This Cloudflare Worker generates a podcast RSS feed based on the contents of a S3-compatible bucket. It is based on worker-signed-s3-template
You have to create your own wrangler.toml
, which you can base on wrangler.toml.example
.
Configuration is done through environment variables:
AWS_DEFAULT_REGION
: the default S3 region. It is blank with some providers (e.g. Backblaze).AWS_S3_BUCKET
: The virtualhost-style S3 endpoint for your bucket. (e.g.mybucket.s3.eu-west-003.backblazeb2.com
ormybucket.s3.eu-south-1.amazonaws.com
).BUCKET_PUBLIC_URL
: The base URL of your bucket. Usually it ishttps://${AWS_S3_BUCKET}
, but sometimes it is useful to change it (for example if you have Cloudflare in front of your S3 storage).SUBFOLDER
: Optional subfolder of the bucket to search files in.PODCAST_TITLE
: The title that appears on the RSS feed.PODCAST_DESCRIPTION
: The description that appears on the RSS feed.PODCAST_IMAGE
: The URL of an image to be used as the podcast album art.PODCAST_IMAGE_WIDTH
: The width ofPODCAST_IMAGE
PODCAST_IMAGE_HEIGHT
: The height ofPODCAST_IMAGE
Generation using wrangler (wrangler generate projectname https://github.com/LucaTNT/worker-feed-from-bucket
) is currently not supported due to this repo using the "main" branch as opposed to the "master" branch that wrangler is expecting.
After you cloned the repo, set the required secrets containing your S3 API Keys:
wrangler secret put AWS_ACCESS_KEY_ID
wrangler secret put AWS_SECRET_ACCESS_KEY
Then you can push it to cloudflare by running wrangler publish
.