# Copyright © Fundacja Nowoczesna Polska. See NOTICE for more information.
#
import os
+import re
+import urllib
+from .utils import XMLNamespace
-class ParseError(Exception):
+
+class UnicodeException(Exception):
def __str__(self):
""" Dirty workaround for Python Unicode handling problems. """
- return self.message.message
+ return unicode(self).encode('utf-8')
def __unicode__(self):
""" Dirty workaround for Python Unicode handling problems. """
- return self.message.message
-
-class ValidationError(Exception):
+ args = self.args[0] if len(self.args) == 1 else self.args
+ try:
+ message = unicode(args)
+ except UnicodeDecodeError:
+ message = unicode(args, encoding='utf-8', errors='ignore')
+ return message
+
+class ParseError(UnicodeException):
pass
-class NoDublinCore(ValidationError):
+class ValidationError(UnicodeException):
pass
-class XMLNamespace(object):
- '''A handy structure to repsent names in an XML namespace.'''
-
- def __init__(self, uri):
- self.uri = uri
-
- def __call__(self, tag):
- return '{%s}%s' % (self.uri, tag)
-
- def __contains__(self, tag):
- return tag.startswith('{' + str(self) + '}')
-
- def __repr__(self):
- return 'XMLNamespace(%r)' % self.uri
-
- def __str__(self):
- return '%s' % self.uri
class EmptyNamespace(XMLNamespace):
def __init__(self):
NCXNS = XMLNamespace("http://www.daisy.org/z3986/2005/ncx/")
OPFNS = XMLNamespace("http://www.idpf.org/2007/opf")
-WLNS = EmptyNamespace()
-
-
-class DocProvider(object):
- """ Base class for a repository of XML files.
- Used for generating joined files, like EPUBs
- """
+SSTNS = XMLNamespace('http://nowoczesnapolska.org.pl/sst#')
- def by_slug(self, slug):
- raise NotImplemented
- def __getitem__(self, slug):
- return self.by_slug(slug)
+class WLURI(object):
+ """Represents a WL URI. Extracts slug from it."""
+ slug = None
- def by_uri(self, uri):
- return self.by_slug(uri.rsplit('/', 1)[1])
+ example = 'http://wolnelektury.pl/katalog/lektura/template/'
+ _re_wl_uri = re.compile(r'http://(www\.)?wolnelektury.pl/katalog/lektura/'
+ '(?P<slug>[-a-z0-9]+)/?$')
+ def __init__(self, uri):
+ uri = unicode(uri)
+ self.uri = uri
+ self.slug = uri.rstrip('/').rsplit('/', 1)[-1]
-class DirDocProvider(DocProvider):
- """ Serve docs from a directory of files in form <slug>.xml """
-
- def __init__(self, dir):
- self.dir = dir
- self.files = {}
-
- def by_slug(self, slug):
- return open(os.path.join(self.dir, '%s.xml' % slug))
-
-
-import lxml.etree as etree
-import dcparser
-
-DEFAULT_BOOKINFO = dcparser.BookInfo(
- { RDFNS('about'): u'http://wiki.wolnepodreczniki.pl/Lektury:Template'}, \
- { DCNS('creator'): [u'Some, Author'],
- DCNS('title'): [u'Some Title'],
- DCNS('subject.period'): [u'Unknown'],
- DCNS('subject.type'): [u'Unknown'],
- DCNS('subject.genre'): [u'Unknown'],
- DCNS('date'): ['1970-01-01'],
- # DCNS('date'): [creation_date],
- DCNS('publisher'): [u"Fundacja Nowoczesna Polska"],
- DCNS('description'):
- [u"""Publikacja zrealizowana w ramach projektu
- Wolne Lektury (http://wolnelektury.pl). Reprodukcja cyfrowa
- wykonana przez Bibliotekę Narodową z egzemplarza
- pochodzącego ze zbiorów BN."""],
- DCNS('identifier.url'):
- [u"http://wolnelektury.pl/katalog/lektura/template"],
- DCNS('rights'):
- [u"Domena publiczna - zm. [OPIS STANU PRAWNEGO TEKSTU]"] })
-
-def xinclude_forURI(uri):
- e = etree.Element(XINS("include"))
- e.set("href", uri)
- return etree.tostring(e, encoding=unicode)
-
-def wrap_text(ocrtext, creation_date, bookinfo=DEFAULT_BOOKINFO):
- """Wrap the text within the minimal XML structure with a DC template."""
- bookinfo.created_at = creation_date
-
- dcstring = etree.tostring(bookinfo.to_etree(), \
- method='xml', encoding=unicode, pretty_print=True)
+ @classmethod
+ def strict(cls, uri):
+ match = cls._re_wl_uri.match(uri)
+ if not match:
+ raise ValidationError(u'Invalid URI (%s). Should match: %s' % (
+ uri, cls._re_wl_uri.pattern))
+ return cls(uri)
- return u'<utwor>\n' + dcstring + u'\n<plain-text>\n' + ocrtext + \
- u'\n</plain-text>\n</utwor>';
+ @classmethod
+ def from_slug(cls, slug):
+ """Contructs an URI from slug.
+ >>> WLURI.from_slug('a-slug').uri
+ u'http://wolnelektury.pl/katalog/lektura/a-slug/'
-def serialize_raw(element):
- b = u'' + (element.text or '')
+ """
+ uri = 'http://wolnelektury.pl/katalog/lektura/%s/' % slug
+ return cls(uri)
- for child in element.iterchildren():
- e = etree.tostring(child, method='xml', encoding=unicode, pretty_print=True)
- b += e
+ def __unicode__(self):
+ return self.uri
- return b
+ def __str__(self):
+ return self.uri
-SERIALIZERS = {
- 'raw': serialize_raw,
-}
+ def __eq__(self, other):
+ return self.slug == other.slug
-def serialize_children(element, format='raw'):
- return SERIALIZERS[format](element)
-def get_resource(path):
- return os.path.join(os.path.dirname(__file__), path)
+class URLOpener(urllib.FancyURLopener):
+ version = 'FNP Librarian (http://git.nowoczesnapolska.org.pl/?p=librarian.git)'
+urllib._urlopener = URLOpener()
fnp / librarian.git / blobdiff