pycaster

A tool for the automated generation of podcast feeds with a built-in upload functionality for new episodes

Features

• Upload your freshly produced podcast episode via an interactive CLI
• Alternatively, use arguments when interacting with the CLI
• Automagically create a podcast feed via episode uploads and keep it updated with each new upload
• Supports AWS S3, Google Cloud Storage, Digital Ocean Spaces, and every other provider that is AWS S3 API-compatible
• Automatically creates backups of the local episode database which is used for generating the feed
• Offers built-in protection against overwriting episodes that were already uploaded in the past
• Allows to use HTML markup in the description of an episode

Motivation

I produce a podcast myself (shameless plug), so naturally I needed a tool to upload new episodes to my hosting provider. As I am not a big fan of Wordpress, I did not want to go down that route that is understandibly quite popular with podcast producers. Instead I wanted something for the command-line, could not find anything and created my own instead.

Configuration

First, duplicate the config.template.json to a file with the name config.json.
Next, populate the config.json file with your information.

Remarks

The hosting object:

• accessKey: Look up the documentation of your hosting provider regarding AWS S3 API compatibility. In AWS S3, it is called aws_access_key_id.
• secret: Look up the documentation of your hosting provider regarding AWS S3 API compatibility. In AWS S3, it is called aws_secret_access_key.
• databasePath, episodePath, feedPath: Correspond to the paths in the bucket of the hosting provider as relative paths from the bucketName root.
• endpointUrl, regionName: Have to be looked up in the hosting providers documentation, this is necessary as per AWS S3 API. Consequently, if your provider does not offer these values, it probably is not AWS S3 API compliant.

The podcast object:

• category: Follows the categories from Apple Podcasts (formerly iTunes Podcasts). A list can be found here.
• language: Has to be a language code out of this list.

Usage

The tool can be used in two modes, both on a command line:

• interactive
• argument-based

Before starting, create a new virtualenv with:

virtualenv -p python3.6 venv

and install the requirements:

venv/bin/pip3 install -r requirements.txt

The interactive mode is just triggered by normally executing the program and following the prompts:

venv/bin/python3 pycaster/pycaster.py

The argument-based accepts the following arguments:

venv/bin/python3 pycaster/pycaster.py \
    --title='Episode #1' \
    --description='Just a test' \
    --duration='160:55' \
    --file='./episode-1.mp3'