X-Git-Url: https://git.mdrn.pl/librarian.git/blobdiff_plain/f8e5f031c04122d65d1066077be3920ae95518ae..e0174e46955329347268e756e2662e77441cf09b:/librarian/parser.py?ds=sidebyside
diff --git a/librarian/parser.py b/librarian/parser.py
index 55b4e4b..341eaf8 100644
--- a/librarian/parser.py
+++ b/librarian/parser.py
@@ -1,6 +1,10 @@
# -*- coding: utf-8 -*-
+#
+# 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, DCNS
+from librarian import RDFNS
from librarian import dcparser
from xml.parsers.expat import ExpatError
@@ -17,9 +21,9 @@ class WLDocument(object):
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)
@@ -28,17 +32,17 @@ class WLDocument(object):
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, swap_endlines=False, parse_dublincore=True):
- return cls.from_file(StringIO(xml), swap_endlines, parse_dublincore=parse_dublincore)
+ 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):
+ def from_file(cls, xmlfile, swap_endlines=False, parse_dublincore=True, preserve_lines=True):
# first, prepare for parsing
if isinstance(xmlfile, basestring):
@@ -53,27 +57,29 @@ class WLDocument(object):
if not isinstance(data, unicode):
data = data.decode('utf-8')
+ data = data.replace(u'\ufeff', '')
+
if swap_endlines:
- data = cls.LINE_SWAP_EXPR.sub(u'
\n', data)
-
+ sub = u'
'
+ if preserve_lines:
+ sub += u'\n'
+ data = cls.LINE_SWAP_EXPR.sub(sub, data)
+
try:
- parser = etree.XMLParser(remove_blank_text=True)
+ 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)
+ raise ParseError(e)
- def part_as_text(self, path):
- # convert the path to XPath
- print "[L] Retrieving part:", path
+ def chunk(self, path):
+ # convert the path to XPath
+ expr = self.path_to_xpath(path)
+ elems = self.edoc.xpath(expr)
- elems = self.edoc.xpath(self.path_to_xpath(path))
- print "[L] xpath", elems
-
if len(elems) == 0:
- return None
-
- return etree.tostring(elems[0], encoding=unicode, pretty_print=True)
-
+ return None
+ else:
+ return elems[0]
def path_to_xpath(self, path):
parts = []
@@ -84,7 +90,7 @@ class WLDocument(object):
parts.append(part)
else:
tag, n = match.groups()
- parts.append("node()[position() = %d and name() = '%s']" % (int(n), tag) )
+ parts.append("*[%d][name() = '%s']" % (int(n)+1, tag) )
if parts[0] == '.':
parts[0] = ''
@@ -95,8 +101,9 @@ class WLDocument(object):
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) )
+ 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()
@@ -108,10 +115,20 @@ class WLDocument(object):
for key, data in chunk_dict.iteritems():
try:
xpath = self.path_to_xpath(key)
- node = self.edoc.xpath(xpath)[0]
- repl = etree.fromstring(data)
+ 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
\ No newline at end of file
+ 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