0
@@ -88,7 +88,8 @@ module FeedParser
0
     "John Beimler <http://john.beimler.org/>",
0
     "Fazal Majid <http://www.majid.info/mylos/weblog/>",
0
     "Aaron Swartz <http://aaronsw.com/>",
0
-    "Kevin Marks <http://epeus.blogspot.com/>"
0
+    "Kevin Marks <http://epeus.blogspot.com/>",
0
+    "Jesse Newland <http://jnewland.com/>"
0
   ]
0
   # HTTP "User-Agent" header to send to servers when downloading feeds.
0
   # If you are embedding feedparser in a larger application, you should
0
@@ -148,16 +149,19 @@ module FeedParser
0
     $compatible = options[:compatible].nil? ? $compatible : options[:compatible]# Use the default compatibility if compatible is nil
0
     strictklass = options[:strict] || StrictFeedParser
0
     looseklass = options[:loose] || LooseFeedParser
0
+    
0
     result = FeedParserDict.new
0
     result['feed'] = FeedParserDict.new
0
     result['entries'] = []
0
+    
0
     if options[:modified]
0
-      options[:modified] = Time.parse(options[:modified]).utc.rfc2822 
0
-      # FIXME this ignores all of our time parsing work.  Does it matter?
0
+      options[:modified] = py2rtime(parse_date(options[:modified])).httpdate
0
     end
0
+    
0
     result['bozo'] = false
0
+    
0
     handlers = options[:handlers]
0
-    if handlers.class != Array # FIXME why does this happen?
0
+    if handlers.class != Array 
0
       handlers = [handlers]
0
     end
0
 
0
@@ -167,40 +171,41 @@ module FeedParser
0
         $stderr << "Opening local file #{furi}\n" if $debug
0
         f = open(parsed_furi.path) # OpenURI doesn't behave well when passing HTTP options to a file.
0
       else
0
-        # And when you do pass them, make sure they aren't just nil (this still true?)
0
-        newd = {}
0
-        newd["If-None-Match"] = options[:etag] unless options[:etag].nil?
0
-        newd["If-Modified-Since"] = options[:modified] unless options[:modified].nil?
0
-        newd["User-Agent"] = (options[:agent] || USER_AGENT).to_s 
0
-        newd["Referer"] = options[:referrer] unless options[:referrer].nil?
0
-        newd["Content-Location"] = options[:content_location] unless options[:content_location].nil?
0
-        newd["Content-Language"] = options[:content_language] unless options[:content_language].nil?                    
0
-        newd["Content-type"] = options[:content_type] unless options[:content_type].nil?
0
         
0
-        f = open(furi, newd)
0
+        hout = {} # Dictionary of headers to send out
0
+        hout["If-None-Match"] = options[:etag] unless options[:etag].nil?
0
+        hout["If-Modified-Since"] = options[:modified] unless options[:modified].nil?
0
+        hout["User-Agent"] = (options[:agent] || USER_AGENT).to_s 
0
+        hout["Referer"] = options[:referrer] unless options[:referrer].nil?
0
+        hout["Content-Location"] = options[:content_location] unless options[:content_location].nil?
0
+        hout["Content-Language"] = options[:content_language] unless options[:content_language].nil?                    
0
+        hout["Content-type"] = options[:content_type] unless options[:content_type].nil?
0
         
0
+        f = open(furi, hout)
0
+        result['status'] = f.status[0]
0
       end
0
 
0
       data = f.read
0
       f.close 
0
       
0
     rescue OpenURI::HTTPError => e
0
-      unless e.to_s[0..2].to_i == 0
0
-        result['status'] = e.to_s[0..2].to_i
0
-        data = ''
0
-        f = nil
0
-      end
0
+      result['status'] = e.io.status[0]
0
+      result['bozo'] = true
0
+      result['bozo_exception'] = e
0
+      data = ''
0
+      f = nil
0
       
