tag disambiguation page updated
[wolnelektury.git] / apps / catalogue / tests.py
index fae4ba2..829f8dc 100644 (file)
 # -*- coding: utf-8 -*-
 from django.test import TestCase
 from catalogue import models, views
 # -*- coding: utf-8 -*-
 from django.test import TestCase
 from catalogue import models, views
+from django.core.files.base import ContentFile
 from django.contrib.auth.models import User, AnonymousUser
 from django.contrib.auth.models import User, AnonymousUser
+from django.test.client import Client
 
 
-class SimpleSearchTest(TestCase):
+from nose.tools import raises
+from StringIO import StringIO
+
+class BasicSearchLogicTests(TestCase):
+
+    def setUp(self):
+        self.author_tag = models.Tag.objects.create(
+                                name=u'Adam Mickiewicz [SubWord]',
+                                category=u'author', slug="one")
+
+        self.unicode_tag = models.Tag.objects.create(
+                                name=u'Tadeusz Żeleński (Boy)',
+                                category=u'author', slug="two")
+
+        self.polish_tag = models.Tag.objects.create(
+                                name=u'ĘÓĄŚŁŻŹĆŃęóąśłżźćń',
+                                category=u'author', slug="three")
+
+    @raises(ValueError)
+    def test_empty_query(self):
+        """ Check that empty queries raise an error. """
+        views.find_best_matches(u'')
+
+    @raises(ValueError)
+    def test_one_letter_query(self):
+        """ Check that one letter queries aren't permitted. """
+        views.find_best_matches(u't')
+
+    def test_match_by_prefix(self):
+        """ Tags should be matched by prefix of words within it's name. """
+        self.assertEqual(views.find_best_matches(u'Ada'), (self.author_tag,))
+        self.assertEqual(views.find_best_matches(u'Mic'), (self.author_tag,))
+        self.assertEqual(views.find_best_matches(u'Mickiewicz'), (self.author_tag,))
+
+    def test_match_case_insensitive(self):
+        """ Tag names should match case insensitive. """
+        self.assertEqual(views.find_best_matches(u'adam mickiewicz'), (self.author_tag,))
+
+    def test_match_case_insensitive_unicode(self):
+        """ Tag names should match case insensitive (unicode). """
+        self.assertEqual(views.find_best_matches(u'tadeusz żeleński (boy)'), (self.unicode_tag,))
+
+    def test_word_boundary(self):
+        self.assertEqual(views.find_best_matches(u'SubWord'), (self.author_tag,))
+        self.assertEqual(views.find_best_matches(u'[SubWord'), (self.author_tag,))
+
+    def test_unrelated_search(self):
+        self.assertEqual(views.find_best_matches(u'alamakota'), tuple())
+        self.assertEqual(views.find_best_matches(u'Adama'), ())
+
+    def test_infix_doesnt_match(self):
+        """ Searching for middle of a word shouldn't match. """
+        self.assertEqual(views.find_best_matches(u'deusz'), tuple())
+
+    def test_diactricts_removal_pl(self):
+        """ Tags should match both with and without national characters. """
+        self.assertEqual(views.find_best_matches(u'ĘÓĄŚŁŻŹĆŃęóąśłżźćń'), (self.polish_tag,))
+        self.assertEqual(views.find_best_matches(u'EOASLZZCNeoaslzzcn'), (self.polish_tag,))
+        self.assertEqual(views.find_best_matches(u'eoaslzzcneoaslzzcn'), (self.polish_tag,))
+
+    def test_diactricts_query_removal_pl(self):
+        """ Tags without national characters shouldn't be matched by queries with them. """
+        self.assertEqual(views.find_best_matches(u'Adąm'), ())
+
+    def test_sloppy(self):
+        self.assertEqual(views.find_best_matches(u'Żelenski'), (self.unicode_tag,))
+        self.assertEqual(views.find_best_matches(u'zelenski'), (self.unicode_tag,))
+
+
+class PersonStub(object):
+
+    def __init__(self, first_names, last_name):
+        self.first_names = first_names
+        self.last_name = last_name
+
+from slughifi import slughifi
+
+class BookInfoStub(object):
+
+    def __init__(self, **kwargs):            
+        self.__dict = kwargs
+
+    def __setattr__(self, key, value):
+        if not key.startswith('_'):
+            self.__dict[key] = value
+        return object.__setattr__(self, key, value)
+
+    def __getattr__(self, key):
+        return self.__dict[key]
+
+    def to_dict(self):
+        return dict((key, unicode(value)) for key, value in self.__dict.items())
+
+def info_args(title):
+    """ generate some keywords for comfortable BookInfoCreation  """
+    slug = unicode(slughifi(title))
+    return {'title': unicode(title),
+            'slug': slug,
+            'url': u"http://wolnelektury.pl/example/%s" % slug,
+            'about': u"http://wolnelektury.pl/example/URI/%s" % slug,
+            }
+
+class BookImportLogicTests(TestCase):
+
+    def setUp(self):
+        self.book_info = BookInfoStub(
+            url=u"http://wolnelektury.pl/example/default_book",
+            about=u"http://wolnelektury.pl/example/URI/default_book",
+            title=u"Default Book",
+            author=PersonStub(("Jim",), "Lazy"),
+            kind="X-Kind",
+            genre="X-Genre",
