X-Git-Url: https://git.mdrn.pl/librarian.git/blobdiff_plain/918a554570309fdddb8158b963fef3b3ae3c2c2b..e741f082e5b4122a5fa0f2c992f3519e573a8c7b:/librarian/__init__.py diff --git a/librarian/__init__.py b/librarian/__init__.py index 52234fe..bf15d13 100644 --- a/librarian/__init__.py +++ b/librarian/__init__.py @@ -8,24 +8,34 @@ from __future__ import with_statement 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 @@ -55,6 +65,7 @@ class EmptyNamespace(XMLNamespace): 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") @@ -66,60 +77,45 @@ WLNS = EmptyNamespace() 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[-a-z]+)(/(?P[a-z]{3})/?)?') + _re_wl_uri = re.compile(r'http://(www\.)?wolnelektury.pl/katalog/lektura/' + '(?P[-a-z0-9]+)/?$') - def __init__(self, uri=None): - if uri is not None: - self.uri = uri - match = self._re_wl_uri.match(uri) - assert match - self.slug = match.group('slug') - self.language = match.group('lang') or self.DEFAULT_LANGUAGE + def __init__(self, uri): + uri = unicode(uri) + self.uri = uri + self.slug = uri.rstrip('/').rsplit('/', 1)[-1] @classmethod - def from_slug_and_lang(cls, slug, lang): - """Contructs an URI from slug and language code. + 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) - >>> 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/' + @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/' """ - if lang is None: - lang = self.DEFAULT_LANGUAGE uri = 'http://wolnelektury.pl/katalog/lektura/%s/' % slug - if lang is not None and lang != cls.DEFAULT_LANGUAGE: - uri += lang + '/' - instance = cls() - instance.slug = slug - instance.language = lang - instance.uri = uri - return instance + 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 __str__(self): + return self.uri - def validate_language(self, language): - if language != self.language: - raise ValidationError("Incorrect language definition in URI") + def __eq__(self, other): + return self.slug == other.slug class DocProvider(object): @@ -128,18 +124,14 @@ 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): @@ -148,10 +140,9 @@ 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)) @@ -166,7 +157,7 @@ DEFAULT_BOOKINFO = dcparser.BookInfo( 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'): @@ -284,3 +275,8 @@ class OutputFile(object): 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()