Support TABLESAMPLE pipe operator

Author: hendrikmakait

src/ast/query.rs

@@ -2680,6 +2680,10 @@ pub enum PipeOperator {
         full_table_exprs: Vec<ExprWithAliasAndOrderBy>,
         group_by_expr: Vec<ExprWithAliasAndOrderBy>,
     },
+    /// Selects a random sample of rows from the input table.
+    /// Syntax: `|> TABLESAMPLE <method> (<size> {ROWS | PERCENT})`
+    /// See more at <https://cloud.google.com/bigquery/docs/reference/standard-sql/pipe-syntax#tablesample_pipe_operator>
+    TableSample { sample: Box <TableSample> },
 }
 
 impl fmt::Display for PipeOperator {
@@ -2731,6 +2735,10 @@ impl fmt::Display for PipeOperator {
             PipeOperator::OrderBy { exprs } => {
                 write!(f, "ORDER BY {}", display_comma_separated(exprs.as_slice()))
             }
+
+            PipeOperator::TableSample { sample } => {
+                write!(f, " {}", sample)
+            }
         }
     }
 }

src/parser/mod.rs

@@ -11047,6 +11047,7 @@ impl<'a> Parser<'a> {
                 Keyword::LIMIT,
                 Keyword::AGGREGATE,
                 Keyword::ORDER,
+                Keyword::TABLESAMPLE,
             ])?;
             match kw {
                 Keyword::SELECT => {
@@ -11109,6 +11110,12 @@ impl<'a> Parser<'a> {
                     let exprs = self.parse_comma_separated(Parser::parse_order_by_expr)?;
                     pipe_operators.push(PipeOperator::OrderBy { exprs })
                 }
+                Keyword::TABLESAMPLE => {
+                    if let Some(sample) = self.maybe_parse_table_sample()? {
+                        pipe_operators.push(PipeOperator::TableSample { sample });
+                    };
+
+                }
                 unhandled => {
                     return Err(ParserError::ParserError(format!(
                     "`expect_one_of_keywords` further up allowed unhandled keyword: {unhandled:?}"

tests/sqlparser_common.rs

@@ -15155,6 +15155,12 @@ fn parse_pipeline_operator() {
     dialects.verified_stmt("SELECT * FROM users |> ORDER BY id DESC");
     dialects.verified_stmt("SELECT * FROM users |> ORDER BY id DESC, name ASC");
 
+    // tablesample pipe operator
+    dialects.verified_stmt("SELECT * FROM tbl AS t TABLESAMPLE BERNOULLI (50)");
+    dialects.verified_stmt("SELECT * FROM tbl AS t TABLESAMPLE SYSTEM (50)");
+    // TODO: Technically, REPEATABLE is not available in BigQuery, but it is used with TABLESAMPLE in other dialects
+    dialects.verified_stmt("SELECT * FROM tbl AS t TABLESAMPLE SYSTEM (50) REPEATABLE (10)");
+
     // 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",