diff --git a/fast_autocomplete/normalize.py b/fast_autocomplete/normalize.py
index 41fd424..0bd3e67 100644
--- a/fast_autocomplete/normalize.py
+++ b/fast_autocomplete/normalize.py
@@ -12,11 +12,14 @@
 
 
 def normalize_node_name(name, extra_chars=None):
+    if name is None:
+        return ''
     name = name[:MAX_WORD_LENGTH]
-    result = _normalized_lfu_cache.get(name)
+    key = name if extra_chars is None else f"{name}{extra_chars}"
+    result = _normalized_lfu_cache.get(key)
     if result == -1:
         result = _get_normalized_node_name(name, extra_chars=extra_chars)
-        _normalized_lfu_cache.set(name, result)
+        _normalized_lfu_cache.set(key, result)
     return result
 
 
@@ -32,6 +35,8 @@ def remove_any_special_character(name):
     """
     Only remove invalid characters from a name. Useful for cleaning the user's original word.
     """
+    if name is None:
+        return ''
     name = name.lower()[:MAX_WORD_LENGTH]
     _remove_invalid_chars.prev_x = ''
 
@@ -43,7 +48,7 @@ def _get_normalized_node_name(name, extra_chars=None):
     result = []
     last_i = None
     for i in name:
-        if i in valid_chars_for_node_name or extra_chars and i in extra_chars:
+        if i in valid_chars_for_node_name or (extra_chars and i in extra_chars):
             if i == '-':
                 i = ' '
             elif (i in valid_chars_for_integer and last_i in valid_chars_for_string) or (i in valid_chars_for_string and last_i in valid_chars_for_integer):
diff --git a/tests/test_normalize.py b/tests/test_normalize.py
index 6f84f7c..60aa51b 100644
--- a/tests/test_normalize.py
+++ b/tests/test_normalize.py
@@ -1,5 +1,5 @@
 import pytest
-from fast_autocomplete.normalize import remove_any_special_character
+from fast_autocomplete.normalize import remove_any_special_character, normalize_node_name
 
 
 class TestMisc:
@@ -9,7 +9,20 @@ class TestMisc:
         ('HONDA and Toyota!', 'honda and toyota'),
         (r'bmw? \#1', 'bmw 1'),
         (r'bmw? \#', 'bmw'),
+        (None, ''),
     ])
-    def test_extend_and_repeat(self, name, expected_result):
+    def test_remove_any_special_character(self, name, expected_result):
         result = remove_any_special_character(name)
         assert expected_result == result
+
+    @pytest.mark.parametrize("name, extra_chars, expected_result", [
+        ('type-r', None, 'type r'),
+        ('HONDA and Toyota!', None, 'honda and toyota'),
+        (r'bmw? \#1', None, 'bmw 1'),
+        (r'bmw? \#', None, 'bmw'),
+        (r'bmw? \#', {'#'}, 'bmw #'),
+        (None, None, ''),
+    ])
+    def test_normalize_node_name(self, name, extra_chars, expected_result):
+        result = normalize_node_name(name, extra_chars=extra_chars)
+        assert expected_result == result