src/librarian/document.py

   1 import gettext
   2 import os
   3 import re
   4 from urllib.request import urlopen
   5 from lxml import etree
   6 from .parser import parser
   7 from . import dcparser, DCNS
   8 from .functions import lang_code_3to2
   9
  10
  11 class WLDocument:
  12     def __init__(self, filename=None, url=None):
  13         source = filename or urlopen(url)
  14         tree = etree.parse(source, parser=parser)
  15         self.tree = tree
  16         tree.getroot().document = self
  17         self.base_meta = dcparser.BookInfo({}, {
  18             DCNS('language'): ["pol"],
  19         }, validate_required=False)
  20
  21     @property
  22     def meta(self):
  23         # Allow metadata of the master element as document meta.
  24         #master = self.tree.getroot()[-1]
  25         return self.tree.getroot().meta
  26         return master.meta
  27
  28     def build(self, builder, **kwargs):
  29         return builder().build(self, **kwargs)
  30
  31     def _compat_assign_ordered_ids(self):
  32         """
  33         Compatibility: ids in document order, to be roughly compatible with legacy
  34         footnote ids. Just for testing consistency, change to some sane identifiers
  35         at convenience.
  36         """
  37         EXPR = re.compile(r'/\s', re.MULTILINE | re.UNICODE)
  38         def _compat_assign_ordered_ids_in_elem(elem, i):
  39             elem.attrib['_compat_ordered_id'] = str(i)
  40             i += 1
  41             if getattr(elem, 'HTML_CLASS', None) == 'stanza':
  42                 if elem.text:
  43                     i += len(EXPR.split(elem.text)) - 1
  44                 for sub in elem:
  45                     i = _compat_assign_ordered_ids_in_elem(sub, i)
  46                     if sub.tail:
  47                         i += len(EXPR.split(sub.tail)) - 1
  48             else:
  49                 if elem.tag in ('uwaga', 'extra'):
  50                     return i
  51                 for sub in elem:
  52                     i = _compat_assign_ordered_ids_in_elem(sub, i)
  53             return i
  54
  55         _compat_assign_ordered_ids_in_elem(self.tree.getroot(), 4)
  56
  57     def _compat_assign_section_ids(self):
  58         """
  59         Ids in master-section order. These need to be compatible with the
  60         #secN anchors used by WL search results page to link to fragments.
  61         """
  62         def _compat_assigns_section_ids_in_elem(elem, prefix='sec'):
  63             for i, child in enumerate(elem):
  64                 idfier = '{}{}'.format(prefix, i + 1)
  65                 child.attrib['_compat_section_id'] = idfier
  66                 _compat_assigns_section_ids_in_elem(child, idfier + '-')
  67         _compat_assigns_section_ids_in_elem(self.tree.getroot().master)
  68