Yet ugly simple script but get the job done. Convert docx to markdown using pandoc.

I have a multi-level folder that contains many mixed doc/docx files along with other file types.

And, I don't wanna play with a word proccessor every time I need to see what it is in the doc/docx file.

Then, I know I have to convert these word files to a text-based format. Yes, markdown, of course, the most widely-used and simplest markup format of humanity :3. And yes, pandoc, a great tool to convert-a-file-format-to-another-file-format.

• Shell that support text substitution.
• GNU sed (-i for inline-replacement). Most Linux distro.
• LibreOffice.
• Perl (> v5.x.x) comes with Unicode module.
• Pandoc, of course.

• Walks through the folder with multi-level sub-folder
• Watches all the files for doc or docx:
  • If the file name contans any white-space, replaces the white-space with hyphen (-). I hate the white-spaces in the path!
  • If the file name contains any latin-based char but not the ascii (I mean the Unicode), converts the char to an us-ascii char.
  • What about non-latin chars, suck as CJK? Eh, I don't get it now.
  • Lower-cases the file name. Yes, non-spaces and lower-cases paths. I love it.
  • If the file is doc, convert it to docx. This stage calls libreoffice (or soffice) directly. You also can use unoconv as an alternative. The doc2docx function. Then calls the docx2md function.
  • Removes the converted doc file.
  • Creates one media folder for every docx file to save every extracted media files (jpeg, png, ...) from the being-converted docx file. Named this folder ${the-docx-file-name}-media.
  • Calls the docx2md function. Converts all the docx files to md files. Converts the absolute to relative paths for extracted media files (if any).
  • Remove the converted docx file.
• Exits the loop.

Unfortunately, you have to delete the empty media folder manually (if any).

That it.

I don't have much time to play with pandoc or Pandoc Template, although I think the out put should be more pleasure.