not everyone uses cmake. it would be proper to at least put the build instructions in the README.md

Hi! I wonder whether the examples are supposed to work in the master branch.
I'm trying the api2-decode-encode-video.cpp

There might be a problem in the Stream::mediaType() method,
since m_raw->codecpar doesn't seem to contain valid data, while m_raw->codec has it, but the former is chosen by preprocessor directive.

Additionally I don't see code where you set the width/height/pix_fmt fields of AVStream's codec/codecpar. Having these unset causes an error like this:
[mp4 @ 000001bae753ba00] dimensions not set
while calling octx.writeHeader(); line of api2-decode-encode-video.cpp

I want to use avcpp in my project and willing to contribute the hardware-accelerated encoding code, but with these basic problems it might not be possible. Am I doing something wrong?

Libavformat version 57.66.102, target platform Windows 10 64-bit (visual studio toolchain is used)

I'm using Debian Stretch amd64, libavformat 57.
These lines block adding a subtitles stream. I can form a valid H264+Vorbis stream packed to MKV but can't add subtitles due to that support check. MKV indeed supports subtitles but the check says otherwise. It could be an ffmpeg bug or something with my local setup. I commented the check out, after that avformat_new_stream successfully adds the subtitles stream and I can encode my subs to it getting a valid MKV file.

The code is very straightforward, like this:

        Codec sub_codec = findEncodingCodec(AV_CODEC_ID_SUBRIP);
        auto sub_stream = ofctx.addStream(sub_codec, ec);
        LOGFATAL("Can't add subtitles stream")
        subctx.openInput(argv[4], ec);
        LOGFATAL("Can't open subtitles")
        subctx.findStreamInfo(ec);
        LOGFATAL("No subtitles streams found")

        ofctx.dump();
        ofctx.writeHeader();
        ofctx.flush();

        while (true) {
            auto pkt = subctx.readPacket(ec);
            if (pkt.isNull()) {
                break;
            }
            LOGERR("Packet reading error")
            pkt.setStreamIndex(2);
            ofctx.writePacket(pkt, ec);
            LOGERR("Packet writing error")
        }

I think we shouldn't explicitly check for stream support and instead rely on avformat_new_stream returning NULL on error.

Hi There,

This project is still using an old implementation of LGTM's automated code review, which has now been disabled. To continue using automated code review, and receive checks on your Pull Requests, please install the GitHub App on this repository.

Thanks,
The LGTM Team

Hello!

I have a question about audio format conversion. Need to create function to convert file from ogg to wav in linux. Could you provide me any instructions how I can do It via avcpp? Thanks!

Nadya

Can library be switched to cmake v3.1 ?

cmake_minimum_required (VERSION 3.1)

and re-use new features like

set(CMAKE_CXX_STANDARD 11)
set(CMAKE_CXX_STANDARD_REQUIRED yes)

etc.

This issue is pretty weird, it happens only when optimizations are on (-O1, -O2, -O3, -Os etc.) and on G++ 6 and up and clang++ 3.5 at least. G++ 5 works fine. So, here's a minimal example:

#include <iostream>
#include <system_error>

using namespace std;

error_code& throws() {
  return *((error_code*)0);
}

void clear_if(error_code& ec) {
  error_code& ec_local = throws();
  cout << &ec << " / " << &throws() << " / " << &ec_local << endl;
  if (&ec != &throws()) {
    cout << "&ec != &throws()" << endl;
  } else {
    cout << "&ec == &throws()" << endl;
  }
  if (&ec != &ec_local) {
    cout << "&ec != &ec_local" << endl;
  } else {
    cout << "&ec == &ec_local" << endl;
  }
  if (&ec_local != &throws()) {
    cout << "&ec_local != &throws()" << endl;
  } else {
    cout << "&ec_local == &throws()" << endl;
  }
  if (&ec == 0) {
    cout << "&ec == 0" << endl;
  } else {
    cout << "&ec != 0" << endl;
  }
  if (&ec != 0) {
    cout << "&ec != 0" << endl;
  } else {
    cout << "&ec == 0" << endl;
  }
  if (&ec_local == 0) {
    cout << "&ec_local == 0" << endl;
  } else {
    cout << "&ec_local != 0" << endl;
  }
  if (&ec_local != 0) {
    cout << "&ec_local != 0" << endl;
  } else {
    cout << "&ec_local == 0" << endl;
  }
}

int main() {
  clear_if(throws());
  return 0;
}

I tried to compile it with the following commands and checked the output:

