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
   7
   8 class WLElement(etree.ElementBase):
   9     TXT_TOP_MARGIN = 0
  10     TXT_BOTTOM_MARGIN = 0
  11     TXT_PREFIX = ""
  12     TXT_SUFFIX = ""
  13
  14     HTML_TAG = None
  15     HTML_ATTR = {}
  16     HTML_CLASS = None
  17     HTML_SECTION = False
  18
  19     CAN_HAVE_TEXT = True
  20     STRIP = False
  21
  22     text_substitutions = [
  23         (u'---', u'—'),
  24         (u'--', u'–'),
  25         (u'...', u'…'),
  26         (u',,', u'„'),
  27         (u'"', u'”'),
  28         ('\ufeff', ''),
  29     ]
  30
  31     @property
  32     def meta_object(self):
  33         if not hasattr(self, '_meta_object'):
  34             elem = self.find(RDFNS('RDF'))
  35             if elem is not None:
  36                 self._meta_object = dcparser.BookInfo.from_element(elem)
  37             else:
  38                 self._meta_object = None
  39         return self._meta_object
  40
  41     @property
  42     def meta(self):
  43         if self.meta_object is not None:
  44             return self.meta_object
  45         else:
  46             if self.getparent() is not None:
  47                 return self.getparent().meta
  48             else:
  49                 return self.document.base_meta
  50
  51     def normalize_text(self, text):
  52         text = text or ''
  53         for e, s in self.text_substitutions:
  54             text = text.replace(e, s)
  55         text = re.sub(r'\s+', ' ', text)
  56         return text
  57
  58     def _build_inner(self, builder, build_method):
  59         child_count = len(self)
  60         if self.CAN_HAVE_TEXT and self.text:
  61             text = self.normalize_text(self.text)
  62             if self.STRIP:
  63                 text = text.lstrip()
  64                 if not child_count:
  65                     text = text.rstrip()
  66             builder.push_text(text)
  67         for i, child in enumerate(self):
  68             if isinstance(child, WLElement):
  69                 getattr(child, build_method)(builder)
  70             if self.CAN_HAVE_TEXT and child.tail:
  71                 text = self.normalize_text(child.tail)
  72                 if self.STRIP and i == child_count - 1:
  73                     text = text.rstrip()
  74                 builder.push_text(text)
  75
  76     def _txt_build_inner(self, builder):
  77         self._build_inner(builder, 'txt_build')
  78
  79     def txt_build(self, builder):
  80         if hasattr(self, 'TXT_LEGACY_TOP_MARGIN'):
  81             builder.push_legacy_margin(self.TXT_LEGACY_TOP_MARGIN)
  82         else:
  83             builder.push_margin(self.TXT_TOP_MARGIN)
  84         builder.push_text(self.TXT_PREFIX, True)
  85         self._txt_build_inner(builder)
  86         builder.push_text(self.TXT_SUFFIX, True)
  87         if hasattr(self, 'TXT_LEGACY_BOTTOM_MARGIN'):
  88             builder.push_legacy_margin(self.TXT_LEGACY_BOTTOM_MARGIN)
  89         else:
  90             builder.push_margin(self.TXT_BOTTOM_MARGIN)
  91
  92     def _html_build_inner(self, builder):
  93         self._build_inner(builder, 'html_build')
  94
  95     def get_html_attr(self, builder):
  96         attr = self.HTML_ATTR.copy()
  97         if self.HTML_CLASS:
  98             attr['class'] = self.HTML_CLASS
  99         # always copy the id attribute (?)
 100         if self.attrib.get('id'):
 101             attr['id'] = self.attrib['id']
 102         return attr
 103
 104     def html_build(self, builder):
 105         if self.HTML_SECTION:
 106             builder.start_element(
 107                 'a', {"name": "f18", "class": "target"}
 108             )
 109             builder.push_text(" ")
 110             builder.end_element()
 111
 112             builder.start_element(
 113                 "a", {"href": "#f18", "class": "anchor"}
 114             )
 115             builder.push_text("18")
 116             builder.end_element()
 117
 118
 119         if self.HTML_TAG:
 120             builder.start_element(
 121                 self.HTML_TAG,
 122                 self.get_html_attr(builder),
 123             )
 124
 125         if self.HTML_SECTION:
 126             builder.start_element(
 127                 "a", {"name": "sec34"}
 128             )
 129             builder.end_element()
 130
 131         self._html_build_inner(builder)
 132         if self.HTML_TAG:
 133             builder.end_element()
 134
 135     def sanitize(self):
 136         # TODO: Remove insanity here.
 137         for e in self:
 138             if isinstance(e, WLElement):
 139                 e.sanitize()