+            epoch="X-Epoch",
+        )
+
+        self.expected_tags = [
+           ('author', 'jim-lazy'),
+           ('genre', 'x-genre'),
+           ('epoch', 'x-epoch'),
+           ('kind', 'x-kind'),
+        ]
+        self.expected_tags.sort()
+
+    def tearDown(self):
+        models.Book.objects.all().delete()
+
+    def test_empty_book(self):
+        BOOK_TEXT = "<utwor />"
+        book = models.Book.from_text_and_meta(ContentFile(BOOK_TEXT), self.book_info)
+
+        self.assertEqual(book.title, "Default Book")
+        self.assertEqual(book.slug, "default_book")
+        self.assert_(book.parent is None)
+        self.assertFalse(book.has_html_file())
+
+        # no fragments generated
+        self.assertEqual(book.fragments.count(), 0)
+
+        # TODO: this should be filled out probably...
+        self.assertEqual(book.wiki_link, '')
+        self.assertEqual(book.gazeta_link, '')
+        self.assertEqual(book._short_html, '')
+        self.assertEqual(book.description, '')
+
+        tags = [ (tag.category, tag.slug) for tag in book.tags ]
+        tags.sort()
+
+        self.assertEqual(tags, self.expected_tags)
+    
+    def test_not_quite_empty_book(self):
+        """ Not empty, but without any real text.
+        
+        Should work like any other non-empty book.
+        """
+        
+        BOOK_TEXT = """<utwor>
+        <liryka_l>
+            <nazwa_utworu>Nic</nazwa_utworu>
+        </liryka_l></utwor>
+        """
+        
+        book = models.Book.from_text_and_meta(ContentFile(BOOK_TEXT), self.book_info)
+        self.assertTrue(book.has_html_file())
+
+    def test_book_with_fragment(self):
+        BOOK_TEXT = """<utwor>
+        <opowiadanie>
+            <akap><begin id="m01" /><motyw id="m01">Love</motyw>Ala ma kota<end id="m01" /></akap>
+        </opowiadanie></utwor>
+        """
+
+        book = models.Book.from_text_and_meta(ContentFile(BOOK_TEXT), self.book_info)
+        self.assertTrue(book.has_html_file())
+
+        self.assertEqual(book.fragments.count(), 1)
+        self.assertEqual(book.fragments.all()[0].text, u'<p class="paragraph">Ala ma kota</p>\n')
+
+        self.assert_(('theme', 'love') in [ (tag.category, tag.slug) for tag in book.tags ])
+
+    def test_book_replace_title(self):
+        BOOK_TEXT = """<utwor />"""
+        self.book_info.title = u"Extraordinary"
+        book = models.Book.from_text_and_meta(ContentFile(BOOK_TEXT), self.book_info)
+
+        tags = [ (tag.category, tag.slug) for tag in book.tags ]
+        tags.sort()
+
+        self.assertEqual(tags, self.expected_tags)
+
+    def test_book_replace_author(self):
+        BOOK_TEXT = """<utwor />"""
+        self.book_info.author = PersonStub(("Hans", "Christian"), "Andersen")
+        book = models.Book.from_text_and_meta(ContentFile(BOOK_TEXT), self.book_info)
+
+        tags = [ (tag.category, tag.slug) for tag in book.tags ]
+        tags.sort()
+
+        self.expected_tags.remove(('author', 'jim-lazy'))
+        self.expected_tags.append(('author', 'hans-christian-andersen'))
+        self.expected_tags.sort()
+
+        self.assertEqual(tags, self.expected_tags)
+
+        # the old tag should disappear 
+        self.assertRaises(models.Tag.DoesNotExist, models.Tag.objects.get,
+                    slug="jim-lazy", category="author")
+
+
+    
+class BooksByTagTests(TestCase):
+    """ tests the /katalog/tag page for found books """
+    
     def setUp(self):
     def setUp(self):
-        t = models.Tag(name=u'Tadeusz Żeleński (Boy)', category=u'author', slug=u'a1')
-        t.save()
-        self.t = t
-        self.user = AnonymousUser()
+        author = PersonStub(("Common",), "Man")
+        tags = dict(genre='G', epoch='E', author=author, kind="K")
+
+        # grandchild
+        kwargs = info_args(u"GChild")
+        kwargs.update(tags)
+        gchild_info = BookInfoStub(**kwargs)
+        # child
+        kwargs = info_args(u"Child")
+        kwargs.update(tags)
+        child_info = BookInfoStub(parts=[gchild_info.url], **kwargs)
+        # other grandchild
+        kwargs = info_args(u"Different GChild")
+        kwargs.update(tags)
+        diffgchild_info = BookInfoStub(**kwargs)
+        # other child
+        kwargs = info_args(u"Different Child")
+        kwargs.update(tags)
+        kwargs['kind'] = 'K2'
+        diffchild_info = BookInfoStub(parts=[diffgchild_info.url], **kwargs)
+        # parent
+        kwargs = info_args(u"Parent")
+        kwargs.update(tags)
+        parent_info = BookInfoStub(parts=[child_info.url, diffchild_info.url], **kwargs)
+
+        # create the books
+        book_file = ContentFile('<utwor />')
+        for info in gchild_info, child_info, diffgchild_info, diffchild_info, parent_info:
+            book = models.Book.from_text_and_meta(book_file, info)
+
+        # useful tags
+        self.author = models.Tag.objects.get(name='Common Man', category='author')
+        tag_empty = models.Tag(name='Empty tag', slug='empty', category='author')
+        tag_empty.save()
+        
+        self.client = Client()
+    
     