g++-5 -O0 -std=c++11 test.cpp

0 / 0 / 0
&ec == &throws()
&ec == &ec_local
&ec_local == &throws()
&ec == 0
&ec == 0
&ec_local == 0
&ec_local == 0

g++-6 -O0 -std=c++11 test.cpp

0 / 0 / 0
&ec == &throws()
&ec == &ec_local
&ec_local == &throws()
&ec == 0
&ec == 0
&ec_local == 0
&ec_local == 0

g++-5 -O1 -std=c++11 test.cpp

0 / 0 / 0
&ec == &throws()
&ec == &ec_local
&ec_local == &throws()
&ec == 0
&ec == 0
&ec_local == 0
&ec_local == 0

g++-6 -O1 -std=c++11 test.cpp

0 / 0 / 0
&ec != &throws()
&ec != &ec_local
&ec_local == &throws()
&ec == 0
&ec != 0
&ec_local == 0
&ec_local == 0

clang++ -O0 -std=c++11 test.cpp

0 / 0 / 0
&ec == &throws()
&ec == &ec_local
&ec_local == &throws()
&ec == 0
&ec == 0
&ec_local == 0
&ec_local == 0

clang++ -O1 -std=c++11 test.cpp

0 / 0 / 0
&ec != &throws()
&ec != &ec_local
&ec_local == &throws()
&ec != 0
&ec != 0
&ec_local != 0
&ec_local != 0

I suspect making a null reference is an undefined behavior and that's why it breaks. Note, that only null reference arguments get borked somehow, local variables are compared just fine. Still, UB is UB and we shouldn't rely on it. As you can see, &ec both equals and not equals zero, it's undefined. The first branch is always executed as the condition is always true.

The exact issue in avcpp is that this line is executed while ec is a null reference that causes a segfault if the library is compiled with G++ 6 and up in a release configuration (which usually assumes -O3).

I'm using avcpp and I really like it!
I'm decoding mp4 (h264+acc) and aac decoder returns AV_SAMPLE_FMT_FLTP (planar floats). In my case I have stereo aac and when I get a decoded audio av::AudioSamples frame values that I get for samples count size etc are all confusing.

these are the values that I have:

aframe.raw.nb_samples 1024
aframe.channelsCount() => 2
aframe.size() => 16384
aframe.sampleFormat().bytesPerSample() => 1 (?!)
aframe.sampleBitDepth() => 8 (?!)
aframe->linesize[0] = 8192

aframe.raw.buf[0] and buf[1] both have size 8192, however, looking at the memory in debugger it seems that samples actually take half of that buffer. Either it's ffmpeg allocates more (padding, rounding) or it's a bug? For 1024 FLTP sample it should be 4K in each plane or total 8K for 2 channels.

IMO, aframe.size() should return size of actual samples stored, and but currently it seem to return size of allocated memory. In any case, bytes per sample seems to be completely wrong there (it should be 32 or 64 if a sample is counted for all channels).

If input context close and Packet still use SEGFAULT occured. Need check.

Add ability to copy metadata from input sources to the output one.

Metadata aware objects:

1. AVFormatContext (FormatContext)
2. AVChapter (#15)
3. AVProgram (#16)
4. AVStream (Stream2)
5. AVFrame (VideoFrame2, AudioSamples2, av_frame_get_metadata(), av_frame_copy_props())

Note, that metadata of the FormatContext and Stream2 can be changed during playback. Refer to the FormatContext::eventFlags() and Stream2::eventFlags() family methods.

What I want to do:

`
Packet pkt = ictx.readPacket(ec);
AudioSamples samples = running->adec->decode(pkt, ec);

resampler1.push(samples);
resampler2.push(samples);
resampler3.push(samples);
`

Where each of those resamplers is configured to a different sampling rate and sample format.
I tried cloning the samples before feeding them into the resampler, still not working.

Is that at all possible?

Hey there! It's really great project.

diff --git a/src/videorescaler.h b/src/videorescaler.h
index 169eb84..085f86e 100644
--- a/src/videorescaler.h
+++ b/src/videorescaler.h
@@ -55,8 +55,8 @@ private:

 };

-typedef std::shared_ptr<VideoRescaler> VideoResamplerPtr;
-typedef std::weak_ptr<VideoRescaler> VideoResamplerWPtr;
+typedef std::shared_ptr<VideoRescaler> VideoRescalerPtr;
+typedef std::weak_ptr<VideoRescaler> VideoRescalerWPtr;

 } // ::av

Hello!
Firstly thanks a lot for a great work.
I have a little remark -- I think git cloning should be made in recursive way to clone catch2, isn't It?
Otherwise during cmake It will crash in some cases and give something like

CMake Error at tests/CMakeLists.txt:20 (include):
include could not find load file:

Catch

Best regards,
Nadya

This field (and AVFrame::nb_extended_buf) allows to store more than AV_NUM_DATA_POINTERS planes for multiplanar audio.

