Transliterate HTML entities decoded after the unidecode pass

slugify/slugify.py

@@ -151,6 +151,7 @@ def slugify(
         text = unicodedata.normalize('NFKC', text)
     else:
         text = unicodedata.normalize('NFKD', text)
+        text = unidecode.unidecode(text)
 
     # make the text lowercase (optional)
     if lowercase:

test.py

@@ -166,6 +166,26 @@ def test_html_decimal_on(self):
         r = slugify(txt, decimal=True)
         self.assertEqual(r, 'z')
 
+    def test_html_decimal_on_transliterated(self):
+        # A decoded numeric entity must be transliterated to ASCII exactly like
+        # the same character passed in directly (docstring: Ž -> Ž -> z).
+        txt = 'ß'  # ß
+        r = slugify(txt, decimal=True)
+        self.assertEqual(r, slugify('ß'))
+        self.assertEqual(r, 'ss')
+
+    def test_html_hexadecimal_on_transliterated(self):
+        txt = 'ß'  # ß
+        r = slugify(txt, hexadecimal=True)
+        self.assertEqual(r, slugify('ß'))
+        self.assertEqual(r, 'ss')
+
+    def test_html_entities_on_transliterated(self):
+        txt = 'ß'  # ß
+        r = slugify(txt, entities=True)
+        self.assertEqual(r, slugify('ß'))
+        self.assertEqual(r, 'ss')
+
     def test_html_decimal_off(self):
         txt = 'Ž'
         r = slugify(txt, entities=False, decimal=False)