Joining similar MKV and TS results in broken video

[GrayFace]

Summary of the bug:
ffmpeg can't join two excerpts of the same live stream if one of them is first converted (-c copy) to MKV format. This happens with any recorded *.ts files. I've attached 2 videos as an example, along with concat list and run.bat.
How to reproduce:

% ffmpeg.exe -nostats -safe 0 -f concat -i "List.txt" -c copy "out.mkv"
ffmpeg version 4.3.1 (from https://www.gyan.dev/ffmpeg/builds/)
built on 2021-01-01

[Carl Eugen Hoyos]

Please use the FFmpeg user mailing list for all usage questions: it is not a limitation of FFmpeg that you cannot join many seemingly similar videos.

[GrayFace]

This is a bug, not a usage question.

[GrayFace]

And they're not "seemingly similar", they can be joined if both are left as TS or both are converted (-c copy) to MKV first.

[Carl Eugen Hoyos]

If you want to make this a valid tickets, please:
Attach your input files (no zip)
Test with current FFmpeg git head, the only version supported here, and post the command line you tested together with the complete, uncut console output

[GrayFace]

ffmpeg version 2021-01-20-git-2021dbe1d6-essentials_build-www.gyan.dev Copyright (c) 2000-2021 the FFmpeg developers
  built with gcc 10.2.0 (Rev6, Built by MSYS2 project)
  configuration: --enable-gpl --enable-version3 --enable-static --disable-w32threads --disable-autodetect --enable-fontconfig --enable-iconv --enable-gnutls --enable-libxml2 --enable-gmp --enable-lzma --enable-zlib --enable-libsrt --enable-libssh --enable-libzmq --enable-avisynth --enable-sdl2 --enable-libwebp --enable-libx264 --enable-libx265 --enable-libxvid --enable-libaom --enable-libopenjpeg --enable-libvpx --enable-libass --enable-libfreetype --enable-libfribidi --enable-libvidstab --enable-libvmaf --enable-libzimg --enable-amf --enable-cuda-llvm --enable-cuvid --enable-ffnvcodec --enable-nvdec --enable-nvenc --enable-d3d11va --enable-dxva2 --enable-libmfx --enable-libgme --enable-libopenmpt --enable-libopencore-amrwb --enable-libmp3lame --enable-libtheora --enable-libvo-amrwbenc --enable-libgsm --enable-libopencore-amrnb --enable-libopus --enable-libspeex --enable-libvorbis --enable-librubberband
  libavutil      56. 63.101 / 56. 63.101
  libavcodec     58.117.101 / 58.117.101
  libavformat    58. 65.101 / 58. 65.101
  libavdevice    58. 11.103 / 58. 11.103
  libavfilter     7. 96.100 /  7. 96.100
  libswscale      5.  8.100 /  5.  8.100
  libswresample   3.  8.100 /  3.  8.100
  libpostproc    55.  8.100 / 55.  8.100
[matroska,webm @ 00000000005fdc80] Auto-inserting h264_mp4toannexb bitstream filter
Input #0, concat, from 'List.txt':
  Duration: N/A, start: 0.000000, bitrate: N/A
    Stream #0:0: Video: h264 (High), yuv420p(tv, smpte170m, progressive), 1920x1080 [SAR 1:1 DAR 16:9], 60 fps, 60 tbr, 1k tbn, 120 tbc
    Metadata:
      DURATION        : 00:00:01.050000000
    Stream #0:1: Audio: aac (LC), 44100 Hz, stereo, fltp
    Metadata:
      DURATION        : 00:00:01.022000000                                 19:50
Output #0, matroska, to 'out.mkv':
  Metadata:
    encoder         : Lavf58.65.101
    Stream #0:0: Video: h264 (High) (H264 / 0x34363248), yuv420p(tv, smpte170m, progressive), 1920x1080 [SAR 1:1 DAR 16:9], q=2-31, 60 fps, 60 tbr, 1k tbn, 1k tbc
    Metadata:
      DURATION        : 00:00:01.050000000
    Stream #0:1: Audio: aac (LC) ([255][0][0][0] / 0x00FF), 44100 Hz, stereo, fltp
    Metadata:
      DURATION        : 00:00:01.022000000
Stream mapping:
  Stream #0:0 -> #0:0 (copy)
  Stream #0:1 -> #0:1 (copy)
Press [q] to stop, [?] for help
[matroska @ 0000000002dde940] Starting new cluster due to timestamp
frame=  114 fps=0.0 q=-1.0 Lsize=    1105kB time=00:03:03.83 bitrate=  49.3kbits/s speed=2.65e+03x
video:1063kB audio:40kB subtitle:0kB other streams:0kB global headers:0kB muxing overhead: 0.239883%

[Carl Eugen Hoyos]

To quote the documentation at https://ffmpeg.org/ffmpeg-formats.html#concat-1
All files must have the same streams (same codecs, same time base, etc.).

[GrayFace]

And they do. It says nothing about "container format", which is the only thing that's different between them.

[GrayFace]

What do you think you're doing? Stop acting maliciously!

[GrayFace]

(edited:removed double post)
All your questions have been clearly answered. Bug reports are for the devs to improve the product, not to fight over nothing making up invalid excuses to block them from seeing the bug.

[Gyan]

Varied container formats are supported by the concat demuxer. However, Matrsoka uses a fixed time_base of 1/1000 and MPEG-TS a fixed time_base of 1/90000, and that is not supported by the demuxer, as documented.

[GrayFace]

Replying to Gyan:

Varied container formats are supported by the concat demuxer. However, Matrsoka uses a fixed time_base of 1/1000 and MPEG-TS a fixed time_base of 1/90000, and that is not supported by the demuxer, as documented.

Thanks for the explanation. Though, ffmpeg producing invalid output and reporting that there were no errors still doesn't appear to be the best behavior.

[Gyan]

The concat demuxer could do with input validation.