Look into AVFrame::extended_data (relative AVFrame::data). Size of the AVFrame::extended_data is AV_NUM_DATA_POINTERS+AVFrame::nb_extended_buf

Hi,
I am upgrading to the latest version. It looks like Timestamp replace TimeStamp. Now I get lipsync problems with my video player.

I've made a simple encoding app, with code very similar to api2-decode-encode-video.cpp.
The only difference is that I don't really need an output Stream, since I want to manually operate on encoded packets.
The first problem that I see is that there is no VideoEncoderContext constructor which doesn't require a Stream parameter (I mean there is, but it will lead to problems, no such example provided). So maybe there should be a constructor for such scenario, shouldn't it?

Even when I create a fake output Stream, avcodec_receive_frame (codeccontext.cpp, int decode(AVCodecContext *avctx, AVFrame *picture, int *got_picture_ptr, const AVPacket *avpkt), returns AVERROR(EAGAIN) during roughly first 50 decode() calls, although I seem to provide correct input data to it.

I assume it should return valid packet every time a new frame is provided to avcodec_send_packet, shouldn't it? Any ideas on that?

Libavcodec version 57.80.100, target platform Windows 10 64-bit (visual studio toolchain is used)

Make more debugable with output to the streams. Process special values like AV_NOPTS_VALUE

Add classess to work with AVDictionary.

Rework all classes that can accept parameters via dictionary. Like: FormatContext::openInput().

Ref: https://www.ffmpeg.org/doxygen/2.7/group__lavu__dict.html

At the moment windows is not tested despite the large developer/user base.
Here are some steps to get basic windows support working.

• figure out how to install windows dependencies
• update Ci to also build for windows
• test meson windows build (should work out of the box)
• test cmake windows build (maybe also works out of the box)

Add support for AVChapter reading and adding to the streams.

Also, add ability "transmux" chapters.

/**
 * the picture in the bitstream
 * - encoding: Set by libavcodec.
 * - decoding: unused
 *
 * @deprecated use the quality factor packet side data instead
 */
attribute_deprecated AVFrame *coded_frame;

Additional info:
https://ffmpeg.org/pipermail/ffmpeg-cvslog/2015-July/092046.html

I am trying to build a tool to convert a bunch of png's into a video. I figured the best place to start was the example api2-decode-encode-video.cpp but I don't understand how to use the example. I compiled it using homebrew ffmpeg 4.0.2 on MacOS and then run with:

./api2-decode-encode-video input.png out.mkv

The line above always crashes when initiating the VideoEncoderContext with an error:

libc++abi.dylib: terminating with uncaught exception of type av::Exception: Codec context invalid media type

I tried other output names but get the same error. How am I supposed to use this example?

I've tried writing a simple video encoder with the source image data being read from a buffer in memory, but I've noticed that it would only encode the very first row of my image data and the rest would be black.
After inspecting the source code, I've noticed this in the constructor for VideoFrame:

VideoFrame::VideoFrame(const uint8_t *data, size_t size, PixelFormat pixelFormat, int width, int height, int align)
{
    [...]
    for (size_t i = 0; i < 4 && buf[i]; ++i) {
        std::copy(buf[i], buf[i]+linesize[i], m_raw->data[i]);
    }
}

I'm new to ffmpeg, but this looks wrong to me. linesize only contains the size of a single row, so this would only copy the very first row from the source buffer, which would explain what I got as a result.

As a test, I've tried overwriting the data manually (memcpy) after creating the frame:

auto width = encoder.width();
auto height = encoder.height();
using Color = std::array<uint8_t,3>;
std::vector<Color> colors {};
colors.resize(width *height,Color{0,255,255});

auto pxFormatRGB = AVPixelFormat::AV_PIX_FMT_RGB24;
uint32_t alignment = 32u;
av::VideoFrame frame{reinterpret_cast<uint8_t*>(colors.data()),colors.size() *sizeof(colors.front()),pxFormatRGB,static_cast<int32_t>(width),static_cast<int32_t>(height),static_cast<int32_t>(alignment)};

memcpy(frame.raw()->data[0],colors.data(),colors.size() *sizeof(colors.front()));
frame.raw()->pts = 0;
frame.raw()->pkt_dts = 0;
frame.setComplete(true);

This is just a dirty work-around (and doesn't take any line padding into account), but it fixed the issue for my test case. I'm still mostly unfamiliar with ffmpeg, so I haven't created a pull request, since I'm not sure what a proper fix would be. Or maybe I just did something wrong altogether and this is not actually a bug?

