Module: Feedbag

Defined in:
lib/feedbag.rb

Defined Under Namespace

Classes: Feed

Class Method Summary collapse

Class Method Details

._is_http_valid(uri, orig_url) ⇒ Object

not used. yet.



168
169
170
171
172
173
174
175
176
177
 
# File 'lib/feedbag.rb', line 168

def self._is_http_valid(uri, orig_url)
	req = Net::HTTP.get_response(uri)
	orig_uri = URI.parse(orig_url)
	case req
		when Net::HTTPSuccess then
			return true
		else
			return false
	end
end

.add_feed(feed_url, orig_url, base_uri = nil, title = "") ⇒ Object 



143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
 
# File 'lib/feedbag.rb', line 143

def self.add_feed(feed_url, orig_url, base_uri = nil, title = "")
	# puts "#{feed_url} - #{orig_url}"
	url = feed_url.sub(/^feed:/, '').strip

	if base_uri
		#	url = base_uri + feed_url
		url = URI.parse(base_uri).merge(feed_url).to_s
	end

	begin
		uri = URI.parse(url)
	rescue
		puts "Error with `#{url}'"
		exit 1
	end
	unless uri.absolute?
		orig = URI.parse(orig_url)
		url = orig.merge(url).to_s
	end

	# verify url is really valid
	$feeds.push(Feed.new(url, title)) unless $feeds.any? { |f| f.url == url }# if self._is_http_valid(URI.parse(url), orig_url)
end

.feed?(url) ⇒ Boolean

Returns:

  • (Boolean) 


40
41
42
43
44
45
46
47
48
49
50
51
52
53
 
# File 'lib/feedbag.rb', line 40

def self.feed?(url)
	# use LWR::Simple.normalize some time
	url_uri = URI.parse(url)
	url = "#{url_uri.scheme or 'http'}://#{url_uri.host}#{url_uri.path}"
	# hack:
	url.sub!(/^feed:\/\//, 'http://')

	res = self.find(url)
	if res.size == 1 and res.first == url
		return true
	else
		return false
	end
end

.find(url, args = {}) ⇒ Object 



55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
 
# File 'lib/feedbag.rb', line 55

def self.find(url, args = {})
	$feeds = []

	url_uri = URI.parse(url)
	url = nil
	if url_uri.scheme.nil?
	  url = "http://#{url_uri.to_s}"
	elsif url_uri.scheme == "feed"
	  return self.add_feed(url_uri.to_s.sub(/^feed:\/\//, 'http://'), nil)
	else
	  url = url_uri.to_s
	end
	#url = "#{url_uri.scheme or 'http'}://#{url_uri.host}#{url_uri.path}"

   return self.add_feed(url, nil) if looks_like_feed? url

	# check if feed_valid is avail
   unless args[:narrow]
     begin
  		require "feed_validator"
	  	v = W3C::FeedValidator.new
		  v.validate_url(url)
 			return self.add_feed(url, nil) if v.valid?
  	rescue LoadError
	  	# scoo
 		rescue REXML::ParseException
  	  # usually indicates timeout
	    # TODO: actually find out timeout. use Terminator?
	    $stderr.puts "Feed looked like feed but might not have passed validation or timed out"
     rescue => ex
  		$stderr.puts "#{ex.class} error ocurred with: `#{url}': #{ex.message}"
	  end
   end

	begin
     Timeout::timeout(10) do
       open(url) do |f|
         if @content_types.include?(f.content_type.downcase)
           return self.add_feed(url, nil)
         end

         doc = Hpricot(f.read)

         if doc.at("base") and doc.at("base")["href"]
           $base_uri = doc.at("base")["href"]
         else
           $base_uri = nil
         end

         # first with links
         (doc/"link").each do |l|
           next unless l["rel"]
           if l["type"] and @content_types.include?(l["type"].downcase.strip) and (l["rel"].downcase =~ /alternate/i or l["rel"] == "service.feed")
             self.add_feed(l["href"], url, $base_uri, l["title"])
           end
         end

         unless args[:narrow]
           (doc/"a").each do |a|
             next unless a["href"]
             if self.looks_like_feed?(a["href"])
               self.add_feed(a["href"], url, $base_uri, a["title"] || a.inner_html || a['alt']) # multiple fallbacks, first title, then the tag content, then the alt tag (in case of image)
             end
           end
         end
       end
     end
	rescue Timeout::Error => err
		$stderr.puts "Timeout error ocurred with `#{url}: #{err}'"
	rescue OpenURI::HTTPError => the_error
		$stderr.puts "Error ocurred with `#{url}': #{the_error}"
	rescue SocketError => err
		$stderr.puts "Socket error ocurred with: `#{url}': #{err}"
	rescue => ex
		$stderr.puts "#{ex.class} error ocurred with: `#{url}': #{ex.message}"
	ensure
		return $feeds
	end
end

.looks_like_feed?(url) ⇒ Boolean

Returns:

  • (Boolean) 


135
136
137
138
139
140
141
 
# File 'lib/feedbag.rb', line 135

def self.looks_like_feed?(url)
	if url =~ /((\.|\/)(rdf|xml|rdf|rss)$|feed=(rss|atom)|(atom|feed)\/?$)/i
		true
	else
		false
	end
end