 +0200) ---------------------------------------------------------------- Thilo Borgmann (1): doc: rewrite developer documentation doc/developer.texi | 1000 ++++++++++++++++++++++++++-------------------------- 1 file changed, 493 insertions(+), 507 deletions(-) diff --git a/doc/developer.texi b/doc/developer.texi index b9951a4..8ac8e66 100644 --- a/doc/developer.texi +++ b/doc/developer.texi @@ -9,76 +9,153 @@ @contents -@chapter Developers Guide +@chapter Preface -@section Notes for external developers +This document is most useful for developers working on the FFmpeg libraries. +Developers using the FFmpeg libraries in their application should +refer to the doxygen @url{http://ffmpeg.org/doxygen/trunk/index.html, API documentation} and +check the examples in @file{doc/examples} and in the source code of +the provided FFmpeg tools to see how the public API is employed. -This document is mostly useful for internal FFmpeg developers. -External developers who need to use the API in their application should -refer to the API doxygen documentation in the public headers, and -check the examples in @file{doc/examples} and in the source code to -see how the public API is employed. - -You can use the FFmpeg libraries in your commercial program, but you -are encouraged to @emph{publish any patch you make}. In this case the -best way to proceed is to send your patches to the ffmpeg-devel -mailing list following the guidelines illustrated in the remainder of -this document. +You can also use the FFmpeg libraries in your commercial program, but +you have to conform to the license of the libraries. +You are also encouraged to @emph{publish any patch you make}. +In this case the best way to proceed is to send your patches to the +@uref{http://lists.ffmpeg.org/mailman/listinfo/ffmpeg-devel, ffmpeg-devel mailing list} +following the guidelines illustrated in the remainder of this document. For more detailed legal information about the use of FFmpeg in external programs read the @file{LICENSE} file in the source tree and -consult @url{http://ffmpeg.org/legal.html}. +consult @url{http://ffmpeg.org/legal.html, FFmpeg License and Legal Considerations}. -@section Contributing -There are 3 ways by which code gets into ffmpeg. -@itemize @bullet -@item Submitting Patches to the main developer mailing list - see @ref{Submitting patches} for details. -@item Directly committing changes to the main tree. -@item Committing changes to a git clone, for example on github.com or - gitorious.org. And asking us to merge these changes. -@end itemize +@chapter Contributing to FFmpeg + +To contribute your work to FFmpeg you usually have to: + +@enumerate +@item @ref{implementation, Implement your contribution according to the coding rules.} +@item @ref{regression, Test for possible regressions and memory leaks and ensure robustness of your modifications.} +@item @ref{submission, Send a submission to the FFmpeg development mailing list.} +@item @ref{review, Iteratively update your submission according to the objections raised by reviewing FFmpeg developers.} +@item @ref{maintainership, Find your submission finally merged into the FFmpeg repository and claim maintainership for new files you created.} +@end enumerate + +The following sections correspond to the list above and form +a detailed guideline to help all contributors to comply to our +rules of development. +Following this guideline as close as possible is in the best interest +of every contributor, developer, maintainer and user of FFmpeg. -Whichever way, changes should be reviewed by the maintainer of the code -before they are committed. And they should follow the @ref{Coding Rules}. -The developer making the commit and the author are responsible for their changes -and should try to fix issues their commit causes. -@anchor{Coding Rules} +@anchor{implementation} @section Coding Rules -@subsection Code formatting conventions +The main priority in FFmpeg is simplicity and small code size in order to +minimize the bug count. +The coding rules help with readability and maintainability of the code base +across heterogeneous systems, editors and developers. + +@subsection Programming Language and Features -There are the following guidelines regarding the indentation in files: +FFmpeg is written in the ISO C90 programming language. +Additionally, the following features from ISO C99 are utilized: @itemize @bullet -@item -Indent size is 4. +@item The @samp{inline} keyword +@item Single-line comments (@samp{//}) +@item Designated struct initializers (@samp{struct s x = @{ .i = 17 @};}) +@item Compound literals (@samp{x = (struct s) @{ 17, 23 @};}) +@end itemize -@item -The TAB character is forbidden outside of Makefiles as is any -form of trailing whitespace. Commits containing either will be -rejected by the git repository. +These features are supported by all compilers we care about, thus we will not +accept patches to remove their use unless they absolutely do not impair +clarity and performance. -@item -You should try to limit your code lines to 80 characters; however, do so if -and only if this improves readability. +All code must compile with recent versions of GCC and a number of other +currently supported compilers. +To ensure compatibility, please do not use additional C99 features or +GCC extensions. +Especially watch out for: + +@itemize @bullet +@item Mixing statements and declarations +@item Avoid @samp{long long}, use @samp{int64_t} instead +@item @samp{__attribute__} not protected by @samp{#ifdef __GNUC__} or similar +@item GCC statement expressions (@samp{(x = (@{ int y = 4; y; @})}) @end itemize -The presentation is one inspired by 'indent -i4 -kr -nut'. -The main priority in FFmpeg is simplicity and small code size in order to -minimize the bug count. +@anchor{licenses} +@subsection Licenses + +Contributions are to be licensed under one of the following licenses in order of preference: +@enumerate +@item @uref{http://www.gnu.org/licenses/lgpl-2.1.html, LGPL 2.1}, + including an "or any later version" clause +@item @uref{http://www.gnu.org/licenses/gpl-2.0.html, GPL 2} including + an "or any later version" clause +@end enumerate + +For creating a new file, the appropriate license header should be copied +from another file of FFmpeg and not randomly copied and pasted from +somewhere else. + +@subsection Code formatting + +@itemize @bullet +@item Indent size is 4 characters. +@item The TAB character is forbidden outside of makefiles. +@item Trailing whitespaces are forbidden +@item You should try to limit your code lines to 80 characters. + However, do so if and only if this improves readability. +@end itemize + +In order to configure Vim to follow our code formatting conventions, paste +the following snippet into your @file{.vimrc}: +@example +" indentation rules for FFmpeg: 4 spaces, no tabs +set expandtab +set shiftwidth=4 +set softtabstop=4 +set cindent +set cinoptions=(0 +" Allow tabs in Makefiles. +autocmd FileType make,automake set noexpandtab shiftwidth=8 softtabstop=8 +" Trailing whitespace and tabs are forbidden, so highlight them. +highlight ForbiddenWhitespace ctermbg=red guibg=red +match ForbiddenWhitespace /\s\+$\|\t/ +" Do not highlight spaces at the end of line while typing on that line. +autocmd InsertEnter * match ForbiddenWhitespace /\t\|\s\+\%#\@@<!$/ +@end example + +For Emacs, add these roughly equivalent lines to your @file{.emacs.d/init.el}: +@example +(c-add-style "ffmpeg" + '("k&r" + (c-basic-offset . 4) + (indent-tabs-mode . nil) + (show-trailing-whitespace . t) + (c-offsets-alist + (statement-cont . (c-lineup-assignments +))) + ) + ) +(setq c-default-style "ffmpeg") +@end example -@subsection 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. -Avoid Qt-style and similar Doxygen syntax with @code{!} in it, i.e. replace -@code{//!} with @code{///} and similar. Also @@ syntax should be employed -for markup commands, i.e. use @code{@@param} and not @code{\param}. +@subsection Documentation + +@itemize @bullet +@item The API documentation is generated automatically using Doxygen format. +@item All functions of the public API should therefore feature a + Doxygen formatted comment covering all parameters and return values. +@item All nontrivial functions should have a comment above them explaining + what the function does, even if it is just one sentence. +@item All structures and their member variables should be documented. +@item Avoid Qt-style and similar Doxygen syntax with @code{!} in it, i.e. replace + @code{//!} with @code{///} and similar. Also @@ syntax should be employed + for markup commands, i.e. use @code{@@param} and not @code{\param}. +@end itemize @example /** @@ -92,12 +169,12 @@ for markup commands, i.e. use @code{@@param} and not @code{\param}. * more text ... * ... */ -typedef struct Foobar@{ +typedef struct FooBar@{ int var1; /**< var1 description */ int var2; ///< var2 description /** var3 description */ int var3; -@} Foobar; +@} FooBar; /** * Summary sentence. @@ -110,82 +187,57 @@ int myfunc(int my_parameter) ... @end example -@subsection C language features - -FFmpeg is programmed in the ISO C90 language with a few additional -features from ISO C99, namely: +@subsection Naming Conventions @itemize @bullet -@item -the @samp{inline} keyword; - -@item -@samp{//} comments; - -@item -designated struct initializers (@samp{struct s x = @{ .i = 17 @};}) - -@item -compound literals (@samp{x = (struct s) @{ 17, 23 @};}) +@item All type names, like structs and enums, should always be written + in CamelCase. +@item All other names should always be composed using underscores (_). @end itemize -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 recent versions of GCC and a number of other -currently supported compilers. To ensure compatibility, please do not use -additional C99 features or GCC extensions. Especially watch out for: - -@itemize @bullet -@item -mixing statements and declarations; +@example +typedef struct FooBar@{ + int video_var1; + int audio_var2; + int subtitle_var3; +@} FooBar; -@item -@samp{long long} (use @samp{int64_t} instead); +int avfilter_get_video_buffer(AVFilterContext *filter_ctx, AVBuffer *out_buf); +@end example -@item -@samp{__attribute__} not protected by @samp{#ifdef __GNUC__} or similar; +@subsection Prefix Conventions -@item -GCC statement expressions (@samp{(x = (@{ int y = 4; y; @})}). +@itemize @bullet +@item Prefix: @code{<none>} +@itemize @bullet + @item Local variables + @item File-scope variables + @item Static functions @end itemize - -@subsection Naming conventions -All names should be composed with underscores (_), not CamelCase. For example, -@samp{avfilter_get_video_buffer} is an acceptable function name and -@samp{AVFilterGetVideo} is not. The exception from this are type names, like -for example structs and enums; they should always be in the CamelCase - -There are the following conventions for naming variables and functions: - +@item Prefix: @code{ff_} @itemize @bullet -@item -For local variables no prefix is required. - -@item -For file-scope variables and functions declared as @code{static}, no prefix -is required. - -@item -For variables and functions visible outside of file scope, but only used -internally by a library, an @code{ff_} prefix should be used, -e.g. @samp{ff_w64_demuxer}. - -@item -For variables and functions visible outside of file scope, used internally -across multiple libraries, use @code{avpriv_} as prefix, for example, -@samp{avpriv_aac_parse_header}. + @item Variables and functions visible outside of the file scope + which are used internally in the respective library, only. +@end itemize +@item Prefix: @code{avpriv_} +@itemize @bullet + @item Variables and functions visible outside of the file scope + which are used internally in multiple libraries. +@end itemize +@item Prefix: @code{av_}, @code{avformat_}, @code{avcodec_}, @code{swr_}, ... +@itemize @bullet +@item Public symbols. + In addition to the prefix @code{av_}, each library + also has its own prefix for public symbols, corresponding to the + name of the library. For example @code{avformat_} for libavformat, + @code{avcodec_} for libavcodec, @code{swr_} for libswresample, etc. + Check the existing code and choose names accordingly. +@end itemize +@end itemize -@item -Each library has its own prefix for public symbols, in addition to the -commonly used @code{av_} (@code{avformat_} for libavformat, -@code{avcodec_} for libavcodec, @code{swr_} for libswresample, etc). -Check the existing code and choose names accordingly. Note that some symbols without these prefixes are also exported for -retro-compatibility reasons. These exceptions are declared in the -@code{lib<name>/lib<name>.v} files. -@end itemize +retro-compatibility reasons. +These exceptions are declared in the @code{lib<name>/lib<name>.v} files. Furthermore, name space reserved for the system should not be invaded. Identifiers ending in @code{_t} are reserved by @@ -198,298 +250,316 @@ symbols. If in doubt, just avoid names starting with @code{_} altogether. @subsection Miscellaneous conventions @itemize @bullet -@item -fprintf and printf are forbidden in libavformat and libavcodec, -please use av_log() instead. +@item Logging +@itemize @bullet + @item Since @code{fprintf} and @code{printf} are forbidden in + libavformat and libavcodec, use @code{av_log} instead. +@end itemize +@item Casting +@itemize @bullet + @item Casts should be avoided. Use only if necessary. +@end itemize +@item Parentheses +@itemize @bullet + @item Unneeded parentheses should be avoided if they don't + make the code easier to understand. +@end itemize +@end itemize -@item -Casts should be used only when necessary. Unneeded parentheses -should also be avoided if they don't make the code easier to understand. + +@anchor{regression} +@section FATE Regression tests and hardening + +@itemize +@item Run FATE to test for regressions introduced by your modifications by + calling @code{make fate}. +@item Optionally, e.g. if your modifications allocate memory, configure FATE + to utilize valgrind for spotting bugs related to memory handling. +@item Optionally, e.g. if your modifications are related to any kind of input + bitstream handling, utilize a fuzzer program for hardening your code. @end itemize -@subsection Editor configuration -In order to configure Vim to follow FFmpeg formatting conventions, paste -the following snippet into your @file{.vimrc}: -@example -" indentation rules for FFmpeg: 4 spaces, no tabs -set expandtab -set shiftwidth=4 -set softtabstop=4 -set cindent -set cinoptions=(0 -" Allow tabs in Makefiles. -autocmd FileType make,automake set noexpandtab shiftwidth=8 softtabstop=8 -" Trailing whitespace and tabs are forbidden, so highlight them. -highlight ForbiddenWhitespace ctermbg=red guibg=red -match ForbiddenWhitespace /\s\+$\|\t/ -" Do not highlight spaces at the end of line while typing on that line. -autocmd InsertEnter * match ForbiddenWhitespace /\t\|\s\+\%#\@@<!