X-Git-Url: https://git.mdrn.pl/librarian.git/blobdiff_plain/ae0e673a17c3edcdca910fafb84eeff9dfe7b588..13480b3da2d3da87f1d99c6d340c1553ca9d89c1:/librarian/__init__.py

diff --git a/librarian/__init__.py b/librarian/__init__.py
index 0074ee9..0616f23 100644
--- a/librarian/__init__.py
+++ b/librarian/__init__.py
@@ -4,33 +4,31 @@
 # Copyright © Fundacja Nowoczesna Polska. See NOTICE for more information.
 #
 import os
+import re
+import urllib
+from .utils import XMLNamespace
 
-class ParseError(Exception):
-    pass
 
-class ValidationError(Exception):
+class UnicodeException(Exception):
+    def __str__(self):
+        """ Dirty workaround for Python Unicode handling problems. """
+        return unicode(self).encode('utf-8')
+
+    def __unicode__(self):
+        """ Dirty workaround for Python Unicode handling problems. """
+        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 ParseError(UnicodeException):
     pass
 
-class NoDublinCore(ValidationError):
+class ValidationError(UnicodeException):
     pass
 
-class XMLNamespace(object):
-    '''A handy structure to repsent names in an XML namespace.'''
-
-    def __init__(self, uri):
-        self.uri = uri
-
-    def __call__(self, tag):
-        return '{%s}%s' % (self.uri, tag)
-
-    def __contains__(self, tag):
-        return tag.startswith('{' + str(self) + '}')
-
-    def __repr__(self):
-        return 'XMLNamespace(%r)' % self.uri
-
-    def __str__(self):
-        return '%s' % self.uri
 
 class EmptyNamespace(XMLNamespace):
     def __init__(self):
@@ -47,86 +45,52 @@ 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 DocProvider(object):
-    """ Base class for a repository of XML files.
-        Used for generating joined files, like EPUBs
-    """
+class WLURI(object):
+    """Represents a WL URI. Extracts slug from it."""
+    slug = None
 
-    def by_slug(self, slug):
-        raise NotImplemented
+    example = 'http://wolnelektury.pl/katalog/lektura/template/'
+    _re_wl_uri = re.compile(r'http://(www\.)?wolnelektury.pl/katalog/lektura/'
+            '(?P<slug>[-a-z0-9]+)/?$')
 
-    def __getitem__(self, slug):
-        return self.by_slug(slug)
-
-    def by_uri(self, uri):
-        return self.by_slug(uri.rsplit('/', 1)[1])
-
-
-class DirDocProvider(DocProvider):
-    """ Serve docs from a directory of files in form <slug>.xml """
-
-    def __init__(self, dir):
-        self.dir = dir
-        self.files = {}
-
-    def by_slug(self, slug):
-        return open(os.path.join(self.dir, '%s.xml' % slug))
-
-
-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('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'):
-            [u"http://wolnelektury.pl/katalog/lektura/template"],
-          DCNS('rights'):
-            [u"Domena publiczna - zm. [OPIS STANU PRAWNEGO TEKSTU]"] })
+    def __init__(self, uri):
+        uri = unicode(uri)
+        self.uri = uri
+        self.slug = uri.rstrip('/').rsplit('/', 1)[-1]
 
-def xinclude_forURI(uri):
-    e = etree.Element(XINS("include"))
-    e.set("href", uri)
-    return etree.tostring(e, encoding=unicode)
+    @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)
 
-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
+    @classmethod
+    def from_slug(cls, slug):
+        """Contructs an URI from slug.
 
-    dcstring = etree.tostring(bookinfo.to_etree(), \
-        method='xml', encoding=unicode, pretty_print=True)
+        >>> WLURI.from_slug('a-slug').uri
+        u'http://wolnelektury.pl/katalog/lektura/a-slug/'
 
-    return u'<utwor>\n' + dcstring + u'\n<plain-text>\n' + ocrtext + \
-        u'\n</plain-text>\n</utwor>';
+        """
+        uri = 'http://wolnelektury.pl/katalog/lektura/%s/' % slug
+        return cls(uri)
 
+    def __unicode__(self):
+        return self.uri
 
-def serialize_raw(element):
-    b = u'' + (element.text or '')
+    def __str__(self):
+        return self.uri
 
-    for child in element.iterchildren():
-        e = etree.tostring(child, method='xml', encoding=unicode, pretty_print=True)
-        b += e
+    def __eq__(self, other):
+        return self.slug == other.slug
 
-    return b
 
-SERIALIZERS = {
-    'raw': serialize_raw,
-}
+class URLOpener(urllib.FancyURLopener):
+    version = 'FNP Librarian (http://git.nowoczesnapolska.org.pl/?p=librarian.git)'
+urllib._urlopener = URLOpener()
 
-def serialize_children(element, format='raw'):
-    return SERIALIZERS[format](element)