It is quite common for people to use ffmpeg to cut videos, as detailed on:
https://trac.ffmpeg.org/wiki/Seeking

It would be awesome for avcpp to support a similar feature. Ideally, the user could just pass the starting point in the video and the duration of the section.

I think should use av_frame_free() instead

I want to play the video at twice the speed. can I use this project?

Good things. But I have no time and ffmpeg-related project for this time to implement it, so any help will be welcomed.

My preferred C++ test framework is Catch2

1. Use current implementation of the https://github.com/h4tr3d/avcpp/tree/master/example/api2-samples as base for tests.
2. Create another repository that can be submoduled via master repo and will contains media data for tests. Good base for media data is Big Buck Bunny.
3. Add tests running to the CI configuration.

I tried to call this function but the linker says symbol not found. I was not able to find the implementation of this function in the cpp file either.

PS. Is this a good workaround?

av::FormatContext inputContext;
inputContext.openInput(&customIO, av::InputFormat());

Hello

As in subject when uninstalled I can't use library, because headers can't be included correctly:

#include <avcpp/av.h>
// or
#include "avcpp/av.h"

I am aware of 3 potential solutions to that, two simple, one not that simple:

• move headers to include/avcpp directory
• move all sources to src/avcpp directory
• copy headers to builddir: my meson workaround, example of samples (breaks cmake)

I can live with my 'copy headers' workaround, didn't care yet to implement it for cmake though.

Current master must be saved as a 'legacy-api' branch only fixes.

Now this API is deprecated.

Hello, I want to compile avcpp in Windows. Can you support the PkgConfig module lookup for Windows FFmpeg? cmake tool reported wrong:

Call Stack (most recent call first):
CMake Error at cmake/FindFFmpeg.cmake:186 (set_target_properties):
set_target_properties called with incorrect number of arguments.
Call Stack (most recent call first):
CMakeLists.txt:20 (find_package)
CMake Error at D:/develop/CMake/share/cmake-3.14/Modules/FindPackageHandleStandardArgs.cmake:137 (message):
Could NOT find FFmpeg (missing: FFMPEG_LIBRARIES FFMPEG_INCLUDE_DIRS
AVCODEC_LIBRARIES AVCODEC_INCLUDE_DIRS AVFORMAT_LIBRARIES
AVFORMAT_INCLUDE_DIRS AVUTIL_LIBRARIES AVUTIL_INCLUDE_DIRS
AVDEVICE_LIBRARIES AVDEVICE_INCLUDE_DIRS AVFILTER_LIBRARIES
AVFILTER_INCLUDE_DIRS SWSCALE_LIBRARIES SWSCALE_INCLUDE_DIRS
SWRESAMPLE_LIBRARIES SWRESAMPLE_INCLUDE_DIRS)
Call Stack (most recent call first):
D:/develop/CMake/share/cmake-3.14/Modules/FindPackageHandleStandardArgs.cmake:378 (_FPHSA_FAILURE_MESSAGE)
cmake/FindFFmpeg.cmake:204 (find_package_handle_standard_args)
CMakeLists.txt:20 (find_package)

I'm doing simple remux and replaced raw ffmpeg with your library but couldnt find a way to do this without accessing raw() ?

        const auto srcVideoStream = formatCtx.stream(videoStream).raw();
        videoOutStream = outCtx.addStream(av::Codec(srcVideoStream->codec->codec));
        if (outCtx.outputFormat().isFlags(AVFMT_GLOBALHEADER))
            videoOutStream.raw()->codec->flags |= AV_CODEC_FLAG_GLOBAL_HEADER;

        videoOutStream.setTimeBase(srcVideoStream->codec->time_base);

Also doing outCtx.dump() gives me Stream #0:0: Unknown: none So not sure if i even did that correctly ?

hey, you wrote a great library which made using ffmpeg a lot easier.
Does this library support hwaccel?

The examples refer to Container.h, StreamCoder.h and similar. Where can I find these files?

Hi,
i found that libavcpp.pc file have a mistake in includedir,it is includedir = ${prefix}/include now.In this situation,we cant include header file directly like the way in example #include "av.h" instead of #include "avcpp/av.h"
could you please fix this problem,or im very glad to fix it
sorry im not a native english speaker.....

Return value, pass ec variable via reference

Note, that this changes breaks API. Code must be adopted to work with new error handling.
Suggest: do not use master branch until this issue closed. You can use stable branch api-2.0 (recommened) or api-legacy (not recommended).

