search

johnb-7 / hdhr-ac4

This project aims to emulate an HDHomerun tuner that supplies ATSC 3.0 programs with AC3 audio.
Apache License 2.0
55 stars 12 forks source link

Trying to figure out how to use binary in other docker containers #3

Closed madasus closed 1 year ago

madasus commented 1 year ago

Hello,

I'd like to be able to use the ffmpeg binary in other docker containers. If all docker containers are using the same OS (ubuntu:latest for example) should I be able to copy the libraries from /usr/lib and the ffmpeg executable from /usr/bin/ffmpeg into the new container?

I did try this and it seemed to run but the AC4 appeared not to be processed when pushing a stream into ffmpeg.

  1. Do you think this is possible? am I missing a step?
  2. I tried moving the Dockerfile to ubuntu:latest to make sure they match but then it complains that python isn't available. I changed the apt-get to use python3 rather than python but now i get other errors that I am unclear how to resolve

checking for xcb-proto >= 1.13... no configure: error: Package requirements (xcb-proto >= 1.13) were not met:

Variable 'PYTHON_PREFIX' not defined in '/opt/ffmpeg/lib/pkgconfig/xcb-proto.pc'

--UPDATE1-- I tried changing the deps line from python to python2 but i get the same error still

Additional question. If i'm trying to do the transcoding in another docker and use this ffmpeg executable is there something obvious i am missing here? I believe the -c:a ac3 is the key component here to convert the ac4 to ac3?

-hide_banner -loglevel error -i [URL] -c:a ac3 -c:v copy -f mpegts pipe:1

appreciate any assistance.

M

madasus commented 1 year ago

Oh no - it looks like the ffmpeg repo for ac-4 has been deleted. Does anyone have it cached?

taegge commented 1 year ago

Try https://github.com/makrsmark/FFmpeg/tree/ac4 for an alternate FFmpeg AC-4 fork.

johnb-7 commented 1 year ago

v1.5.0rc now copies all the needed binary files from Emby. No need for a full ffmpeg build anymore. You can easily see what I am copying from Emby and do something similar.