0
     rescue => e
0
-      $stderr << "Rescued in parse: "+e.to_s+"\n" if $debug # My addition
0
+      $stderr << "Rescued in parse: "+e.to_s+"\n" if $debug
0
       result['bozo'] = true
0
       result['bozo_exception'] = e
0
       data = ''
0
       f = nil
0
     end
0
+    
0
     if f.respond_to?(:meta)
0
       result['etag'] = f.meta['etag']
0
-      result['modified'] = f.meta['modified']
0
+      result['modified'] = f.meta['last-modified']
0
       result['url'] = f.base_uri.to_s
0
       result['status'] ||= f.status[0].to_i
0
       result['headers'] = f.meta

0
@@ -1,4 +1,4 @@
0
-#!/usr/bin/ruby
0
+#!/usr/bin/env ruby
0
 
0
 module FeedParserUtilities
0
   # Adapted from python2.4's encodings/aliases.py

0
@@ -1,4 +1,4 @@
0
-#!/usr/bin/ruby
0
+#!/usr/bin/env ruby
0
 
0
 # Add some helper methods to make AttributeList (all of those damn attrs
0
 # and attrsD used by StrictFeedParser) act more like a Hash.

0
@@ -1,4 +1,4 @@
0
-#!/usr/bin/ruby
0
+#!/usr/bin/env ruby
0
 
0
 
0
 class BetterSGMLParserError < Exception; end;

0
@@ -1,4 +1,4 @@
0
-#!/usr/bin/ruby
0
+#!/usr/bin/env ruby
0
 
0
 module FeedParserUtilities
0
 
0
@@ -54,7 +54,7 @@ module FeedParserUtilities
0
     begin 
0
       if xml_data[0..3] == "\x4c\x6f\xa7\x94"
0
         # EBCDIC
0
-        xml_data = _ebcdic_to_ascii(xml_data)
0
+        xml_data = __ebcdic_to_ascii(xml_data)
0
       elsif xml_data[0..3] == "\x00\x3c\x00\x3f"
0
         # UTF-16BE
0
         sniffed_xml_encoding = 'utf-16be'

0
@@ -1,4 +1,4 @@
0
-#!/usr/bin/ruby
0
+#!/usr/bin/env ruby
0
 module FeedParser
0
   class FeedParserDict < Hash 
0
 =begin

0
@@ -1,4 +1,4 @@
0
-#!/usr/bin/ruby
0
+#!/usr/bin/env ruby
0
 # From Robert Aman's GentleCMS URI.
0
 # GentleCMS, Copyright (c) 2006 Robert Aman
0
 #

0
@@ -1,4 +1,4 @@
0
-#!/usr/bin/ruby
0
+#!/usr/bin/env ruby
0
 module FeedParserUtilities
0
   def stripDoctype(data)
0
     #Strips DOCTYPE from XML document, returns (rss_version, stripped_data)

0
@@ -1,4 +1,4 @@
0
-#!/usr/bin/ruby
0
+#!/usr/bin/env ruby
0
 module FeedParser
0
 module FeedParserMixin
0
   include FeedParserUtilities

0
@@ -1,4 +1,4 @@
0
-#!/usr/bin/ruby
0
+#!/usr/bin/env ruby
0
 
0
 module FeedParser
0
   class StrictFeedParser < XML::SAX::HandlerBase # expat

0
@@ -1,4 +1,4 @@
0
-#!/usr/bin/ruby
0
+#!/usr/bin/env ruby
0
 # This used to be based on Michael Moen's Hpricot#scrub, but that seems to 
0
 # have only been part of its evolution. Hpricot#scrub is cool code, though.
0
 # http://underpantsgnome.com/2007/01/20/hpricot-scrub