Ref: www.cplusplus.com/reference/system_error/error_code

In Ubuntu 18.04.4 Desktop I installed avcpp following these indications: https://github.com/h4tr3d/avcpp#ubuntu-and-linux-mint

(base) marco@pc01:~$ git clone --recurse-submodules https://github.com/h4tr3d/avcpp.git avcpp-
git
Cloning into 'avcpp-git'...
remote: Enumerating objects: 45, done.
remote: Counting objects: 100% (45/45), done.
remote: Compressing objects: 100% (35/35), done.
remote: Total 1605 (delta 16), reused 14 (delta 6), pack-reused 1560
Receiving objects: 100% (1605/1605), 523.42 KiB | 1.56 MiB/s, done.
Resolving deltas: 100% (1110/1110), done.
Submodule 'catch2' (https://github.com/catchorg/Catch2.git) registered for path 'catch2'
Cloning into '/home/marco/avcpp-git/catch2'...
remote: Enumerating objects: 110, done.
remote: Counting objects: 100% (110/110), done.
remote: Compressing objects: 100% (92/92), done.
remote: Total 26279 (delta 34), reused 43 (delta 18), pack-reused 26169
Receiving objects: 100% (26279/26279), 15.91 MiB | 9.42 MiB/s, done.
Resolving deltas: 100% (18294/18294), done.
Submodule path 'catch2': checked out 'f0b7b0ca119d8cf58c2b89dea663f921dbdbd332'

(base) marco@pc01:~/avcpp-git$ mkdir build
(base) marco@pc01:~/avcpp-git$ cd build
(base) marco@pc01:~/avcpp-git/build$ cmake ..
-- The CXX compiler identification is GNU 9.2.1
-- Check for working CXX compiler: /usr/bin/c++
-- Check for working CXX compiler: /usr/bin/c++ -- works
-- Detecting CXX compiler ABI info
-- Detecting CXX compiler ABI info - done
-- Detecting CXX compile features
-- Detecting CXX compile features - done
-- Looking for C++ include pthread.h
-- Looking for C++ include pthread.h - found
-- Performing Test CMAKE_HAVE_LIBC_PTHREAD
-- Performing Test CMAKE_HAVE_LIBC_PTHREAD - Failed
-- Check if compiler accepts -pthread
-- Check if compiler accepts -pthread - yes
-- Found Threads: TRUE
-- Found PkgConfig: /usr/bin/pkg-config (found version "0.29.1")
-- Checking for module 'libavcodec'
--   Found libavcodec, version 58.54.100
-- Checking for module 'libavformat'
--   Found libavformat, version 58.29.100
-- Checking for module 'libavdevice'
--   Found libavdevice, version 58.8.100
-- Checking for module 'libavutil'
--   Found libavutil, version 56.31.100
-- Checking for module 'libavfilter'
--   Found libavfilter, version 7.57.100
-- Checking for module 'libswscale'
--   Found libswscale, version 5.5.100
-- Checking for module 'libpostproc'
--   Found libpostproc, version 55.5.100
-- Checking for module 'libswresample'
--   Found libswresample, version 3.5.100
-- Found FFmpeg: /usr/lib/x86_64-linux-gnu/libavcodec.so;/usr/lib/x86_64-linux-gnu/libavformat.so;
/usr/lib/x86_64-linux-gnu/libavutil.so;/usr/lib/x86_64-linux-gnu/libavdevice.so;/usr/lib/x86_64-linux-
gnu$
-- /home/marco/avcpp-git/catch2/contrib
-- Configuring done
-- Generating done
-- Build files have been written to: /home/marco/avcpp-git/build
(base) marco@pc01:~/avcpp-git/build$ sudo make -j8
Scanning dependencies of target test_main
Scanning dependencies of target avcpp
Scanning dependencies of target avcpp_static
[  1%] Building CXX object tests/CMakeFiles/test_main.dir/test-main.cpp.o
[  4%] Building CXX object src/CMakeFiles/avcpp.dir/averror.cpp.o
....
[100%] Linking CXX executable test_executor
/usr/bin/ld: warning: libavutil.so.55, needed by //usr/lib/x86_64-linux-gnu/libchromaprint.so.1, may 
conflict with libavutil.so.56
[100%] Built target test_executor
(base) marco@pc01:~/avcpp-git/build$

