Don't make parsing of square bracket identifiers too greedy (fixes andialbrecht#583).

andialbrecht · andialbrecht · commit cd4a723ee509 · 2020-09-30T08:11:53.000+02:00
diff --git a/CHANGELOG b/CHANGELOG
@@ -20,6 +20,7 @@ Bug Fixes
 
 * Improved parsing of IN(...) statements (issue566, pr567 by hurcy).
 * Preserve line breaks when removing comments (issue484).
+* Fix parsing error when using square bracket notation (issue583).
 
 
 Release 0.3.1 (Feb 29, 2020)
diff --git a/sqlparse/keywords.py b/sqlparse/keywords.py
@@ -73,7 +73,7 @@ def is_keyword(value):
         # sqlite names can be escaped with [square brackets]. left bracket
         # cannot be preceded by word character or a right bracket --
         # otherwise it's probably an array index
-        (r'(?<![\w\])])(\[[^\]]+\])', tokens.Name),
+        (r'(?<![\w\])])(\[[^\]\[]+\])', tokens.Name),
         (r'((LEFT\s+|RIGHT\s+|FULL\s+)?(INNER\s+|OUTER\s+|STRAIGHT\s+)?'
          r'|(CROSS\s+|NATURAL\s+)?)?JOIN\b', tokens.Keyword),
         (r'END(\s+IF|\s+LOOP|\s+WHILE)?\b', tokens.Keyword),
diff --git a/tests/test_parse.py b/tests/test_parse.py
@@ -100,6 +100,12 @@ def test_parse_square_brackets_notation_isnt_too_greedy():
     assert t[0].tokens[-1].get_real_name() == '[bar]'
 
 
+def test_parse_square_brackets_notation_isnt_too_greedy2():
+    # see issue583
+    t = sqlparse.parse('[(foo[i])]')[0].tokens
+    assert isinstance(t[0], sql.SquareBrackets)  # not Identifier!
+
+
 def test_parse_keyword_like_identifier():
     # see issue47
     t = sqlparse.parse('foo.key')[0].tokens
