From b8fd9d05eaf2f7b47bb12429d8eba51e16f1ca59 Mon Sep 17 00:00:00 2001
From: eeshsaxena <eeshsaxena@gmail.com>
Date: Wed, 12 Aug 2026 08:23:14 +0530
Subject: [PATCH] Don't crash on malformed SQL that makes sqlglot raise
 mid-parse

DialectParser runs sqlglot in WARN mode so it returns a best-effort AST instead
of raising. For a handful of short inputs like "{ =" sqlglot instead raises an
AttributeError while assembling that partial tree (a node whose key is None).
That was neither ParseError nor TokenError, so it escaped _try_dialects and
every public accessor (tables, columns, query_type, ...) crashed with a raw
AttributeError.

Catch any such sqlglot-side failure in _parse_with_dialect and treat it as 'this
dialect produced no result', so the query is reported through the normal
InvalidQueryDefinition path. ParseError/TokenError are still re-raised so the
existing syntax-error handling is unchanged.
---
 sql_metadata/dialect_parser.py | 12 ++++++++++++
 test/test_malformed_input.py   | 27 +++++++++++++++++++++++++++
 2 files changed, 39 insertions(+)
 create mode 100644 test/test_malformed_input.py

diff --git a/sql_metadata/dialect_parser.py b/sql_metadata/dialect_parser.py
index 87807458..10eeecd0 100644
--- a/sql_metadata/dialect_parser.py
+++ b/sql_metadata/dialect_parser.py
@@ -249,6 +249,18 @@ def _parse_with_dialect(clean_sql: str, dialect: Any) -> exp.Expression | None:
                 dialect=dialect,
                 error_level=sqlglot.ErrorLevel.WARN,
             )
+        except (ParseError, TokenError):
+            # Re-raise so _try_dialects can report a real syntax error on the
+            # last dialect (see its except clause).
+            raise
+        except Exception:
+            # WARN mode is supposed to return a best-effort AST instead of
+            # raising, but sqlglot can still blow up while assembling that tree,
+            # e.g. an AttributeError on a node whose key is None for input like
+            # "{ =". Treat any such failure as "this dialect produced nothing"
+            # so the query is reported as invalid rather than crashing the
+            # public accessors with a raw sqlglot exception.
+            return None
         finally:
             logger.setLevel(old_level)
 
diff --git a/test/test_malformed_input.py b/test/test_malformed_input.py
new file mode 100644
index 00000000..34a68fd1
--- /dev/null
+++ b/test/test_malformed_input.py
@@ -0,0 +1,27 @@
+"""Regression tests for malformed input that used to crash the parser.
+
+sqlglot in best-effort (WARN) mode can build a partial AST and then raise while
+assembling it, e.g. an ``AttributeError`` on a node whose key is ``None`` for a
+few-character string like ``"{ ="``. That escaped DialectParser and every public
+accessor crashed with a raw ``AttributeError`` instead of reporting an invalid
+query.
+"""
+
+import pytest
+
+from sql_metadata import InvalidQueryDefinition, Parser
+
+
+@pytest.mark.parametrize("query", ["{ =", "SELECT { =", "x { ="])
+def test_bracket_equals_does_not_crash(query):
+    # query_type / tables validate the AST, so they surface the invalid query
+    # as InvalidQueryDefinition rather than an AttributeError.
+    with pytest.raises(InvalidQueryDefinition):
+        Parser(query).query_type
+
+    with pytest.raises(InvalidQueryDefinition):
+        Parser(query).tables
+
+    # The best-effort accessors must simply come back empty, not crash.
+    assert Parser(query).columns == []
+    assert Parser(query).columns_dict == {}