(base) marco@pc01:~/avcpp-git/build$ sudo make install
[ 33%] Built target avcpp_static
[ 67%] Built target avcpp
[ 70%] Built target api2-decode
[ 72%] Built target api2-decode-audio
[ 75%] Built target api2-scale-video
[ 78%] Built target api2-remux
[ 81%] Built target api2-decode-rasample-audio
[ 83%] Built target api2-decode-encode-audio
[ 86%] Built target api2-demux-seek
[ 89%] Built target api2-decode-encode-video
[ 91%] Built target api2-dict-basic
[ 94%] Built target api2-timestamp
[ 97%] Built target test_main
[100%] Built target test_executor
Install the project...
-- Install configuration: ""
-- Installing: /usr/local/lib/libavcpp.so
-- Installing: /usr/local/lib/libavcpp.a
-- Installing: /usr/local/include/avcpp
-- Installing: /usr/local/include/avcpp/stream.h
-- Installing: /usr/local/include/avcpp/packet.h
-- Installing: /usr/local/include/avcpp/dictionary.h
-- Installing: /usr/local/include/avcpp/sampleformat.h
-- Installing: /usr/local/include/avcpp/pixelformat.h
-- Installing: /usr/local/include/avcpp/codeccontext.h
-- Installing: /usr/local/include/avcpp/rational.h
-- Installing: /usr/local/include/avcpp/timestamp.h
-- Installing: /usr/local/include/avcpp/ffmpeg.h
-- Installing: /usr/local/include/avcpp/frame.h
-- Installing: /usr/local/include/avcpp/videorescaler.h
-- Installing: /usr/local/include/avcpp/avtime.h
-- Installing: /usr/local/include/avcpp/filters
-- Installing: /usr/local/include/avcpp/filters/filterpad.h
-- Installing: /usr/local/include/avcpp/filters/filtercontext.h
-- Installing: /usr/local/include/avcpp/filters/buffersink.h
-- Installing: /usr/local/include/avcpp/filters/filter.h
-- Installing: /usr/local/include/avcpp/filters/buffersrc.h
-- Installing: /usr/local/include/avcpp/filters/filtergraph.h
-- Installing: /usr/local/include/avcpp/formatcontext.h
-- Installing: /usr/local/include/avcpp/av.h
-- Installing: /usr/local/include/avcpp/avutils.h
-- Installing: /usr/local/include/avcpp/audioresampler.h
-- Installing: /usr/local/include/avcpp/linkedlistutils.h
-- Installing: /usr/local/include/avcpp/format.h
-- Installing: /usr/local/include/avcpp/avlog.h
-- Installing: /usr/local/include/avcpp/codec.h
-- Installing: /usr/local/include/avcpp/rect.h
-- Installing: /usr/local/include/avcpp/averror.h
-- Installing: /usr/local/lib/pkgconfig/libavcpp.pc
-- Installing: /usr/local/lib/cmake/Catch2/Catch2Targets.cmake
-- Up-to-date: /usr/local/include
-- Installing: /usr/local/include/catch2
-- Installing: /usr/local/include/catch2/catch.hpp
-- Installing: /usr/local/include/catch2/catch_reporter_teamcity.hpp
-- Installing: /usr/local/include/catch2/catch_reporter_automake.hpp
-- Installing: /usr/local/include/catch2/catch_reporter_tap.hpp
-- Installing: /usr/local/lib/cmake/Catch2/Catch2Config.cmake
-- Installing: /usr/local/lib/cmake/Catch2/Catch2ConfigVersion.cmake
-- Installing: /usr/local/share/doc/Catch2
-- Installing: /usr/local/share/doc/Catch2/ci-and-misc.md
-- Installing: /usr/local/share/doc/Catch2/configuration.md
-- Installing: /usr/local/share/doc/Catch2/limitations.md
-- Installing: /usr/local/share/doc/Catch2/release-notes.md
-- Installing: /usr/local/share/doc/Catch2/test-cases-and-sections.md
-- Installing: /usr/local/share/doc/Catch2/command-line.md
-- Installing: /usr/local/share/doc/Catch2/tutorial.md
-- Installing: /usr/local/share/doc/Catch2/logging.md
-- Installing: /usr/local/share/doc/Catch2/opensource-users.md
-- Installing: /usr/local/share/doc/Catch2/test-fixtures.md
-- Installing: /usr/local/share/doc/Catch2/release-process.md
-- Installing: /usr/local/share/doc/Catch2/cmake-integration.md
-- Installing: /usr/local/share/doc/Catch2/benchmarks.md
-- Installing: /usr/local/share/doc/Catch2/slow-compiles.md
-- Installing: /usr/local/share/doc/Catch2/own-main.md
-- Installing: /usr/local/share/doc/Catch2/assertions.md
-- Installing: /usr/local/share/doc/Catch2/deprecations.md
-- Installing: /usr/local/share/doc/Catch2/reporters.md
-- Installing: /usr/local/share/doc/Catch2/event-listeners.md
-- Installing: /usr/local/share/doc/Catch2/generators.md
-- Installing: /usr/local/share/doc/Catch2/commercial-users.md
-- Installing: /usr/local/share/doc/Catch2/tostring.md
-- Installing: /usr/local/share/doc/Catch2/Readme.md
-- Installing: /usr/local/share/doc/Catch2/other-macros.md
-- Installing: /usr/local/share/doc/Catch2/matchers.md
-- Installing: /usr/local/share/doc/Catch2/list-of-examples.md
-- Installing: /usr/local/share/doc/Catch2/why-catch.md
-- Installing: /usr/local/share/doc/Catch2/contributing.md
-- Installing: /usr/local/lib/cmake/Catch2/ParseAndAddCatchTests.cmake
-- Installing: /usr/local/lib/cmake/Catch2/Catch.cmake
-- Installing: /usr/local/lib/cmake/Catch2/CatchAddTests.cmake
-- Installing: /usr/local/share/Catch2/gdbinit
-- Installing: /usr/local/share/Catch2/lldbinit
-- Installing: /usr/local/share/pkgconfig/catch2.pc

