fix expected html output to contain secX anchors

[librarian.git] / librarian / __init__.py

diff --git a/librarian/__init__.py b/librarian/__init__.py
index 32a3af0..8a69d00 100644 (file)
--- a/librarian/__init__.py
+++ b/librarian/__init__.py
@@ -9,7 +9,7 @@ import os
 import re
 import shutil
 
-class ParseError(Exception):
+class UnicodeException(Exception):
     def __str__(self):
         """ Dirty workaround for Python Unicode handling problems. """
         return self.message
@@ -18,14 +18,17 @@ class ParseError(Exception):
         """ Dirty workaround for Python Unicode handling problems. """
         return self.message
 
-class ValidationError(Exception):
+class ParseError(UnicodeException):
+    pass
+
+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
 
@@ -66,42 +69,35 @@ WLNS = EmptyNamespace()
 
 
 class WLURI(object):
-    """Represents a WL URI. Extracts slug and language from it."""
-    DEFAULT_LANGUAGE = u'pol'
-
+    """Represents a WL URI. Extracts slug from it."""
     slug = None
-    language = None
 
     example = 'http://wolnelektury.pl/katalog/lektura/template/'
-    _re_wl_uri = re.compile('http://wolnelektury.pl/katalog/lektura/'
-            '(?P<slug>[-a-z0-9]+)(/(?P<lang>[a-z]{3}))?/?$')
-
-    def __init__(self, uri=None):
-        if uri is not None:
-            uri = unicode(uri)
-            self.uri = uri
-            match = self._re_wl_uri.match(uri)
-            if not match:
-                raise ValueError('Supplied URI (%s) does not match '
-                        'the WL document URI template.' % uri)
-            self.slug = match.group('slug')
-            self.language = match.group('lang') or self.DEFAULT_LANGUAGE
+    _re_wl_uri = re.compile(r'http://(www\.)?wolnelektury.pl/katalog/lektura/'
+            '(?P<slug>[-a-z0-9]+)/?$')
+
+    def __init__(self, uri):
+        uri = unicode(uri)
+        self.uri = uri
+        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_and_lang(cls, slug, lang):
-        """Contructs an URI from slug and language code.
+    def from_slug(cls, slug):
+        """Contructs an URI from slug.
 
-        >>> WLURI.from_slug_and_lang('a-slug', WLURI.DEFAULT_LANGUAGE).uri
+        >>> WLURI.from_slug('a-slug').uri
         u'http://wolnelektury.pl/katalog/lektura/a-slug/'
-        >>> WLURI.from_slug_and_lang('a-slug', 'deu').uri
-        u'http://wolnelektury.pl/katalog/lektura/a-slug/deu/'
 
         """
-        if lang is None:
-            lang = cls.DEFAULT_LANGUAGE
         uri = 'http://wolnelektury.pl/katalog/lektura/%s/' % slug
-        if lang is not None and lang != cls.DEFAULT_LANGUAGE:
-            uri += lang + '/'
         return cls(uri)
 
     def __unicode__(self):
@@ -111,17 +107,7 @@ class WLURI(object):
         return self.uri
 
     def __eq__(self, other):
-        return self.slug, self.language == other.slug, other.language
-
-    def filename_stem(self):
-        stem = self.slug
-        if self.language != self.DEFAULT_LANGUAGE:
-            stem += '_' + self.language
-        return stem
-
-    def validate_language(self, language):
-        if language != self.language:
-            raise ValidationError("Incorrect language definition in URI")
+        return self.slug == other.slug
 
 
 class DocProvider(object):
@@ -130,18 +116,14 @@ 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):
@@ -150,10 +132,9 @@ 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 = WLURI.from_slug_and_lang(slug, lang).filename_stem() + '.xml'
+    def by_slug(self, slug):
+        fname = slug + '.xml'
         return open(os.path.join(self.dir, fname))
 
 
@@ -168,7 +149,7 @@ DEFAULT_BOOKINFO = dcparser.BookInfo(
           DCNS('subject.type'): [u'Unknown'],
           DCNS('subject.genre'): [u'Unknown'],
           DCNS('date'): ['1970-01-01'],
-          DCNS('language'): [WLURI.DEFAULT_LANGUAGE],
+          DCNS('language'): [u'pol'],
           # DCNS('date'): [creation_date],
           DCNS('publisher'): [u"Fundacja Nowoczesna Polska"],
           DCNS('description'):