     def tearDown(self):
     
     def tearDown(self):
-        self.t.delete()
+        models.Book.objects.all().delete()
+
     
     
-    def test_too_short(self):
-        self.assertRaises(ValueError, views.find_best_matches, u'', self.user)
-        self.assertRaises(ValueError, views.find_best_matches, u't', self.user)
+    def test_nonexistent_tag(self):
+        """ Looking for a non-existent tag should yield 404 """
+        self.assertEqual(404, self.client.get('/katalog/czeslaw_milosz/').status_code)
         
         
-    def test_match_beginning(self):
-        self.assertEqual(views.find_best_matches(u'Tad', self.user), (self.t,))
+    def test_book_tag(self):
+        """ Looking for a book tag isn't permitted """
+        self.assertEqual(404, self.client.get('/katalog/parent/').status_code)
+    
+    def test_tag_empty(self):
+        """ Tag with no books should return no books """
+        context = self.client.get('/katalog/empty/').context
+        self.assertEqual(0, len(context['object_list']))
     
     
-    def test_match_case(self):
-        self.assertEqual(views.find_best_matches(u'TAD', self.user), (self.t,))
+    def test_tag_common(self):
+        """ Filtering by tag should only yield top-level books. """
+        context = self.client.get('/katalog/%s/' % self.author.slug).context
+        self.assertEqual([book.title for book in context['object_list']],
+                         ['Parent'])
 