0
@@ -1,4 +1,4 @@
0
-#!/usr/bin/ruby
0
+#!/usr/bin/env ruby
0
 require 'time'
0
 
0
 module FeedParser

0
@@ -1,119 +1,10 @@
0
-#!/usr/bin/ruby
0
-# This is the same server code that runs in rfeedparsertest.rb, but split
0
-# off so that we can fully check each test individually (i.e. get the HTTP
0
-# headers right).
0
-# This really needs to be merged into rfeedparsertest.rb
0
-require 'rubygems'
0
-gem 'mongrel'
0
-require 'mongrel'
0
-require File.dirname(__FILE__)+'/../lib/rfeedparser'
0
+#!/usr/bin/env ruby
0
 
0
-$PORT = 8097
0
-
0
-def uconvert(one, two, three); FeedParser::uconvert(one, two, three); end
0
-def _ebcdic_to_ascii(one); FeedParser::_ebcdic_to_ascii(one); end
0
-
0
-def translate_data(data)
0
-  if data[0..3] == "\x4c\x6f\xa7\x94"
0
-    # EBCDIC
0
-    data = _ebcdic_to_ascii(data)
0
-  elsif data[0..3] == "\x00\x3c\x00\x3f"
0
-    # UTF-16BE
0
-    data = uconvert(data, 'utf-16be', 'utf-8')
0
-  elsif data.size >= 4 and data[0..1] == "\xfe\xff" and data[2..3] != "\x00\x00"
0
-    # UTF-16BE with BOM
0
-    data = uconvert(data[2..-1], 'utf-16be', 'utf-8')
0
-  elsif data[0..3] == "\x3c\x00\x3f\x00"
0
-    # UTF-16LE
0
-    data = uconvert(data, 'utf-16le', 'utf-8')
0
-  elsif data.size >=4 and data[0..1] == "\xff\xfe" and data[2..3] != "\x00\x00"
0
-    # UTF-16LE with BOM
0
-    data = uconvert(data[2..-1], 'utf-16le', 'utf-8')
0
-  elsif data[0..3] == "\x00\x00\x00\x3c"
0
-    # UTF-32BE
0
-    data = uconvert(data, 'utf-32be', 'utf-8')
0
-  elsif data[0..3] == "\x3c\x00\x00\x00"
0
-    # UTF-32LE
0
-    data = uconvert(data, 'utf-32le', 'utf-8')
0
-  elsif data[0..3] == "\x00\x00\xfe\xff"
0
-    # UTF-32BE with BOM
0
-    data = uconvert(data[4..-1], 'utf-32BE', 'utf-8')
0
-  elsif data[0..3] == "\xef\xfe\x00\x00"
0
-    # UTF-32LE with BOM
0
-    data = uconvert(data[4..-1], 'utf-32le', 'utf-8')
0
-  elsif data[0..2] == "\xef\xbb\xbf"
0
-    # UTF-8 with BOM
0
-    data = data[3..-1]
0
-  end
0
-  return data
0
-end
0
-
0
-def scrape_headers(xmlfile)
0
-  # Called by the server
0
-  xm = open(xmlfile)
0
-  data = xm.read
0
-  htaccess = File.dirname(xmlfile)+"/.htaccess"
0
-  xml_headers = {}
0
-  server_headers = {}
0
-  the_type = nil
0
-  if File.exists? htaccess
0
