X-Git-Url: https://git.mdrn.pl/librarian.git/blobdiff_plain/6642c1c71c5c6ce6ef3401c8c9da84cf076b018b..d2997b73a50a2885a993408d277d511f98320ed3:/librarian/parser.py

diff --git a/librarian/parser.py b/librarian/parser.py
index 595dd97..7b48624 100644
--- a/librarian/parser.py
+++ b/librarian/parser.py
@@ -1,70 +1,30 @@
 # -*- coding: utf-8 -*-
-from librarian import ValidationError, NoDublinCore, dcparser, ParseError
-from xml.parsers.expat import ExpatError
+#
+# This file is part of Librarian, licensed under GNU Affero GPLv3 or later.
+# Copyright © Fundacja Nowoczesna Polska. See NOTICE for more information.
+#
 from lxml import etree
-from lxml.etree import XMLSyntaxError
-
-import re
-from StringIO import StringIO
-
-class WLDocument(object):
-    LINE_SWAP_EXPR = re.compile(r'/\s', re.MULTILINE | re.UNICODE);
-
-    def __init__(self, edoc):
-        self.edoc = edoc
-
-        root_elem = edoc.getroot()
-        rdf_ns = dcparser.BookInfo.RDF
-        dc_path = './/' + rdf_ns('RDF')
-        
-        if root_elem.tag != 'utwor':
-            raise ValidationError("Invalid root element. Found '%s', should be 'utwor'" % root_elem.tag)
-
-        self.rdf_elem = root_elem.find(dc_path)
-
-        if self.rdf_elem is None:
-            raise NoDublinCore('Document has no DublinCore - which is required.')
-
-        self.book_info = dcparser.BookInfo.from_element(self.rdf_elem)
-
-    @classmethod
-    def from_string(cls, xml, swap_endlines=False):
-        return cls.from_file(StringIO(xml), swap_endlines)
-
-    @classmethod
-    def from_file(cls, xmlfile, swap_endlines=False):
-
-        # first, prepare for parsing
-        if isinstance(xmlfile, basestring):
-            file = open(xmlfile, 'rb')
-            try:
-                data = file.read()
-            finally:
-                file.close()
-        else:
-            data = xmlfile.read()
-
-        if not isinstance(data, unicode):
-            data = data.decode('utf-8')
-
-        if swap_endlines:
-            data = cls.LINE_SWAP_EXPR.sub(u'<br />\n', data)
-    
-        try:
-            parser = etree.XMLParser(remove_blank_text=True)
-            return cls( etree.parse(StringIO(data), parser) )
-        except XMLSyntaxError, e:
-             raise ParseError(e.message)            
-        except ExpatError, e:
-            raise ParseError(e.message)            
-
-    def transform(self, stylesheet, **options):
-        return self.edoc.xslt(stylesheet, **options)
-
-    def update_dc(self):
-        parent = self.rdf_elem.getparent()
-        parent.replace( self.rdf_elem, self.book_info.to_etree(parent) )
-
-    def serialize(self):
-        self.update_dc()
-        return etree.tostring(self.edoc, encoding=unicode, pretty_print=True)
+from . import DCNS, SSTNS
+from . import core, meta
+
+
+class SSTParser(etree.XMLParser):
+    """ XML parser using relevant element classes. """
+    def __init__(self):
+        super(SSTParser, self).__init__()
+        lookup = etree.ElementNamespaceClassLookup()
+        self.set_element_class_lookup(lookup)
+
+        # Define core language tags.
+        sst_ns = lookup.get_namespace(SSTNS.uri)
+        sst_ns['aside'] = core.Aside
+        sst_ns['div'] = core.Div
+        sst_ns['header'] = core.Header
+        sst_ns['section'] = core.Section
+        sst_ns['span'] = core.Span
+        sst_ns['metadata'] = meta.Metadata
+
+        # Define any special metadata.
+        dc_ns = lookup.get_namespace(DCNS.uri)
+        dc_ns['creator'] = meta.Person
+        dc_ns['identifier'] = meta.Identifier