Word docx files suck!

docx_to_markup.rb

#!/usr/bin/env ruby

# Converts those nasty xml documents word likes to create into something
# remotely resembling markdown
# Fun side effect: you'll get every revision of the document!

abort "No file name given" unless filename = ARGV.first

remove_tags = [
  '\?xml',
  "w:rFonts",
  "w:cols",
  "w:docGrid",
  "w:szCs",
  "w:sz",
  "w:p",
  "w:proofErr",
  "w:pgMar",
  "w:body",
  "/w:body",
  "w:sectPr",
  "/w:sectPr",
  "w:pgSz",
  "w:document",
  "/w:document",
  "w:r",
  "/w:r",
  "w:t",
  "/w:t",
  "w:rPr",
  "/w:rPr",
  "w:pPr",
  "/w:pPr",
]
gsubs = [
  [/<\/w:p>/, "\n\n"],
  [/<w:br\/>/, "  \n"],
  [/[“”]+/, '"'],
  [/<w:i\/>(\w+) /, '*\1* '],
  [/\n\n+/, "\n\n"],
  [/\n\n+\Z/m, "\n"],
]

f = File.read(filename)
out = f
remove_tags.each {|tag| out = out.gsub(/<#{tag}( .*?)?\/?>/, '')}
gsubs.each {|r,s| out = out.gsub(r,s)}
puts out