[pull] main from apache:main

benchmarks/queries/clickbench/README.md

@@ -155,7 +155,7 @@ WHERE
         THEN split_part(split_part("URL", 'resolution=', 2), '&', 1)::INT 
         ELSE 0 
     END > 1920 -- Extract and validate resolution parameter
-    AND levenshtein("UTMSource", "UTMCampaign") < 3 -- Verify UTM parameter similarity
+    AND levenshtein(CAST("UTMSource" AS STRING), CAST("UTMCampaign" AS STRING)) < 3 -- Verify UTM parameter similarity
 ```
 Result is empty,Since it has already been filtered by `"SocialAction" = 'share'`.
 

benchmarks/queries/clickbench/extended.sql

@@ -4,4 +4,4 @@ SELECT "BrowserCountry",  COUNT(DISTINCT "SocialNetwork"), COUNT(DISTINCT "HitCo
 SELECT "SocialSourceNetworkID", "RegionID", COUNT(*), AVG("Age"), AVG("ParamPrice"), STDDEV("ParamPrice") as s, VAR("ParamPrice")  FROM hits GROUP BY "SocialSourceNetworkID", "RegionID" HAVING s IS NOT NULL ORDER BY s DESC LIMIT 10;
 SELECT "ClientIP", "WatchID",  COUNT(*) c, MIN("ResponseStartTiming") tmin, MEDIAN("ResponseStartTiming") tmed, MAX("ResponseStartTiming") tmax FROM hits WHERE "JavaEnable" = 0  GROUP BY  "ClientIP", "WatchID" HAVING c > 1 ORDER BY tmed DESC LIMIT 10;
 SELECT "ClientIP", "WatchID",  COUNT(*) c, MIN("ResponseStartTiming") tmin, APPROX_PERCENTILE_CONT("ResponseStartTiming", 0.95) tp95, MAX("ResponseStartTiming") tmax FROM 'hits' WHERE "JavaEnable" = 0 GROUP BY  "ClientIP", "WatchID" HAVING c > 1 ORDER BY tp95 DESC LIMIT 10;
-SELECT COUNT(*) AS ShareCount FROM hits WHERE "IsMobile" = 1 AND "MobilePhoneModel" LIKE 'iPhone%' AND "SocialAction" = 'share' AND "SocialSourceNetworkID" IN (5, 12) AND "ClientTimeZone" BETWEEN -5 AND 5 AND regexp_match("Referer", '\/campaign\/(spring|summer)_promo') IS NOT NULL AND CASE WHEN split_part(split_part("URL", 'resolution=', 2), '&', 1) ~ '^\d+$' THEN split_part(split_part("URL", 'resolution=', 2), '&', 1)::INT ELSE 0 END > 1920 AND levenshtein("UTMSource", "UTMCampaign") < 3;
+SELECT COUNT(*) AS ShareCount FROM hits WHERE "IsMobile" = 1 AND "MobilePhoneModel" LIKE 'iPhone%' AND "SocialAction" = 'share' AND "SocialSourceNetworkID" IN (5, 12) AND "ClientTimeZone" BETWEEN -5 AND 5 AND regexp_match("Referer", '\/campaign\/(spring|summer)_promo') IS NOT NULL AND CASE WHEN split_part(split_part("URL", 'resolution=', 2), '&', 1) ~ '^\d+$' THEN split_part(split_part("URL", 'resolution=', 2), '&', 1)::INT ELSE 0 END > 1920 AND levenshtein(CAST("UTMSource" AS STRING), CAST("UTMCampaign" AS STRING)) < 3;

datafusion/sql/src/parser.rs

@@ -20,6 +20,7 @@
 //! This parser implements DataFusion specific statements such as
 //! `CREATE EXTERNAL TABLE`
 
+use datafusion_common::config::SqlParserOptions;
 use datafusion_common::DataFusionError;
 use datafusion_common::{sql_err, Diagnostic, Span};
 use sqlparser::ast::{ExprWithAlias, OrderByOptions};
@@ -284,6 +285,7 @@ fn ensure_not_set<T>(field: &Option<T>, name: &str) -> Result<(), DataFusionErro
 /// [`Statement`] for a list of this special syntax
 pub struct DFParser<'a> {
     pub parser: Parser<'a>,
+    options: SqlParserOptions,
 }
 
 /// Same as `sqlparser`
@@ -366,6 +368,10 @@ impl<'a> DFParserBuilder<'a> {
             parser: Parser::new(self.dialect)
                 .with_tokens_with_locations(tokens)
                 .with_recursion_limit(self.recursion_limit),
+            options: SqlParserOptions {
+                recursion_limit: self.recursion_limit,
+                ..Default::default()
+            },
         })
     }
 }
@@ -471,9 +477,7 @@ impl<'a> DFParser<'a> {
                         if let Token::Word(w) = self.parser.peek_nth_token(1).token {
