This is designed to take an RSS/atom feed and filter things in/out of it. YML files are plucked out of s3 (or via http). These contain the configuration for each feed.
An include filter means only items that match something on theh list will be kept. An exclude means matching items will be removed.
Individual matches are done against the 'content', 'title', 'summary', and 'link' fields. Text is lowercased. Entries can be string matches or a regular expression with slashes at either end. This is useful for URLs.
In my version, output URLs are prepended with tedder.me/rss/. This should probably be configurable in the future.
The following examples show my personal usage. I've removed a private URL.
- url: "privateurl"
filter:
- include:
- soup
- /daily.show/
- "category: top gear (uk)"
- mythbusters
- cats
exclude:
- /8.out.of.10.cats/
output: ted/feed.rss
- url: http://gdata.youtube.com/feeds/base/users/BuzzFeedVideo/uploads?alt=rss&v=2&orderby=published&client=ytapi-youtube-profile
filter:
- include:
- /dear.kitten/
output: ted/buzzfeed_dearkitten.rss