$/ -@end example +For details about how to set up and run FATE, please refer to @url{fate.html}. -For Emacs, add these roughly equivalent lines to your @file{.emacs.d/init.el}: -@example -(c-add-style "ffmpeg" - '("k&r" - (c-basic-offset . 4) - (indent-tabs-mode . nil) - (show-trailing-whitespace . t) - (c-offsets-alist - (statement-cont . (c-lineup-assignments +))) - ) - ) -(setq c-default-style "ffmpeg") -@end example +Of course, some modifications may change the results of the regression tests. +In this case, the reference results of the regression tests shall be modified +accordingly. -@section Development Policy +Also, some modifications may require a new test sample to be included into the +FATE-suite. +This is usually the case if there is no muxer or encoder available to generate +an appropriate sample. +For that purpose, please create a sample that is as small as possible +while still testing all your modifications sufficiently. -@enumerate -@item -Contributions should be licensed under the -@uref{http://www.gnu.org/licenses/lgpl-2.1.html, LGPL 2.1}, -including an "or any later version" clause, or, if you prefer -a gift-style license, the -@uref{http://www.isc.org/software/license/, ISC} or -@uref{http://mit-license.org/, MIT} license. -@uref{http://www.gnu.org/licenses/gpl-2.0.html, GPL 2} including -an "or any later version" clause is also acceptable, but LGPL is -preferred. +Once you have created a working fate test and fate sample, please include a +direct download link to your sample in your submission so that it can easily +be included. -@item -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. +For hardening your code, you might utilize @code{tools/trasher}, the noise +bitstream filter, or other fuzzer programs like +@uref{http://caca.zoy.org/wiki/zzuf, zzuf}. +Your code should not crash, end in a (near) infinite loop, or allocate ridiculous +amounts of memory when fed with damaged data. -@item -The commit message should have a short first line in the form of -a @samp{topic: short description} as a header, separated by a newline -from the body consisting of an explanation of why the change is necessary. -If the commit fixes a known bug on the bug tracker, the commit message -should include its bug ID. Referring to the issue on the bug tracker does -not exempt you from writing an excerpt of the bug in the commit message. -@item -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. -@item -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. +@anchor{submission} +@section Submitting your contribution -@item -Do not change behavior of the programs (renaming options etc) or public -API or ABI without first discussing it on the ffmpeg-devel mailing list. -Do not remove functionality from the code. Just improve! +@itemize @bullet +@item Clone a local or remote git repository. +@item Commit your modifications in seperate, logically self-contained parts + and compose descriptive commit messages. +@item Generate patches or a pull-request based on your commits. +@item Verify the contents of your patches or pull-request. +@item Send your patches or pull-request to the + @uref{http://lists.ffmpeg.org/mailman/listinfo/ffmpeg-devel, ffmpeg-devel} + mailing list. +@end itemize -Note: Redundant code can be removed. +Cloning a local or remote git repository to generate patches +or pull-requests using the git tools helps us the most to review and +merge your work. +Any other form is most likely to be rejected. + +Committing your modifications in seperate, logically self-contained parts +does not mean splitting file by file. +Instead, make each commit 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 modifications much easier for us and greatly increases +the chances of getting your work merged. + +Composing descriptive commit messages includes to point out as clearly as possible +what this commit does (for example 'replaces lrint by lrintf'), and why it does +so (for example '*BSD isn't C99 compliant and has no lrint()'). +Long commit messages should include a header in the form of +a @samp{topic: short description}, separated by a blank line +from the body consisting of the detailed description. +If a known bug is fixed, the commit message should also include its bug ID as +defined in the @url{https://trac.ffmpeg.org, bug tracker}. + +The following real-world commit message serves as a good example: +@example +configure: Only build ffserver if SA_RESAMPLE is available. -@item -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. +QNX does not define SA_RESAMPLE (because it is not implemented). +Fixes ticket #2780. +@end example -@item -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. +Once your modifications are commited into your local or remote repository, you +can generate patches or a pull-request based on your commits. +In case you want to create a pull-request, ensure that you have also pushed your +commits into your remote repository.@* +In order to create patches, please use: @code{git format-patch}@* +In order to create a pull-request, please use: @code{git request-pull -p} -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 +For verifying your submission, you can use @code{tools/patcheck} to check your +patches, if applicable. +At least, you should visually inspect any submission before actually sending it. -@item -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. -Recommended format: -area changed: Short 1 line description +In order to send patches, please use: @code{git send-email} or @code{git imap-send}@* +In order to send a pull-request or cannot use the commands above, send +your submissions as base64-encoded attachments. This ensures that your +submission is not trashed during transmission. -details describing what and why and giving references. +If you are about to send several submissions, ensure to send each one as a +separate mail. Do not attach several unrelated submissions to the same mail. -@item -Make sure the author of the commit is set correctly. (see git commit --author) -If you apply a patch, send an -answer to ffmpeg-devel (or wherever you got the patch from) saying that -you applied the patch. +We appreciate the use of the following prefixes for better orientation on the +mailing list.@* +To indicate that your submission shall be merged, please use: @code{[PATCH]}@* +To indicate that your submission shall be the base of a topic +related discussion (a Request For Comments), please use: @code{[RFC]}. -@item -When applying patches that have been discussed (at length) on the mailing -list, reference the thread in the log message. -@item -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! +@anchor{review} +@section Reviewing process -@item -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. +Every submission will be reviewed on the +@uref{http://lists.ffmpeg.org/mailman/listinfo/ffmpeg-devel, ffmpeg-devel} +mailing list, unless they contain a +clear note that the patch is not for the git master branch. +You will likely be asked to make some changes and are expected to send in an +improved version that incorporates the requests from the review. +This process may go through several iterations. +Once your patch is deemed good enough, some developer will pick it up and +commit it to the official FFmpeg repository. -@item -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. +Give us a few days to react upon your submission. +But if some time has passed without reaction, feel free to send a reminder. -@item -Try to keep important discussions and requests (also) on the public -developer mailing list, so that all developers can benefit from them. +Sometimes we are quite busy so especially for large patches this reviewing +process can take several weeks. -@item -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. +When resubmitting, please do not make any significant changes not related to +the comments received during review. +Such submission will likely be rejected because nobody can follow for review. +Instead, submit significant changes or new features as separate patches. -@item -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). The third -component always starts at 100 to distinguish FFmpeg from Libav. -@item -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. +@anchor{maintainership} +@section Maintainership -@item -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. -@end enumerate +The developer who pushed your submission as well as the author are +responsible for their changes and should try to fix any issues their commit +might cause. -We think our rules are not too hard. If you have comments, contact us. +In the case your contributions are the origin of new files, you are +encouraged to claim maintainership of your files. -@anchor{Submitting patches} -@section Submitting patches +A maintainer's is responsibility includes fixing bugs and regressions, reviewing +corresponding submissions, etc. in order to keep the maintained parts of FFmpeg +as functional and bugfree as possible. +Also, implementing new features, optimizing for speed or simplicity and all +other improvements are strongly encouraged. -First, read the @ref{Coding Rules} above if you did not yet, in particular -the rules regarding patch submission. +Usually, maintainership demands to be subscribed to our mailing lists and +create an account for our @url{https://trac.ffmpeg.org, bug tracker}. -When you submit your patch, please use @code{git format-patch} or -@code{git send-email}. We cannot read other diffs :-) +Maintainership also increases the possibility to be granted write access to our +repository. +However, write access is not mendatory for maintainership. -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. +@c MOVE TO FATE.HTML +@c @subsection Visualizing Test Coverage -Use the patcheck tool of FFmpeg to check your patch. -The tool is located in the tools directory. +@c The FFmpeg build system allows visualizing the test coverage in an easy +@c manner with the coverage tools @code{gcov}/@code{lcov}. This involves +@c the following steps: -Run the @ref{Regression tests} before submitting a patch in order to verify -it does not cause unexpected problems. +@c @enumerate +@c @item +@c Configure to compile with instrumentation enabled: +@c @code{configure --toolchain=gcov}. -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()') +@c @item +@c Run your test case, either manually or via FATE. This can be either +@c the full FATE regression suite, or any arbitrary invocation of any +@c front-end tool provided by FFmpeg, in any combination. -Also please if you send several patches, send each patch as a separate mail, -do not attach several unrelated patches to the same mail. +@c @item +@c Run @code{make lcov} to generate coverage data in HTML format. -Patches should be posted to the -@uref{http://lists.ffmpeg.org/mailman/listinfo/ffmpeg-devel, ffmpeg-devel} -mailing list. Use @code{git send-email} when possible since it will properly -send patches without requiring extra care. If you cannot, then send patches -as base64-encoded attachments, so your patch is not trashed during -transmission. +@c @item +@c View @code{lcov/index.html} in your preferred HTML viewer. +@c @end enumerate -Your patch will be reviewed on the mailing list. You will likely be asked -to make some changes and are expected to send in an improved version that -incorporates the requests from the review. This process may go through -several iterations. Once your patch is deemed good enough, some developer -will pick it up and commit it to the official FFmpeg tree. +@c You can use the command @code{make lcov-reset} to reset the coverage +@c measurements. You will need to rerun @code{make lcov} after running a +@c new test. -Give us a few days to react. But if some time passes without reaction, -send a reminder by email. Your patch should eventually be dealt with. +@c @subsection Using Valgrind +@c The configure script provides a shortcut for using valgrind to spot bugs +@c related to memory handling. Just add the option +@c @code{--toolchain=valgrind-memcheck} or @code{--toolchain=valgrind-massif} +@c to your configure line, and reasonable defaults will be set for running +@c FATE under the supervision of either the @strong{memcheck} or the +@c @strong{massif} tool of the valgrind suite. -@section New codecs or formats checklist +@c In