rough draft of a Markdown converter for .doc files

doc2mmd.rb

#!/usr/bin/ruby
# doc2mmd.rb
# rough draft of a Markdown converter for .doc files
# requires python html2text, edit path in last shell call

def e_sh(str)
	str.to_s.gsub(/(?=["\\])/, '\\')
end

ARGV.each {|f|
	input = %x{/usr/bin/textutil -convert html -stdout "#{f}"}
	input.gsub!(/<p.*?><br><\/p>/,'')

	sizes = input.scan(/p.(p\d+) \{.*?font: ([\d\.]+)px.*?\}/)
	
	sorted = sizes.sort_by{|a| a[1]}.reverse
	count = sorted.length > 3 ? sorted.length - 3 : sorted.length - 1
	sorted = sorted[0,count]
	
	conversions = {}

	sorted.each_with_index { |e,i|
		conversions[e[0]] = "h#{i+1}"
	}
	
	conversions.each {|k,v|
		input.gsub!(/<p class="#{k}">(.*?)<\/p>/,"<#{v}>\\1</#{v}>")
	}
	puts %x{echo "#{e_sh input}"|/Users/ttscoff/scripts/html2text}
	
}