Elixir Feed Parser
Switch branches/tags
Nothing to show
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Permalink
Type Name Latest commit message Commit time
Failed to load latest commit information.
config
lib
test
.gitignore
.travis.yml
HISTORY.md
LICENSE.md
README.md
mix.exs
mix.lock

README.md

Elixir Feed Parser

Build Status

Elixir Atom/RSS Parser. Depends on Erlang's xmerl xml parser.

It is basically a port of the excellent Ruby Feed Jira gem.

Supports specifically Feedburner Atom/RSS2 feeds, iTunes Podcast and Google Docs feeds.

Setup

Add elixir-feed-parser to your mix dependencies and application.

def application do
  [applications: [:"elixir_feed_parser"]]
end

defp deps do
  [{:"elixir_feed_parser", "~> 0.0.1"}]
end

Then run mix deps.get to install it.

Parsing feeds

feed = ElixirFeedParser.parse(xml_string)
title = feed.title

The feed map provides the following properties which is a normalization based on the Atom and RSS2 standards. Whenever an attribute is normalized the original attribute is provided with a corresponding namespace.

For example the url attribute is actually a link element as defined by the RSS2 standard, therefore we additionally expose the rss2:link attribute.

List of feed properties

  • title
  • id (a unique identifier)
  • description
  • url
  • links (array of strings)
  • feed_url (the canonical link to the feed)
  • updated (most recent update)
  • authors
  • language
  • icon
  • logo
  • copyright
  • generator
  • categories (an array of strings)

List of article properties

  • title
  • id (a unique identifier)
  • description
  • url
  • links (array of strings)
  • updated (most recent update)
  • published
  • authors
  • categories (an array of strings)
  • source

Contribute

I appreciate any contribution to elixir-feed-parser. Open a pull request or file issues if you got feedback!

I'm new to Elixir and would love to get some feedback on how to improve this codebase.

Credits

Since this is my first Elixir project I've taken lots of ideas from existing projects. In particular I'd like to mention FeederEx, Feedme and Quinn.