(base) marco@pc01:/usr/local/include$ sudo mv avcpp av

But when trying to compile some examples I get "No such file or directory":

(base) marco@pc01:~/avcpp-git/example/test-basic$ g++ -std=c++17 -I /usr/local/include/av/ av-
main.cpp -L /usr/local/lib/libavcpp.so -o av-main
av-main.cpp:7:10: fatal error: av/containerformat.h: No such file or directory
    7 | #include "av/containerformat.h"
av-main.cpp:8:10: fatal error: av/container.h: No such file or directory
av-main.cpp:10:10: fatal error: av/streamcoder.h: No such file or directory

The same problem for another example:

(base) marco@pc01:~/avcpp-git/example/sample-view$ g++ -std=c++17 -I /usr/local/include/av/    
av-view.cpp -L /usr/local/lib/libavcpp.so -o av-view
av-view.cpp:5:10: fatal error: SDL/SDL.h: No such file or directory
    5 | #include <SDL/SDL.h>
       |          ^~~~~~~~~~~


(base) marco@pc01:~/avcpp-git/example/sample-view$ cd build
(base) marco@pc01:~/avcpp-git/example/sample-view/build$ cmake .. 
CMake Warning (dev) in CMakeLists.txt:
  No project() command is present.  The top-level CMakeLists.txt file must
  contain a literal, direct call to the project() command.  Add a line of
  code such as

    project(ProjectName)

  near the top of the file, but after cmake_minimum_required().

  CMake is pretending there is a "project(Project)" command on the first
  line.
This warning is for project developers.  Use -Wno-dev to suppress it.

-- The C compiler identification is GNU 9.2.1
-- The CXX compiler identification is GNU 9.2.1
-- Check for working C compiler: /usr/bin/cc
-- Check for working C compiler: /usr/bin/cc -- works
-- Detecting C compiler ABI info
-- Detecting C compiler ABI info - done
-- Detecting C compile features
-- Detecting C compile features - done
-- Check for working CXX compiler: /usr/bin/c++
-- Check for working CXX compiler: /usr/bin/c++ -- works
-- Detecting CXX compiler ABI info
-- Detecting CXX compiler ABI info - done
-- Detecting CXX compile features
-- Detecting CXX compile features - done
CMake Error at CMakeLists.txt:13 (pkg_search_module):
  Unknown CMake command "pkg_search_module".

CMake Warning (dev) in CMakeLists.txt:
  No cmake_minimum_required command is present.  A line of code such as

    cmake_minimum_required(VERSION 3.16)

  should be added at the top of the file.  The version specified may be lower
  if you wish to support older CMake versions for this project.  For more
  information run "cmake --help-policy CMP0000".
This warning is for project developers.  Use -Wno-dev to suppress it.

-- Configuring incomplete, errors occurred!
See also "/home/marco/avcpp-git/example/sample-view/build/CMakeFiles/CMakeOutput.log".

How to solve the problem?

Stacktrace:

Encoder function body:
https://pastebin.com/JwHeTAc3

If you need more info, i'm ready to provide it

Add support for AVProgram reading and adding to the streams.

Also, add ability "transmux" programs.

Add classes to work with FFmpeg filters. Build graphs.

One of the most important things is missing at the moment and that is some at least half decent documentation.

