Podcast Archiver

A fast and simple command line client to archive all episodes from your favorite podcasts.

Podcast Archiver takes the feed URLs of your favorite podcasts and downloads all available episodes for you—even those "hidden" in paged feeds. You'll end up with a complete archive of your shows. The archiver also supports updating an existing archive, so that it lends itself to be set up as a cronjob.

Setup

Install via pipx:

pipx install podcast-archiver

Install via brew:

brew install janw/tap/podcast-archiver

Or use it via Docker:

docker run --tty --rm ghcr.io/janw/podcast-archiver --help

By default, the docker image downloads episodes to a volume mounted at /archive.

Usage

Run podcast-archiver --help for details on how to use it:

Example invocation

podcast-archiver -d ~/Music/Podcasts \
    -f http://logbuch-netzpolitik.de/feed/m4a \
    -f http://raumzeit-podcast.de/feed/m4a/ \
    -f https://feeds.lagedernation.org/feeds/ldn-mp3.xml

This way, you can easily add and remove feeds to the list and let the archiver fetch the newest episodes for example by adding it to your crontab.

Feeds can also be "fetched" from a local file:

podcast-archiver -f file:/Users/janw/downloaded_feed.xml

Continuous mode

When the --sleep-seconds option is set to a non-zero value, Podcast Archiver operates in continuous mode. After successfully populating the archive, it will not terminate but rather sleep for the given number of seconds until it refreshes the feeds again and downloads episodes that have been published in the meantime.

If no new episodes have been published, no download attempts will be made, and the archiver will go to sleep again. This mode of operation is ideal to be run in a containerized setup, for example using docker compose:

services:
  podcast-archiver:
    restart: always
    image: ghcr.io/janw/podcast-archiver
    volumes:
      - ./archive:/archive
    command:
      - --sleep-seconds=3600  # sleep for 1 hour between updates
      - --feed=https://feeds.feedburner.com/TheAnthropoceneReviewed
      - --feed=https://feeds.megaphone.fm/heavyweight-spot

Changing the filename format

Podcast Archiver has a --filename-template option that allows you to change the particular naming scheme of the archive. The default value for --filename-template. is shown in podcast-archiver --help, as well as all the available variables. The basic ones are:

• Episode: episode.title, episode.subtitle, episode.author, episode.published_time, episode.original_filename
• Podcast: show.title, show.subtitle, show.author, show.language

Note here that episode.published_time is a Python-native datetime, so its exact format can be adjusted further a la {episode.published_time:%Y-%m-%d %H%M%S} using strftime-placeholders. By default it uses %Y-%m-%d (e.g. 2024-12-31).

Examples

• More precise published time

  {show.title}/{episode.published_time:%Y-%m-%d %H%M%S %Z} - {episode.title}.{ext}

  Results in …/That Show/2023-03-12 123456 UTC - Some Episode.mp3

• Using the original filename (roughly equivalent to pre-1.0 --subdirs)

  {show.title}/{episode.original_filename}

  Results in …/That Show/ts001-episodefilename.mp3

• Using the original filename (roughly equivalent to pre-1.0 --subdirs + --date-prefix)

  {show.title}/{episode.published_time} {episode.original_filename}

  Results in …/That Show/2023-03-12 ts001-episodefilename.mp3

Using a config file

Command line arguments can be replaced with entries in a YAML configuration file. An example config can be generated with

podcast-archiver --config-generate > config.yaml

After modifying the settings to your liking, podcast-archiver can be run with

podcast-archiver --config config.yaml

Alternatively (for example, if you're running podcast-archiver in Docker), you may point it to the config file using the PODCAST_ARCHIVER_CONFIG=path/to/config.yaml environment variable.

Using environment variables

Some settings of Podcast Archiver are available as environment variables, too. Check podcast-archiver --help for options with env var: … next to them.