Method: WWMD::Page#html2text

Defined in:
lib/wwmd/page/html2text_hpricot.rb,
lib/wwmd/page/html2text_nokogiri.rb

#html2text ⇒ Object

def lookup_named_char(s)

  c = Hpricot::NamedCharacters[s[1...-1]]
  c.chr if c
end



58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
 
# File 'lib/wwmd/page/html2text_hpricot.rb', line 58

def html2text
  doc = self.scrape.hdoc
  text = node_to_text(doc)
#      text.gsub!(NamedCharRegex){|s| "#{lookup_named_char(s)}"}
  # clean up white space
  text.gsub!("\r"," ")
  text.squeeze!(" ")
  text.strip!
  ret = ''
  text.split(/\n/).each do |l|
    l.strip!
    next if l == ''
    next if l =~ /^\?+$/
    ret += "#{l}\n"
  end
  return ret
end