-    fn = xm.path.split(File::Separator)[-1] # I can't find the right method for this
0
-    ht_file = open(htaccess)
0
-    type_match = ht_file.read.match(/^\s*<Files\s+#{fn}>\s*\n\s*AddType\s+(.*?)\s+.xml/m)
0
-    the_type = type_match[1].strip.gsub(/^("|')/,'').gsub(/("|')$/,'').strip if type_match and type_match[1]
0
-    if type_match and the_type
0
-      #content_type, charset = type_match[1].split(';')
0
-      server_headers["Content-Type"] = the_type
0
-    end
0
-  end
0
-  data = translate_data(data)
0
-  da = data.scan /^Header:\s*([^:]+):(.+)\s$/
0
-  unless da.nil? or da.empty?
0
-    da.flatten!
0
-    da.each{|e| e.strip!;e.gsub!(/(Content-type|content-type|content-Type)/, "Content-Type")}
0
-    xml_headers = Hash[*da] # Asterisk magic!
0
-  end
0
-  return xml_headers.merge(server_headers)
0
-end
0
-
0
-
0
-class FeedParserTestRequestHandler < Mongrel::DirHandler
0
-  def process(request, response)
0
-    req_method = request.params[Mongrel::Const::REQUEST_METHOD] || Mongrel::Const::GET
0
-    req_path = can_serve request.params[Mongrel::Const::PATH_INFO]
0
-    if not req_path
0
-      # not found, return a 404
0
-      response.start(404) do |head, out|
0
-        out << "File not found"
0
-      end
0
-    else
0
-      begin
0
-        if File.directory? req_path
0
-          send_dir_listing(request.params[Mongrel::Const::REQUEST_URI], req_path, response)
0
-        elsif req_method == Mongrel::Const::HEAD
0
-          response.start do |head,out| 
0
-            xml_head = scape_headers(req_path)
0
-            xml_head.each_key{|k| head[k] = xml_head[k] }
0
-          end
0
-          send_file(req_path, request, response, true)
0
-        elsif req_method == Mongrel::Const::GET
0
-          response.start do |head,out| 
0
-            xml_head = scrape_headers(req_path)
0
-            xml_head.each_key{|k| head[k] = xml_head[k] }
0
-          end
0
-          send_file(req_path, request, response, false)
0
-        else
0
-          response.start(403) {|head,out| out.write(ONLY_HEAD_GET) }
0
-        end
0
-      rescue => details
0
-        STDERR.puts "MON Error sending file #{req_path}: #{details}"
0
-      end
0
-    end
0
-  end
0
-end
0
+require File.join(File.dirname(__FILE__), 'rfeedparser_test_helper')
0
 
