feat: Implement dql Distinct

Author: KKould

README.md

@@ -54,7 +54,7 @@ Storage Support:
 - DQL
   - [x] Select
   - [x] Where
-  - [ ] Distinct
+  - [x] Distinct
   - [x] Alias
   - [x] Aggregation: count()/sum()/avg()/min()/max()
   - [ ] Subquery

src/binder/distinct.rs

@@ -0,0 +1,15 @@
+use crate::binder::Binder;
+use crate::expression::ScalarExpression;
+use crate::planner::LogicalPlan;
+use crate::planner::operator::aggregate::AggregateOperator;
+use crate::storage::Storage;
+
+impl<S: Storage> Binder<S> {
+    pub fn bind_distinct(
+        &mut self,
+        children: LogicalPlan,
+        select_list: Vec<ScalarExpression>,
+    ) -> LogicalPlan {
+        AggregateOperator::new(children, vec![], select_list)
+    }
+}

src/binder/mod.rs

@@ -7,6 +7,7 @@ mod update;
 mod delete;
 mod drop_table;
 mod truncate;
+mod distinct;
 
 use std::collections::BTreeMap;
 use sqlparser::ast::{Ident, ObjectName, ObjectType, SetExpr, Statement};

src/binder/select.rs

@@ -20,10 +20,7 @@ use super::Binder;
 use crate::catalog::{ColumnCatalog, DEFAULT_DATABASE_NAME, DEFAULT_SCHEMA_NAME, TableCatalog, TableName};
 use itertools::Itertools;
 use sqlparser::ast;
-use sqlparser::ast::{
-    Expr, Ident, Join, JoinConstraint, JoinOperator, Offset, OrderByExpr, Query, Select,
-    SelectItem, SetExpr, TableFactor, TableWithJoins,
-};
+use sqlparser::ast::{Distinct, Expr, Ident, Join, JoinConstraint, JoinOperator, Offset, OrderByExpr, Query, Select, SelectItem, SetExpr, TableFactor, TableWithJoins};
 use crate::binder::BindError;
 use crate::execution::executor::dql::join::joins_nullable;
 use crate::expression::BinaryOperator;
@@ -98,9 +95,9 @@ impl<S: Storage> Binder<S> {
             plan = self.bind_having(plan, having)?;
         }
 
-        // if select.distinct {
-        //     plan = self.bind_distinct(plan, select_list.clone())?;
-        // }
+        if let Some(Distinct::Distinct) = select.distinct {
+            plan = self.bind_distinct(plan, select_list.clone());
+        }
 
         if let Some(orderby) = having_orderby.1 {
             plan = self.bind_sort(plan, orderby);

src/db.rs

@@ -182,9 +182,9 @@ mod test {
     async fn test_crud_sql() -> Result<(), DatabaseError> {
         let temp_dir = TempDir::new().expect("unable to create temporary working directory");
         let kipsql = Database::with_kipdb(temp_dir.path()).await?;
-        let _ = kipsql.run("create table t1 (a int primary key, b int)").await?;
+        let _ = kipsql.run("create table t1 (a int primary key, b int, k int)").await?;
         let _ = kipsql.run("create table t2 (c int primary key, d int unsigned null, e datetime)").await?;
-        let _ = kipsql.run("insert into t1 (a, b) values (1, 1), (4, 3), (5, 2)").await?;
+        let _ = kipsql.run("insert into t1 (a, b, k) values (1, 1, 1), (4, 2, 2), (5, 2, 2)").await?;
         let _ = kipsql.run("insert into t2 (d, c, e) values (2, 1, '2021-05-20 21:00:00'), (3, 4, '2023-09-10 00:00:00')").await?;
 
         println!("full t1:");
@@ -265,6 +265,10 @@ mod test {
 
         assert!(kipsql.run("select max(d) from t2 group by c").await.is_err());
 
+        println!("distinct t1:");
+        let tuples_distinct_t1 = kipsql.run("select distinct b, k from t1").await?;
+        println!("{}", create_table(&tuples_distinct_t1));
+
         println!("update t1 with filter:");
         let _ = kipsql.run("update t1 set a = 0 where b > 1").await?;
         println!("after t1:");