hack for pdf from collection

[librarian.git] / librarian / __init__.py

diff --git a/librarian/__init__.py b/librarian/__init__.py
index fdd6b55..9a9e23e 100644 (file)
--- a/librarian/__init__.py
+++ b/librarian/__init__.py
@@ -3,27 +3,41 @@
 # 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
+
 import os
 import re
 import shutil
+import urllib
+
+from util import makedirs
+
 
-class ParseError(Exception):
+class UnicodeException(Exception):
     def __str__(self):
         """ Dirty workaround for Python Unicode handling problems. """
-        return self.message
+        return unicode(self).encode('utf-8')
 
     def __unicode__(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 = unicode(args)
+        except UnicodeDecodeError:
+            message = unicode(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
 
@@ -53,31 +67,58 @@ class EmptyNamespace(XMLNamespace):
         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()
 
 
 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 = unicode(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.
+
+        >>> WLURI.from_slug('a-slug').uri
+        u'http://wolnelektury.pl/katalog/lektura/a-slug/'
+
+        """
+        uri = 'http://wolnelektury.pl/katalog/lektura/%s/' % slug
+        return cls(uri)
+
+    def __unicode__(self):
+        return self.uri
+
+    def __str__(self):
+        return self.uri
+
+    def __eq__(self, other):
+        return self.slug == other.slug
 
 
 class DocProvider(object):
@@ -86,18 +127,14 @@ 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):
@@ -106,10 +143,9 @@ 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 "")
+    def by_slug(self, slug):
+        fname = slug + '.xml'
         return open(os.path.join(self.dir, fname))
 
 
@@ -132,8 +168,7 @@ DEFAULT_BOOKINFO = dcparser.BookInfo(
              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]"] })
 
@@ -240,6 +275,10 @@ class OutputFile(object):
         """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(urllib.FancyURLopener):
+    version = 'FNP Librarian (http://github.com/fnp/librarian)'
+urllib._urlopener = URLOpener()