htmlclipping 0.1.0 → 0.1.1

data/lib/htmlclipping.rb

@@ -33,7 +33,7 @@
 require 'iconv'
 
 class HtmlClipping
-	Version = '0.1.0'
+	Version = '0.1.1'
 	
 	# html:: The HTML of the referring web page.
 	# referred_uri:: The URI that is being referred to.
@@ -173,6 +173,7 @@ class HtmlClipping
 			substitute_links
 			compact_brs
 			@detokenized.gsub!( /\s+/, ' ' )
+			@detokenized.gsub!( /&([^\s;]*\s)/, '&\1' )
 			@detokenized
 		end
 		

data/lib/htmlclipping.rb~

@@ -1,8 +1,43 @@
+# HtmlClipping generates excerpts from an HTML page that has a link pointing to 
+# a particular URI. It removes most HTML markup, bolds the link text, and 
+# trims the resulting text to a fixed number of characters. I developed it to 
+# help me track referers to my website, though I suppose it might have other 
+# uses.
+#
+# For example, the following script gets the HTML at http://rubyforge.org/credits/, and forms an excerpt around the link to http://www.rubycentral.org/pledge/.
+#
+#   require 'htmlclipping'
+#   require 'net/http'
+#
+#   contents = ''
+#   Net::HTTP.start( 'rubyforge.org' ) do |http|
+#     response = http.get '/credits/'
+#     contents = response.body
+#   end
+#   clipping = HtmlClipping.new(
+#     contents, 'http://www.rubycentral.org/pledge/', 500
+#   )
+#   puts clipping.to_s
+#
+#   => "… RubyForge takes time, effort, and money. Many thanks to the
+#      folks listed below who are making it possible! <br /> If RubyForge has
+#      been helpful to you, and you want to give something back to the Ruby
+#      community, please consider supporting <strong>RubyCentral</strong>.
+#      Thanks! <br /> InfoEther, Inc purchased the RubyForge hardware and
+#      provides system administration support. <br /> Several folks provide
+#      file mirrors to help share the bandwidth load: <br /> Evan Webb <br />
+#      Dennis Oelkers <br /> Austin &#8230;"
+#
+# The Rubyforge project page can be found at http://rubyforge.org/projects/htmlclipping.
+
 require 'iconv'
 
 class HtmlClipping
 	Version = '0.1.0'
 	
+	# html:: The HTML of the referring web page.
+	# referred_uri:: The URI that is being referred to.
+	# excerpt_limit:: The maximum size of the resulting clipping
 	def initialize( html, referred_uri, excerpt_limit )
 		@contents = html
 		if @contents =~ %r{<meta.*text/html; charset=(.*?)('|")}i
@@ -12,10 +47,11 @@ class HtmlClipping
 		@excerpt_limit = excerpt_limit
 	end
 	
-	def convert( str )
+	def convert( str ) # :nodoc:
 		@converter ? @converter.iconv( str ) : str
 	end
 	
+	# Returns the clipping as a string suitable for use as XML text.
 	def to_s
 		if @contents =~ %r{<body[^>]*>(.*)</body>}mi
 			to_detokenize = $1
@@ -28,9 +64,8 @@ class HtmlClipping
 		excerpt.gsub( /[\200-\377]/ ) { |c| "&#%04d;" % c[0] }
 		convert( excerpt )
 	end
-	
 
-	class ArrayOfWordsAndBrTags < Array
+	class ArrayOfWordsAndBrTags < Array # :nodoc:
 		def initialize( text )
 			super()
 			text.split( %r{(<br />)} ).each { |br_or_between_br|
@@ -43,7 +78,7 @@ class HtmlClipping
 		end
 	end
 
-	class Clipper
+	class Clipper # :nodoc:
 		def initialize( excerpt, excerpt_limit )
 			@excerpt = excerpt
 			@excerpt_limit = excerpt_limit
@@ -112,7 +147,7 @@ class HtmlClipping
 		end
 	end
 
-	class Detokenizer
+	class Detokenizer # :nodoc:
 		def initialize( contents, referred_uri )
 			@contents, @referred_uri = contents, referred_uri;
 		end
@@ -157,7 +192,7 @@ class HtmlClipping
 		end
 	end
 
-	class HtmlWithFixedAttributes < String
+	class HtmlWithFixedAttributes < String # :nodoc:
 		def initialize( contents )
 			super( '' )
 			script_stack = []

metadata

@@ -3,8 +3,8 @@ rubygems_version: 0.8.6
 specification_version: 1
 name: htmlclipping
 version: !ruby/object:Gem::Version 
-  version: 0.1.0
-date: 2005-05-15
+  version: 0.1.1
+date: 2005-06-26
 summary: HtmlClipping generates excerpts from an HTML page that has a link pointing to a particular URI.
 require_paths: 
   - lib