Allow setting the recursion limit for sql parsing

Author: cetra3

datafusion/common/src/config.rs

@@ -256,6 +256,9 @@ config_namespace! {
         /// query (i.e. [`Span`](sqlparser::tokenizer::Span)) will be collected
         /// and recorded in the logical plan nodes.
         pub collect_spans: bool, default = false
+
+        /// Specifies the recursion depth limit when parsing complex SQL Queries
+        pub recursion_limit: usize, default = 50
     }
 }
 

datafusion/core/src/execution/session_state.rs

@@ -483,12 +483,21 @@ impl SessionState {
                      MsSQL, ClickHouse, BigQuery, Ansi, DuckDB, Databricks."
             )
         })?;
-        let mut statements = DFParser::parse_sql_with_dialect(sql, dialect.as_ref())?;
+
+        let recursion_limit = self.config.options().sql_parser.recursion_limit;
+
+        let mut statements = DFParser::parse_sql_with_dialect_limit(
+            sql,
+            dialect.as_ref(),
+            recursion_limit,
+        )?;
+
         if statements.len() > 1 {
             return not_impl_err!(
                 "The context currently only supports a single SQL statement"
             );
         }
+
         let statement = statements.pop_front().ok_or_else(|| {
             plan_datafusion_err!("No SQL statements were provided in the query string")
         })?;
@@ -522,7 +531,12 @@ impl SessionState {
             )
         })?;
 
-        let expr = DFParser::parse_sql_into_expr_with_dialect(sql, dialect.as_ref())?;
+        let recursion_limit = self.config.options().sql_parser.recursion_limit;
+        let expr = DFParser::parse_sql_into_expr_with_dialect_limit(
+            sql,
+            dialect.as_ref(),
+            recursion_limit,
+        )?;
 
         Ok(expr)
     }

datafusion/sql/src/parser.rs

@@ -257,6 +257,9 @@ fn ensure_not_set<T>(field: &Option<T>, name: &str) -> Result<(), ParserError> {
     Ok(())
 }
 
+/// Same as `sqlparser`
+const DEFAULT_RECURSION_LIMIT: usize = 50;
+
 /// DataFusion SQL Parser based on [`sqlparser`]
 ///
 /// Parses DataFusion's SQL dialect, often delegating to [`sqlparser`]'s [`Parser`].
@@ -282,20 +285,31 @@ impl<'a> DFParser<'a> {
     pub fn new_with_dialect(
         sql: &str,
         dialect: &'a dyn Dialect,
+    ) -> Result<Self, ParserError> {
+        DFParser::new_with_dialect_limit(sql, dialect, DEFAULT_RECURSION_LIMIT)
+    }
+    /// Create a new parser for the specified tokens with the
+    /// specified dialect and recursion limit
+    pub fn new_with_dialect_limit(
+        sql: &str,
+        dialect: &'a dyn Dialect,
+        recursion_limit: usize,
     ) -> Result<Self, ParserError> {
         let mut tokenizer = Tokenizer::new(dialect, sql);
         let tokens = tokenizer.tokenize_with_location()?;
 
         Ok(DFParser {
-            parser: Parser::new(dialect).with_tokens_with_locations(tokens),
+            parser: Parser::new(dialect)
+                .with_tokens_with_locations(tokens)
+                .with_recursion_limit(recursion_limit),
         })
     }
 
     /// Parse a sql string into one or [`Statement`]s using the
     /// [`GenericDialect`].
     pub fn parse_sql(sql: &str) -> Result<VecDeque<Statement>, ParserError> {
         let dialect = &GenericDialect {};
-        DFParser::parse_sql_with_dialect(sql, dialect)
+        DFParser::parse_sql_with_dialect_limit(sql, dialect, DEFAULT_RECURSION_LIMIT)
     }
 
