X-Git-Url: https://git.mdrn.pl/librarian.git/blobdiff_plain/fefdce4e24f9e397df5538fe6e7f54b5ece4d841..88f39c4f90a67164b2c3da7c7a41df420f3b17ab:/src/librarian/__init__.py diff --git a/src/librarian/__init__.py b/src/librarian/__init__.py index 119b6b1..0ed908c 100644 --- a/src/librarian/__init__.py +++ b/src/librarian/__init__.py @@ -1,47 +1,50 @@ -# -*- 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. +# Copyright © Fundacja Wolne Lektury. See NOTICE for more information. # -from __future__ import print_function, unicode_literals - +import io import os import re import shutil from tempfile import NamedTemporaryFile import urllib from lxml import etree -import six -from six.moves.urllib.request import FancyURLopener +from urllib.request import FancyURLopener from .util import makedirs +# Compatibility imports. +from .meta.types.wluri import WLURI + -@six.python_2_unicode_compatible class UnicodeException(Exception): def __str__(self): """ Dirty workaround for Python Unicode handling problems. """ args = self.args[0] if len(self.args) == 1 else self.args try: - message = six.text_type(args) + message = str(args) except UnicodeDecodeError: - message = six.text_type(args, encoding='utf-8', errors='ignore') + message = str(args, encoding='utf-8', errors='ignore') return message + class ParseError(UnicodeException): pass + class ValidationError(UnicodeException): pass + class NoDublinCore(ValidationError): """There's no DublinCore section, and it's required.""" pass + class NoProvider(UnicodeException): """There's no DocProvider specified, and it's needed.""" pass -class XMLNamespace(object): + +class XMLNamespace: '''A handy structure to repsent names in an XML namespace.''' def __init__(self, uri): @@ -59,6 +62,7 @@ class XMLNamespace(object): def __str__(self): return '%s' % self.uri + class EmptyNamespace(XMLNamespace): def __init__(self): super(EmptyNamespace, self).__init__('') @@ -66,60 +70,18 @@ class EmptyNamespace(XMLNamespace): def __call__(self, tag): 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") 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") PLMETNS = XMLNamespace("http://dl.psnc.pl/schemas/plmet/") WLNS = EmptyNamespace() -@six.python_2_unicode_compatible -class WLURI(object): - """Represents a WL URI. Extracts slug from it.""" - slug = None - - example = 'http://wolnelektury.pl/katalog/lektura/template/' - _re_wl_uri = re.compile(r'http://(www\.)?wolnelektury.pl/katalog/lektur[ay]/' - '(?P[-a-z0-9]+)/?$') - - def __init__(self, uri): - uri = six.text_type(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(cls, slug): - """Contructs an URI from slug. - - >>> print(WLURI.from_slug('a-slug').uri) - http://wolnelektury.pl/katalog/lektura/a-slug/ - - """ - uri = 'http://wolnelektury.pl/katalog/lektura/%s/' % slug - return cls(uri) - - def __str__(self): - return self.uri - - def __eq__(self, other): - return self.slug == other.slug - - -class DocProvider(object): +class DocProvider: """Base class for a repository of XML files. Used for generating joined files, like EPUBs. @@ -129,11 +91,6 @@ class DocProvider(object): """Should return a file-like object with a WL document XML.""" raise NotImplementedError - 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(wluri.slug) - class DirDocProvider(DocProvider): """ Serve docs from a directory of files in form .xml """ @@ -147,66 +104,11 @@ class DirDocProvider(DocProvider): return open(os.path.join(self.dir, fname), 'rb') -from . 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'): [u'pol'], - # 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): +class OutputFile: """Represents a file returned by one of the converters.""" _bytes = None @@ -248,7 +150,7 @@ class OutputFile(object): """Get file as a file-like object.""" if self._bytes is not None: - return six.BytesIO(self._bytes) + return io.BytesIO(self._bytes) elif self._filename is not None: return open(self._filename, 'rb') @@ -275,5 +177,7 @@ class OutputFile(object): class URLOpener(FancyURLopener): - version = 'FNP Librarian (http://github.com/fnp/librarian)' + version = 'WL Librarian (http://github.com/fnp/librarian)' + + urllib._urlopener = URLOpener()