case you need finer control over how valgrind is invoked, use the +@c @code{--target-exec='valgrind <your_custom_valgrind_options>} option in +@c your configure line instead. + + +@anchor{Release process} +@chapter FFmpeg releases + +FFmpeg maintains a set of release branches, which are the +recommended deliverable for system integrators and distributors (such as +Linux distributions, etc.). At regular times, a release +manager prepares, tests and publishes tarballs on the +@url{http://ffmpeg.org} website. + +There are two kinds of releases: @enumerate -@item -Did you use av_cold for codec initialization and close functions? +@item Major releases@* + Always including the latest and greatest features and functionality. +@item Point releases@* + These are cut from release branches, which are named @code{release/X}, + with @code{X} being the release version number. +@end enumerate -@item -Did you add a long_name under NULL_IF_CONFIG_SMALL to the AVCodec or -AVInputFormat/AVOutputFormat struct? +Note that we promise to our users that shared libraries from any FFmpeg +release never break programs that have been compiled against +previous versions of the same release series in any case! -@item -Did you bump the minor version number (and reset the micro version -number) in @file{libavcodec/version.h} or @file{libavformat/version.h}? +However, from time to time, we do make API changes that require adaptations +in applications. Such changes are only allowed in (new) major releases and +require further steps such as bumping library version numbers and/or +adjustments to the symbol versioning file. Please discuss such changes +on the ffmpeg-devel mailing list in time to allow forward planning. -@item -Did you register it in @file{allcodecs.c} or @file{allformats.c}? +@anchor{Criteria for Point Releases} +@section Criteria for Point Releases -@item -Did you add the AVCodecID to @file{avcodec.h}? -When adding new codec IDs, also add an entry to the codec descriptor -list in @file{libavcodec/codec_desc.c}. +Changes that match the following criteria are valid candidates for +inclusion into a point release: -@item -If it has a FourCC, did you add it to @file{libavformat/riff.c}, -even if it is only a decoder? +@enumerate +@item Fixes a security issue, preferably identified by a CVE + number issued by @url{http://cve.mitre.org/}. -@item -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. +@item Fixes a documented bug in @url{https://trac.ffmpeg.org}. -@item -Did you add an entry to the table of supported formats or codecs in -(a)file{doc/general.texi}? +@item Improves the included documentation. -@item -Did you add an entry in the Changelog? +@item Retains both source code and binary compatibility with previous + point releases of the same release branch. +@end enumerate -@item -If it depends on a parser or a library, did you add that dependency in -configure? +The order for checking the rules is: @code{(1 OR 2 OR 3) AND 4} -@item -Did you @code{git add} the appropriate files before committing? -@item -Did you make sure it compiles standalone, i.e. with -@code{configure --disable-everything --enable-decoder=foo} -(or @code{--enable-demuxer} or whatever your component is)? +@chapter Checklists + +@section General checklist + +@enumerate +@item We reject submissions of mixed cosmetical and functional changes.@* + 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, separate them strictly from functional + 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 +@item 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. +@item Try to keep important discussions and requests (also) on the public + developer mailing list, so that all developers can benefit from them. +@item Never write to unallocated memory +@item Never write over the end of arrays +@item Always check values read from some untrusted source before using them + as array index or other risky things. +@item Remember to check if you need to bump versions.@* + For each specific part of libav* (libavutil, libavcodec, libavformat) + you are changing, you need to update 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). The third + component always starts at 100 to distinguish FFmpeg from Libav. +@item 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. @end enumerate -@section patch submission checklist +@anchor{General submission checklist} +@section General submission checklist @enumerate @item @@ -598,195 +668,111 @@ Test your code with valgrind and or Address Sanitizer to ensure it's free of leaks, out of array accesses, etc. @end enumerate -@section Patch review process - -All patches posted to ffmpeg-devel will be reviewed, unless they contain a -clear note that the patch is not for the git master branch. -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. - -If you feel that the review process is too slow and you are willing to try to -take over maintainership of the area of code you change then just clone -git master and maintain the area of code there. We will merge each area from -where its best maintained. - -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. - -@anchor{Regression tests} -@section Regression tests - -Before submitting a patch (or committing to the repository), you should at least -test that you did not break anything. - -Running 'make fate' accomplishes this, please see @url{fate.html} for details. -[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]. - -@subsection Adding files to the fate-suite dataset - -When there is no muxer or encoder available to generate test media for a -specific test then the media has to be inlcuded in the fate-suite. -First please make sure that the sample file is as small as possible to test the -respective decoder or demuxer sufficiently. Large files increase network -bandwidth and disk space requirements. -Once you have a working fate test and fate sample, provide in the commit -message or introductionary message for the patch series that you post to -the ffmpeg-devel mailing list, a direct link to download the sample media. - - -@subsection Visualizing Test Coverage - -The FFmpeg build system allows visualizing the test coverage in an easy -manner with the coverage tools @code{gcov}/@code{lcov}. This involves -the following steps: +@anchor{New codecs or formats checklist} +@section New codecs or formats checklist @enumerate @item - Configure to compile with instrumentation enabled: - @code{configure --toolchain=gcov}. +Did you use av_cold for codec initialization and close functions? @item - Run your test case, either manually or via FATE. This can be either - the full FATE regression suite, or any arbitrary invocation of any - front-end tool provided by FFmpeg, in any combination. +Did you add a long_name under NULL_IF_CONFIG_SMALL to the AVCodec or +AVInputFormat/AVOutputFormat struct? @item - Run @code{make lcov} to generate coverage data in HTML format. +Did you bump the minor version number (and reset the micro version +number) in @file{libavcodec/version.h} or @file{libavformat/version.h}? @item - View @code{lcov/index.html} in your preferred HTML viewer. -@end enumerate - -You can use the command @code{make lcov-reset} to reset the coverage -measurements. You will need to rerun @code{make lcov} after running a -new test. - -@subsection Using Valgrind - -The configure script provides a shortcut for using valgrind to spot bugs -related to memory handling. Just add the option -@code{--toolchain=valgrind-memcheck} or @code{--toolchain=valgrind-massif} -to your configure line, and reasonable defaults will be set for running -FATE under the supervision of either the @strong{memcheck} or the -@strong{massif} tool of the valgrind suite. - -In case you need finer control over how valgrind is invoked, use the -@code{--target-exec='valgrind <your_custom_valgrind_options>} option in -your configure line instead. - -@anchor{Release process} -@section Release process - -FFmpeg maintains a set of @strong{release branches}, which are the -recommended deliverable for system integrators and distributors (such as -Linux distributions, etc.). At regular times, a @strong{release -manager} prepares, tests and publishes tarballs on the -@url{http://ffmpeg.org} website. - -There are two kinds of releases: +Did you register it in @file{allcodecs.c} or @file{allformats.c}? -@enumerate @item -@strong{Major releases} always include the latest and greatest -features and functionality. +Did you add the AVCodecID to @file{avcodec.h}? +When adding new codec IDs, also add an entry to the codec descriptor +list in @file{libavcodec/codec_desc.c}. @item -@strong{Point releases} are cut from @strong{release} branches, -which are named @code{release/X}, with @code{X} being the release -version number. -@end enumerate - -Note that we promise to our users that shared libraries from any FFmpeg -release never break programs that have been @strong{compiled} against -previous versions of @strong{the same release series} in any case! - -However, from time to time, we do make API changes that require adaptations -in applications. Such changes are only allowed in (new) major releases and -require further steps such as bumping library version numbers and/or -adjustments to the symbol versioning file. Please discuss such changes -on the @strong{ffmpeg-devel} mailing list in time to allow forward planning. +If it has a FourCC, did you add it to @file{libavformat/riff.c}, +even if it is only a decoder? -@anchor{Criteria for Point Releases} -@subsection Criteria for Point Releases +@item +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. -Changes that match the following criteria are valid candidates for -inclusion into a point release: +@item +Did you add an entry to the table of supported formats or codecs in +(a)file{doc/general.texi}? -@enumerate @item -Fixes a security issue, preferably identified by a @strong{CVE -number} issued by @url{http://cve.mitre.org/}. +Did you add an entry in the Changelog? @item -Fixes a documented bug in @url{https://trac.ffmpeg.org}. +If it depends on a parser or a library, did you add that dependency in +configure? @item -Improves the included documentation. +Did you @code{git add} the appropriate files before committing? @item -Retains both source code and binary compatibility with previous -point releases of the same release branch. +Did you make sure it compiles standalone, i.e. with +@code{configure --disable-everything --enable-decoder=foo} +(or @code{--enable-demuxer} or whatever your component is)? @end enumerate -The order for checking the rules is (1 OR 2 OR 3) AND 4. - -@subsection Release Checklist -The release process involves the following steps: +@section Push checklist @enumerate -@item -Ensure that the @file{RELEASE} file contains the version number for -the upcoming release. - -@item -Add the release at @url{https://trac.ffmpeg.org/admin/ticket/versions}. - -@item -Announce the intent to do a release to the mailing list. - -@item -Make sure all relevant security fixes have been backported. See -@url{https://ffmpeg.org/security.html}. - -@item -Ensure that the FATE regression suite still passes in the release -branch on at least @strong{i386} and @strong{amd64} -(cf. @ref{Regression tests}). +@item Do not push anything without prior discussion on the + @uref{http://lists.ffmpeg.org/mailman/listinfo/ffmpeg-devel, ffmpeg-devel} + mailing list that affects one of the following: +@itemize @bullet +@item The public API or ABI +@item The behavior of the programs (renaming options, etc.) +@item The build systems behaviour +@item Compiler warning fixes +@item Functionality of the code +@item Code maintained by others (send a usual submission in that case) +@end itemize +@item Before pushing, ensure the following: +@itemize +@item Grand credits to the author of the submission (@code{git commit --author}) +@end itemize +@item After pushing, ensure the following: +@itemize +@item Send an answer to the + @uref{http://lists.ffmpeg.org/mailman/listinfo/ffmpeg-devel, ffmpeg-devel} + (or wherever you got the patch from) saying that you have applied the submission. +@end itemize +@end enumerate -@item -Prepare the release tarballs in @code{bz2} and @code{gz} formats, and -supplementing files that contain @code{gpg} signatures -@item -Publish the tarballs at @url{http://ffmpeg.org/releases}. Create and -push an annotated tag in the form @code{nX}, with @code{X} -containing the version number. +@section Release checklist -@item -Propose and send a patch to the @strong{ffmpeg-devel} mailing list -with a news entry for the website. +@enumerate +@item Ensure that the @file{RELEASE} file contains the version number for + the upcoming release. +@item Add the release at @url{https://trac.ffmpeg.org/admin/ticket/versions}. +@item Announce the intent to do a release to the mailing list. +@item Make sure all relevant security fixes have been backported. See + @url{https://ffmpeg.org/security.html}. +@item Ensure that the FATE regression suite still passes in the release + branch on at least i386 and amd64 + (cf. @ref{regression}). +@item Prepare the release tarballs in @code{bz2} and @code{gz} formats, and + supplementing files that contain @code{gpg} signatures +@item Publish the tarballs at @url{http://ffmpeg.org/releases}. Create and + push an annotated tag in the form @code{nX}, with @code{X} + containing the version number. +@item Propose and send a patch to the ffmpeg-devel mailing list + with a news entry for the website. +@item Publish the news entry. +@item Send announcement to the mailing list. +@end enumerate -@item -Publish the news entry. -@item -Send announcement to the mailing list. -@end enumerate @bye

