librarian/parser.py

   1 # -*- coding: utf-8 -*-
   2 from librarian import ValidationError, NoDublinCore, dcparser, ParseError
   3 from xml.parsers.expat import ExpatError
   4 from lxml import etree
   5 from lxml.etree import XMLSyntaxError
   6
   7 import re
   8 from StringIO import StringIO
   9
  10 class WLDocument(object):
  11     LINE_SWAP_EXPR = re.compile(r'/\s', re.MULTILINE | re.UNICODE);
  12
  13     def __init__(self, edoc):
  14         self.edoc = edoc
  15
  16         root_elem = edoc.getroot()
  17         rdf_ns = dcparser.BookInfo.RDF
  18         dc_path = './/' + rdf_ns('RDF')
  19
  20         if root_elem.tag != 'utwor':
  21             raise ValidationError("Invalid root element. Found '%s', should be 'utwor'" % root_elem.tag)
  22
  23         self.rdf_elem = root_elem.find(dc_path)
  24
  25         if self.rdf_elem is None:
  26             raise NoDublinCore('Document has no DublinCore - which is required.')
  27
  28         self.book_info = dcparser.BookInfo.from_element(self.rdf_elem)
  29
  30     @classmethod
  31     def from_string(cls, xml, swap_endlines=False):
  32         return cls.from_file(StringIO(xml), swap_endlines)
  33
  34     @classmethod
  35     def from_file(cls, xmlfile, swap_endlines=False):
  36
  37         # first, prepare for parsing
  38         if isinstance(xmlfile, basestring):
  39             file = open(xmlfile, 'rb')
  40             try:
  41                 data = file.read()
  42             finally:
  43                 file.close()
  44         else:
  45             data = xmlfile.read()
  46
  47         if not isinstance(data, unicode):
  48             data = data.decode('utf-8')
  49
  50         if swap_endlines:
  51             data = cls.LINE_SWAP_EXPR.sub(u'<br />\n', data)
  52
  53         try:
  54             parser = etree.XMLParser(remove_blank_text=True)
  55             return cls( etree.parse(StringIO(data), parser) )
  56         except XMLSyntaxError, e:
  57              raise ParseError(e.message)
  58         except ExpatError, e:
  59             raise ParseError(e.message)
  60
  61     def transform(self, stylesheet, **options):
  62         return self.edoc.xslt(stylesheet, **options)
  63
  64     def update_dc(self):
  65         parent = self.rdf_elem.getparent()
  66         parent.replace( self.rdf_elem, self.book_info.to_etree(parent) )
  67
  68     def serialize(self):
  69         self.update_dc()
  70         return etree.tostring(self.edoc, encoding=unicode, pretty_print=True)