0
 # Start up the mongrel server and tell it how to send the tests
0
 server = Mongrel::HttpServer.new("0.0.0.0", $PORT)
0
 Mongrel::DirHandler::add_mime_type('.xml','application/xml')
0
+Mongrel::DirHandler::add_mime_type('.xml_redirect','application/xml')
0
 server.register("/", FeedParserTestRequestHandler.new('.'))
0
 server.run.join

0
@@ -1,170 +1,9 @@
0
 #!/usr/bin/env ruby
0
 # This is based off of Sam Ruby's xml_filetest.rb
0
-# I've adapted it for feedparser.rb
0
+# I've adapted it for rfeedparser
0
 # http://intertwingly.net/blog/2005/10/30/Testing-FeedTools-Dynamically/
0
 
0
-require 'test/unit'
0
-require File.dirname(__FILE__)+'/../lib/rfeedparser'
0
-
0
-begin 
0
-  require 'rubygems'
0
-  gem 'mongrel'
0
-  require 'mongrel'
0
-rescue => details
0
-  STDERR.puts "Whoops, had an error with loading mongrel as a gem. Trying just 'require'. Mongrel is required for testing."
0
-  require 'mongrel'
0
-end
0
-
0
-def uconvert(one, two, three); FeedParser::uconvert(one, two, three); end
0
-def _ebcdic_to_ascii(one); FeedParser::_ebcdic_to_ascii(one); end
0
-
0
-$PORT = 8097 # Not configurable, hard coded in the xml files
0
-
0
-def translate_data(data)
0
-  if data[0..3] == "\x4c\x6f\xa7\x94"
0
-    # EBCDIC
0
-    data = _ebcdic_to_ascii(data)
0
-  elsif data[0..3] == "\x00\x3c\x00\x3f"
0
-    # UTF-16BE
0
-    data = uconvert(data, 'utf-16be', 'utf-8')
0
-  elsif data.size >= 4 and data[0..1] == "\xfe\xff" and data[2..3] != "\x00\x00"
0
-    # UTF-16BE with BOM
0
-    data = uconvert(data[2..-1], 'utf-16be', 'utf-8')
0
-  elsif data[0..3] == "\x3c\x00\x3f\x00"
0
-    # UTF-16LE
0
-    data = uconvert(data, 'utf-16le', 'utf-8')
0
-  elsif data.size >=4 and data[0..1] == "\xff\xfe" and data[2..3] != "\x00\x00"
0
-    # UTF-16LE with BOM
0
-    data = uconvert(data[2..-1], 'utf-16le', 'utf-8')
0
-  elsif data[0..3] == "\x00\x00\x00\x3c"
0
-    # UTF-32BE
0
-    data = uconvert(data, 'utf-32be', 'utf-8')
0
-  elsif data[0..3] == "\x3c\x00\x00\x00"
0
-    # UTF-32LE
0
-    data = uconvert(data, 'utf-32le', 'utf-8')
0
-  elsif data[0..3] == "\x00\x00\xfe\xff"
0
-    # UTF-32BE with BOM
0
-    data = uconvert(data[4..-1], 'utf-32BE', 'utf-8')
0
-  elsif data[0..3] == "\xff\xfe\x00\x00"
0
-    # UTF-32LE with BOM
0
-    data = uconvert(data[4..-1], 'utf-32LE', 'utf-8')
0
-  elsif data[0..2] == "\xef\xbb\xbf"
0
-    # UTF-8 with BOM
0
-    data = data[3..-1]
0
-  else
0
-    # ASCII-compatible
0
-  end
0
-  return data
0
-end
0
-
0
-def scrape_headers(xmlfile)
0
-  # Called by the server
0
-  xm = open(xmlfile)
0
-  data = xm.read
0
-  htaccess = File.dirname(xmlfile)+"/.htaccess"
0
-  xml_headers = {}
0
-  server_headers = {}
0
-  the_type = nil
0
-  if File.exists? htaccess
0
-    fn = xm.path.split(File::Separator)[-1] # I can't find the right method for this
0
-    ht_file = open(htaccess)
0
-    type_match = ht_file.read.match(/^\s*<Files\s+#{fn}>\s*\n\s*AddType\s+(.*?)\s+.xml/m)
0
-    the_type = type_match[1].strip.gsub(/^("|')/,'').gsub(/("|')$/,'').strip if type_match and type_match[1]
0
-    if type_match and the_type
0
-      #content_type, charset = type_match[1].split(';')
0
-      server_headers["Content-Type"] = the_type
0
-    end
0
-  end
0
-  data = translate_data(data)
0
-  da = data.scan /^Header:\s*([^:]+):(.+)\s$/
0
-  unless da.nil? or da.empty?
0
-    da.flatten!
0
-    da.each{|e| e.strip!;e.gsub!(/(Content-type|content-type|content-Type)/, "Content-Type")}
0
-    xml_headers = Hash[*da] # Asterisk magic!
0
-  end
0
-  return xml_headers.merge(server_headers)
0
-end
0
-
0
-def scrape_assertion_strings(xmlfile)
0
-  # Called by the testing client
0
-  data = open(xmlfile).read
