feedjira 0.9.0

data/lib/feedjira/feed_entry_utilities.rb

@@ -0,0 +1,66 @@
+module Feedjira
+  module FeedEntryUtilities
+
+    include Enumerable
+
+    def published
+      @published ||= @updated
+    end
+
+    def parse_datetime(string)
+      begin
+        DateTime.parse(string).feed_utils_to_gm_time
+      rescue
+        warn "Failed to parse date #{string.inspect}"
+        nil
+      end
+    end
+
+    ##
+    # Returns the id of the entry or its url if not id is present, as some formats don't support it
+    def id
+      @entry_id ||= @url
+    end
+
+    ##
+    # Writer for published. By default, we keep the "oldest" publish time found.
+    def published=(val)
+      parsed = parse_datetime(val)
+      @published = parsed if !@published || parsed < @published
+    end
+
+    ##
+    # Writer for updated. By default, we keep the most recent update time found.
+    def updated=(val)
+      parsed = parse_datetime(val)
+      @updated = parsed if !@updated || parsed > @updated
+    end
+
+    def sanitize!
+      %w[title author summary content image].each do |name|
+        if self.respond_to?(name) && self.send(name).respond_to?(:sanitize!)
+          self.send(name).send :sanitize!
+        end
+      end
+    end
+
+    alias_method :last_modified, :published
+
+    def each
+      @rss_fields ||= self.instance_variables
+
+      @rss_fields.each do |field|
+        yield(field.to_s.sub('@', ''), self.instance_variable_get(field))
+      end
+    end
+
+    def [](field)
+      self.instance_variable_get("@#{field.to_s}")
+    end
+
+    def []=(field, value)
+      self.instance_variable_set("@#{field.to_s}", value)
+    end
+
+  end
+end

data/lib/feedjira/feed_utilities.rb

@@ -0,0 +1,103 @@
+module Feedjira
+  module FeedUtilities
+    UPDATABLE_ATTRIBUTES = %w(title feed_url url last_modified etag)
+
+    attr_writer   :new_entries, :updated, :last_modified
+    attr_accessor :etag
+
+    def self.included(base)
+      base.extend ClassMethods
+    end
+
+    module ClassMethods
+      def parse(xml, &block)
+        xml = preprocess(xml) if preprocess_xml
+        super xml.lstrip, &block
+      end
+
+      def preprocess(xml)
+        # noop
+        xml
+      end
+
+      def preprocess_xml=(value)
+        @preprocess_xml = value
+      end
+
+      def preprocess_xml
+        @preprocess_xml
+      end
+    end
+
+    def last_modified
+      @last_modified ||= begin
+        entry = entries.reject {|e| e.published.nil? }.sort_by { |entry| entry.published if entry.published }.last
+        entry ? entry.published : nil
+      end
+    end
+
+    def updated?
+      @updated || false
+    end
+
+    def new_entries
+      @new_entries ||= []
+    end
+
+    def has_new_entries?
+      new_entries.size > 0
+    end
+
+    def update_from_feed(feed)
+      self.new_entries += find_new_entries_for(feed)
+      self.entries.unshift(*self.new_entries)
+
+      @updated = false
+
+      UPDATABLE_ATTRIBUTES.each do |name|
+        @updated ||= update_attribute(feed, name)
+      end
+    end
+
+    def update_attribute(feed, name)
+      old_value, new_value = send(name), feed.send(name)
+
+      if old_value != new_value
+        send("#{name}=", new_value)
+        true
+      else
+        false
+      end
+    end
+
+    def sanitize_entries!
+      entries.each {|entry| entry.sanitize!}
+    end
+
+    private
+
+    def find_new_entries_for(feed)
+      # this implementation is a hack, which is why it's so ugly.
+      # it's to get around the fact that not all feeds have a published date.
+      # however, they're always ordered with the newest one first.
+      # So we go through the entries just parsed and insert each one as a new entry
+      # until we get to one that has the same id as the the newest for the feed
+      return feed.entries if self.entries.length == 0
+      latest_entry = self.entries.first
+      found_new_entries = []
+      feed.entries.each do |entry|
+        if entry.entry_id.nil? && latest_entry.entry_id.nil?
+          break if entry.url == latest_entry.url
+        else
+          break if entry.entry_id == latest_entry.entry_id || entry.url == latest_entry.url
+        end
+        found_new_entries << entry
+      end
+      found_new_entries
+    end
+
+    def existing_entry?(test_entry)
+      entries.any? { |entry| entry.id == test_entry.id }
+    end
+  end
+end

