support union

Author: simonvandel

src/ast/query.rs

@@ -2690,6 +2690,15 @@ pub enum PipeOperator {
     ///
     /// See more at <https://cloud.google.com/bigquery/docs/reference/standard-sql/pipe-syntax#rename_pipe_operator>
     Rename { mappings: Vec<IdentWithAlias> },
+    /// Combines the input table with one or more tables using UNION.
+    ///
+    /// Syntax: `|> UNION [ALL|DISTINCT] (<query>), (<query>), ...`
+    ///
+    /// See more at <https://cloud.google.com/bigquery/docs/reference/standard-sql/pipe-syntax#union_pipe_operator>
+    Union { 
+        set_quantifier: SetQuantifier,
+        queries: Vec<Box<Query>>,
+    },
 }
 
 impl fmt::Display for PipeOperator {
@@ -2748,6 +2757,25 @@ impl fmt::Display for PipeOperator {
             PipeOperator::Rename { mappings } => {
                 write!(f, "RENAME {}", display_comma_separated(mappings))
             }
+            PipeOperator::Union { set_quantifier, queries } => {
+                write!(f, "UNION")?;
+                match set_quantifier {
+                    SetQuantifier::All => write!(f, " ALL")?,
+                    SetQuantifier::Distinct => write!(f, " DISTINCT")?,
+                    SetQuantifier::None => {},
+                    _ => {
+                        write!(f, " {}", set_quantifier)?;
+                    }
+                }
+                write!(f, " ")?;
+                for (i, query) in queries.iter().enumerate() {
+                    if i > 0 {
+                        write!(f, ", ")?;
+                    }
+                    write!(f, "({})", query)?;
+                }
+                Ok(())
+            }
         }
     }
 }

src/parser/mod.rs

@@ -11084,6 +11084,7 @@ impl<'a> Parser<'a> {
                 Keyword::ORDER,
                 Keyword::TABLESAMPLE,
                 Keyword::RENAME,
+                Keyword::UNION,
             ])?;
             match kw {
                 Keyword::SELECT => {
@@ -11154,6 +11155,19 @@ impl<'a> Parser<'a> {
                     let mappings = self.parse_comma_separated(Parser::parse_identifier_with_optional_alias)?;
                     pipe_operators.push(PipeOperator::Rename { mappings });
                 }
+                Keyword::UNION => {
+                    // Reuse existing set quantifier parser for consistent BY NAME support
+                    let set_quantifier = self.parse_set_quantifier(&Some(SetOperator::Union));
+                    // BigQuery UNION pipe operator requires parentheses around queries
+                    // Parse comma-separated list of parenthesized queries
+                    let queries = self.parse_comma_separated(|parser| {
+                        parser.expect_token(&Token::LParen)?;
+                        let query = parser.parse_query()?;
+                        parser.expect_token(&Token::RParen)?;
+                        Ok(query)
+                    })?;
+                    pipe_operators.push(PipeOperator::Union { set_quantifier, queries });
+                }
                 unhandled => {
                     return Err(ParserError::ParserError(format!(
                     "`expect_one_of_keywords` further up allowed unhandled keyword: {unhandled:?}"

tests/sqlparser_common.rs

@@ -15201,6 +15201,24 @@ fn parse_pipeline_operator() {
         "SELECT * FROM users |> RENAME id AS user_id",
     );
 
+    // union pipe operator
+    dialects.verified_stmt("SELECT * FROM users |> UNION ALL (SELECT * FROM admins)");
+    dialects.verified_stmt("SELECT * FROM users |> UNION DISTINCT (SELECT * FROM admins)");
+    dialects.verified_stmt("SELECT * FROM users |> UNION (SELECT * FROM admins)");
+    
+    // union pipe operator with multiple queries
+    dialects.verified_stmt("SELECT * FROM users |> UNION ALL (SELECT * FROM admins), (SELECT * FROM guests)");
+    dialects.verified_stmt("SELECT * FROM users |> UNION DISTINCT (SELECT * FROM admins), (SELECT * FROM guests), (SELECT * FROM employees)");
+    dialects.verified_stmt("SELECT * FROM users |> UNION (SELECT * FROM admins), (SELECT * FROM guests)");
+    
+    // union pipe operator with BY NAME modifier
+    dialects.verified_stmt("SELECT * FROM users |> UNION BY NAME (SELECT * FROM admins)");
+    dialects.verified_stmt("SELECT * FROM users |> UNION ALL BY NAME (SELECT * FROM admins)");
+    dialects.verified_stmt("SELECT * FROM users |> UNION DISTINCT BY NAME (SELECT * FROM admins)");
+    
+    // union pipe operator with BY NAME and multiple queries
+    dialects.verified_stmt("SELECT * FROM users |> UNION BY NAME (SELECT * FROM admins), (SELECT * FROM guests)");
+
     // many pipes
     dialects.verified_stmt(
         "SELECT * FROM CustomerOrders |> AGGREGATE SUM(cost) AS total_cost GROUP BY customer_id, state, item_type |> EXTEND COUNT(*) OVER (PARTITION BY customer_id) AS num_orders |> WHERE num_orders > 1 |> AGGREGATE AVG(total_cost) AS average GROUP BY state DESC, item_type ASC",