# -*- 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 xml.parsers.expat import ExpatError
from datetime import date
import time
-from librarian import ValidationError, NoDublinCore, ParseError, DCNS, RDFNS
+from librarian import (ValidationError, NoDublinCore, ParseError, DCNS, RDFNS,
+ WLURI)
import lxml.etree as etree # ElementTree API using libxml2
from lxml.etree import XMLSyntaxError
def __init__(self, last_name, *first_names):
self.last_name = last_name
self.first_names = first_names
-
+
@classmethod
def from_text(cls, text):
parts = [ token.strip() for token in text.split(',') ]
raise ValueError("Found a comma, but no names given: \"%s\" -> %r." % (text, parts))
names = [ name for name in parts[1].split() if len(name) ] # all non-whitespace tokens
return cls(surname, *names)
-
+
+ def readable(self):
+ return u" ".join(self.first_names + (self.last_name,))
+
def __eq__(self, right):
return self.last_name == right.last_name and self.first_names == right.first_names
-
-
+
+ def __cmp__(self, other):
+ return cmp((self.last_name, self.first_names), (other.last_name, other.first_names))
+
+ def __hash__(self):
+ return hash((self.last_name, self.first_names))
+
def __unicode__(self):
if len(self.first_names) > 0:
return '%s, %s' % (self.last_name, ' '.join(self.first_names))
else:
return self.last_name
-
+
def __repr__(self):
return 'Person(last_name=%r, first_names=*%r)' % (self.last_name, self.first_names)
return val
return [ self.validator(v) if v is not None else v for v in val ]
elif len(val) > 1:
- raise ValidationError("Mulitply values not allowed for field '%s'" % self.uri)
+ raise ValidationError("Multiple values not allowed for field '%s'" % self.uri)
elif len(val) == 0:
raise ValidationError("Field %s has no value to assign. Check your defaults." % self.uri)
else:
-class BookInfo(object):
+class BookInfo(object):
FIELDS = (
Field( DCNS('creator'), 'author', as_person),
Field( DCNS('title'), 'title'),
Field( DCNS('subject.period'), 'epochs', salias='epoch', multiple=True),
Field( DCNS('subject.type'), 'kinds', salias='kind', multiple=True),
Field( DCNS('subject.genre'), 'genres', salias='genre', multiple=True),
+ Field( DCNS('audience'), 'audiences', salias='audience', multiple=True,
+ required=False),
Field( DCNS('date'), 'created_at', as_date),
Field( DCNS('date.pd'), 'released_to_public_domain_at', as_date, required=False),
Field( DCNS('contributor.editor'), 'editors', \
Field( DCNS('publisher'), 'publisher'),
Field( DCNS('source'), 'source_name', required=False),
Field( DCNS('source.URL'), 'source_url', required=False),
- Field( DCNS('identifier.url'), 'url'),
- Field( DCNS('relation.hasPart'), 'parts', multiple=True, required=False),
+ Field( DCNS('identifier.url'), 'url', WLURI),
+ Field( DCNS('relation.hasPart'), 'parts', WLURI, multiple=True, required=False),
Field( DCNS('rights.license'), 'license', required=False),
Field( DCNS('rights'), 'license_description'),
+ Field( DCNS('language'), 'language'),
+ Field( DCNS('description'), 'description', required=False),
)
@classmethod
def from_string(cls, xml):
from StringIO import StringIO
return cls.from_file(StringIO(xml))
-
+
@classmethod
def from_file(cls, xmlfile):
- desc_tag = None
+ desc_tag = None
try:
- iter = etree.iterparse(xmlfile, ['start', 'end'])
+ iter = etree.iterparse(xmlfile, ['start', 'end'])
for (event, element) in iter:
if element.tag == RDFNS('RDF') and event == 'start':
desc_tag = element
break
# if there is no end, Expat should yell at us with an ExpatError
-
+
# extract data from the element and make the info
return cls.from_element(desc_tag)
except XMLSyntaxError, e:
# the tree is already parsed, so we don't need to worry about Expat errors
field_dict = {}
desc = rdf_tag.find(".//" + RDFNS('Description') )
-
+
if desc is None:
raise NoDublinCore("No DublinCore section found.")
fv = field_dict.get(e.tag, [])
fv.append(e.text)
field_dict[e.tag] = fv
-
+
return cls( desc.attrib, field_dict )
def __init__(self, rdf_attrs, dc_fields):
"""rdf_attrs should be a dictionary-like object with any attributes of the RDF:Description.
- dc_fields - dictionary mapping DC fields (with namespace) to list of text values for the
+ dc_fields - dictionary mapping DC fields (with namespace) to list of text values for the
given field. """
self.about = rdf_attrs.get(RDFNS('about'))
self.fmap[field.name] = field
if field.salias: self.fmap[field.salias] = field
+ self.validate()
+
+ def validate(self):
+ self.url.validate_language(self.language)
+
def __getattribute__(self, name):
try:
field = object.__getattribute__(self, 'fmap')[name]
else: # singular alias
if not field.multiple:
raise "OUCH!! for field %s" % name
-
+
return value[0]
except (KeyError, AttributeError):
return object.__getattribute__(self, name)
return object.__setattr__(self, name, newvalue)
def update(self, field_dict):
- """Update using field_dict. Verify correctness, but don't check if all
+ """Update using field_dict. Verify correctness, but don't check if all
required fields are present."""
for field in self.FIELDS:
if field_dict.has_key(field.name):
"""XML representation of this object."""
#etree._namespace_map[str(self.RDF)] = 'rdf'
#etree._namespace_map[str(self.DC)] = 'dc'
-
+
if parent is None:
root = etree.Element(RDFNS('RDF'))
else:
root = parent.makeelement(RDFNS('RDF'))
description = etree.SubElement(root, RDFNS('Description'))
-
+
if self.about:
description.set(RDFNS('about'), self.about)
-
+
for field in self.FIELDS:
v = getattr(self, field.name, None)
if v is not None:
if len(v) == 0: continue
for x in v:
e = etree.Element(field.uri)
- e.text = unicode(x)
+ if x is not None:
+ e.text = unicode(x)
description.append(e)
else:
e = etree.Element(field.uri)
e.text = unicode(v)
description.append(e)
-
+
return root
if v is not None:
if field.multiple:
if len(v) == 0: continue
- v = [ unicode(x) for x in v if v is not None ]
+ v = [ unicode(x) for x in v if x is not None ]
else:
v = unicode(v)
-
+
dc[field.name] = {'uri': field.uri, 'value': v}
rdf['fields'] = dc
return rdf
if v is not None:
if field.multiple:
if len(v) == 0: continue
- v = [ unicode(x) for x in v if v is not None ]
+ v = [ unicode(x) for x in v if x is not None ]
else:
v = unicode(v)
result[field.name] = v
if field.salias:
v = getattr(self, field.salias)
if v is not None: result[field.salias] = unicode(v)
-
+
return result
def parse(file_name):
fnp / librarian.git / blobdiff