X-Git-Url: https://git.mdrn.pl/librarian.git/blobdiff_plain/be641fe5a98107edfe4fa423337c0bf8a06a9a4b..30a74b436b8a81adbdbeea5a9b2f99dfc2e4949f:/librarian/parser.py?ds=inline

diff --git a/librarian/parser.py b/librarian/parser.py
index 3f9f710..7b48624 100644
--- a/librarian/parser.py
+++ b/librarian/parser.py
@@ -3,130 +3,28 @@
 # This file is part of Librarian, licensed under GNU Affero GPLv3 or later.
 # Copyright © Fundacja Nowoczesna Polska. See NOTICE for more information.
 #
-from librarian import ValidationError, NoDublinCore,  ParseError
-from librarian import RDFNS
-from librarian import dcparser
-
-from xml.parsers.expat import ExpatError
 from lxml import etree
-from lxml.etree import XMLSyntaxError, XSLTApplyError
-
-import re
-from StringIO import StringIO
-
-class WLDocument(object):
-    LINE_SWAP_EXPR = re.compile(r'/\s', re.MULTILINE | re.UNICODE);
-
-    def __init__(self, edoc, parse_dublincore=True):
-        self.edoc = edoc
-
-        root_elem = edoc.getroot()
-
-        dc_path = './/' + RDFNS('RDF')
-
-        if root_elem.tag != 'utwor':
-            raise ValidationError("Invalid root element. Found '%s', should be 'utwor'" % root_elem.tag)
-
-        if parse_dublincore:
-            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)
-        else:
-            self.book_info = None
-
-    @classmethod
-    def from_string(cls, xml, *args, **kwargs):
-        return cls.from_file(StringIO(xml), *args, **kwargs)
-
-    @classmethod
-    def from_file(cls, xmlfile, swap_endlines=False, parse_dublincore=True, preserve_lines=True):
-
-        # 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:
-            sub = u'<br/>'
-            if preserve_lines:
-                sub += u'\n'
-            data = cls.LINE_SWAP_EXPR.sub(sub, data)
-
-        try:
-            parser = etree.XMLParser(remove_blank_text=False)
-            return cls(etree.parse(StringIO(data), parser), parse_dublincore=parse_dublincore)
-        except (ExpatError, XMLSyntaxError, XSLTApplyError), e:
-            raise ParseError(e)
-
-    def chunk(self, path):
-        # convert the path to XPath
-        expr = self.path_to_xpath(path)
-        elems = self.edoc.xpath(expr)
-
-        if len(elems) == 0:
-            return None
-        else:
-            return elems[0]
-
-    def path_to_xpath(self, path):
-        parts = []
-
-        for part in path.split('/'):
-            match = re.match(r'([^\[]+)\[(\d+)\]', part)
-            if not match:
-                parts.append(part)
-            else:
-                tag, n = match.groups()
-                parts.append("*[%d][name() = '%s']" % (int(n)+1, tag) )
-
-        if parts[0] == '.':
-            parts[0] = ''
-
-        return '/'.join(parts)
-
-    def transform(self, stylesheet, **options):
-        return self.edoc.xslt(stylesheet, **options)
-
-    def update_dc(self):
-        if self.book_info:
-            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)
-
-    def merge_chunks(self, chunk_dict):
-        unmerged = []
-
-        for key, data in chunk_dict.iteritems():
-            try:
-                xpath = self.path_to_xpath(key)
-                node = self.edoc.xpath(xpath)[0]
-                repl = etree.fromstring(u"<%s>%s</%s>" %(node.tag, data, node.tag) )
-                node.getparent().replace(node, repl);
-            except Exception, e:
-                unmerged.append( repr( (key, xpath, e) ) )
-
-        return unmerged
-
-    def clean_ed_note(self):
-        """ deletes forbidden tags from nota_red """
-
-        for node in self.edoc.xpath('|'.join('//nota_red//%s' % tag for tag in
-                    ('pa', 'pe', 'pr', 'pt', 'begin', 'end', 'motyw'))):
-            tail = node.tail
-            node.clear()
-            node.tag = 'span'
-            node.tail = tail
+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