Here are a few suggestions on what should be done, feel free to propose additions in the comments:

• adding a doxyfile to build html documentation (#69)
• build and publish the documentation to github pages or something different (github actions or maybe travis)
• document all functions (javadoc style)
• document every type, enum and class (javadoc style)
• properly document the examples
• write some kind of getting started/beginners guide

I noticed you require swscale in your cmake, and also include its header at one point. swscale is effectively gpl, so that conflicts with the proposed bsd licensing.

I have a simple app with this decoding method:

      ```
      static int decode(AVCodecContext *avctx, AVFrame *frame, int *got_frame, AVPacket *pkt)
      {
      int ret;
      *got_frame = 0;
      if (pkt) 
      {
                ret = avcodec_send_packet(avctx, pkt);
                if (ret < 0) 
                {
                          return ret == AVERROR_EOF ? 0 : ret;
                }
      }
      ret = avcodec_receive_frame(avctx, frame);
      if (ret < 0 && ret != AVERROR(EAGAIN) && ret != AVERROR_EOF) 
      {
                return ret;
      }
      if (ret >= 0)
      {
                *got_frame = 1;
      }
      return 0;
      }

I use this method to store video clip but there are a few frames in the video clip with errors or 'miss frames' from previous clip.

avcodec_receive_frame(avctx, frame); - returns code '-11' that stands for 'other negative values: legitimate decoding errors'. What the 'other negative values' are?

What's the best way to use your library from another cmake project ?

I'm a beginner at cpp, but from what i understood of cmake, shouldn't I be able to find either a FindAVCPP.cmake or AVCPPConfig.cmake file ?

Thanks.

Hello,
I hope someone could help me. The background is: I use the xupnpd2-mediaserver (https://github.com/clark15b/xupnpd2) on my router to display some hls-streams on the TV-device. I try to receive the hls-streams from https://www.mall.tv/zive. I solved the missing https-support in xupnpd2 by using haproxy as reverse proxy. I'm not a developer, but I think there is an "error" or missing part in the m3u8 parsing/decoding (this file: https://github.com/clark15b/xupnpd2/blob/master/plugin_hls_common.cpp), because the stream https://www.mall.tv/planespotting is 4 hours behind the actual time.

I contacted the xupnpd2's original-author several times to extend/fix the hls/m3u8-decoding, but never got a reply.

My idea is to create an hls/m3u8-playlist decoder for xupnpd2 using a c++ ffmpeg wrapper. I hope you could help me, because I'm not a developer.

Best regards,
Thomas

Access to the objects (AVClass related) options. Rework all related classes.

Code:
Dictionary::RawStringPtr Dictionary::toRawStringPtr(const char keyValSep, const char pairsSep, OptionalErrorCode ec) const
{
RawStringPtr str;
}

\MSVC\14.20.27508\include\memory(1934): error C2664: 'void av::Dictionary::RawStringDeleter::operator ()(char *&)': cannot convert argument 1 from 'char *' to 'char *&'
\MSVC\14.20.27508\include\memory(1932): note: while compiling class template member function 'std::unique_ptr<char,av::Dictionary::RawStringDeleter>::~unique_ptr(void) noexcept'
src\dictionary.cpp(229): note: see reference to function template instantiation 'std::unique_ptr<char,av::Dictionary::RawStringDeleter>::~unique_ptr(void) noexcept' being compiled
src\dictionary.cpp(228): note: see reference to class template instantiation 'std::unique_ptr<char,av::Dictionary::RawStringDeleter>' being compiled
1>format.cpp

/Users/jeroen/workspace/avcpp/src/codec.cpp:33:28: warning: implicit conversion of nullptr constant to 'bool' [-Wnull-conversion]
    return RAW_GET(decode, nullptr);
    ~~~~~~                 ^~~~~~~
                           false
1 warning generated.

And also

/Users/jeroen/workspace/avcpp/example/api2-samples/api2-decode-rasample-audio.cpp:168:30: warning: using the result of an assignment as a condition without parentheses
      [-Wparentheses]
            while (ouSamples = resampler.pop(samples.samplesCount()/2, ec)) {
                   ~~~~~~~~~~^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
/Users/jeroen/workspace/avcpp/example/api2-samples/api2-decode-rasample-audio.cpp:168:30: note: place parentheses around the assignment to silence this warning
            while (ouSamples = resampler.pop(samples.samplesCount()/2, ec)) {
                             ^
                   (                                                      )
/Users/jeroen/workspace/avcpp/example/api2-samples/api2-decode-rasample-audio.cpp:168:30: note: use '==' to turn this assignment into an equality comparison
            while (ouSamples = resampler.pop(samples.samplesCount()/2, ec)) {
                             ^
                             ==
1 warning generated.