4 6

[PATCH] avcodec/jpeg2000dec: check transform equality in MCT
by Michael Niedermayer 13 Oct '13

13 Oct '13

Fixes null pointer dereference Fixes Ticket2843 Signed-off-by: Michael Niedermayer <michaelni(a)gmx.at> --- libavcodec/jpeg2000dec.c | 6 ++++++ 1 file changed, 6 insertions(+) diff --git a/libavcodec/jpeg2000dec.c b/libavcodec/jpeg2000dec.c index 3c14a13..00473b6 100644 --- a/libavcodec/jpeg2000dec.c +++ b/libavcodec/jpeg2000dec.c @@ -1126,6 +1126,12 @@ static void mct_decode(Jpeg2000DecoderContext *s, Jpeg2000Tile *tile) int32_t *src[3], i0, i1, i2; float *srcf[3], i0f, i1f, i2f; + for (i = 1; i < 3; i++) + if (tile->codsty[0].transform != tile->codsty[i].transform) { + av_log(s->avctx, AV_LOG_ERROR, "Transforms mismatch, MCT not supported\n"); + return; + } + for (i = 0; i < 3; i++) if (tile->codsty[0].transform == FF_DWT97) srcf[i] = tile->comp[i].f_data; -- 1.7.9.5

1 1

[PATCH] exr: piz decompression
by Paul B Mahol 11 Oct '13

11 Oct '13

Signed-off-by: Paul B Mahol <onemda(a)gmail.com> --- libavcodec/exr.c | 477 +++++++++++++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 477 insertions(+) diff --git a/libavcodec/exr.c b/libavcodec/exr.c index 7e9e68c..a65e6ba 100644 --- a/libavcodec/exr.c +++ b/libavcodec/exr.c @@ -32,6 +32,7 @@ #include <zlib.h> +#include "get_bits.h" #include "avcodec.h" #include "bytestream.h" #include "mathops.h" @@ -54,6 +55,9 @@ typedef struct EXRThreadData { uint8_t *tmp; int tmp_size; + + uint8_t *bitmap; + uint16_t *lut; } EXRThreadData; typedef struct EXRContext { @@ -259,6 +263,471 @@ static int rle_uncompress(const uint8_t *src, int compressed_size, return 0; } +#define USHORT_RANGE (1 << 16) +#define BITMAP_SIZE (1 << 13) + +static uint16_t reverse_lut(const uint8_t *bitmap, uint16_t *lut) +{ + int i, k = 0; + + for (i = 0; i < USHORT_RANGE; ++i) { + if ((i == 0) || (bitmap[i >> 3] & (1 << (i & 7)))) + lut[k++] = i; + } + + i = k - 1; + + memset(lut + k, 0, (USHORT_RANGE - k) * 2); + + return i; +} + +static void apply_lut(const uint16_t *lut, uint16_t *dst, int dsize) +{ + int i; + + for (i = 0; i < dsize; ++i) + dst[i] = lut[dst[i]]; +} + +#define HUF_ENCBITS 16 // literal (value) bit length +#define HUF_DECBITS 14 // decoding bit size (>= 8) + +#define HUF_ENCSIZE ((1 << HUF_ENCBITS) + 1) // encoding table size +#define HUF_DECSIZE (1 << HUF_DECBITS) // decoding table size +#define HUF_DECMASK (HUF_DECSIZE - 1) + +typedef struct HufDec { + int len:8; + int lit:24; + int *p; +} HufDec; + +static void huf_canonical_code_table(uint64_t *hcode) +{ + uint64_t c, n[59] = { 0 }; + int i; + + /* + * For each i from 0 through 58, count the + * number of different codes of length i, and + * store the count in n[i]. + */ + for (i = 0; i < HUF_ENCSIZE; ++i) + n[hcode[i]] += 1; + + /* + * For each i from 58 through 1, compute the + * numerically lowest code with length i, and + * store that code in n[i]. + */ + c = 0; + for (i = 58; i > 0; --i) { + uint64_t nc = ((c + n[i]) >> 1); + n[i] = c; + c = nc; + } + + /* + * hcode[i] contains the length, l, of the + * code for symbol i. Assign the next available + * code of length l to the symbol and store both + * l and the code in hcode[i]. + */ + for (i = 0; i < HUF_ENCSIZE; ++i) { + int l = hcode[i]; + + if (l > 0) + hcode[i] = l | (n[l]++ << 6); + } +} + +#define SHORT_ZEROCODE_RUN 59 +#define LONG_ZEROCODE_RUN 63 +#define SHORTEST_LONG_RUN (2 + LONG_ZEROCODE_RUN - SHORT_ZEROCODE_RUN) +#define LONGEST_LONG_RUN (255 + SHORTEST_LONG_RUN) + +static int huf_unpack_enc_table(GetByteContext *gb, + int32_t im, int32_t iM, uint64_t *hcode) +{ + GetBitContext gbit; + + init_get_bits8(&gbit, gb->buffer, bytestream2_get_bytes_left(gb)); + + for (; im <= iM; im++) { + uint64_t l = hcode[im] = get_bits(&gbit, 6); + + if (l == LONG_ZEROCODE_RUN) { + int zerun = get_bits(&gbit, 8) + SHORTEST_LONG_RUN; + + if (im + zerun > iM + 1) + return AVERROR_INVALIDDATA; + + while (zerun--) + hcode[im++] = 0; + + im--; + } else if (l >= (uint64_t) SHORT_ZEROCODE_RUN) { + int zerun = l - SHORT_ZEROCODE_RUN + 2; + + if (im + zerun > iM + 1) + return AVERROR_INVALIDDATA; + + while (zerun--) + hcode[im++] = 0; + + im--; + } + } + + bytestream2_skip(gb, (get_bits_count(&gbit) + 7) / 8); + huf_canonical_code_table(hcode); + + return 0; +} + +static int huf_build_dec_table(const uint64_t *hcode, int im, + int iM, HufDec *hdecod) +{ + for (; im <= iM; im++) { + uint64_t c = hcode[im] >> 6; + int i, l = hcode[im] & 63; + + if (c >> l) + return AVERROR_INVALIDDATA; + + if (l > HUF_DECBITS) { + HufDec *pl = hdecod + (c >> (l - HUF_DECBITS)); + if (pl->len) + return AVERROR_INVALIDDATA; + + pl->lit++; + + if (pl->p) + pl->p = av_realloc_f(pl->p, pl->lit, sizeof(int)); + else + pl->p = av_malloc(sizeof(int)); + if (!pl->p) + return AVERROR(ENOMEM); + + pl->p[pl->lit - 1] = im; + } else if (l) { + HufDec *pl = hdecod + (c << (HUF_DECBITS - l)); + + for (i = 1 << (HUF_DECBITS - l); i > 0; i--, pl++) { + if (pl->len || pl->p) + return AVERROR_INVALIDDATA; + pl->len = l; + pl->lit = im; + } + } + } + + return 0; +} + +#define get_char(c, lc, gb) { \ + c = (c << 8) | bytestream2_get_byte(gb); \ + lc += 8; \ +} + +#define get_code(po, rlc, c, lc, gb, out, oe) { \ + if (po == rlc) { \ + if (lc < 8) \ + get_char(c, lc, gb); \ + lc -= 8; \ + \ + uint8_t cs = (c >> lc); \ + \ + if (out + cs > oe) \ + return AVERROR_INVALIDDATA; \ + \ + uint16_t s = out[-1]; \ + \ + while (cs-- > 0) \ + *out++ = s; \ + } else if (out < oe) { \ + *out++ = po; \ + } else { \ + return AVERROR_INVALIDDATA; \ + } \ +} + +static int huf_decode(const uint64_t *hcode, const HufDec *hdecod, + GetByteContext *gb, int nbits, + int rlc, int no, uint16_t *out) +{ + uint64_t c = 0; + int lc = 0; + uint16_t *outb = out; + uint16_t *oe = out + no; + const uint8_t *ie = gb->buffer + (nbits + 7) / 8; // input byte size + + while (gb->buffer < ie) { + get_char(c, lc, gb); + + while (lc >= HUF_DECBITS) { + const HufDec pl = hdecod[(c >> (lc-HUF_DECBITS)) & HUF_DECMASK]; + + if (pl.len) { + lc -= pl.len; + get_code(pl.lit, rlc, c, lc, gb, out, oe); + } else { + int j; + + if (!pl.p) + return AVERROR_INVALIDDATA; + + for (j = 0; j < pl.lit; j++) { + int l = hcode[pl.p[j]] & 63; + + while (lc < l && bytestream2_get_bytes_left(gb) > 0) + get_char(c, lc, gb); + + if (lc >= l) { + if ((hcode[pl.p[j]] >> 6) == + ((c >> (lc - l)) & ((1LL << l) - 1))) { + lc -= l; + get_code(pl.p[j], rlc, c, lc, gb, out, oe); + break; + } + } + } + + if (j == pl.lit) + return AVERROR_INVALIDDATA; + } + } + } + + int i = (8 - nbits) & 7; + c >>= i; + lc -= i; + + while (lc > 0) { + const HufDec pl = hdecod[(c << (HUF_DECBITS - lc)) & HUF_DECMASK]; + + if (pl.len) { + lc -= pl.len; + get_code(pl.lit, rlc, c, lc, gb, out, oe); + } else { + return AVERROR_INVALIDDATA; + } + } + + if (out - outb != no) + return AVERROR_INVALIDDATA; + return 0; +} + +static int huf_uncompress(GetByteContext *gb, + uint16_t *dst, int dst_size) +{ + int32_t src_size, im, iM; + uint32_t tableLength, nBits; + uint64_t *freq; + HufDec *hdec; + int ret, i; + + src_size = bytestream2_get_le32(gb); + im = bytestream2_get_le32(gb); + iM = bytestream2_get_le32(gb); + tableLength = bytestream2_get_le32(gb); + nBits = bytestream2_get_le32(gb); + if (im < 0 || im >= HUF_ENCSIZE || + iM < 0 || iM >= HUF_ENCSIZE || + src_size < 0) + return AVERROR_INVALIDDATA; + + bytestream2_skip(gb, 4); + + freq = av_malloc(HUF_ENCSIZE * sizeof(uint64_t)); + hdec = av_mallocz(HUF_DECSIZE * sizeof(HufDec)); + if (!freq || !hdec) + return AVERROR(ENOMEM); + + if ((ret = huf_unpack_enc_table(gb, im, iM, freq)) < 0) + return ret; + + if (nBits > 8 * bytestream2_get_bytes_left(gb)) + return AVERROR_INVALIDDATA; + + if ((ret = huf_build_dec_table(freq, im, iM, hdec)) < 0) + return ret; + if ((ret = huf_decode(freq, hdec, gb, nBits, iM, dst_size, dst)) < 0) + return ret; + + for (i = 0; i < HUF_DECSIZE; i++) { + if (hdec[i].p) + av_free(hdec[i].p); + } + + av_free(freq); + av_free(hdec); + + return 0; +} + +static inline void wdec14(uint16_t l, uint16_t h, uint16_t *a, uint16_t *b) +{ + int16_t ls = l; + int16_t hs = h; + int hi = hs; + int ai = ls + (hi & 1) + (hi >> 1); + int16_t as = ai; + int16_t bs = ai - hi; + + *a = as; + *b = bs; +} + +#define NBITS 16 +#define A_OFFSET (1 << (NBITS - 1)) +#define MOD_MASK ((1 << NBITS) - 1) + +static inline void wdec16(uint16_t l, uint16_t h, uint16_t *a, uint16_t *b) +{ + int m = l; + int d = h; + int bb = (m - (d >> 1)) & MOD_MASK; + int aa = (d + bb - A_OFFSET) & MOD_MASK; + *b = bb; + *a = aa; +} + +static void wav_decode(uint16_t *in, int nx, int ox, + int ny, int oy, uint16_t mx) +{ + int w14 = (mx < (1 << 14)); + int n = (nx > ny) ? ny: nx; + int p = 1; + int p2; + + while (p <= n) + p <<= 1; + + p >>= 1; + p2 = p; + p >>= 1; + + while (p >= 1) { + uint16_t *py = in; + uint16_t *ey = in + oy * (ny - p2); + uint16_t i00, i01, i10, i11; + int oy1 = oy * p; + int oy2 = oy * p2; + int ox1 = ox * p; + int ox2 = ox * p2; + + for (; py <= ey; py += oy2) { + uint16_t *px = py; + uint16_t *ex = py + ox * (nx - p2); + + + for (; px <= ex; px += ox2) { + uint16_t *p01 = px + ox1; + uint16_t *p10 = px + oy1; + uint16_t *p11 = p10 + ox1; + + if (w14) { + wdec14(*px, *p10, &i00, &i10); + wdec14(*p01, *p11, &i01, &i11); + wdec14(i00, i01, px, p01); + wdec14(i10, i11, p10, p11); + } else { + wdec16(*px, *p10, &i00, &i10); + wdec16(*p01, *p11, &i01, &i11); + wdec16(i00, i01, px, p01); + wdec16(i10, i11, p10, p11); + } + } + + if (nx & p) { + uint16_t *p10 = px + oy1; + + if (w14) + wdec14(*px, *p10, &i00, p10); + else + wdec16(*px, *p10, &i00, p10); + + *px= i00; + } + } + + if (ny & p) { + uint16_t *px = py; + uint16_t *ex = py + ox * (nx - p2); + + for (; px <= ex; px += ox2) { + uint16_t *p01 = px + ox1; + + if (w14) + wdec14(*px, *p01, &i00, p01); + else + wdec16(*px, *p01, &i00, p01); + + *px = i00; + } + } + + p2 = p; + p >>= 1; + } +} + +static int piz_uncompress(EXRContext *s, const uint8_t *src, int ssize, int dsize, EXRThreadData *td) +{ + GetByteContext gb; + uint16_t maxval, min_non_zero, max_non_zero; + uint16_t *ptr, *tmp = (uint16_t *)td->tmp; + int8_t *out; + int ret, i, ysize; + + if (!td->bitmap) + td->bitmap = av_malloc(BITMAP_SIZE); + if (!td->lut) + td->lut = av_malloc(1 << 17); + if (!td->bitmap || !td->lut) + return AVERROR(ENOMEM); + + bytestream2_init(&gb, src, ssize); + min_non_zero = bytestream2_get_le16(&gb); + max_non_zero = bytestream2_get_le16(&gb); + + if (max_non_zero >= BITMAP_SIZE) + return AVERROR_INVALIDDATA; + + memset(td->bitmap, 0, FFMIN(min_non_zero, BITMAP_SIZE)); + if (min_non_zero <= max_non_zero) + bytestream2_get_buffer(&gb, td->bitmap + min_non_zero, max_non_zero - min_non_zero + 1); + + maxval = reverse_lut(td->bitmap, td->lut); + + ret = huf_uncompress(&gb, tmp, dsize / sizeof(int16_t)); + if (ret) + return ret; + + ysize = dsize / (s->xdelta * 2 * s->desc->nb_components); + + ptr = tmp; + for (i = 0; i < s->desc->nb_components; i++, ptr += s->xdelta * ysize) + wav_decode(ptr, s->xdelta, 1, ysize, s->xdelta, maxval); + + apply_lut(td->lut, tmp, dsize / sizeof(int16_t)); + + out = td->uncompressed_data; + for (i = 0; i < ysize; i++) { + int j; + for (j = 0; j < s->desc->nb_components; j++) { + uint16_t *in = tmp + j * s->xdelta * ysize + i * s->xdelta; + memcpy(out, in, s->xdelta * 2); + out += s->xdelta * 2; + } + } + + return 0; +} + static int decode_block(AVCodecContext *avctx, void *tdata, int jobnr, int threadnr) { @@ -310,6 +779,9 @@ static int decode_block(AVCodecContext *avctx, void *tdata, case EXR_ZIP16: ret = zip_uncompress(src, data_size, uncompressed_size, td); break; + case EXR_PIZ: + ret = piz_uncompress(s, src, data_size, uncompressed_size, td); + break; case EXR_RLE: ret = rle_uncompress(src, data_size, uncompressed_size, td); } @@ -626,6 +1098,9 @@ static int decode_frame(AVCodecContext *avctx, case EXR_ZIP16: s->scan_lines_per_block = 16; break; + case EXR_PIZ: + s->scan_lines_per_block = 32; + break; default: av_log(avctx, AV_LOG_ERROR, "Compression type %d is not supported\n", s->compr); return AVERROR_PATCHWELCOME; @@ -721,6 +1196,8 @@ static av_cold int decode_end(AVCodecContext *avctx) EXRThreadData *td = &s->thread_data[i]; av_free(td->uncompressed_data); av_free(td->tmp); + av_free(td->bitmap); + av_free(td->lut); } av_freep(&s->thread_data); -- 1.7.11.4

