Do not ignore unexpected tags in <aktywnosc>

[librarian.git] / librarian / __init__.py

diff --git a/librarian/__init__.py b/librarian/__init__.py
index dd09ce4..215cacb 100644 (file)
--- a/librarian/__init__.py
+++ b/librarian/__init__.py
@@ -8,24 +8,34 @@ from __future__ import with_statement
 import os
 import re
 import shutil
+import urllib
 
-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
 
@@ -69,8 +79,8 @@ class WLURI(object):
     """Represents a WL URI. Extracts slug from it."""
     slug = None
 
-    example = 'http://wolnelektury.pl/katalog/lektura/template/'
-    _re_wl_uri = re.compile('http://wolnelektury.pl/katalog/lektura/'
+    example = 'http://edukacjamedialna.edu.pl/'
+    _re_wl_uri = re.compile(r'http://(www\.)?edukacjamedialna.edu.pl/(lekcje/)?'
             '(?P<slug>[-a-z0-9]+)/?$')
 
     def __init__(self, uri):
@@ -82,8 +92,8 @@ class WLURI(object):
     def strict(cls, uri):
         match = cls._re_wl_uri.match(uri)
         if not match:
-            raise ValueError('Supplied URI (%s) does not match '
-                'the template: %s.' % (uri, cls._re_wl_uri))
+            raise ValidationError(u'Invalid URI (%s). Should match: %s' % (
+                        uri, cls._re_wl_uri.pattern))
         return cls(uri)
 
     @classmethod
@@ -94,7 +104,7 @@ class WLURI(object):
         u'http://wolnelektury.pl/katalog/lektura/a-slug/'
 
         """
-        uri = 'http://wolnelektury.pl/katalog/lektura/%s/' % slug
+        uri = 'http://prawokultury.pl/publikacje/%s/' % slug
         return cls(uri)
 
     def __unicode__(self):
@@ -132,7 +142,7 @@ class DirDocProvider(DocProvider):
 
     def by_slug(self, slug):
         fname = slug + '.xml'
-        return open(os.path.join(self.dir, fname))
+        return IOFile.from_filename(os.path.join(self.dir, fname))
 
 
 import lxml.etree as etree
@@ -140,7 +150,10 @@ import dcparser
 
 DEFAULT_BOOKINFO = dcparser.BookInfo(
         { RDFNS('about'): u'http://wiki.wolnepodreczniki.pl/Lektury:Template'},
-        { DCNS('creator'): [u'Some, Author'],
+        { 
+          DCNS('creator.expert'): [u'Some, Author'],
+          DCNS('creator.scenario'): [u'Some, Author'],
+          DCNS('creator.textbook'): [u'Some, Author'],
           DCNS('title'): [u'Some Title'],
           DCNS('subject.period'): [u'Unknown'],
           DCNS('subject.type'): [u'Unknown'],
@@ -195,32 +208,35 @@ def get_resource(path):
     return os.path.join(os.path.dirname(__file__), path)
 
 
-class OutputFile(object):
-    """Represents a file returned by one of the converters."""
-
+class IOFile(object):
+    """ Represents a file fed as input or returned as a result. """
     _string = None
     _filename = None
+    _filename_tmp = False
+
+    def __init__(self, attachments=None):
+        self.attachments = attachments or {}
 
     def __del__(self):
-        if self._filename:
+        if self._filename_tmp:
             os.unlink(self._filename)
 
     def __nonzero__(self):
         return self._string is not None or self._filename is not None
 
     @classmethod
-    def from_string(cls, string):
+    def from_string(cls, string, *args, **kwargs):
         """Converter returns contents of a file as a string."""
 
-        instance = cls()
+        instance = cls(*args, **kwargs)
         instance._string = string
         return instance
 
     @classmethod
-    def from_filename(cls, filename):
+    def from_filename(cls, filename, *args, **kwargs):
         """Converter returns contents of a file as a named file."""
 
-        instance = cls()
+        instance = cls(*args, **kwargs)
         instance._filename = filename
         return instance
 
@@ -253,6 +269,7 @@ class OutputFile(object):
             temp.write(self._string)
             temp.close()
             self._filename = temp.name
+            self._filename_tmp = True
             return self._filename
         else:
             return None
@@ -264,3 +281,25 @@ class OutputFile(object):
         if not os.path.isdir(dirname):
             os.makedirs(dirname)
         shutil.copy(self.get_filename(), path)
+
+    def dump_to(self, path, directory=None):
+        """ Path should be name for main file. """
+        self.save_as(path)
+        dirname = os.path.dirname(os.path.abspath(path))
+        for filename, attachment in self.attachments.items():
+            attachment.save_as(os.path.join(dirname, filename))
+
+
+class Format(object):
+    """ Generic format class. """
+    def __init__(self, wldoc, **kwargs):
+        self.wldoc = wldoc
+        self.customization = kwargs
+
+    def build(self):
+        raise NotImplementedError
+
+
+class URLOpener(urllib.FancyURLopener):
+    version = 'FNP Librarian (http://github.com/fnp/librarian)'
+urllib._urlopener = URLOpener()