data/lib/feedjira/parser.rb

@@ -0,0 +1,20 @@
+module Feedjira
+  module Parser
+    autoload :RSS,                'feedjira/parser/rss'
+    autoload :RSSEntry,           'feedjira/parser/rss_entry'
+    autoload :RSSFeedBurner,      'feedjira/parser/rss_feed_burner'
+    autoload :RSSFeedBurnerEntry, 'feedjira/parser/rss_feed_burner_entry'
+
+    autoload :ITunesRSS,      'feedjira/parser/itunes_rss'
+    autoload :ITunesRSSItem,  'feedjira/parser/itunes_rss_item'
+    autoload :ITunesRSSOwner, 'feedjira/parser/itunes_rss_owner'
+
+    autoload :GoogleDocsAtom,      'feedjira/parser/google_docs_atom'
+    autoload :GoogleDocsAtomEntry, 'feedjira/parser/google_docs_atom_entry'
+
+    autoload :Atom,                'feedjira/parser/atom'
+    autoload :AtomEntry,           'feedjira/parser/atom_entry'
+    autoload :AtomFeedBurner,      'feedjira/parser/atom_feed_burner'
+    autoload :AtomFeedBurnerEntry, 'feedjira/parser/atom_feed_burner_entry'
+  end
+end

data/lib/feedjira/parser/atom.rb

@@ -0,0 +1,61 @@
+module Feedjira
+  module Parser
+    # Parser for dealing with Atom feeds.
+    class Atom
+      include SAXMachine
+      include FeedUtilities
+      element :title
+      element :subtitle, :as => :description
+      element :link, :as => :url, :value => :href, :with => {:type => "text/html"}
+      element :link, :as => :feed_url, :value => :href, :with => {:type => "application/atom+xml"}
+      elements :link, :as => :links, :value => :href
+      elements :link, :as => :hubs, :value => :href, :with => {:rel => "hub"}
+      elements :entry, :as => :entries, :class => AtomEntry
+
+      def self.able_to_parse?(xml) #:nodoc:
+        /\<feed[^\>]+xmlns\s?=\s?[\"|\'](http:\/\/www\.w3\.org\/2005\/Atom|http:\/\/purl\.org\/atom\/ns\#)[\"|\'][^\>]*\>/ =~ xml
+      end
+
+      def url
+        @url || links.last
+      end
+
+      def feed_url
+        @feed_url ||= links.first
+      end
+
+      def self.preprocess(xml)
+        Preprocessor.new(xml).to_xml
+      end
+
+      class Preprocessor
+        def initialize(xml)
+          @xml = xml
+        end
+
+        def to_xml
+          process_nodes
+          doc.to_xml
+        end
+
+        private
+
+        def process_nodes
+          nodes.each { |node| node.content = raw_html(node) unless node.cdata? }
+        end
+
+        def nodes
+          doc.search 'entry > content[type="xhtml"]'
+        end
+
+        def raw_html(node)
+          CGI.unescape_html node.inner_html
+        end
+
+        def doc
+          @doc ||= Nokogiri::XML @xml
+        end
+      end
+    end
+  end
+end

data/lib/feedjira/parser/atom_entry.rb

@@ -0,0 +1,34 @@
+module Feedjira
+
+  module Parser
+    # Parser for dealing with Atom feed entries.
+    class AtomEntry
+      include SAXMachine
+      include FeedEntryUtilities
+
+      element :title
+      element :link, :as => :url, :value => :href, :with => {:type => "text/html", :rel => "alternate"}
+      element :name, :as => :author
+      element :content
+      element :summary
+
+      element :"media:content", :as => :image, :value => :url
+      element :enclosure, :as => :image, :value => :href
+
+      element :published
+      element :id, :as => :entry_id
+      element :created, :as => :published
+      element :issued, :as => :published
+      element :updated
+      element :modified, :as => :updated
+      elements :category, :as => :categories, :value => :term
+      elements :link, :as => :links, :value => :href
+
+      def url
+        @url ||= links.first
+      end
+    end
+
+  end
+
+end

data/lib/feedjira/parser/atom_feed_burner.rb

@@ -0,0 +1,22 @@
+module Feedjira
+
+  module Parser
+    # Parser for dealing with Feedburner Atom feeds.
+    class AtomFeedBurner
+      include SAXMachine
+      include FeedUtilities
+      element :title
+      element :subtitle, :as => :description
+      element :link, :as => :url, :value => :href, :with => {:type => "text/html"}
+      element :link, :as => :feed_url, :value => :href, :with => {:type => "application/atom+xml"}
+      elements :"atom10:link", :as => :hubs, :value => :href, :with => {:rel => "hub"}
+      elements :entry, :as => :entries, :class => AtomFeedBurnerEntry
+
+      def self.able_to_parse?(xml) #:nodoc:
+        ((/Atom/ =~ xml) && (/feedburner/ =~ xml) && !(/\<rss|\<rdf/ =~ xml)) || false
+      end
+    end
+
+  end
+
+end

