bin/bookfragments.py

   1 #!/usr/bin/env python
   2 # -*- coding: utf-8 -*-
   3 import optparse
   4 import os
   5 import copy
   6
   7 from lxml import etree
   8
   9
  10 # Parse args
  11 usage = """Usage: %prog [options] SOURCE [SOURCE...]
  12 Extract theme fragments from SOURCE."""
  13
  14 parser = optparse.OptionParser(usage=usage)
  15
  16 parser.add_option('-v', '--verbose', action='store_true', dest='verbose', default=False,
  17     help='print status messages to stdout')
  18
  19 options, input_filenames = parser.parse_args()
  20
  21 if len(input_filenames) < 1:
  22     parser.print_help()
  23     exit(1)
  24
  25
  26 class Fragment(object):
  27     def __init__(self, id, themes):
  28         super(Fragment, self).__init__()
  29         self.id = id
  30         self.themes = themes
  31         self.events = []
  32
  33     def append(self, event, element):
  34         self.events.append((event, element))
  35
  36     def closed_events(self):
  37         stack = []
  38         for event, element in self.events:
  39             if event == 'start':
  40                 stack.append(('end', element))
  41             elif event == 'end':
  42                 try:
  43                     stack.pop()
  44                 except IndexError:
  45                     print 'CLOSED NON-OPEN TAG:', element
  46
  47         stack.reverse()
  48         return self.events + stack
  49
  50     def to_string(self):
  51         result = []
  52         for event, element in self.closed_events():
  53             if event == 'start':
  54                 result.append(u'<%s %s>' % (element.tag, ' '.join('%s="%s"' % (k, v) for k, v in element.attrib.items())))
  55                 if element.text:
  56                     result.append(element.text)
  57             elif event == 'end':
  58                 result.append(u'</%s>' % element.tag)
  59                 if element.tail:
  60                     result.append(element.tail)
  61             else:
  62                 result.append(element)
  63
  64         return ''.join(result)
  65
  66     def __unicode__(self):
  67         return self.to_string()
  68
  69
  70 # Do some real work
  71 for input_filename in input_filenames:
  72     if options.verbose:
  73         print input_filename
  74
  75     output_filename = os.path.splitext(input_filename)[0] + '.fragments.html'
  76
  77     open_fragments = {}
  78     closed_fragments = {}
  79     lost_text = []
  80
  81     for event, element in etree.iterparse(input_filename, events=('start', 'end')):
  82
  83         # Process begin and end elements
  84         if element.tag == 'span' and element.get('class', '') in ('theme-begin', 'theme-end'):
  85             if not event == 'end': continue # Process elements only once, on end event
  86
  87             # Open new fragment
  88             if element.get('class', '') == 'theme-begin':
  89                 fragment = Fragment(id=element.get('fid'), themes=element.text)
  90
  91                 # Append parents
  92                 if element.getparent().tag != 'body':
  93                     parents = [element.getparent()]
  94                     while parents[-1].getparent().tag != 'body':
  95                         parents.append(parents[-1].getparent())
  96
  97                     parents.reverse()
  98                     for parent in parents:
  99                         fragment.append('start', parent)
 100
 101                 open_fragments[fragment.id] = fragment
 102
 103             # Close existing fragment
 104             else:
 105                 try:
 106                     fragment = open_fragments[element.get('fid')]
 107                 except KeyError:
 108                     print '%s:closed not open fragment #%s' % (input_filename, element.get('fid'))
 109                 else:
 110                     closed_fragments[fragment.id] = fragment
 111                     del open_fragments[fragment.id]
 112
 113             # Append element tail to lost_text (we don't want to lose any text)
 114             if element.tail:
 115                 for fragment_id in open_fragments:
 116                     open_fragments[fragment_id].append('text', element.tail)
 117
 118
 119         # Process all elements except begin and end
 120         else:
 121             # Omit annotation tags
 122             if len(element.get('name', '')) or element.get('class', '') == 'annotation':
 123                 if event == 'end' and element.tail:
 124                     for fragment_id in open_fragments:
 125                         open_fragments[fragment_id].append('text', element.tail)
 126             else:
 127                 for fragment_id in open_fragments:
 128                     open_fragments[fragment_id].append(event, copy.copy(element))
 129
 130
 131     for fragment_id in open_fragments:
 132         print '%s:warning:unclosed fragment #%s' % (input_filename, fragment_id)
 133
 134     output_file = open(output_filename, 'w')
 135     output_file.write("""
 136         <!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.1//EN" "http://www.w3.org/TR/xhtml11/DTD/xhtml11.dtd">
 137         <html><head>
 138             <title>bookfragments output</title>
 139             <meta http-equiv="content-type" content="text/html;charset=utf-8"/>
 140             <link rel="stylesheet" href="master.css" type="text/css" media="screen" charset="utf-8" />
 141         </head>
 142         <body>""")
 143     for fragment in closed_fragments.values():
 144         html = u'<div class="fragment"><h3>[#%s] %s</h3>%s</div>' % (fragment.id, fragment.themes, fragment)
 145         output_file.write(html.encode('utf-8'))
 146     output_file.write('</body></html>')
 147     output_file.close()
 148