X-Git-Url: https://git.mdrn.pl/librarian.git/blobdiff_plain/4077ffb3cbd868df95239898563508b64e6d6ecf..8287173da88bfaa3a8b90a39768198dceba2e25f:/librarian/__init__.py diff --git a/librarian/__init__.py b/librarian/__init__.py index 32a3af0..a0d70e7 100644 --- a/librarian/__init__.py +++ b/librarian/__init__.py @@ -3,49 +3,42 @@ # This file is part of Librarian, licensed under GNU Affero GPLv3 or later. # Copyright © Fundacja Nowoczesna Polska. See NOTICE for more information. # -from __future__ import with_statement - -import os import re -import shutil +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 + 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 ValidationError(Exception): - pass -class NoDublinCore(ValidationError): - """There's no DublinCore section, and it's required.""" +class ParseError(UnicodeException): pass -class NoProvider(Exception): - """There's no DocProvider specified, and it's needed.""" - pass -class XMLNamespace(object): - '''A handy structure to repsent names in an XML namespace.''' +class ValidationError(UnicodeException): + pass - def __init__(self, uri): - self.uri = uri - def __call__(self, tag): - return '{%s}%s' % (self.uri, tag) +# was deleted, but still used??? +class NoDublinCore(ValidationError): + pass - def __contains__(self, tag): - return tag.startswith('{' + str(self) + '}') - def __repr__(self): - return 'XMLNamespace(%r)' % self.uri +class BuildError(Exception): + pass - def __str__(self): - return '%s' % self.uri class EmptyNamespace(XMLNamespace): def __init__(self): @@ -62,46 +55,38 @@ XHTMLNS = XMLNamespace("http://www.w3.org/1999/xhtml") NCXNS = XMLNamespace("http://www.daisy.org/z3986/2005/ncx/") OPFNS = XMLNamespace("http://www.idpf.org/2007/opf") -WLNS = EmptyNamespace() +SSTNS = XMLNamespace('http://nowoczesnapolska.org.pl/sst#') 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-z0-9]+)(/(?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: - 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 + 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) + + @classmethod + def from_slug(cls, slug): + """Contructs an URI from slug. - >>> WLURI.from_slug_and_lang('a-slug', WLURI.DEFAULT_LANGUAGE).uri + >>> WLURI.from_slug('a-slug').uri u'http://wolnelektury.pl/katalog/lektura/a-slug/' - >>> WLURI.from_slug_and_lang('a-slug', 'deu').uri - u'http://wolnelektury.pl/katalog/lektura/a-slug/deu/' """ - 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): @@ -111,178 +96,9 @@ class WLURI(object): 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") - - -class DocProvider(object): - """Base class for a repository of XML files. - - 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) - - def by_uri(self, uri): - """Should return a file-like object with a WL document XML.""" - wluri = WLURI(uri) - return self.by_slug_and_lang(wluri.slug, wluri.language) - - -class DirDocProvider(DocProvider): - """ Serve docs from a directory of files in form .xml """ - - 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' - return open(os.path.join(self.dir, fname)) - - -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('language'): [WLURI.DEFAULT_LANGUAGE], - # 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'): [WLURI.example], - 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) - - return u'\n' + dcstring + u'\n\n' + ocrtext + \ - u'\n\n' - - -def serialize_raw(element): - b = u'' + (element.text or '') - - for child in element.iterchildren(): - e = etree.tostring(child, method='xml', encoding=unicode, - pretty_print=True) - b += e - - return b - -SERIALIZERS = { - 'raw': serialize_raw, -} - -def serialize_children(element, format='raw'): - return SERIALIZERS[format](element) - -def get_resource(path): - return os.path.join(os.path.dirname(__file__), path) - - -class OutputFile(object): - """Represents a file returned by one of the converters.""" - - _string = None - _filename = None - - def __del__(self): - if self._filename: - os.unlink(self._filename) - - def __nonzero__(self): - return self._string is not None or self._filename is not None - - @classmethod - def from_string(cls, string): - """Converter returns contents of a file as a string.""" - - instance = cls() - instance._string = string - return instance - - @classmethod - def from_filename(cls, filename): - """Converter returns contents of a file as a named file.""" - - instance = cls() - instance._filename = filename - return instance - - def get_string(self): - """Get file's contents as a string.""" - - if self._filename is not None: - with open(self._filename) as f: - return f.read() - else: - return self._string - - def get_file(self): - """Get file as a file-like object.""" - - if self._string is not None: - from StringIO import StringIO - return StringIO(self._string) - elif self._filename is not None: - return open(self._filename) - - def get_filename(self): - """Get file as a fs path.""" - - if self._filename is not None: - return self._filename - elif self._string is not None: - from tempfile import NamedTemporaryFile - temp = NamedTemporaryFile(prefix='librarian-', delete=False) - temp.write(self._string) - temp.close() - self._filename = temp.name - return self._filename - else: - return None + return self.slug == other.slug - def save_as(self, path): - """Save file to a path. Create directories, if necessary.""" - dirname = os.path.dirname(os.path.abspath(path)) - if not os.path.isdir(dirname): - os.makedirs(dirname) - shutil.copy(self.get_filename(), path) +class URLOpener(urllib.FancyURLopener): + version = 'FNP Librarian (http://git.nowoczesnapolska.org.pl/?p=librarian.git)' +urllib._urlopener = URLOpener()