 
-    def test_word_boundary(self):
-        self.assertEqual(views.find_best_matches(u'Boy', self.user), (self.t,))
-        self.assertEqual(views.find_best_matches(u'(Boy', self.user), (self.t,))
+    def test_tag_child(self):
+        """ Filtering by child's tag should yield the child """
+        context = self.client.get('/katalog/k2/').context
+        self.assertEqual([book.title for book in context['object_list']],
+                         ['Different Child'])
+
+    def test_tag_child_jump(self):
+        """ Of parent and grandchild, only parent should be returned. """
+        context = self.client.get('/katalog/k/').context
+        self.assertEqual([book.title for book in context['object_list']],
+                         ['Parent'])
+        
+
+class TagRelatedTagsTests(TestCase):
+    """ tests the /katalog/tag/ page for related tags """
+    
+    def setUp(self):
+        author = PersonStub(("Common",), "Man")
+
+        gchild_info = BookInfoStub(author=author, genre="GchildGenre", epoch='Epoch', kind="Kind", 
+                                   **info_args(u"GChild"))
+        child1_info = BookInfoStub(author=author, genre="ChildGenre", epoch='Epoch', kind="ChildKind",
+                                   parts=[gchild_info.url],
+                                   **info_args(u"Child1"))
+        child2_info = BookInfoStub(author=author, genre="ChildGenre", epoch='Epoch', kind="ChildKind",
+                                   **info_args(u"Child2"))
+        parent_info = BookInfoStub(author=author, genre="Genre", epoch='Epoch', kind="Kind", 
+                                   parts=[child1_info.url, child2_info.url],
+                                   **info_args(u"Parent"))
+        
+        for info in gchild_info, child1_info, child2_info, parent_info:
+            book_text = """<utwor><opowiadanie><akap>
+                <begin id="m01" />
+                    <motyw id="m01">Theme, %sTheme</motyw>
+                    Ala ma kota
+                <end id="m01" />
+                </akap></opowiadanie></utwor>
+                """ % info.title.encode('utf-8')
+            book = models.Book.from_text_and_meta(ContentFile(book_text), info)
+            book.save()
+        
+        tag_empty = models.Tag(name='Empty tag', slug='empty', category='author')
+        tag_empty.save()
 
 
-    def test_not_found(self):
-        self.assertEqual(views.find_best_matches(u'andrzej', self.user), ())
-        self.assertEqual(views.find_best_matches(u'deusz', self.user), ())
+        self.client = Client()
     
     
-    def test_locale(self):
-        self.assertEqual(views.find_best_matches(u'ele', self.user), ())
-        self.assertEqual(views.find_best_matches(u'Żele', self.user), (self.t,))
-        self.assertEqual(views.find_best_matches(u'żele', self.user), (self.t,))
     
     
-    def test_sloppy(self):
-        self.assertEqual(views.find_best_matches(u'Żelenski', self.user), (self.t,))
-        self.assertEqual(views.find_best_matches(u'zelenski', self.user), (self.t,))
+    def tearDown(self):
+        models.Book.objects.all().delete()
+    
+    
+    def test_empty(self):
+        """ empty tag should have no related tags """
         
         
+        cats = self.client.get('/katalog/empty/').context['categories']
+        self.assertEqual(cats, {}, 'tags related to empty tag')
+    
+    
+    def test_has_related(self):
+        """ related own and descendants' tags should be generated """
+        
+        cats = self.client.get('/katalog/kind/').context['categories']
+        self.assertTrue('Common Man' in [tag.name for tag in cats['author']],
+                        'missing `author` related tag')
+        self.assertTrue('Epoch' in [tag.name for tag in cats['epoch']],
+                        'missing `epoch` related tag')
+        self.assertTrue("ChildKind" in [tag.name for tag in cats['kind']],
+                        "missing `kind` related tag")
+        self.assertTrue("Genre" in [tag.name for tag in cats['genre']],
+                        'missing `genre` related tag')
+        self.assertTrue("ChildGenre" in [tag.name for tag in cats['genre']],
+                        "missing child's related tag")
+        self.assertTrue("GchildGenre" in [tag.name for tag in cats['genre']],
+                        "missing grandchild's related tag")
+        self.assertTrue('Theme' in [tag.name for tag in cats['theme']],
+                        "missing related theme")
+        self.assertTrue('Child1Theme' in [tag.name for tag in cats['theme']],
+                        "missing child's related theme")
+        self.assertTrue('GChildTheme' in [tag.name for tag in cats['theme']],
+                        "missing grandchild's related theme")
+    
+    
+    def test_related_differ(self):
+        """ related tags shouldn't include filtering tags """
+        
+        cats = self.client.get('/katalog/kind/').context['categories']
+        self.assertFalse('Kind' in [tag.name for tag in cats['kind']],
+                         'filtering tag wrongly included in related')
+        cats = self.client.get('/katalog/theme/').context['categories']
+        self.assertFalse('Theme' in [tag.name for tag in cats['theme']],
+                         'filtering theme wrongly included in related')
+    
+    
+    def test_parent_tag_once(self):
+        """ if parent and descendants have a common tag, count it only once """
+
+        cats = self.client.get('/katalog/kind/').context['categories']
+        self.assertEqual([(tag.name, tag.count) for tag in cats['epoch']],
+                         [('Epoch', 1)],
+                         'wrong related tag epoch tag on tag page')
+    
+    
+    def test_siblings_tags_add(self):
+        """ if children have tags and parent hasn't, count the children """
+        
+        cats = self.client.get('/katalog/epoch/').context['categories']
+        self.assertTrue(('ChildKind', 2) in [(tag.name, tag.count) for tag in cats['kind']],
+                    'wrong related kind tags on tag page')
+    
+    def test_themes_add(self):
+        """ all occurencies of theme should be counted """
+
+        cats = self.client.get('/katalog/epoch/').context['categories']
+        self.assertTrue(('Theme', 4) in [(tag.name, tag.count) for tag in cats['theme']],
+                    'wrong related theme count')
+    
 
