Short python script to recursively process html files with mark2text.py (https://github.com/aaronsw/html2text) thus converting them to MD docs instead (e.g. a Tomboy notes extracted collection). Written for OSX, YMMV.

recursivehtml2text.py

import os
import subprocess
import sys
import string
import re

for dir, subdir, files in os.walk( sys.argv[1] ):
	for filename in files:
		# ignore OSX fs file
		if( filename != ".DS_Store"):
			# remove all special chars from filename if there are any,
		  	# and clip the file extension
			f = re.sub('\W', '', string.split(filename, '.')[0]) + '.md'
			# open the output file and pipe it to the stdout of html2text
			with open(dir + '/' + f, 'w') as fout:
				cmd = ["python", "html2text.py", dir + '/' + filename]
				subprocess.call(cmd, stdout=fout)