doc/markupSanitizer.py

   1 #!/usr/bin/env python
   2
   3 from bs4 import BeautifulSoup,NavigableString
   4 import sys
   5 import os
   6
   7 if sys.version < '3':
   8     import codecs
   9     def u(x):
  10         return codecs.unicode_escape_decode(x)[0]
  11 else:
  12     def u(x):
  13         return x
  14
  15 # Accept filename as user input
  16 argc = len( sys.argv )
  17 if (argc < 2): raise Exception
  18 fileName = sys.argv[1];
  19
  20 # Construct a DOM object
  21 soup = BeautifulSoup(open(fileName), "lxml")
  22
  23 # Assuming, tt tags are not spewed recklessly by latex2html,
  24 # replace them with code tags
  25 for t in soup('tt'):
  26     t.wrap( soup.new_tag('code') )
  27
  28     # Remove first space to fix wrong indentation
  29     tmp = t.contents[0]
  30     if isinstance(tmp, NavigableString) and tmp.startswith('\n'):
  31         t.contents[0].replace_with(soup.new_string(tmp.replace('\n ', '\n', 1)))
  32
  33     t.unwrap()
  34
  35 # Rewrap all div class=alltt blocks in pre tags
  36 for d in soup('div','alltt'):
  37     d.wrap( soup.new_tag('pre') )
  38     d.unwrap()
  39
  40 # Remove br and span tags from within pre sections
  41 for p in soup('pre'):
  42     for b in p('br'):
  43         b.extract()
  44     for s in p('span'):
  45         s.unwrap()
  46
  47 # Remove all useless class 'arabic' spans
  48 for s in soup('span','arabic'):
  49     s.unwrap()
  50
  51 # Extract the navigation bar
  52 navmenu = soup.find('div', 'navigation')
  53 if navmenu:
  54     navmenu.extract()
  55
  56 # Wrap the remaining contents within a div
  57 if not soup.find('div', id='maincontainer'):
  58     soup.body['id'] = 'maincontainer'
  59     soup.body.name = 'div'
  60     soup.find('div', id='maincontainer').wrap( soup.new_tag('body') )
  61
  62 if navmenu:
  63     # If this navmenu doesn't already have a TOC, insert one
  64     if not navmenu.find('ul','manual-toc'):
  65         # Add a toc within the navmenu
  66         navmenuTOC = BeautifulSoup(open("tmp-navmenu.html"), "lxml")
  67         navmenuTOC = navmenuTOC.find('ul','manual-toc').extract()
  68         navmenuTOC.append(BeautifulSoup("".join([
  69             '<li><a href="http://charm.cs.illinois.edu">PPL Homepage</a></li>',
  70             '<li><a href="http://charm.cs.illinois.edu/help">Other Manuals</a></li>'
  71         ]), "lxml"))
  72         navmenu.append(navmenuTOC)
  73
  74     # Insert navigation symbols to prev and next links
  75     prevsymbol = soup.new_tag('span')
  76     prevsymbol['class'] = 'navsymbol'
  77     prevsymbol.string = u('\xab')
  78     prv = navmenu.find('li',id='nav-prev')
  79     if prv:
  80         prv.find('a').insert(0, prevsymbol)
  81
  82     nextsymbol = soup.new_tag('span')
  83     nextsymbol['class'] = 'navsymbol'
  84     nextsymbol.string = u('\xbb')
  85     nxt = navmenu.find('li',id='nav-next')
  86     if nxt:
  87         nxt.find('a').append(nextsymbol)
  88
  89     # Reinsert the navigation bar at the end
  90     soup.body.append(navmenu)
  91
  92 # Extract the title
  93 titl = soup.find('title')
  94
  95 # Replace the head section with the user-supplied head markup
  96 soup.find('head').extract()
  97 newhead = BeautifulSoup(open("../assets/head.html"), "lxml")
  98 newhead = newhead.find('head').extract()
  99 newhead.append(titl)
 100 soup.html.body.insert_before(newhead)
 101
 102 # Print cleaned up markup to stdout
 103 print( soup.encode("utf-8") )