Tidy.py

#!/usr/bin/env python
from __future__ import with_statement
from BeautifulSoup import BeautifulSoup
import subprocess as sub
import re
import glob
import sys

repls = [
 (re.compile(r'^[Mm][Oo][Dd][Ee][Ll][Ii][Cc][Aa]://Modelica/'),'../../'),
 (re.compile(r'^[Mm][Oo][Dd][Ee][Ll][Ii][Cc][Aa]://ModelicaReference/'),'../../'),
 (re.compile(r'^[Mm][Oo][Dd][Ee][Ll][Ii][Cc][Aa]://ModelicaServices/'),'../../'),
 (re.compile(r'.*/omlibrary/'), ''),
 (re.compile(r'[Mm][Oo][Dd][Ee][Ll][Ii][Cc][Aa]://([A-Za-z0-9.\'()_]*#)'), r'\1.html#'),
 (re.compile(r'[Mm][Oo][Dd][Ee][Ll][Ii][Cc][Aa]://([A-Za-z0-9.\'()_]*)'), r'\1.html'),
]

def linkreplace(link):
  for (regex,repl) in repls:
    link = regex.sub(repl,link)
  return link

for filepath in sorted(glob.glob('/home/dietmarw/.workspace/Modelica.git/Modelica/Resources/help/*.html')):
  tag = '[Checking file %s]:\n' % filepath
  sys.stdout.write(tag)
  sys.stderr.write(tag)
  pid = sub.call(['tidy', '-modify', '-quiet', filepath])
  with open(filepath,'r') as html_file:
    soup = BeautifulSoup(html_file)
    for a in soup.findAll('a'):
      try:
        a['href'] = linkreplace(a['href'])
      except:
        pass
      for img in soup.findAll('img'):
        try:
          img['src'] = linkreplace(img['src'])
        except:
          pass
  with open(filepath,'w') as html_file:
    html_file.write(soup.__str__())