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
12 class ParseError(Exception):
14 """ Dirty workaround for Python Unicode handling problems. """
17 def __unicode__(self):
18 """ Dirty workaround for Python Unicode handling problems. """
21 class ValidationError(Exception):
24 class NoDublinCore(ValidationError):
25 """There's no DublinCore section, and it's required."""
28 class NoProvider(Exception):
29 """There's no DocProvider specified, and it's needed."""
32 class XMLNamespace(object):
33 '''A handy structure to repsent names in an XML namespace.'''
35 def __init__(self, uri):
38 def __call__(self, tag):
39 return '{%s}%s' % (self.uri, tag)
41 def __contains__(self, tag):
42 return tag.startswith('{' + str(self) + '}')
45 return 'XMLNamespace(%r)' % self.uri
48 return '%s' % self.uri
50 class EmptyNamespace(XMLNamespace):
52 super(EmptyNamespace, self).__init__('')
54 def __call__(self, tag):
57 # some common namespaces we use
58 RDFNS = XMLNamespace('http://www.w3.org/1999/02/22-rdf-syntax-ns#')
59 DCNS = XMLNamespace('http://purl.org/dc/elements/1.1/')
60 XINS = XMLNamespace("http://www.w3.org/2001/XInclude")
61 XHTMLNS = XMLNamespace("http://www.w3.org/1999/xhtml")
62 NCXNS = XMLNamespace("http://www.daisy.org/z3986/2005/ncx/")
63 OPFNS = XMLNamespace("http://www.idpf.org/2007/opf")
65 WLNS = EmptyNamespace()
69 """Represents a WL URI. Extracts slug and language from it."""
74 _re_wl_uri = re.compile('http://wolnelektury.pl/katalog/lektura/'
75 '(?P<slug>[-a-z]+)(/(?P<lang>[a-z]{3})/?)?')
77 def __init__(self, uri):
79 match = self._re_wl_uri.match(uri)
81 self.slug = match.group('slug')
82 self.language = match.group('lang')
85 class DocProvider(object):
86 """Base class for a repository of XML files.
88 Used for generating joined files, like EPUBs.
91 def by_slug_and_lang(self, slug, lang=None):
92 """Should return a file-like object with a WL document XML."""
93 raise NotImplementedError
95 def by_slug(self, slug):
96 """Should return a file-like object with a WL document XML."""
97 return self.by_slug_and_lang(slug)
99 def by_uri(self, uri):
100 """Should return a file-like object with a WL document XML."""
102 return self.by_slug_and_lang(wluri.slug, wluri.language)
105 class DirDocProvider(DocProvider):
106 """ Serve docs from a directory of files in form <slug>.xml """
108 def __init__(self, dir_):
111 return super(DirDocProvider, self).__init__()
113 def by_slug_and_lang(self, slug, lang=None):
114 fname = "%s%s.xml" % (slug, ".%s" % lang if lang else "")
115 return open(os.path.join(self.dir, fname))
118 import lxml.etree as etree
121 DEFAULT_BOOKINFO = dcparser.BookInfo(
122 { RDFNS('about'): u'http://wiki.wolnepodreczniki.pl/Lektury:Template'},
123 { DCNS('creator'): [u'Some, Author'],
124 DCNS('title'): [u'Some Title'],
125 DCNS('subject.period'): [u'Unknown'],
126 DCNS('subject.type'): [u'Unknown'],
127 DCNS('subject.genre'): [u'Unknown'],
128 DCNS('date'): ['1970-01-01'],
129 DCNS('language'): [u'pol'],
130 # DCNS('date'): [creation_date],
131 DCNS('publisher'): [u"Fundacja Nowoczesna Polska"],
133 [u"""Publikacja zrealizowana w ramach projektu
134 Wolne Lektury (http://wolnelektury.pl). Reprodukcja cyfrowa
135 wykonana przez Bibliotekę Narodową z egzemplarza
136 pochodzącego ze zbiorów BN."""],
137 DCNS('identifier.url'):
138 [u"http://wolnelektury.pl/katalog/lektura/template"],
140 [u"Domena publiczna - zm. [OPIS STANU PRAWNEGO TEKSTU]"] })
142 def xinclude_forURI(uri):
143 e = etree.Element(XINS("include"))
145 return etree.tostring(e, encoding=unicode)
147 def wrap_text(ocrtext, creation_date, bookinfo=DEFAULT_BOOKINFO):
148 """Wrap the text within the minimal XML structure with a DC template."""
149 bookinfo.created_at = creation_date
151 dcstring = etree.tostring(bookinfo.to_etree(), \
152 method='xml', encoding=unicode, pretty_print=True)
154 return u'<utwor>\n' + dcstring + u'\n<plain-text>\n' + ocrtext + \
155 u'\n</plain-text>\n</utwor>'
158 def serialize_raw(element):
159 b = u'' + (element.text or '')
161 for child in element.iterchildren():
162 e = etree.tostring(child, method='xml', encoding=unicode,
169 'raw': serialize_raw,
172 def serialize_children(element, format='raw'):
173 return SERIALIZERS[format](element)
175 def get_resource(path):
176 return os.path.join(os.path.dirname(__file__), path)
179 class OutputFile(object):
180 """Represents a file returned by one of the converters."""
187 os.unlink(self._filename)
189 def __nonzero__(self):
190 return self._string is not None or self._filename is not None
193 def from_string(cls, string):
194 """Converter returns contents of a file as a string."""
197 instance._string = string
201 def from_filename(cls, filename):
202 """Converter returns contents of a file as a named file."""
205 instance._filename = filename
208 def get_string(self):
209 """Get file's contents as a string."""
211 if self._filename is not None:
212 with open(self._filename) as f:
218 """Get file as a file-like object."""
220 if self._string is not None:
221 from StringIO import StringIO
222 return StringIO(self._string)
223 elif self._filename is not None:
224 return open(self._filename)
226 def get_filename(self):
227 """Get file as a fs path."""
229 if self._filename is not None:
230 return self._filename
231 elif self._string is not None:
232 from tempfile import NamedTemporaryFile
233 temp = NamedTemporaryFile(prefix='librarian-', delete=False)
234 temp.write(self._string)
236 self._filename = temp.name
237 return self._filename
241 def save_as(self, path):
242 """Save file to a path. Create directories, if necessary."""
244 dirname = os.path.dirname(os.path.abspath(path))
245 if not os.path.isdir(dirname):
247 shutil.copy(self.get_filename(), path)