4 15

[PATCH] avcodec/jpeg2000dec: Zero code and quant style
by Michael Niedermayer 03 Oct '13

03 Oct '13

This prevents future frames from using half initialized/inconsistent structures Fixes assertion failure Fixes Ticket2897 Signed-off-by: Michael Niedermayer <michaelni(a)gmx.at> --- libavcodec/jpeg2000dec.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/libavcodec/jpeg2000dec.c b/libavcodec/jpeg2000dec.c index 3f22ead..6e8f0d6 100644 --- a/libavcodec/jpeg2000dec.c +++ b/libavcodec/jpeg2000dec.c @@ -1369,6 +1369,8 @@ static void jpeg2000_dec_cleanup(Jpeg2000DecoderContext *s) } } av_freep(&s->tile); + memset(s->codsty, 0, sizeof(s->codsty)); + memset(s->qntsty, 0, sizeof(s->qntsty)); s->numXtiles = s->numYtiles = 0; } -- 1.7.9.5

2 2

[PATCH 1/7] lavc/libopenjpegdec: support interlaced layout
by Matthieu Bouron 28 Sep '13

28 Sep '13

Fixes tickets #1102. Interlaced layout is supported thought use of codec extradata: avctx->extradata[0] enables interlaced layout, avctx->extradata[1] stores field order. --- libavcodec/libopenjpegdec.c | 99 ++++++++++++++++++++++++++++++++++++--------- 1 file changed, 80 insertions(+), 19 deletions(-) diff --git a/libavcodec/libopenjpegdec.c b/libavcodec/libopenjpegdec.c index f8b6165..4ea0489 100644 --- a/libavcodec/libopenjpegdec.c +++ b/libavcodec/libopenjpegdec.c @@ -68,6 +68,12 @@ typedef struct { AVClass *class; opj_dparameters_t dec_params; int lowqual; + AVFrame* picture; + uint64_t curpict; + int separate_fields; + int field_order; + int top_field_coded_first; + int width, height; } LibOpenJPEGContext; static inline int libopenjpeg_matches_pix_fmt(const opj_image_t *image, enum AVPixelFormat pix_fmt) @@ -148,12 +154,13 @@ static inline int libopenjpeg_ispacked(enum AVPixelFormat pix_fmt) return 1; } -static inline void libopenjpeg_copy_to_packed8(AVFrame *picture, opj_image_t *image) { +static inline void libopenjpeg_copy_to_packed8(AVFrame *picture, opj_image_t *image, int separate_fields, int field_pos) { uint8_t *img_ptr; - int index, x, y, c; + int index, x, y, c, line; for (y = 0; y < picture->height; y++) { index = y*picture->width; - img_ptr = picture->data[0] + y*picture->linesize[0]; + line = separate_fields * (field_pos + y) + y; + img_ptr = picture->data[0] + line*picture->linesize[0]; for (x = 0; x < picture->width; x++, index++) { for (c = 0; c < image->numcomps; c++) { *img_ptr++ = image->comps[c].data[index]; @@ -162,16 +169,17 @@ static inline void libopenjpeg_copy_to_packed8(AVFrame *picture, opj_image_t *im } } -static inline void libopenjpeg_copy_to_packed16(AVFrame *picture, opj_image_t *image) { +static inline void libopenjpeg_copy_to_packed16(AVFrame *picture, opj_image_t *image, int separate_fields, int field_pos) { uint16_t *img_ptr; - int index, x, y, c; + int index, x, y, c, line; int adjust[4]; for (x = 0; x < image->numcomps; x++) adjust[x] = FFMAX(FFMIN(16 - image->comps[x].prec, 8), 0); for (y = 0; y < picture->height; y++) { index = y*picture->width; - img_ptr = (uint16_t*) (picture->data[0] + y*picture->linesize[0]); + line = separate_fields * (field_pos + y) + y; + img_ptr = (uint16_t*) (picture->data[0] + line*picture->linesize[0]); for (x = 0; x < picture->width; x++, index++) { for (c = 0; c < image->numcomps; c++) { *img_ptr++ = image->comps[c].data[index] << adjust[c]; @@ -180,7 +188,7 @@ static inline void libopenjpeg_copy_to_packed16(AVFrame *picture, opj_image_t *i } } -static inline void libopenjpeg_copyto8(AVFrame *picture, opj_image_t *image) { +static inline void libopenjpeg_copyto8(AVFrame *picture, opj_image_t *image, int separate_fields, int field_pos) { int *comp_data; uint8_t *img_ptr; int index, x, y; @@ -188,7 +196,8 @@ static inline void libopenjpeg_copyto8(AVFrame *picture, opj_image_t *image) { for (index = 0; index < image->numcomps; index++) { comp_data = image->comps[index].data; for (y = 0; y < image->comps[index].h; y++) { - img_ptr = picture->data[index] + y * picture->linesize[index]; + int line = separate_fields * (field_pos + y) + y; + img_ptr = picture->data[index] + line * picture->linesize[index]; for (x = 0; x < image->comps[index].w; x++) { *img_ptr = (uint8_t) *comp_data; img_ptr++; @@ -198,14 +207,15 @@ static inline void libopenjpeg_copyto8(AVFrame *picture, opj_image_t *image) { } } -static inline void libopenjpeg_copyto16(AVFrame *picture, opj_image_t *image) { +static inline void libopenjpeg_copyto16(AVFrame *picture, opj_image_t *image, int separate_fields, int field_pos) { int *comp_data; uint16_t *img_ptr; int index, x, y; for (index = 0; index < image->numcomps; index++) { comp_data = image->comps[index].data; for (y = 0; y < image->comps[index].h; y++) { - img_ptr = (uint16_t*) (picture->data[index] + y * picture->linesize[index]); + int line = separate_fields * (field_pos + y) + y; + img_ptr = (uint16_t*) (picture->data[index] + line * picture->linesize[index]); for (x = 0; x < image->comps[index].w; x++) { *img_ptr = *comp_data; img_ptr++; @@ -219,10 +229,27 @@ static av_cold int libopenjpeg_decode_init(AVCodecContext *avctx) { LibOpenJPEGContext *ctx = avctx->priv_data; + if (avctx->extradata_size > 2) { + ctx->picture = av_frame_alloc(); + ctx->separate_fields = avctx->extradata[0]; + if (ctx->separate_fields) { + avctx->field_order = ctx->field_order = avctx->extradata[1]; + ctx->top_field_coded_first = (ctx->field_order == AV_FIELD_TT || + ctx->field_order == AV_FIELD_TB); + } + } + opj_set_default_decoder_parameters(&ctx->dec_params); return 0; } +static av_cold int libopenjpeg_decode_uninit(AVCodecContext *avctx) +{ + LibOpenJPEGContext *ctx = avctx->priv_data; + av_frame_free(&ctx->picture); + return 0; +} + static int libopenjpeg_decode_frame(AVCodecContext *avctx, void *data, int *got_frame, AVPacket *avpkt) @@ -239,9 +266,22 @@ static int libopenjpeg_decode_frame(AVCodecContext *avctx, int width, height, ret = -1; int pixel_size = 0; int ispacked = 0; - int i; + int realloc_il_buffer = 0; + int i, field_pos, field_nb; *got_frame = 0; + if (ctx->separate_fields) { + picture = ctx->picture; + frame.f = picture; + } + + field_nb = ctx->curpict % 2; + field_pos = field_nb ^ !ctx->top_field_coded_first; + + if (ctx->separate_fields && avctx->thread_count > 1) { + av_log(avctx, AV_LOG_ERROR, "Separate fields layout does not support threading. Please specify -threads 1 before input.\n"); + return AVERROR(EINVAL); + } // Check if input is a raw jpeg2k codestream or in jp2 wrapping if ((AV_RB32(buf) == 12) && @@ -287,12 +327,21 @@ static int libopenjpeg_decode_frame(AVCodecContext *avctx, width = image->x1 - image->x0; height = image->y1 - image->y0; + if (ctx->separate_fields) + height *= 2; + if (av_image_check_size(width, height, 0, avctx) < 0) { av_log(avctx, AV_LOG_ERROR, "%dx%d dimension invalid.\n", width, height); goto done; } + if (ctx->width != width || ctx->height != height) { + ctx->width = width; + ctx->height = height; + realloc_il_buffer = 1; + } + avcodec_set_dimensions(avctx, width, height); if (avctx->pix_fmt != AV_PIX_FMT_NONE) @@ -310,8 +359,13 @@ static int libopenjpeg_decode_frame(AVCodecContext *avctx, if (image->comps[i].prec > avctx->bits_per_raw_sample) avctx->bits_per_raw_sample = image->comps[i].prec; - if (ff_thread_get_buffer(avctx, &frame, 0) < 0) - goto done; + if (ctx->separate_fields && realloc_il_buffer && picture->data[0]) + av_frame_unref(picture); + + if (!ctx->separate_fields || + (ctx->separate_fields && (!ctx->curpict || realloc_il_buffer))) + if (ff_thread_get_buffer(avctx, &frame, 0) < 0) + goto done; ctx->dec_params.cp_limit_decoding = NO_LIMITATION; ctx->dec_params.cp_reduce = avctx->lowres; @@ -341,28 +395,28 @@ static int libopenjpeg_decode_frame(AVCodecContext *avctx, switch (pixel_size) { case 1: if (ispacked) { - libopenjpeg_copy_to_packed8(picture, image); + libopenjpeg_copy_to_packed8(picture, image, ctx->separate_fields, field_pos); } else { - libopenjpeg_copyto8(picture, image); + libopenjpeg_copyto8(picture, image, ctx->separate_fields, field_pos); } break; case 2: if (ispacked) { - libopenjpeg_copy_to_packed8(picture, image); + libopenjpeg_copy_to_packed8(picture, image, ctx->separate_fields, field_pos); } else { - libopenjpeg_copyto16(picture, image); + libopenjpeg_copyto16(picture, image, ctx->separate_fields, field_pos); } break; case 3: case 4: if (ispacked) { - libopenjpeg_copy_to_packed8(picture, image); + libopenjpeg_copy_to_packed8(picture, image, ctx->separate_fields, field_pos); } break; case 6: case 8: if (ispacked) { - libopenjpeg_copy_to_packed16(picture, image); + libopenjpeg_copy_to_packed16(picture, image, ctx->separate_fields, field_pos); } break; default: @@ -373,7 +427,13 @@ static int libopenjpeg_decode_frame(AVCodecContext *avctx, *got_frame = 1; ret = buf_size; + if (ctx->separate_fields) + if (field_nb) + av_frame_ref(data, ctx->picture); + else + *got_frame = 0; done: + ctx->curpict++; opj_image_destroy(image); opj_destroy_decompress(dec); return ret; @@ -401,6 +461,7 @@ AVCodec ff_libopenjpeg_decoder = { .priv_data_size = sizeof(LibOpenJPEGContext), .init = libopenjpeg_decode_init, .decode = libopenjpeg_decode_frame, + .close = libopenjpeg_decode_uninit, .capabilities = CODEC_CAP_DR1 | CODEC_CAP_FRAME_THREADS, .max_lowres = 31, .long_name = NULL_IF_CONFIG_SMALL("OpenJPEG JPEG 2000"), -- 1.8.2

