ariddell/podcatcher repository overview

⁠Podcatcher

Simple containerized podcatcher (a podcast-retriever) using gPodder⁠ which downloads new episodes daily.

If present, gPodder will subscribe to all the feeds in subscriptions.opml. (Export existing feeds with gpo export subscriptions.opml.)

podcatcher needs a data container at /podcatcher-data, so execute the following sequence of commands:

docker build -t podcatcher .
docker run -v /data --name podcatcher-data podcatcher echo Data-only container for podcatcher
docker run -d --volumes-from podcatcher-data --name podcatcher-worker podcatcher

⁠Backup and restore data volume

There is no way to "mark all as read" from the gPodder command line interface. This means that by default it will download all podcasts. To avoid this you likely want to restore an existing gPodder database. Doing this with docker is not too painful (for details, see Backup, restore, or migrate data volumes⁠).

Backing up the mutable part of this setup, the /data directory can be accomplished with:

docker run --rm --volumes-from podcatcher-data -v $(pwd):/backup busybox tar cvf /backup/backup.tar /data

Restoring /data from backup may be accomplished with:

docker run --rm --volumes-from podcatcher-data -v $(pwd):/backup busybox tar xvf /backup/backup.tar

⁠Getting podcasts out of the data container

The tutum/ubuntu container runs an ssh server that can allow access on an arbitrary port:

docker run -d --volumes-from podcatcher-data --name podcatcher-ssh -p 0.0.0.0:2222:22 tutum/ubuntu:trusty

⁠Subscribing to new podcasts

docker run --rm --volumes-from podcatcher-data podcatcher gpo subscribe http://99percentinvisible.org/feed/

⁠TODO

Get FEEDS from environment variables and do a "mark all as read" on initial run