 
 
 
-class SetSearchTest(TestCase):
+class CleanTagRelationTests(TestCase):
+    """ tests for tag relations cleaning after deleting things """
+    
     def setUp(self):
     def setUp(self):
-        self.me = User(name='me')
-        self.other = User(name='other')
+        author = PersonStub(("Common",), "Man")
+
+        book_info = BookInfoStub(author=author, genre="G", epoch='E', kind="K", 
+                                   **info_args(u"Book"))
+        book_text = """<utwor><opowiadanie><akap>
+            <begin id="m01" /><motyw id="m01">Theme</motyw>Ala ma kota
+            <end id="m01" />
+            </akap></opowiadanie></utwor>
+            """
+        book = models.Book.from_text_and_meta(ContentFile(book_text), book_info)
+        book.save()
+        
+        self.client = Client()
+    
     
     def tearDown(self):
     
     def tearDown(self):
-        self.me.delete()
-        self.other.delete()
+        models.Book.objects.all().delete()
     
     
+    
+    def test_delete_objects(self):
+        """ there should be no related tags left after deleting some objects """
+        
+        models.Book.objects.all().delete()
+        cats = self.client.get('/katalog/k/').context['categories']
+        self.assertEqual({}, cats)
+
+
+    def test_deleted_tag(self):
+        """ there should be no tag relations left after deleting tags """
+        
+        models.Tag.objects.all().delete()
+        cats = self.client.get('/katalog/lektura/book/').context['categories']
+        self.assertEqual(cats, {})
+
+
+class TestIdenticalTag(TestCase):
+    
+    def setUp(self):
+        author = PersonStub(("A",), "B")
+
+        book_info = BookInfoStub(author=author, genre="A B", epoch='A B', kind="A B", 
+                                   **info_args(u"A B"))
+        book_text = """<utwor><opowiadanie><akap>
+            <begin id="m01" /><motyw id="m01">A B</motyw>Ala ma kota
+            <end id="m01" />
+            </akap></opowiadanie></utwor>
+            """
+        book = models.Book.from_text_and_meta(ContentFile(book_text), book_info)
+        book.save()
+        
+        self.client = Client()
+    
+    
+    def tearDown(self):
+        models.Book.objects.all().delete()
+    
+    
+    def test_book_tags(self):
+        """ there should be all related tags in relevant categories """
+        
+        cats = self.client.get('/katalog/lektura/a-b/').context['categories']
+        for category in 'author', 'kind', 'genre', 'epoch', 'theme':
+            self.assertTrue('A B' in [tag.name for tag in cats[category]],
+                            'missing related tag for %s' % category)
+
+    def test_qualified_url(self):
+        categories = {'author': 'autor', 'theme': 'motyw', 'epoch': 'epoka', 'kind':'rodzaj', 'genre':'gatunek'}
+        for cat, localcat in categories.iteritems():
+            context = self.client.get('/katalog/%s/a-b/' % localcat).context
+            self.assertEqual(1, len(context['object_list']))
+            self.assertNotEqual({}, context['categories'])
+            self.assertFalse(cat in context['categories'])
+