4 14

[PATCH 1/5] doc/encoders: Add libxvid doc
by Timothy Gu 26 Sep '13

26 Sep '13

--- doc/encoders.texi | 129 ++++++++++++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 129 insertions(+) diff --git a/doc/encoders.texi b/doc/encoders.texi index fdf054c..c703000 100644 --- a/doc/encoders.texi +++ b/doc/encoders.texi @@ -1212,6 +1212,135 @@ no-fast-pskip=1:subq=6:8x8dct=0:trellis=0 OUTPUT Encoding ffpresets for common usages are provided so they can be used with the general presets system (e.g. passing the @option{pre} option). +@section libxvid + +Xvid MPEG-4 Part 2 encoder wrapper. + +Requires the presence of the libxvidcore headers and library during +configuration. You need to explicitly configure the build with +@code{--enable-libxvid}. + +An native MPEG-4 Part 2 encoder exists, so users can encode to this +format without this library. + +@subsection Options + +The following options are supported by the libxvid wrapper: + +@table @option +@item b +Set bit rate in bits/second. + +@item g +Set group of picture (GOP) size. Default value is 12. + +@item qmin +Set minimum quantizer scale. Must be included between -1 and 69, default +value is 2. + +@item qmax +Set maximum quantizer scale. Must be included between -1 and 1024, +default value is 31. + +@item mpeg_quant +Enable MPEG quantizer instead of H.263 when set to 1. + +@item threads +Set multi-threading thread count. Possible values are integers greater or +equal to 0, or @samp{auto}. 0 or @samp{auto} means that it automatically +detects a good number of threads. 1 means to disable multi-threading. + +@item bf +Set maximum number of B-frames. + +@item b_qfactor +Set quantization parameter (QP) factor between P- and B-frames. Value +is a float number. + +@item b_qoffset +Set QP offset between P- and B-frames. Value is a float number. + +@item flags +Set specific encoding flags. Possible values: + +@table @samp + +@item mv4 +Use four motion vector by macroblock. + +@item aic +Enable high quality AC prediction. + +@item gray +Only encode grayscale. + +@item gmc +Enable the use of global motion compensation (GMC). + +@item qpel +Enable quarter-pixel motion compensation. + +@item cgop +Enable closed GOP. + +@item global_header +Place global headers in extradata instead of every keyframe. + +@end table + +@item trellis +Enable trellis-based rate-distortion optimal quantization when set to +1. Default is 0 (disabled). + +@item me_method +Set motion estimation method. Possible values in the decreasing order +of speed and increasing order of quality: + +@table @samp + +@item zero +Use no motion estimation (default). + +@item phods +@item x1 +@item log +Enable advanced diamond zonal search for 16x16 blocks and half-pixel +refinement for 16x16 blocks. @samp{x1} and @samp{log} are aliases for +@samp{phods}. + +@item epzs +Enable all of the things described above, plus advanced diamond zonal +search for 8x8 blocks, half-pixel refinement for 8x8 blocks, and motion +estimation on chroma planes. + +@item full +Enable all of the things described above, plus extended 16x16 and 8x8 +blocks search. + +@end table + +@item mbd +Set macroblock decision algorithm. Possible values in the increasing +order of quality: + +@table @samp + +@item simple +Use macroblock comparing function algorithm (default). + +@item bits +Enable rate distortion-based half pixel and quarter pixel refinement for +16x16 blocks. + +@item rd +Enable all of the things described above, plus rate distortion-based +half pixel and quarter pixel refinement for 8x8 blocks, and rate +distortion-based search using square pattern. + +@end table + +@end table + @section png PNG image encoder. -- 1.8.1.2

4 33

[WIP] [PATCH 1/2] lavfi: add an API to synchronize multiple video inputs.
by Nicolas George 23 Sep '13

