-
-
Save ttscoff/3861434 to your computer and use it in GitHub Desktop.
| #!/usr/bin/ruby | |
| =begin | |
| Usage: rtftomarkdown.rb FILENAME.rtf | |
| Uses textutil, available on Mac only (installed by default) | |
| Outputs to STDOUT | |
| Notes: | |
| Links are replaced with Markdown references (duplicate links combined). | |
| This works fine on RTF files, but the markup that the Word/textutil | |
| combination produces seems impossible to work with. Most links disappear | |
| when converting from a DOC or DOCX file, and even Word's RTF export is | |
| unworkable. | |
| List levels converted by textutil can not be changed at a given depth. | |
| If you start a second-level nested list as an ordered list, the next | |
| second-level list will be ordered. It's a textutil/RTF thing. | |
| =end | |
| write_file = false # for Services set to true to write out files with .md extension | |
| if ARGV.length == 0 | |
| puts "#{__FILE__} expects an input file (RTF or DOC) as an argument" | |
| exit | |
| end | |
| def remove_empty(input) | |
| input.gsub!(/(<(\w+)( class=".*?")?>\s*<\/\2>)/,'') | |
| input = remove_empty(input) if input =~ /(<(\w+)( class=".*")?>\s*<\/\2>)/ | |
| return input.strip | |
| end | |
| ARGV.each do |infile| | |
| file = infile.sub(/\/$/,'') | |
| if File.exists?(File.expand_path(file)) | |
| ext = file.match(/\.(\w+)$/)[1] | |
| input = %x{/usr/bin/textutil -convert html -stdout "#{file}"}.strip | |
| input.gsub!(/.*?<body>(.*?)<\/body>.*/m,"\\1") | |
| # remove span/br tags, unneccessary | |
| input.gsub!(/<br>/,'') | |
| input.gsub!(/<\/?span( class=".*?")?>/,'') | |
| # substitute headers | |
| input.gsub!(/<p class="p1">(?:<b>)?(.+?)(?:<\/b>)?<\/p>/,'# \\1') | |
| input.gsub!(/<p class="p2"><b>(.+?)<\/b><\/p>/,'## \\1') | |
| input.gsub!(/<p class="p3"><b>(.+?)<\/b><\/p>/,'## \\1') | |
| input.gsub!(/<p class="p4"><b>(.+?)<\/b><\/p>/,'### \\1') | |
| input.gsub!(/<p class="p5"><b>(.+?)<\/b><\/p>/,'### \\1') | |
| input = input.split("\n").map { |line| | |
| remove_empty(line) | |
| }.join("\n") | |
| # remove paragraph tags | |
| input.gsub!(/<p class="p\d">(.*?)<\/p>/,'\\1') | |
| # emphasis | |
| input.gsub!(/<\/?b>/,'**') | |
| input.gsub!(/<\/?i>/,'*') | |
| # links | |
| links = {} | |
| footer = '' | |
| input.gsub!(/<a href="(.*?)">(.*?)<\/a>/) do |match| | |
| if links.has_key? $1 | |
| marker = links[$1] | |
| else | |
| links[$1] = links.length + 1 | |
| footer += "\n[#{links[$1]}]: #{$1}" | |
| end | |
| "[#{$2}][#{links[$1]}]" | |
| end | |
| input = input.split("\n").map { |line| | |
| line.strip | |
| }.join("\n") | |
| # handle lists | |
| list_level = 0 | |
| list_type = [] | |
| input = input.split("\n").map { |line| | |
| if line =~ /<([uo])l.*?>/ | |
| list_level += 1 | |
| list_type[list_level] = $1 =~ /u/ ? "*" : "1." | |
| "*REMOVEME" | |
| elsif line =~ /<\/[uo]l>/ | |
| list_level -= 1 | |
| "*REMOVEME" | |
| else | |
| indent = "" | |
| (list_level -1).times do indent += " " end | |
| line.gsub(/<li.*?>(.*?)<\/li>/,"#{indent}#{list_type[list_level]} \\1") | |
| end | |
| }.delete_if {|line| | |
| line =~ /\*REMOVEME/ | |
| }.join("\n") | |
| if write_file | |
| open(file+".md", 'w+') { |f| | |
| f.puts input | |
| f.puts footer | |
| } | |
| else | |
| puts input | |
| puts footer | |
| end | |
| else | |
| puts "File not found: #{file}" | |
| end | |
| end |
Nice, thanks! For more complicated documents (e.g. with math) I've made this repo, hope this could be useful for some: https://github.com/Versal/word2markdown
Hi,
Thanks for your nice sharing. But using code to deal with the related Word Doc converting projects is too complicated for me. As for myself, I prefer to do it with the help of some 3rd party Word Converters. It can make all the converting process much simpler and faster. You can also google it and select one to help you with the related Word Docs converting work. It will be better if it is totally manual and can be customized by users according to our own favors. Remember to check its free trial package first if possible. I hope you success. Good luck.
Best regards,
Arron
Okay, this is probably the best thing that I've found on the internet in a very long time. So very useful. Thank you!
+1 would execute again