(datafusion-sqlparser-rs) branch main updated: Only parse FROM identifier in CTE if using Hive (#2241)

[gi...@apache.org]

This is an automated email from the ASF dual-hosted git repository.

github-bot pushed a commit to branch main
in repository https://gitbox.apache.org/repos/asf/datafusion-sqlparser-rs.git


The following commit(s) were added to refs/heads/main by this push:
     new 93bc80c9 Only parse FROM identifier in CTE if using Hive (#2241)
93bc80c9 is described below

commit 93bc80c946ff32aa065ddaafbbef62ac69c43909
Author: Victorien <65...@users.noreply.github.com>
AuthorDate: Wed Apr 15 06:12:44 2026 +0200

    Only parse FROM identifier in CTE if using Hive (#2241)
---
 src/dialect/hive.rs       |  7 +++++++
 src/dialect/mod.rs        | 15 +++++++++++++++
 src/parser/mod.rs         |  2 +-
 tests/sqlparser_common.rs | 29 ++++++++++++++++++++++++++++-
 4 files changed, 51 insertions(+), 2 deletions(-)

diff --git a/src/dialect/hive.rs b/src/dialect/hive.rs
index b39232ad..085d1598 100644
--- a/src/dialect/hive.rs
+++ b/src/dialect/hive.rs
@@ -72,4 +72,11 @@ impl Dialect for HiveDialect {
     fn supports_group_by_with_modifier(&self) -> bool {
         true
     }
+
+    // TODO: The parsing of the FROM keyword seems wrong, as it happens within the CTE.
+    // See https://github.com/apache/datafusion-sqlparser-rs/issues/2236 for more details.
+    /// See <https://hive.apache.org/docs/latest/language/common-table-expression/>
+    fn supports_from_first_insert(&self) -> bool {
+        true
+    }
 }
diff --git a/src/dialect/mod.rs b/src/dialect/mod.rs
index 1aa5aeac..6439ef2c 100644
--- a/src/dialect/mod.rs
+++ b/src/dialect/mod.rs
@@ -658,6 +658,21 @@ pub trait Dialect: Debug + Any {
         false
     }
 
+    /// Return true if the dialect supports "FROM-first" inserts.
+    ///
+    /// Example:
+    /// ```sql
+    /// WITH cte AS (SELECT key FROM src)
+    /// FROM cte
+    /// INSERT OVERWRITE table my_table
+    /// SELECT *
+    ///
+    /// See <https://hive.apache.org/docs/latest/language/common-table-expression/>
+    /// ```
+    fn supports_from_first_insert(&self) -> bool {
+        false
+    }
+
     /// Return true if the dialect supports pipe operator.
     ///
     /// Example:
diff --git a/src/parser/mod.rs b/src/parser/mod.rs
index 69a3605f..2b2aef89 100644
--- a/src/parser/mod.rs
+++ b/src/parser/mod.rs
@@ -14511,7 +14511,7 @@ impl<'a> Parser<'a> {
             materialized: is_materialized,
             closing_paren_token: closing_paren_token.into(),
         };
-        if self.parse_keyword(Keyword::FROM) {
+        if self.dialect.supports_from_first_insert() && self.parse_keyword(Keyword::FROM) {
             cte.from = Some(self.parse_identifier()?);
         }
         Ok(cte)
diff --git a/tests/sqlparser_common.rs b/tests/sqlparser_common.rs
index 80c37588..d516afe2 100644
--- a/tests/sqlparser_common.rs
+++ b/tests/sqlparser_common.rs
@@ -16218,10 +16218,37 @@ fn test_select_from_first() {
             pipe_operators: vec![],
         };
         assert_eq!(expected, ast);
-        assert_eq!(ast.to_string(), q);
     }
 }
 
+#[test]
+fn test_select_from_first_with_cte() {
+    let dialects = all_dialects_where(|d| d.supports_from_first_select());
+    let q = "WITH test AS (FROM t SELECT a) FROM test SELECT 1";
+
+    let ast = dialects.verified_query(q);
+
+    let ast_select = ast.body.as_select().unwrap();
+
+    let expected_body_select_projection =
+        vec![SelectItem::UnnamedExpr(Expr::Value(ValueWithSpan {
+            value: test_utils::number("1"),
+            span: Span::empty(),
+        }))];
+
+    let expected_body_from = vec![TableWithJoins {
+        relation: table_from_name(ObjectName::from(vec![Ident {
+            value: "test".to_string(),
+            quote_style: None,
+            span: Span::empty(),
+        }])),
+        joins: vec![],
+    }];
+
+    assert_eq!(ast_select.projection, expected_body_select_projection);
+    assert_eq!(ast_select.from, expected_body_from);
+}
+
 #[test]
 fn test_geometric_unary_operators() {
     // Number of points in path or polygon


---------------------------------------------------------------------
To unsubscribe, e-mail: commits-unsubscribe@datafusion.apache.org
For additional commands, e-mail: commits-help@datafusion.apache.org