From 3310076dbe781e0554519fba5c4a2585a463250f Mon Sep 17 00:00:00 2001
From: Galina Shalygina <galashalygina@gmail.com>
Date: Fri, 25 Aug 2017 19:06:13 +0200
Subject: [PATCH] Optimization that transforms IN-predicate in IN-subselect
 made.

Trasformation goes that way:
a in (1,2) ->
a in (select * from (values ((1),(2))) as new_tvc)

Special variable that controlls optimization added.
Now optimization works only in several cases.
New tests added.

Old tests corrected. Now with TVC explain can be used. TVC also can be used in recursive CTEs.
---
 mysql-test/r/opt_tvc.result | 520 ++++++++++++++++++++++++++++++++++++
 mysql-test/t/opt_tvc.test   | 124 +++++++++
 2 files changed, 644 insertions(+)
 create mode 100644 mysql-test/r/opt_tvc.result
 create mode 100644 mysql-test/t/opt_tvc.test

diff --git a/mysql-test/r/opt_tvc.result b/mysql-test/r/opt_tvc.result
new file mode 100644
index 00000000000..f3a5f182e61
--- /dev/null
+++ b/mysql-test/r/opt_tvc.result
@@ -0,0 +1,520 @@
+create table t1 (a int, b int);
+insert into t1 
+values (1,2), (4,6), (9,7),
+(1,1), (2,5), (7,8);
+create table t2 (a int, b int, c int);
+insert into t2 
+values (1,2,3), (5,1,2), (4,3,7), 
+(8,9,0), (10,7,1), (5,5,1);
+# optimization is not used
+select * from t1 where a in (1,2);
+a	b
+1	2
+1	1
+2	5
+explain select * from t1 where a in (1,2);
+id	select_type	table	type	possible_keys	key	key_len	ref	rows	Extra
+1	SIMPLE	t1	ALL	NULL	NULL	NULL	NULL	6	Using where
+explain format=json select * from t1 where a in (1,2);
+EXPLAIN
+{
+  "query_block": {
+    "select_id": 1,
+    "table": {
+      "table_name": "t1",
+      "access_type": "ALL",
+      "rows": 6,
+      "filtered": 100,
+      "attached_condition": "t1.a in (1,2)"
+    }
+  }
+}
+# set minimum number of values in VALUEs list when optimization works to 2
+set @@in_subquery_conversion_threshold= 2;
+# single IN-predicate in WHERE-part
+select * from t1 where a in (1,2);
+a	b
+1	2
+1	1
+2	5
+select * from t1 
+where a in 
+(
+select * 
+from (values (1),(2)) as new_tvc
+);
+a	b
+1	2
+1	1
+2	5
+explain select * from t1 where a in (1,2);
+id	select_type	table	type	possible_keys	key	key_len	ref	rows	Extra
+1	PRIMARY	<subquery2>	ALL	distinct_key	NULL	NULL	NULL	2	
+1	PRIMARY	t1	ALL	NULL	NULL	NULL	NULL	6	Using where; Using join buffer (flat, BNL join)
+2	MATERIALIZED	<derived3>	ALL	NULL	NULL	NULL	NULL	2	
+3	DERIVED	NULL	NULL	NULL	NULL	NULL	NULL	NULL	No tables used
+explain select * from t1 
+where a in 
+(
+select * 
+from (values (1),(2)) as new_tvc
+);
+id	select_type	table	type	possible_keys	key	key_len	ref	rows	Extra
+1	PRIMARY	<subquery2>	ALL	distinct_key	NULL	NULL	NULL	2	
+1	PRIMARY	t1	ALL	NULL	NULL	NULL	NULL	6	Using where; Using join buffer (flat, BNL join)
+2	MATERIALIZED	<derived3>	ALL	NULL	NULL	NULL	NULL	2	
+3	DERIVED	NULL	NULL	NULL	NULL	NULL	NULL	NULL	No tables used
+explain format=json select * from t1 where a in (1,2);
+EXPLAIN
+{
+  "query_block": {
+    "select_id": 1,
+    "table": {
+      "table_name": "<subquery2>",
+      "access_type": "ALL",
+      "possible_keys": ["distinct_key"],
+      "rows": 2,
+      "filtered": 100,
+      "materialized": {
+        "unique": 1,
+        "query_block": {
+          "select_id": 2,
+          "table": {
+            "table_name": "<derived3>",
+            "access_type": "ALL",
+            "rows": 2,
+            "filtered": 100,
+            "materialized": {
+              "query_block": {
+                "union_result": {
+                  "table_name": "<unit3>",
+                  "access_type": "ALL",
+                  "query_specifications": [
+                    {
+                      "query_block": {