data/lib/feedjira/parser/atom_feed_burner_entry.rb

@@ -0,0 +1,35 @@
+module Feedjira
+
+  module Parser
+    # Parser for dealing with Feedburner Atom feed entries.
+    class AtomFeedBurnerEntry
+      include SAXMachine
+      include FeedEntryUtilities
+
+      element :title
+      element :name, :as => :author
+      element :link, :as => :url, :value => :href, :with => {:type => "text/html", :rel => "alternate"}
+      element :"feedburner:origLink", :as => :url
+      element :summary
+      element :content
+
+      element :"media:content", :as => :image, :value => :url
+      element :enclosure, :as => :image, :value => :href
+
+      element :published
+      element :id, :as => :entry_id
+      element :issued, :as => :published
+      element :created, :as => :published
+      element :updated
+      element :modified, :as => :updated
+      elements :category, :as => :categories, :value => :term
+      elements :link, :as => :links, :value => :href
+
+      def url
+        @url ||= links.first
+      end
+
+    end
+  end
+
+end

data/lib/feedjira/parser/google_docs_atom.rb

@@ -0,0 +1,28 @@
+require File.expand_path('./atom', File.dirname(__FILE__))
+
+module Feedjira
+  module Parser
+    class GoogleDocsAtom
+      include SAXMachine
+      include FeedUtilities
+      element :title
+      element :subtitle, :as => :description
+      element :link, :as => :url, :value => :href, :with => {:type => "text/html"}
+      element :link, :as => :feed_url, :value => :href, :with => {:type => "application/atom+xml"}
+      elements :link, :as => :links, :value => :href
+      elements :entry, :as => :entries, :class => GoogleDocsAtomEntry
+
+      def url
+        @url ||= links.first
+      end
+
+      def self.able_to_parse?(xml) #:nodoc:
+        %r{<id>https?://docs.google.com/.*\</id\>} =~ xml
+      end
+
+      def feed_url
+        @feed_url ||= links.first
+      end
+    end
+  end
+end

data/lib/feedjira/parser/google_docs_atom_entry.rb

@@ -0,0 +1,29 @@
+module Feedjira
+  module Parser
+    class GoogleDocsAtomEntry
+      include SAXMachine
+      include FeedEntryUtilities
+
+      element :title
+      element :link, :as => :url, :value => :href, :with => {:type => "text/html", :rel => "alternate"}
+      element :name, :as => :author
+      element :content
+      element :summary
+      element :published
+      element :id, :as => :entry_id
+      element :created, :as => :published
+      element :issued, :as => :published
+      element :updated
+      element :modified, :as => :updated
+      elements :category, :as => :categories, :value => :term
+      elements :link, :as => :links, :value => :href
+      element :"docs:md5Checksum", :as => :checksum
+      element :"docs:filename", :as => :original_filename
+      element :"docs:suggestedFilename", :as => :suggested_filename
+
+      def url
+        @url ||= links.first
+      end
+    end
+  end
+end

data/lib/feedjira/parser/itunes_rss.rb

@@ -0,0 +1,50 @@
+module Feedjira
+
+  module Parser
+    # iTunes is RSS 2.0 + some apple extensions
+    # Source: http://www.apple.com/itunes/whatson/podcasts/specs.html
+    class ITunesRSS
+      include SAXMachine
+      include FeedUtilities
+
+      attr_accessor :feed_url
+
+      # RSS 2.0 elements that need including
+      element :copyright
+      element :description
+      element :language
+      element :managingEditor
+      element :title
+      element :link, :as => :url
+
+      # If author is not present use managingEditor on the channel
+      element :"itunes:author", :as => :itunes_author
+      element :"itunes:block", :as => :itunes_block
+      element :"itunes:image", :value => :href, :as => :itunes_image
+      element :"itunes:explicit", :as => :itunes_explicit
+      element :"itunes:keywords", :as => :itunes_keywords
+      # New URL for the podcast feed
+      element :"itunes:new-feed-url", :as => :itunes_new_feed_url
+      element :"itunes:subtitle", :as => :itunes_subtitle
+      # If summary is not present, use the description tag
+      element :"itunes:summary", :as => :itunes_summary
+
+      # iTunes RSS feeds can have multiple main categories...
+      # ...and multiple sub-categories per category
+      # TODO subcategories not supported correctly - they are at the same level
+      #   as the main categories
+      elements :"itunes:category", :as => :itunes_categories, :value => :text
+
+      elements :"itunes:owner", :as => :itunes_owners, :class => ITunesRSSOwner
+
+      elements :item, :as => :entries, :class => ITunesRSSItem
+
+      def self.able_to_parse?(xml)
+        /xmlns:itunes\s?=\s?\"http:\/\/www.itunes.com\/dtds\/podcast-1.0.dtd\"/i =~ xml
+      end
+
+    end
+
+  end
+
+end