                             // use native parser for COPY INTO
                             if w.keyword == Keyword::INTO {
-                                return Ok(Statement::Statement(Box::from(
-                                    self.parser.parse_statement()?,
-                                )));
+                                return self.parse_and_handle_statement();
                             }
                         }
                         self.parser.next_token(); // COPY
@@ -485,17 +489,13 @@ impl<'a> DFParser<'a> {
                     }
                     _ => {
                         // use sqlparser-rs parser
-                        Ok(Statement::Statement(Box::from(
-                            self.parser.parse_statement()?,
-                        )))
+                        self.parse_and_handle_statement()
                     }
                 }
             }
             _ => {
                 // use the native parser
-                Ok(Statement::Statement(Box::from(
-                    self.parser.parse_statement()?,
-                )))
+                self.parse_and_handle_statement()
             }
         }
     }
@@ -513,6 +513,23 @@ impl<'a> DFParser<'a> {
         Ok(self.parser.parse_expr_with_alias()?)
     }
 
+    /// Helper method to parse a statement and handle errors consistently, especially for recursion limits
+    fn parse_and_handle_statement(&mut self) -> Result<Statement, DataFusionError> {
+        self.parser
+            .parse_statement()
+            .map(|stmt| Statement::Statement(Box::from(stmt)))
+            .map_err(|e| match e {
+                ParserError::RecursionLimitExceeded => DataFusionError::SQL(
+                    ParserError::RecursionLimitExceeded,
+                    Some(format!(
+                        " (current limit: {})",
+                        self.options.recursion_limit
+                    )),
+                ),
+                other => DataFusionError::SQL(other, None),
+            })
+    }
+
     /// Parse a SQL `COPY TO` statement
     pub fn parse_copy(&mut self) -> Result<Statement, DataFusionError> {
         // parse as a query
@@ -1760,6 +1777,9 @@ mod tests {
             .parse_statements()
             .unwrap_err();
 
-        assert_contains!(err.to_string(), "SQL error: RecursionLimitExceeded");
+        assert_contains!(
+            err.to_string(),
+            "SQL error: RecursionLimitExceeded (current limit: 1)"
+        );
     }
 }

datafusion/sqllogictest/test_files/aggregate.slt

@@ -4998,20 +4998,21 @@ create table d as values
   (NULL, NULL, NULL, NULL, 1),
   (NULL, NULL, NULL, NULL, 2);
 
-query I?
+
+query I? rowsort
 SELECT column5, avg(column1) FROM d GROUP BY column5;
 ----
-2 0 days 0 hours 0 mins 15 secs
 1 0 days 0 hours 0 mins 6 secs
+2 0 days 0 hours 0 mins 15 secs
 
-query I??
-SELECT column5, column1, avg(column1) OVER (PARTITION BY column5 ORDER BY column1 ROWS BETWEEN 1 PRECEDING AND CURRENT ROW) as window_avg 
+query I?? rowsort
+SELECT column5, column1, avg(column1) OVER (PARTITION BY column5 ORDER BY column1 ROWS BETWEEN 1 PRECEDING AND CURRENT ROW) as window_avg
 FROM d WHERE column1 IS NOT NULL;
 ----
-2 0 days 0 hours 0 mins 5 secs 0 days 0 hours 0 mins 5 secs
-2 0 days 0 hours 0 mins 25 secs 0 days 0 hours 0 mins 15 secs
 1 0 days 0 hours 0 mins 1 secs 0 days 0 hours 0 mins 1 secs
 1 0 days 0 hours 0 mins 11 secs 0 days 0 hours 0 mins 6 secs
+2 0 days 0 hours 0 mins 25 secs 0 days 0 hours 0 mins 15 secs
+2 0 days 0 hours 0 mins 5 secs 0 days 0 hours 0 mins 5 secs
 
 # Cumulative average window function
 query I??