+                        "select_id": 3,
+                        "table": {
+                          "message": "No tables used"
+                        }
+                      }
+                    }
+                  ]
+                }
+              }
+            }
+          }
+        }
+      }
+    },
+    "block-nl-join": {
+      "table": {
+        "table_name": "t1",
+        "access_type": "ALL",
+        "rows": 6,
+        "filtered": 100
+      },
+      "buffer_type": "flat",
+      "buffer_size": "256Kb",
+      "join_type": "BNL",
+      "attached_condition": "t1.a = new_tvc.`1`"
+    }
+  }
+}
+# AND-condition with IN-predicates in WHERE-part
+select * from t1 
+where a in (1,2) and 
+b in (1,5);
+a	b
+1	1
+2	5
+select * from t1 
+where a in 
+(
+select * 
+from (values (1),(2)) as new_tvc
+) 
+and b in
+(
+select * 
+from (values (1),(5)) as new_tvc
+);
+a	b
+1	1
+2	5
+explain select * from t1 
+where a in (1,2) and 
+b in (1,5);
+id	select_type	table	type	possible_keys	key	key_len	ref	rows	Extra
+1	PRIMARY	<subquery4>	ALL	distinct_key	NULL	NULL	NULL	2	
+1	PRIMARY	t1	ALL	NULL	NULL	NULL	NULL	6	Using where; Using join buffer (flat, BNL join)
+1	PRIMARY	<subquery2>	eq_ref	distinct_key	distinct_key	4	func	1	
+4	MATERIALIZED	<derived5>	ALL	NULL	NULL	NULL	NULL	2	
+2	MATERIALIZED	<derived3>	ALL	NULL	NULL	NULL	NULL	2	
+5	DERIVED	NULL	NULL	NULL	NULL	NULL	NULL	NULL	No tables used
+3	DERIVED	NULL	NULL	NULL	NULL	NULL	NULL	NULL	No tables used
+explain select * from t1 
+where a in 
+(
+select * 
+from (values (1),(2)) as new_tvc
+) 
+and b in
+(
+select * 
+from (values (1),(5)) as new_tvc
+);
+id	select_type	table	type	possible_keys	key	key_len	ref	rows	Extra
+1	PRIMARY	<subquery4>	ALL	distinct_key	NULL	NULL	NULL	2	
+1	PRIMARY	t1	ALL	NULL	NULL	NULL	NULL	6	Using where; Using join buffer (flat, BNL join)
+1	PRIMARY	<subquery2>	eq_ref	distinct_key	distinct_key	4	func	1	
+4	MATERIALIZED	<derived5>	ALL	NULL	NULL	NULL	NULL	2	
+2	MATERIALIZED	<derived3>	ALL	NULL	NULL	NULL	NULL	2	
+5	DERIVED	NULL	NULL	NULL	NULL	NULL	NULL	NULL	No tables used
+3	DERIVED	NULL	NULL	NULL	NULL	NULL	NULL	NULL	No tables used
+explain format=json select * from t1 
+where a in (1,2) and 
+b in (1,5);
+EXPLAIN
+{
+  "query_block": {
+    "select_id": 1,
+    "table": {
+      "table_name": "<subquery4>",
+      "access_type": "ALL",
+      "possible_keys": ["distinct_key"],
+      "rows": 2,
+      "filtered": 100,
+      "materialized": {
+        "unique": 1,
+        "query_block": {
+          "select_id": 4,
+          "table": {
+            "table_name": "<derived5>",
+            "access_type": "ALL",
+            "rows": 2,
+            "filtered": 100,
+            "materialized": {
+              "query_block": {
+                "union_result": {
+                  "table_name": "<unit5>",
+                  "access_type": "ALL",
+                  "query_specifications": [
+                    {
+                      "query_block": {
+                        "select_id": 5,
+                        "table": {
+                          "message": "No tables used"
+                        }
+                      }
+                    }
+                  ]
+                }
+              }
+            }
+          }
+        }
+      }
+    },
+    "block-nl-join": {
+      "table": {
+        "table_name": "t1",
+        "access_type": "ALL",
+        "rows": 6,
+        "filtered": 100
+      },
+      "buffer_type": "flat",
+      "buffer_size": "256Kb",
+      "join_type": "BNL",
+      "attached_condition": "t1.b = new_tvc.`1`"
+    },
+    "table": {
+      "table_name": "<subquery2>",
