Avoid the chardet.universaldetector deprecation warning

AUTHORS.rst

@@ -63,4 +63,5 @@ Patches and suggestions
 - Ville Skyttä
 - Hugo van Kemenade
 - Mark Vasilkov
+- Afonso Januário
 

CHANGES.rst

@@ -20,6 +20,9 @@ Bug fixes:
 
 * The sanitizer now permits ``<summary>`` tags. It used to allow ``<details>``
   already. (#423)
+* Encoding detection no longer triggers a ``DeprecationWarning`` under
+  chardet 7+, which moved ``UniversalDetector`` out of
+  ``chardet.universaldetector``. (#601)
 
 1.1
 ~~~

html5lib/_inputstream.py

@@ -483,10 +483,16 @@ def determineEncoding(self, chardet=True):
         # Guess with chardet, if available
         if chardet:
             try:
-                from chardet.universaldetector import UniversalDetector
+                # chardet 7+ moved UniversalDetector here; the older
+                # chardet.universaldetector path still works but now raises
+                # a DeprecationWarning.
+                from chardet.detector import UniversalDetector
             except ImportError:
-                pass
-            else:
+                try:
+                    from chardet.universaldetector import UniversalDetector
+                except ImportError:
+                    UniversalDetector = None
+            if UniversalDetector is not None:
                 buffers = []
                 detector = UniversalDetector()
                 while not detector.done: