1 # -*- coding: utf-8 -*-
3 # This file is part of Librarian, licensed under GNU Affero GPLv3 or later.
4 # Copyright © Fundacja Nowoczesna Polska. See NOTICE for more information.
6 from __future__ import with_statement
14 class UnicodeException(Exception):
16 """ Dirty workaround for Python Unicode handling problems. """
17 return unicode(self).encode('utf-8')
19 def __unicode__(self):
20 """ Dirty workaround for Python Unicode handling problems. """
21 args = self.args[0] if len(self.args) == 1 else self.args
23 message = unicode(args)
24 except UnicodeDecodeError:
25 message = unicode(args, encoding='utf-8', errors='ignore')
28 class ParseError(UnicodeException):
31 class ValidationError(UnicodeException):
34 class NoDublinCore(ValidationError):
35 """There's no DublinCore section, and it's required."""
38 class NoProvider(UnicodeException):
39 """There's no DocProvider specified, and it's needed."""
42 class XMLNamespace(object):
43 '''A handy structure to repsent names in an XML namespace.'''
45 def __init__(self, uri):
48 def __call__(self, tag):
49 return '{%s}%s' % (self.uri, tag)
51 def __contains__(self, tag):
52 return tag.startswith('{' + str(self) + '}')
55 return 'XMLNamespace(%r)' % self.uri
58 return '%s' % self.uri
60 class EmptyNamespace(XMLNamespace):
62 super(EmptyNamespace, self).__init__('')
64 def __call__(self, tag):
67 # some common namespaces we use
68 RDFNS = XMLNamespace('http://www.w3.org/1999/02/22-rdf-syntax-ns#')
69 DCNS = XMLNamespace('http://purl.org/dc/elements/1.1/')
70 XINS = XMLNamespace("http://www.w3.org/2001/XInclude")
71 XHTMLNS = XMLNamespace("http://www.w3.org/1999/xhtml")
72 NCXNS = XMLNamespace("http://www.daisy.org/z3986/2005/ncx/")
73 OPFNS = XMLNamespace("http://www.idpf.org/2007/opf")
75 WLNS = EmptyNamespace()
79 """Represents a WL URI. Extracts slug from it."""
82 example = 'http://wolnelektury.pl/katalog/lektura/template/'
83 _re_wl_uri = re.compile(r'http://(www\.)?wolnelektury.pl/katalog/lektura/'
84 '(?P<slug>[-a-z0-9]+)/?$')
86 def __init__(self, uri):
89 self.slug = uri.rstrip('/').rsplit('/', 1)[-1]
93 match = cls._re_wl_uri.match(uri)
95 raise ValidationError(u'Invalid URI (%s). Should match: %s' % (
96 uri, cls._re_wl_uri.pattern))
100 def from_slug(cls, slug):
101 """Contructs an URI from slug.
103 >>> WLURI.from_slug('a-slug').uri
104 u'http://wolnelektury.pl/katalog/lektura/a-slug/'
107 uri = 'http://wolnelektury.pl/katalog/lektura/%s/' % slug
110 def __unicode__(self):
116 def __eq__(self, other):
117 return self.slug == other.slug
120 class DocProvider(object):
121 """Base class for a repository of XML files.
123 Used for generating joined files, like EPUBs.
126 def by_slug(self, slug):
127 """Should return a file-like object with a WL document XML."""
128 raise NotImplementedError
130 def by_uri(self, uri, wluri=WLURI):
131 """Should return a file-like object with a WL document XML."""
133 return self.by_slug(wluri.slug)
136 class DirDocProvider(DocProvider):
137 """ Serve docs from a directory of files in form <slug>.xml """
139 def __init__(self, dir_):
143 def by_slug(self, slug):
144 fname = slug + '.xml'
145 return open(os.path.join(self.dir, fname))
148 import lxml.etree as etree
151 DEFAULT_BOOKINFO = dcparser.BookInfo(
152 { RDFNS('about'): u'http://wiki.wolnepodreczniki.pl/Lektury:Template'},
153 { DCNS('creator'): [u'Some, Author'],
154 DCNS('title'): [u'Some Title'],
155 DCNS('subject.period'): [u'Unknown'],
156 DCNS('subject.type'): [u'Unknown'],
157 DCNS('subject.genre'): [u'Unknown'],
158 DCNS('date'): ['1970-01-01'],
159 DCNS('language'): [u'pol'],
160 # DCNS('date'): [creation_date],
161 DCNS('publisher'): [u"Fundacja Nowoczesna Polska"],
163 [u"""Publikacja zrealizowana w ramach projektu
164 Wolne Lektury (http://wolnelektury.pl). Reprodukcja cyfrowa
165 wykonana przez Bibliotekę Narodową z egzemplarza
166 pochodzącego ze zbiorów BN."""],
167 DCNS('identifier.url'): [WLURI.example],
169 [u"Domena publiczna - zm. [OPIS STANU PRAWNEGO TEKSTU]"] })
171 def xinclude_forURI(uri):
172 e = etree.Element(XINS("include"))
174 return etree.tostring(e, encoding=unicode)
176 def wrap_text(ocrtext, creation_date, bookinfo=DEFAULT_BOOKINFO):
177 """Wrap the text within the minimal XML structure with a DC template."""
178 bookinfo.created_at = creation_date
180 dcstring = etree.tostring(bookinfo.to_etree(), \
181 method='xml', encoding=unicode, pretty_print=True)
183 return u'<utwor>\n' + dcstring + u'\n<plain-text>\n' + ocrtext + \
184 u'\n</plain-text>\n</utwor>'
187 def serialize_raw(element):
188 b = u'' + (element.text or '')
190 for child in element.iterchildren():
191 e = etree.tostring(child, method='xml', encoding=unicode,
198 'raw': serialize_raw,
201 def serialize_children(element, format='raw'):
202 return SERIALIZERS[format](element)
204 def get_resource(path):
205 return os.path.join(os.path.dirname(__file__), path)
208 class OutputFile(object):
209 """Represents a file returned by one of the converters."""
216 os.unlink(self._filename)
218 def __nonzero__(self):
219 return self._string is not None or self._filename is not None
222 def from_string(cls, string):
223 """Converter returns contents of a file as a string."""
226 instance._string = string
230 def from_filename(cls, filename):
231 """Converter returns contents of a file as a named file."""
234 instance._filename = filename
237 def get_string(self):
238 """Get file's contents as a string."""
240 if self._filename is not None:
241 with open(self._filename) as f:
247 """Get file as a file-like object."""
249 if self._string is not None:
250 from StringIO import StringIO
251 return StringIO(self._string)
252 elif self._filename is not None:
253 return open(self._filename)
255 def get_filename(self):
256 """Get file as a fs path."""
258 if self._filename is not None:
259 return self._filename
260 elif self._string is not None:
261 from tempfile import NamedTemporaryFile
262 temp = NamedTemporaryFile(prefix='librarian-', delete=False)
263 temp.write(self._string)
265 self._filename = temp.name
266 return self._filename
270 def save_as(self, path):
271 """Save file to a path. Create directories, if necessary."""
273 dirname = os.path.dirname(os.path.abspath(path))
274 if not os.path.isdir(dirname):
276 shutil.copy(self.get_filename(), path)
279 class URLOpener(urllib.FancyURLopener):
280 version = 'FNP Librarian (http://github.com/fnp/librarian)'
281 urllib._urlopener = URLOpener()