23 Sep '13

Compared to dualinput, this API can handle more than two inputs and can generate frames synchronized to any or all input streams. Signed-off-by: Nicolas George <nicolas.george(a)normalesup.org> --- libavfilter/framesync.c | 222 +++++++++++++++++++++++++++++++++++++++++++ libavfilter/framesync.h | 242 +++++++++++++++++++++++++++++++++++++++++++++++ 2 files changed, 464 insertions(+) create mode 100644 libavfilter/framesync.c create mode 100644 libavfilter/framesync.h Note: this is work in progress, but I do not know when I will be able to work on it again. It came as a reaction to mail on the user mailing list: how to overlay a video on top of a constant background. The problem is that the overlay filter, and the dualinput API that was extracted from it to serve for other filters, can only synchronize on the background video: since the background is a constant image, it does not move. If the video on top has constant frame rate, the problem can be solved by setting vf_fps on the background. Otherwise, the suggested solution was to make a sandwich: overlay the completely opaque background on top of the video to get the background with the video timestamps, and then overlay the video. That works (except if the background itself has alpha), but that is ugly and inefficient. The obvious quick-and-dirty solution would be to add sync=second option to the dualinput API, but that is just that: quick and dirty, it does not cover other uses: combining R, G, B input planes (three inputs), computing the PSNR of videos with the same frame rate but different skipped frames (must generate an event for frames on any input). This API aims to handle all the cases. diff --git a/libavfilter/framesync.c b/libavfilter/framesync.c new file mode 100644 index 0000000..2e0b1db --- /dev/null +++ b/libavfilter/framesync.c @@ -0,0 +1,222 @@ +/* + * Copyright (c) 2013 Nicolas George + * + * This file is part of FFmpeg. + * + * FFmpeg is free software; you can redistribute it and/or + * modify it under the terms of the GNU Lesser General Public License + * as published by the Free Software Foundation; either + * version 2.1 of the License, or (at your option) any later version. + * + * FFmpeg is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU Lesser General Public License for more details. + * + * You should have received a copy of the GNU Lesser General Public License + * along with FFmpeg; if not, write to the Free Software Foundation, Inc., + * 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA + */ + +#include "libavutil/avassert.h" +#include "avfilter.h" +#include "bufferqueue.h" +#include "framesync.h" +#include "internal.h" + +/* + * TODO + * Actually compute time base, den = GCM(dens) and num = LCD(nums) + * Some more testing (especially wrt write status when cloning frame). + * Uninit structure and check for leaks. + */ + +#define OFFSET(member) offsetof(FFFrameSync, member) + +static const char *framesync_name(void *ptr) +{ + return "framesync"; +} + +static const AVClass framesync_class = { + .version = LIBAVUTIL_VERSION_INT, + .class_name = "framesync", + .item_name = framesync_name, + .category = AV_CLASS_CATEGORY_FILTER, + .option = NULL, + .parent_log_context_offset = OFFSET(parent), +}; + +enum { + STATE_BOF, + STATE_RUN, + STATE_EOF, +}; + +void ff_framesync_init(FFFrameSync *fs, void *parent, unsigned nb_in) +{ + fs->class = &framesync_class; + fs->parent = parent; + fs->nb_in = nb_in; +} + +int ff_framesync_configure(FFFrameSync *fs) +{ + unsigned i; + + if (!fs->time_base.num) { + for (i = 0; i < fs->nb_in; i++) { + if (fs->in[i].sync) { + if (fs->time_base.num) { + //av_assert0(!"todo"); + } else { + fs->time_base = fs->in[i].time_base; + } + } + } + if (!fs->time_base.num) { + av_log(fs, AV_LOG_ERROR, "Impossible to set time base\n"); + return AVERROR(EINVAL); + } + } + + for (i = 0; i < fs->nb_in; i++) + fs->in[i].pts = fs->in[i].pts_next = AV_NOPTS_VALUE; + + return 0; +} + +static void framesync_advance(FFFrameSync *fs) +{ + int latest; + unsigned i; + int64_t pts; + + if (fs->eof) + return; + while (!fs->frame_ready) { + latest = -1; + for (i = 0; i < fs->nb_in; i++) { + if (!fs->in[i].have_next) { + if (latest < 0 || fs->in[i].pts < fs->in[latest].pts) + latest = i; + } + } + if (latest >= 0) { + fs->in_request = latest; + break; + } + + pts = fs->in[0].pts_next; + for (i = 1; i < fs->nb_in; i++) + if (fs->in[i].pts_next < pts) + pts = fs->in[i].pts_next; + if (pts == INT64_MAX) { + fs->eof = 1; + break; + } + for (i = 0; i < fs->nb_in; i++) { + if (fs->in[i].pts_next == pts || + (fs->in[i].before == EXT_INFINITY && + fs->in[i].state == STATE_BOF)) { + av_frame_free(&fs->in[i].frame); + fs->in[i].frame = fs->in[i].frame_next; + fs->in[i].pts = fs->in[i].pts_next; + fs->in[i].frame_next = NULL; + fs->in[i].pts_next = AV_NOPTS_VALUE; + fs->in[i].have_next = 0; + fs->in[i].state = fs->in[i].frame ? STATE_RUN : STATE_EOF; + if (fs->in[i].sync && fs->in[i].frame) + fs->frame_ready = 1; + if (fs->in[i].state == STATE_EOF && + fs->in[i].after == EXT_NULL) { + fs->eof = 1; + } + } + } + if (fs->eof) + fs->frame_ready = 0; + if (fs->frame_ready) + for (i = 0; i < fs->nb_in; i++) + if ((fs->in[i].state == STATE_BOF && + fs->in[i].before == EXT_NULL)) + fs->frame_ready = 0; + fs->pts = pts; + } +} + +static void framesync_inject_frame(FFFrameSync *fs, unsigned in, AVFrame *frame) +{ + int64_t pts; + + av_assert0(!fs->in[in].have_next); + if (frame) { + pts = av_rescale_q(frame->pts, fs->in[in].time_base, fs->time_base); + frame->pts = pts; + } else { + pts = fs->in[in].state != STATE_RUN || fs->in[in].after == EXT_INFINITY ? INT64_MAX : + fs->in[in].pts + 1; /* TODO extrapolate */ + } + fs->in[in].frame_next = frame; + fs->in[in].pts_next = pts; + fs->in[in].have_next = 1; +} + +int ff_framesync_add_frame(FFFrameSync *fs, unsigned in, AVFrame *frame) +{ + av_assert1(in < fs->nb_in); + if (!fs->in[in].have_next) + framesync_inject_frame(fs, in, frame); + else + ff_bufqueue_add(fs, &fs->in[in].queue, frame); + return 0; +} + +void ff_framesync_next(FFFrameSync *fs) +{ + unsigned i; + + av_assert0(!fs->frame_ready); + for (i = 0; i < fs->nb_in; i++) + if (!fs->in[i].have_next && fs->in[i].queue.available) + framesync_inject_frame(fs, i, ff_bufqueue_get(&fs->in[i].queue)); + fs->frame_ready = 0; + framesync_advance(fs); +} + +void ff_framesync_drop(FFFrameSync *fs) +{ + fs->frame_ready = 0; +} + +int ff_framesync_get_frame(FFFrameSync *fs, unsigned in, AVFrame **rframe, + unsigned get) +{ + AVFrame *frame; + unsigned need_copy = 0, i; + int64_t pts_next; + + if (!fs->in[in].frame) { + *rframe = NULL; + return 0; + } + frame = fs->in[in].frame; + if (get) { + /* Find out if we need to copy the frame: is there another sync + stream, and do we know if its current frame will outlast this one? */ + pts_next = fs->in[in].have_next ? fs->in[in].pts_next : INT64_MAX; + for (i = 0; i < fs->nb_in && !need_copy; i++) + if (i != in && fs->in[i].sync && + (!fs->in[i].have_next || fs->in[i].pts_next < pts_next)) + need_copy = 1; + if (need_copy) { + if (!(frame = av_frame_clone(frame))) + return AVERROR(ENOMEM); + } else { + fs->in[in].frame = NULL; + } + fs->frame_ready = 0; + } + *rframe = frame; + return 0; +} diff --git a/libavfilter/framesync.h b/libavfilter/framesync.h new file mode 100644 index 0000000..ecbb88c --- /dev/null +++ b/libavfilter/framesync.h @@ -0,0 +1,242 @@ +/* + * Copyright (c) 2013 Nicolas George + * + * This file is part of FFmpeg. + * + * FFmpeg is free software; you can redistribute it and/or + * modify it under the terms of the GNU Lesser General Public License + * as published by the Free Software Foundation; either + * version 2.1 of the License, or (at your option) any later version. + * + * FFmpeg is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU Lesser General Public License for more details. + * + * You should have received a copy of the GNU Lesser General Public License + * along with FFmpeg; if not, write to the Free Software Foundation, Inc., + * 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA + */ + +#ifndef AVFILTER_FRAMESYNC_H +#define AVFILTER_FRAMESYNC_H + +#include "bufferqueue.h" + +/* + * TODO + * Callback-based API similar to dualinput. + * Rename ExtModes. + * Export convenient options. + */ + +/** + * This API is intended as a helper for filters that have several video + * input and need to combine them somehow. If the inputs have different or + * variable frame rate, getting the input frames to match requires a rather + * complex logic and a few user-tunable options. + * + * In this API, when a set of synchronized input frames is ready to be + * procesed is called a frame event. Frame event can be generated in + * response to input frames on any or all inputs and the handling of + * situations where some stream extend beyond the beginning or the end of + * others can be configured. + * + * The basic working of this API is the following: + * + * - When a frame is available on any input, add it using + * ff_framesync_add_frame(). + * + * - When a frame event is ready to be processed (i.e. after adding a frame + * or when requested on input): + * - call ff_framesync_next(); + * - if fs->frame_ready is true, process the frames; + * - call ff_framesync_drop(). + */ + +/** + * Stream extrapolation mode + * + * Describe how the frames of a stream are extrapolated before the first one + * and after EOF to keep sync with possibly longer other streams. + */ +enum FFFrameSyncExtMode { + /** + * Completely stop all streams with this one. + */ + EXT_STOP, + + /** + * Ignore this stream and continue processing the other ones. + */ + EXT_NULL, + + /** + * Extend the frame to infinity. + */ + EXT_INFINITY, +}; + +/** + * Input stream structure + */ +typedef struct FFFrameSyncIn { + + /** + * Queue of incoming AVFrame, and NULL to mark EOF + */ + struct FFBufQueue queue; + + /** + * Extrapolation mode for timestamps before the first frame + */ + enum FFFrameSyncExtMode before; + + /** + * Extrapolation mode for timestamps after the last frame + */ + enum FFFrameSyncExtMode after; + + /** + * Time base for the incoming frames + */ + AVRational time_base; + + /** + * Current frame, may be NULL before the first one or after EOF + */ + AVFrame *frame; + + /** + * Next frame, for internal use + */ + AVFrame *frame_next; + + /** + * PTS of the current frame + */ + int64_t pts; + + /** + * PTS of the next frame, for internal use + */ + int64_t pts_next; + + /** + * Boolean flagging the next frame, for internal use + */ + uint8_t have_next; + + /** + * State: before first, in stream or after EOF, for internal use + */ + uint8_t state; + + /** + * Synchronization flag: if not zero, then frames on this stream cause + * frames on the output. + */ + unsigned sync; + +} FFFrameSyncIn; + +/** + * Frame sync structure. + */ +typedef struct FFFrameSync { + const AVClass *class; + void *parent; + + /** + * Number of input streams + */ + unsigned nb_in; + + /** + * Time base for the output events + */ + AVRational time_base; + + /** + * Timestamp of the current event + */ + int64_t pts; + + /** + * Index of the input that requires a request + */ + unsigned in_request; + + /** + * Flag indicating that a frame event is ready + */ + uint8_t frame_ready; + + /** + * Flag indicating that output has reached EOF. + */ + uint8_t eof; + + /** + * Array of inputs; all inputs must be in consecutive memory + */ + FFFrameSyncIn in[1]; /* must be last field */ +} FFFrameSync; + +/** + * Initialize a frame sync structure. + * + * The entire structure is expected to be already set to 0. + * + * @param fs frame sync structure to initialize + * @param parent parent object, used for logging + * @param nb_in number of inputs + */ +void ff_framesync_init(FFFrameSync *fs, void *parent, unsigned nb_in); + +/** + * Configure a frame sync structure. + * + * Must be called after all options are set but before all use. + * + * @return >= 0 for success or a negative error code + */ +int ff_framesync_configure(FFFrameSync *fs); + +/** + * Add a frame to an input + * + * Typically called from the filter_frame() method. + * + * @param fs frame sync structure + * @param in index of the input + * @param frame input frame, or NULL for EOF + */ +int ff_framesync_add_frame(FFFrameSync *fs, unsigned in, AVFrame *frame); + +/** + * Prepare the next frame event. + * + * The status of the operation can be found in fs->frame_ready and fs->eof. + */ +void ff_framesync_next(FFFrameSync *fs); + +/** + * Drop the current frame event. + */ +void ff_framesync_drop(FFFrameSync *fs); + +/** + * Get the current frame in an input. + * + * @param fs frame sync structure + * @param in index of the input + * @param rframe used to return the current frame (or NULL) + * @param get if not zero, the calling code needs to get ownership of + * the returned frame; the current frame will either be + * duplicated or removed from the framesync structure + */ +int ff_framesync_get_frame(FFFrameSync *fs, unsigned in, AVFrame **rframe, + unsigned get); + +#endif /* AVFILTER_FRAMESYNC_H */ -- 1.7.10.4

4 11