Podsync

Podsync - is a simple, free service that lets you listen to any YouTube / Vimeo channels, playlists or user videos in podcast format.

Podcast applications have a rich functionality for content delivery - automatic download of new episodes, remembering last played position, sync between devices and offline listening. This functionality is not available on YouTube and Vimeo. So the aim of Podsync is to make your life easier and enable you to view/listen to content on any device in podcast client.

Features

• Works with YouTube and Vimeo.
• Supports feeds configuration: video/audio, high/low quality, max video height, etc.
• mp3 encoding
• Update scheduler supports cron expressions
• Episodes filtering (match by title, duration).
• Feeds customizations (custom artwork, category, language, etc).
• OPML export.
• Supports episodes cleanup (keep last X episodes).
• One-click deployment for AWS.
• Runs on Windows, Mac OS, Linux, and Docker.
• Supports ARM.
• Automatic youtube-dl self update.
• Supports API keys rotation.

Dependencies

If you're running the CLI as binary (e.g. not via Docker), you need to make sure that dependencies are available on your system. Currently, Podsync depends on youtube-dl , ffmpeg, and go.

On Mac you can install those with brew:

brew install youtube-dl ffmpeg go

Documentation

• How to get Vimeo API token
• How to get YouTube API Key
• Podsync on QNAP NAS Guide
• Schedule updates with cron

Nightly builds

Nightly builds uploaded every midnight from the main branch and available for testing:

$ docker run -it --rm ghcr.io/mxpv/podsync:nightly

Access tokens

In order to query YouTube or Vimeo API you have to obtain an API token first.

• How to get YouTube API key
• Generate an access token for Vimeo

Configuration

You need to create a configuration file (for instance config.toml) and specify the list of feeds that you're going to host. See config.toml.example for all possible configuration keys available in Podsync.

Minimal configuration would look like this:

[server]
port = 8080

[storage]
  [storage.local]
  data_dir = "/data/podsync/"

[tokens]
youtube = "PASTE YOUR API KEY HERE"

[feeds]
    [feeds.ID1]
    url = "https://www.youtube.com/channel/UCxC5Ls6DwqV0e-CYcAKkExQ"

If you want to hide Podsync behind reverse proxy like nginx, you can use hostname field:

[server]
port = 8080
hostname = "https://my.test.host:4443"

[feeds]
  [feeds.ID1]
  ...

Server will be accessible from http://localhost:8080, but episode links will point to https://my.test.host:4443/ID1/...

One click deployment

How to run

Build and run as binary:

Make sure you have created the file config.toml. Also note the location of the data_dir. Depending on the operating system, you may have to choose a different location since /app/data might be not writable.

$ git clone https://github.com/mxpv/podsync
$ cd podsync
$ make
$ ./bin/podsync --config config.toml

How to debug

Use the editor Visual Studio Code and install the official Go extension. Afterwards you can execute "Run & Debug" ▶︎ "Debug Podsync" to debug the application. The required configuration is already prepared (see .vscode/launch.json).

Run via Docker:

$ docker pull mxpv/podsync:latest
$ docker run \
    -p 8080:8080 \
    -v $(pwd)/data:/app/data/ \
    -v $(pwd)/config.toml:/app/config.toml \
    mxpv/podsync:latest

Run via Docker Compose:

$ docker-compose up

How to make a release

Just push a git tag. CI will do the rest.