     /// Parse a SQL string and produce one or more [`Statement`]s with
@@ -304,7 +318,17 @@ impl<'a> DFParser<'a> {
         sql: &str,
         dialect: &dyn Dialect,
     ) -> Result<VecDeque<Statement>, ParserError> {
-        let mut parser = DFParser::new_with_dialect(sql, dialect)?;
+        DFParser::parse_sql_with_dialect_limit(sql, dialect, DEFAULT_RECURSION_LIMIT)
+    }
+
+    /// Parse a SQL string and produce one or more [`Statement`]s with
+    /// with the specified dialect and recursion limit
+    pub fn parse_sql_with_dialect_limit(
+        sql: &str,
+        dialect: &dyn Dialect,
+        recursion_limit: usize,
+    ) -> Result<VecDeque<Statement>, ParserError> {
+        let mut parser = DFParser::new_with_dialect_limit(sql, dialect, recursion_limit)?;
         let mut stmts = VecDeque::new();
         let mut expecting_statement_delimiter = false;
         loop {
@@ -331,7 +355,19 @@ impl<'a> DFParser<'a> {
         sql: &str,
         dialect: &dyn Dialect,
     ) -> Result<ExprWithAlias, ParserError> {
-        let mut parser = DFParser::new_with_dialect(sql, dialect)?;
+        DFParser::parse_sql_into_expr_with_dialect_limit(
+            sql,
+            dialect,
+            DEFAULT_RECURSION_LIMIT,
+        )
+    }
+
+    pub fn parse_sql_into_expr_with_dialect_limit(
+        sql: &str,
+        dialect: &dyn Dialect,
+        recursion_limit: usize,
+    ) -> Result<ExprWithAlias, ParserError> {
+        let mut parser = DFParser::new_with_dialect_limit(sql, dialect, recursion_limit)?;
         parser.parse_expr()
     }
 

datafusion/sqllogictest/test_files/information_schema.slt

@@ -263,6 +263,7 @@ datafusion.sql_parser.dialect generic
 datafusion.sql_parser.enable_ident_normalization true
 datafusion.sql_parser.enable_options_value_normalization false
 datafusion.sql_parser.parse_float_as_decimal false
+datafusion.sql_parser.recursion_limit 50
 datafusion.sql_parser.support_varchar_with_length true
 
 # show all variables with verbose
@@ -359,6 +360,7 @@ datafusion.sql_parser.dialect generic Configure the SQL dialect used by DataFusi
 datafusion.sql_parser.enable_ident_normalization true When set to true, SQL parser will normalize ident (convert ident to lowercase when not quoted)
 datafusion.sql_parser.enable_options_value_normalization false When set to true, SQL parser will normalize options value (convert value to lowercase). Note that this option is ignored and will be removed in the future. All case-insensitive values are normalized automatically.
 datafusion.sql_parser.parse_float_as_decimal false When set to true, SQL parser will parse float as decimal type
+datafusion.sql_parser.recursion_limit 50 Specifies the recursion depth limit when parsing complex SQL Queries
 datafusion.sql_parser.support_varchar_with_length true If true, permit lengths for `VARCHAR` such as `VARCHAR(20)`, but ignore the length. If false, error if a `VARCHAR` with a length is specified. The Arrow type system does not have a notion of maximum string length and thus DataFusion can not enforce such limits.
 
 # show_variable_in_config_options

docs/source/user-guide/configs.md

@@ -128,3 +128,4 @@ Environment variables are read during `SessionConfig` initialisation so they mus
 | datafusion.sql_parser.dialect                                           | generic                   | Configure the SQL dialect used by DataFusion's parser; supported values include: Generic, MySQL, PostgreSQL, Hive, SQLite, Snowflake, Redshift, MsSQL, ClickHouse, BigQuery, Ansi, DuckDB and Databricks.                                                                                                                                                                                                                                                                                                                                                                |
 | datafusion.sql_parser.support_varchar_with_length                       | true                      | If true, permit lengths for `VARCHAR` such as `VARCHAR(20)`, but ignore the length. If false, error if a `VARCHAR` with a length is specified. The Arrow type system does not have a notion of maximum string length and thus DataFusion can not enforce such limits.                                                                                                                                                                                                                                                                                                    |
 | datafusion.sql_parser.collect_spans                                     | false                     | When set to true, the source locations relative to the original SQL query (i.e. [`Span`](sqlparser::tokenizer::Span)) will be collected and recorded in the logical plan nodes.                                                                                                                                                                                                                                                                                                                                                                                          |
+| datafusion.sql_parser.recursion_limit                                   | 50                        | Specifies the recursion depth limit when parsing complex SQL Queries                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                     |