import os
import re
import shutil
+import urllib
-class ParseError(Exception):
+
+class UnicodeException(Exception):
def __str__(self):
""" Dirty workaround for Python Unicode handling problems. """
- return self.message
+ return unicode(self).encode('utf-8')
def __unicode__(self):
""" Dirty workaround for Python Unicode handling problems. """
- return self.message
+ 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 ValidationError(Exception):
+class ValidationError(UnicodeException):
pass
class NoDublinCore(ValidationError):
"""There's no DublinCore section, and it's required."""
pass
-class NoProvider(Exception):
+class NoProvider(UnicodeException):
"""There's no DocProvider specified, and it's needed."""
pass
return tag
# some common namespaces we use
+XMLNS = XMLNamespace('http://www.w3.org/XML/1998/namespace')
RDFNS = XMLNamespace('http://www.w3.org/1999/02/22-rdf-syntax-ns#')
DCNS = XMLNamespace('http://purl.org/dc/elements/1.1/')
XINS = XMLNamespace("http://www.w3.org/2001/XInclude")
class WLURI(object):
- """Represents a WL URI. Extracts slug and language from it."""
- DEFAULT_LANGUAGE = u'pol'
-
+ """Represents a WL URI. Extracts slug from it."""
slug = None
- language = None
example = 'http://wolnelektury.pl/katalog/lektura/template/'
- _re_wl_uri = re.compile('http://wolnelektury.pl/katalog/lektura/'
- '(?P<slug>[-a-z0-9]+)(/(?P<lang>[a-z]{3}))?/?$')
-
- def __init__(self, uri=None):
- if uri is not None:
- uri = unicode(uri)
- self.uri = uri
- match = self._re_wl_uri.match(uri)
- if not match:
- raise ValueError('Supplied URI (%s) does not match '
- 'the WL document URI template.' % uri)
- self.slug = match.group('slug')
- self.language = match.group('lang') or self.DEFAULT_LANGUAGE
+ _re_wl_uri = re.compile(r'http://(www\.)?wolnelektury.pl/katalog/lektura/'
+ '(?P<slug>[-a-z0-9]+)/?$')
+
+ def __init__(self, uri):
+ uri = unicode(uri)
+ self.uri = uri
+ self.slug = uri.rstrip('/').rsplit('/', 1)[-1]
+
+ @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)
@classmethod
- def from_slug_and_lang(cls, slug, lang):
- """Contructs an URI from slug and language code.
+ def from_slug(cls, slug):
+ """Contructs an URI from slug.
- >>> WLURI.from_slug_and_lang('a-slug', WLURI.DEFAULT_LANGUAGE).uri
- 'http://wolnelektury.pl/katalog/lektura/a-slug/'
- >>> WLURI.from_slug_and_lang('a-slug', 'deu').uri
- 'http://wolnelektury.pl/katalog/lektura/a-slug/deu/'
+ >>> WLURI.from_slug('a-slug').uri
+ u'http://wolnelektury.pl/katalog/lektura/a-slug/'
"""
- if lang is None:
- lang = cls.DEFAULT_LANGUAGE
uri = 'http://wolnelektury.pl/katalog/lektura/%s/' % slug
- if lang is not None and lang != cls.DEFAULT_LANGUAGE:
- uri += lang + '/'
return cls(uri)
def __unicode__(self):
return self.uri
def __eq__(self, other):
- return self.slug, self.language == other.slug, other.language
-
- def filename_stem(self):
- stem = self.slug
- if self.language != self.DEFAULT_LANGUAGE:
- stem += '_' + self.language
- return stem
-
- def validate_language(self, language):
- if language != self.language:
- raise ValidationError("Incorrect language definition in URI")
+ return self.slug == other.slug
class DocProvider(object):
Used for generating joined files, like EPUBs.
"""
- def by_slug_and_lang(self, slug, lang=None):
- """Should return a file-like object with a WL document XML."""
- raise NotImplementedError
-
def by_slug(self, slug):
"""Should return a file-like object with a WL document XML."""
- return self.by_slug_and_lang(slug)
+ raise NotImplementedError
- def by_uri(self, uri):
+ def by_uri(self, uri, wluri=WLURI):
"""Should return a file-like object with a WL document XML."""
- wluri = WLURI(uri)
- return self.by_slug_and_lang(wluri.slug, wluri.language)
+ wluri = wluri(uri)
+ return self.by_slug(wluri.slug)
class DirDocProvider(DocProvider):
def __init__(self, dir_):
self.dir = dir_
self.files = {}
- return super(DirDocProvider, self).__init__()
- def by_slug_and_lang(self, slug, lang=None):
- fname = WLURI.from_slug_and_lang(slug, lang).filename_stem() + '.xml'
+ def by_slug(self, slug):
+ fname = slug + '.xml'
return open(os.path.join(self.dir, fname))
DCNS('subject.type'): [u'Unknown'],
DCNS('subject.genre'): [u'Unknown'],
DCNS('date'): ['1970-01-01'],
- DCNS('language'): [WLURI.DEFAULT_LANGUAGE],
+ DCNS('language'): [u'pol'],
# DCNS('date'): [creation_date],
DCNS('publisher'): [u"Fundacja Nowoczesna Polska"],
DCNS('description'):
if not os.path.isdir(dirname):
os.makedirs(dirname)
shutil.copy(self.get_filename(), path)
+
+
+class URLOpener(urllib.FancyURLopener):
+ version = 'FNP Librarian (http://github.com/fnp/librarian)'
+urllib._urlopener = URLOpener()
fnp / librarian.git / blobdiff