# 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
+from __future__ import print_function, unicode_literals
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 .util import makedirs
-class ParseError(Exception):
- def __str__(self):
- """ Dirty workaround for Python Unicode handling problems. """
- return self.message
- def __unicode__(self):
+@six.python_2_unicode_compatible
+class UnicodeException(Exception):
+ def __str__(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 = six.text_type(args)
+ except UnicodeDecodeError:
+ message = six.text_type(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
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 and language from it."""
-
+ """Represents a WL URI. Extracts slug from it."""
slug = None
- language = None
- _re_wl_uri = re.compile('http://wolnelektury.pl/katalog/lektura/'
- '(?P<slug>[-a-z]+)(/(?P<lang>[a-z]{3})/?)?')
+ example = 'http://wolnelektury.pl/katalog/lektura/template/'
+ _re_wl_uri = re.compile(r'http://(www\.)?wolnelektury.pl/katalog/lektur[ay]/'
+ '(?P<slug>[-a-z0-9]+)/?$')
def __init__(self, uri):
+ uri = six.text_type(uri)
self.uri = uri
- match = self._re_wl_uri.match(uri)
- assert match
- self.slug = match.group('slug')
- self.language = match.group('lang')
+ 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):
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):
def __init__(self, dir_):
self.dir = dir_
self.files = {}
- return super(DirDocProvider, self).__init__()
- def by_slug_and_lang(self, slug, lang=None):
- fname = "%s%s.xml" % (slug, ".%s" % lang if lang else "")
- return open(os.path.join(self.dir, fname))
+ def by_slug(self, slug):
+ fname = slug + '.xml'
+ return open(os.path.join(self.dir, fname), 'rb')
-import lxml.etree as etree
-import dcparser
+from . import dcparser
DEFAULT_BOOKINFO = dcparser.BookInfo(
{ RDFNS('about'): u'http://wiki.wolnepodreczniki.pl/Lektury:Template'},
Wolne Lektury (http://wolnelektury.pl). Reprodukcja cyfrowa
wykonana przez Bibliotekę Narodową z egzemplarza
pochodzącego ze zbiorów BN."""],
- DCNS('identifier.url'):
- [u"http://wolnelektury.pl/katalog/lektura/template"],
+ 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)
+ 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)
+ method='xml', encoding='unicode', pretty_print=True)
return u'<utwor>\n' + dcstring + u'\n<plain-text>\n' + ocrtext + \
u'\n</plain-text>\n</utwor>'
b = u'' + (element.text or '')
for child in element.iterchildren():
- e = etree.tostring(child, method='xml', encoding=unicode,
+ e = etree.tostring(child, method='xml', encoding='unicode',
pretty_print=True)
b += e
class OutputFile(object):
"""Represents a file returned by one of the converters."""
- _string = None
+ _bytes = None
_filename = None
def __del__(self):
os.unlink(self._filename)
def __nonzero__(self):
- return self._string is not None or self._filename is not None
+ return self._bytes is not None or self._filename is not None
@classmethod
- def from_string(cls, string):
+ def from_bytes(cls, bytestring):
"""Converter returns contents of a file as a string."""
instance = cls()
- instance._string = string
+ instance._bytes = bytestring
return instance
@classmethod
instance._filename = filename
return instance
- def get_string(self):
- """Get file's contents as a string."""
+ def get_bytes(self):
+ """Get file's contents as a bytestring."""
if self._filename is not None:
- with open(self._filename) as f:
+ with open(self._filename, 'rb') as f:
return f.read()
else:
- return self._string
+ return self._bytes
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)
+ if self._bytes is not None:
+ return six.BytesIO(self._bytes)
elif self._filename is not None:
- return open(self._filename)
+ return open(self._filename, 'rb')
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
+ elif self._bytes is not None:
temp = NamedTemporaryFile(prefix='librarian-', delete=False)
- temp.write(self._string)
+ temp.write(self._bytes)
temp.close()
self._filename = temp.name
return self._filename
"""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)
+ makedirs(dirname)
shutil.copy(self.get_filename(), path)
+
+
+class URLOpener(FancyURLopener):
+ version = 'FNP Librarian (http://github.com/fnp/librarian)'
+urllib._urlopener = URLOpener()
fnp / librarian.git / blobdiff