src/librarian/elements/base.py

   1 # -*- coding: utf-8
   2
   3 import re
   4 from lxml import etree
   5 from librarian import dcparser, RDFNS
   6 from librarian.html import raw_printable_text
   7 from librarian.util import get_translation
   8
   9
  10 class WLElement(etree.ElementBase):
  11     SECTION_PRECEDENCE = None
  12
  13     TXT_TOP_MARGIN = 0
  14     TXT_BOTTOM_MARGIN = 0
  15     TXT_PREFIX = ""
  16     TXT_SUFFIX = ""
  17
  18     HTML_TAG = None
  19     HTML_ATTR = {}
  20     HTML_CLASS = None
  21
  22     EPUB_TAG = None
  23     EPUB_ATTR = {}
  24     EPUB_CLASS = None
  25     EPUB_START_CHUNK = False
  26
  27     CAN_HAVE_TEXT = True
  28     STRIP = False
  29
  30     text_substitutions = [
  31         (u'---', u'—'),
  32         (u'--', u'–'),
  33         #(u'...', u'…'),  # Temporary turnoff for epub
  34         (u',,', u'„'),
  35         (u'"', u'”'),
  36         ('\ufeff', ''),
  37
  38         ("'", "\u2019"),    # This was enabled for epub.
  39     ]
  40
  41     @property
  42     def meta_object(self):
  43         if not hasattr(self, '_meta_object'):
  44             elem = self.find(RDFNS('RDF'))
  45             if elem is not None:
  46                 self._meta_object = dcparser.BookInfo.from_element(elem)
  47             else:
  48                 self._meta_object = None
  49         return self._meta_object
  50
  51     @property
  52     def meta(self):
  53         if self.meta_object is not None:
  54             return self.meta_object
  55         else:
  56             if self.getparent() is not None:
  57                 return self.getparent().meta
  58             else:
  59                 return self.document.base_meta
  60
  61     @property
  62     def gettext(self):
  63         return get_translation(self.meta.language).gettext
  64
  65     def raw_printable_text(self):
  66         # TODO: podtagi, wyroznienia, etc
  67         t = ''
  68         t += self.normalize_text(self.text)
  69         for c in self:
  70             if c.tag not in ('pe', 'pa', 'pt', 'pr', 'motyw'):
  71                 t += c.raw_printable_text()
  72             t += self.normalize_text(c.tail)
  73         return t
  74
  75     def normalize_text(self, text):
  76         text = text or ''
  77         for e, s in self.text_substitutions:
  78             text = text.replace(e, s)
  79             # FIXME: TEmporary turnoff
  80 #        text = re.sub(r'\s+', ' ', text)
  81 ### TODO: Added now for epub
  82         text = re.sub(r'(?<=\s\w)\s+', u'\u00A0', text)
  83
  84         return text
  85
  86     def _build_inner(self, builder, build_method):
  87         child_count = len(self)
  88         if self.CAN_HAVE_TEXT and self.text:
  89             text = self.normalize_text(self.text)
  90             if self.STRIP:
  91                 text = text.lstrip()
  92                 if not child_count:
  93                     text = text.rstrip()
  94             builder.push_text(text)
  95         for i, child in enumerate(self):
  96             if isinstance(child, WLElement):
  97                 getattr(child, build_method)(builder)
  98             if self.CAN_HAVE_TEXT and child.tail:
  99                 text = self.normalize_text(child.tail)
 100                 if self.STRIP and i == child_count - 1:
 101                     text = text.rstrip()
 102                 builder.push_text(text)
 103
 104     def _txt_build_inner(self, builder):
 105         self._build_inner(builder, 'txt_build')
 106
 107     def txt_build(self, builder):
 108         if hasattr(self, 'TXT_LEGACY_TOP_MARGIN'):
 109             builder.push_legacy_margin(self.TXT_LEGACY_TOP_MARGIN)
 110         else:
 111             builder.push_margin(self.TXT_TOP_MARGIN)
 112         builder.push_text(self.TXT_PREFIX, True)
 113         self._txt_build_inner(builder)
 114         builder.push_text(self.TXT_SUFFIX, True)
 115         if hasattr(self, 'TXT_LEGACY_BOTTOM_MARGIN'):
 116             builder.push_legacy_margin(self.TXT_LEGACY_BOTTOM_MARGIN)
 117         else:
 118             builder.push_margin(self.TXT_BOTTOM_MARGIN)
 119
 120     def _html_build_inner(self, builder):
 121         self._build_inner(builder, 'html_build')
 122
 123     def get_html_attr(self, builder):
 124         attr = self.HTML_ATTR.copy()
 125         if self.HTML_CLASS:
 126             attr['class'] = self.HTML_CLASS
 127         # always copy the id attribute (?)
 128         if self.attrib.get('id'):
 129             attr['id'] = self.attrib['id']
 130         elif '_compat_section_id' in self.attrib:
 131             attr['id'] = self.attrib['_compat_section_id']
 132         return attr
 133
 134     def html_build(self, builder):
 135         if self.HTML_TAG:
 136             builder.start_element(
 137                 self.HTML_TAG,
 138                 self.get_html_attr(builder),
 139             )
 140
 141         self._html_build_inner(builder)
 142         if self.HTML_TAG:
 143             builder.end_element()
 144
 145     def _epub_build_inner(self, builder):
 146         self._build_inner(builder, 'epub_build')
 147
 148     def get_epub_attr(self, builder):
 149         attr = self.EPUB_ATTR.copy()
 150         if self.EPUB_CLASS:
 151             attr['class'] = self.EPUB_CLASS
 152         return attr
 153
 154     def epub_build(self, builder):
 155         # TEMPORARY
 156         self.CAN_HAVE_TEXT = True
 157         self.STRIP = False
 158
 159         if self.EPUB_START_CHUNK:
 160             builder.start_chunk()
 161
 162         fragment = None
 163         if self.SECTION_PRECEDENCE:
 164             if not self.EPUB_START_CHUNK:
 165                 fragment = 'sub%d' % builder.assign_section_number()
 166                 self.attrib['id'] = fragment
 167
 168             builder.add_toc_entry(
 169                 fragment,
 170                 self.raw_printable_text(),
 171                 self.SECTION_PRECEDENCE
 172             )
 173
 174         if self.EPUB_TAG:
 175             attr = self.get_epub_attr(builder)
 176             if fragment:
 177                 attr['id'] = fragment
 178             builder.start_element(
 179                 self.EPUB_TAG,
 180                 attr
 181             )
 182
 183         self._epub_build_inner(builder)
 184         if self.EPUB_TAG:
 185             builder.end_element()
 186
 187     def sanitize(self):
 188         # TODO: Remove insanity here.
 189         for e in self:
 190             if isinstance(e, WLElement):
 191                 e.sanitize()