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.util import get_translation
   7
   8
   9 class WLElement(etree.ElementBase):
  10     TXT_TOP_MARGIN = 0
  11     TXT_BOTTOM_MARGIN = 0
  12     TXT_PREFIX = ""
  13     TXT_SUFFIX = ""
  14
  15     HTML_TAG = None
  16     HTML_ATTR = {}
  17     HTML_CLASS = None
  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     @property
  52     def gettext(self):
  53         return get_translation(self.meta.language).gettext
  54
  55     def normalize_text(self, text):
  56         text = text or ''
  57         for e, s in self.text_substitutions:
  58             text = text.replace(e, s)
  59         text = re.sub(r'\s+', ' ', text)
  60         return text
  61
  62     def _build_inner(self, builder, build_method):
  63         child_count = len(self)
  64         if self.CAN_HAVE_TEXT and self.text:
  65             text = self.normalize_text(self.text)
  66             if self.STRIP:
  67                 text = text.lstrip()
  68                 if not child_count:
  69                     text = text.rstrip()
  70             builder.push_text(text)
  71         for i, child in enumerate(self):
  72             if isinstance(child, WLElement):
  73                 getattr(child, build_method)(builder)
  74             if self.CAN_HAVE_TEXT and child.tail:
  75                 text = self.normalize_text(child.tail)
  76                 if self.STRIP and i == child_count - 1:
  77                     text = text.rstrip()
  78                 builder.push_text(text)
  79
  80     def _txt_build_inner(self, builder):
  81         self._build_inner(builder, 'txt_build')
  82
  83     def txt_build(self, builder):
  84         if hasattr(self, 'TXT_LEGACY_TOP_MARGIN'):
  85             builder.push_legacy_margin(self.TXT_LEGACY_TOP_MARGIN)
  86         else:
  87             builder.push_margin(self.TXT_TOP_MARGIN)
  88         builder.push_text(self.TXT_PREFIX, True)
  89         self._txt_build_inner(builder)
  90         builder.push_text(self.TXT_SUFFIX, True)
  91         if hasattr(self, 'TXT_LEGACY_BOTTOM_MARGIN'):
  92             builder.push_legacy_margin(self.TXT_LEGACY_BOTTOM_MARGIN)
  93         else:
  94             builder.push_margin(self.TXT_BOTTOM_MARGIN)
  95
  96     def _html_build_inner(self, builder):
  97         self._build_inner(builder, 'html_build')
  98
  99     def get_html_attr(self, builder):
 100         attr = self.HTML_ATTR.copy()
 101         if self.HTML_CLASS:
 102             attr['class'] = self.HTML_CLASS
 103         # always copy the id attribute (?)
 104         if self.attrib.get('id'):
 105             attr['id'] = self.attrib['id']
 106         elif '_compat_section_id' in self.attrib:
 107             attr['id'] = self.attrib['_compat_section_id']
 108         return attr
 109
 110     def html_build(self, builder):
 111         if self.HTML_TAG:
 112             builder.start_element(
 113                 self.HTML_TAG,
 114                 self.get_html_attr(builder),
 115             )
 116
 117         self._html_build_inner(builder)
 118         if self.HTML_TAG:
 119             builder.end_element()
 120
 121     def sanitize(self):
 122         # TODO: Remove insanity here.
 123         for e in self:
 124             if isinstance(e, WLElement):
 125                 e.sanitize()