0
-  data = translate_data(data)
0
-  test = data.scan /Description:\s*(.*?)\s*Expect:\s*(.*)\s*-->/
0
-  description, evalString = test.first.map{ |s| s.strip }
0
-
0
-  # Here we translate the expected values in Python to Ruby
0
-  evalString.gsub!(/\bu'(.*?)'/) do |m|     
0
-    esc = $1.to_s.dup
0
-    esc.gsub!(/\\u([0-9a-fA-F]{4})/){ |m| [$1.hex].pack('U*') }
0
-    " '"+esc+"'"
0
-  end 
0
-  evalString.gsub!(/\bu"(.*?)"/) do |m|
0
-    esc = $1.to_s.dup
0
-    esc.gsub!(/\\u([0-9a-fA-F]{4})/){ |m| [$1.hex].pack('U*') }
0
-    " \""+esc+"\""
0
-  end
0
-  # The above does the following:               u'string' => 'string'
0
-  #                                             u'ba\u20acha' => 'ba€ha' # Same for double quoted strings
0
-
0
-  evalString.gsub!(/\\x([0-9a-fA-F]{2})/){ |m| [$1.hex].pack('U*') } # "ba\xa3la" => "ba£la"
0
-  evalString.gsub! /'\s*:\s+/, "' => "        # {'foo': 'bar'} => {'foo' => 'bar'}
0
-  evalString.gsub! /"\s*:\s+/, "\" => "       # {"foo": 'bar'} => {"foo" => 'bar'}
0
-  evalString.gsub! /\=\s*\((.*?)\)/, '= [\1]' # = (2004, 12, 4) => = [2004, 12, 4]
0
-  evalString.gsub!(/"""(.*?)"""/) do          # """<a b="foo">""" => "<a b="foo">"
0
-    "\""+$1.gsub!(/"/,"\\\"")+"\"" # haha, ugly!
0
-  end
0
-  evalString.gsub! /(\w|\])\s*\=\= 0\s*$/, '\1 == false'   # ] == 0 => ] == false
0
-  evalString.gsub! /(\w|\])\s*\=\= 1\s*$/, '\1 == true'    # ] == 1 => ] == true
0
-  evalString.gsub! /len\((.*?)\)\s*\=\=\s*(\d{1,3})/, '\1.length == \2' # len(ary) == 1 => ary.length == 1
0
-  evalString.gsub! /None/, "nil" # None => nil # well, duh
0
-  return description, evalString
0
-end
0
-
0
-class FeedParserTestRequestHandler < Mongrel::DirHandler 
0
-  def process(request, response)
0
-    req_method = request.params[Mongrel::Const::REQUEST_METHOD] || Mongrel::Const::GET
0
-    req_path = can_serve request.params[Mongrel::Const::PATH_INFO]
0
-    if not req_path
0
-      # not found, return a 404
0
-      response.start(404) do |head, out|
0
-        out << "File not found"
0
-      end
0
-    else
0
-      begin
0
-        if File.directory? req_path
0
-          send_dir_listing(request.params[Mongrel::Const::REQUEST_URI], req_path, response)
0
-        elsif req_method == Mongrel::Const::HEAD
0
-          response.start do |head,out| 
0
-            xml_head = scrape_headers(req_path)
0
-            xml_head.each_key{|k| head[k] = xml_head[k] }
0
-          end
0
-
0
-          send_file(req_path, request, response, true)
0
-        elsif req_method == Mongrel::Const::GET
0
-          response.start do |head,out| 
0
-            xml_head = scrape_headers(req_path)
0
-            xml_head.each_key{|k| head[k] = xml_head[k] }
0
-          end
0
-
0
-          send_file(req_path, request, response, false)
0
-        else
0
-          response.start(403) {|head,out| out.write(ONLY_HEAD_GET) }
0
-        end
0
-      rescue => details
0
-        STDERR.puts "Error sending file #{req_path}: #{details}"
0
-      end
0
-    end
0
-  end
0
-end
0
-
0
-
0
-class XMLTests < Test::Unit::TestCase
0
-  # Empty, but here for clarity
0
-  def setup
0
-  end
0
-  def teardown
0
-  end
0
-end
0
+require File.join(File.dirname(__FILE__),'rfeedparser_test_helper')
0
 
0
 # default methods to be public
0
 XMLTests.send(:public)
0
@@ -194,5 +33,6 @@ end
0
 # Start up the mongrel server and tell it how to send the tests
0
 server = Mongrel::HttpServer.new("0.0.0.0",$PORT)
0
 Mongrel::DirHandler::add_mime_type('.xml','application/xml')
0
+Mongrel::DirHandler::add_mime_type('.xml_redirect','application/xml')
0
 server.register("/", FeedParserTestRequestHandler.new("."))
0
 server.run