+      "access_type": "eq_ref",
+      "possible_keys": ["distinct_key"],
+      "key": "distinct_key",
+      "key_length": "4",
+      "used_key_parts": ["1"],
+      "ref": ["func"],
+      "rows": 1,
+      "filtered": 100,
+      "materialized": {
+        "unique": 1,
+        "query_block": {
+          "select_id": 2,
+          "table": {
+            "table_name": "<derived3>",
+            "access_type": "ALL",
+            "rows": 2,
+            "filtered": 100,
+            "materialized": {
+              "query_block": {
+                "union_result": {
+                  "table_name": "<unit3>",
+                  "access_type": "ALL",
+                  "query_specifications": [
+                    {
+                      "query_block": {
+                        "select_id": 3,
+                        "table": {
+                          "message": "No tables used"
+                        }
+                      }
+                    }
+                  ]
+                }
+              }
+            }
+          }
+        }
+      }
+    }
+  }
+}
+# OR-condition with IN-predicates in WHERE-part
+select * from t1 
+where a in (1,2) or 
+b in (4,5);
+a	b
+1	2
+1	1
+2	5
+select * from t1 
+where a in 
+(
+select * 
+from (values (1),(2)) as new_tvc
+) 
+or b in
+(
+select * 
+from (values (4),(5)) as new_tvc
+);
+a	b
+1	2
+1	1
+2	5
+explain select * from t1 
+where a in (1,2) or 
+b in (4,5);
+id	select_type	table	type	possible_keys	key	key_len	ref	rows	Extra
+1	PRIMARY	t1	ALL	NULL	NULL	NULL	NULL	6	Using where
+4	MATERIALIZED	<derived5>	ALL	NULL	NULL	NULL	NULL	2	
+5	DERIVED	NULL	NULL	NULL	NULL	NULL	NULL	NULL	No tables used
+2	MATERIALIZED	<derived3>	ALL	NULL	NULL	NULL	NULL	2	
+3	DERIVED	NULL	NULL	NULL	NULL	NULL	NULL	NULL	No tables used
+explain select * from t1 
+where a in 
+(
+select * 
+from (values (1),(2)) as new_tvc
+) 
+or b in
+(
+select * 
+from (values (4),(5)) as new_tvc
+);
+id	select_type	table	type	possible_keys	key	key_len	ref	rows	Extra
+1	PRIMARY	t1	ALL	NULL	NULL	NULL	NULL	6	Using where
+4	MATERIALIZED	<derived5>	ALL	NULL	NULL	NULL	NULL	2	
+5	DERIVED	NULL	NULL	NULL	NULL	NULL	NULL	NULL	No tables used
+2	MATERIALIZED	<derived3>	ALL	NULL	NULL	NULL	NULL	2	
+3	DERIVED	NULL	NULL	NULL	NULL	NULL	NULL	NULL	No tables used
+explain format=json select * from t1 
+where a in (1,2) or 
+b in (4,5);
+EXPLAIN
+{
+  "query_block": {
+    "select_id": 1,
+    "table": {
+      "table_name": "t1",
+      "access_type": "ALL",
+      "rows": 6,
+      "filtered": 100,
+      "attached_condition": "<in_optimizer>(t1.a,t1.a in (subquery#2)) or <in_optimizer>(t1.b,t1.b in (subquery#4))"
+    },
+    "subqueries": [
+      {
+        "query_block": {
+          "select_id": 4,
+          "table": {
+            "table_name": "<derived5>",
+            "access_type": "ALL",
+            "rows": 2,
+            "filtered": 100,
+            "materialized": {
+              "query_block": {
+                "union_result": {
+                  "table_name": "<unit5>",
+                  "access_type": "ALL",
+                  "query_specifications": [
+                    {
+                      "query_block": {
+                        "select_id": 5,
+                        "table": {
+                          "message": "No tables used"
+                        }
+                      }
+                    }
+                  ]
+                }
+              }
+            }
+          }
+        }
+      },
+      {
+        "query_block": {
+          "select_id": 2,
+          "table": {
+            "table_name": "<derived3>",
+            "access_type": "ALL",
+            "rows": 2,
+            "filtered": 100,
+            "materialized": {
+              "query_block": {
+                "union_result": {
+                  "table_name": "<unit3>",
+                  "access_type": "ALL",
+                  "query_specifications": [
+                    {
+                      "query_block": {
+                        "select_id": 3,
+                        "table": {
+                          "message": "No tables used"
+                        }
+                      }
+                    }
+                  ]
+                }
+              }
+            }
+          }
+        }
+      }
+    ]
+  }
+}
+# subquery with IN-predicate
+select * from t1 
+where a in 
+(
+select a 
+from t2 where b in (3,4)
+)
+;
+a	b
+4	6
+select * from t1 
+where a in (
+select a from t2 
+where b in 
+(
+select * 
+from (values (3),(4)) as new_tvc)
+)
+;
+a	b
+4	6
+explain select * from t1 
+where a in 
+(
+select a 
+from t2 where b in (3,4)
+)
+;
+id	select_type	table	type	possible_keys	key	key_len	ref	rows	Extra
+1	PRIMARY	t1	ALL	NULL	NULL	NULL	NULL	6	
+1	PRIMARY	<subquery2>	eq_ref	distinct_key	distinct_key	4	func	1	
+2	MATERIALIZED	<derived4>	ALL	NULL	NULL	NULL	NULL	2	
+2	MATERIALIZED	t2	ALL	NULL	NULL	NULL	NULL	6	Using where; Using join buffer (flat, BNL join)
+4	DERIVED	NULL	NULL	NULL	NULL	NULL	NULL	NULL	No tables used
+explain select * from t1 
+where a in (
+select a from t2 
+where b in 
+(
+select * 
+from (values (3),(4)) as new_tvc)
+)
+;
+id	select_type	table	type	possible_keys	key	key_len	ref	rows	Extra
+1	PRIMARY	t1	ALL	NULL	NULL	NULL	NULL	6	
+1	PRIMARY	<subquery2>	eq_ref	distinct_key	distinct_key	4	func	1	
+2	MATERIALIZED	<derived4>	ALL	NULL	NULL	NULL	NULL	2	
+2	MATERIALIZED	t2	ALL	NULL	NULL	NULL	NULL	6	Using where; Using join buffer (flat, BNL join)
+4	DERIVED	NULL	NULL	NULL	NULL	NULL	NULL	NULL	No tables used
+explain format=json select * from t1 
+where a in 
+(
+select a 
+from t2 where b in (3,4)
+)
+;
+EXPLAIN
+{
+  "query_block": {
+    "select_id": 1,
+    "table": {
+      "table_name": "t1",
+      "access_type": "ALL",
+      "rows": 6,
+      "filtered": 100
+    },
+    "table": {
+      "table_name": "<subquery2>",
+      "access_type": "eq_ref",
+      "possible_keys": ["distinct_key"],
+      "key": "distinct_key",
+      "key_length": "4",
+      "used_key_parts": ["a"],
+      "ref": ["func"],
+      "rows": 1,
+      "filtered": 100,
+      "materialized": {
+        "unique": 1,
+        "query_block": {
+          "select_id": 2,
+          "table": {
+            "table_name": "<derived4>",
+            "access_type": "ALL",
+            "rows": 2,
+            "filtered": 100,
+            "materialized": {
+              "query_block": {
+                "union_result": {
+                  "table_name": "<unit4>",
+                  "access_type": "ALL",
+                  "query_specifications": [
+                    {
+                      "query_block": {
+                        "select_id": 4,
+                        "table": {
+                          "message": "No tables used"
+                        }
+                      }
+                    }
+                  ]
+                }
+              }
+            }
+          },
+          "block-nl-join": {
+            "table": {
+              "table_name": "t2",
+              "access_type": "ALL",
+              "rows": 6,
+              "filtered": 100
+            },
+            "buffer_type": "flat",
+            "buffer_size": "256Kb",
+            "join_type": "BNL",
+            "attached_condition": "t2.b = new_tvc.`3`"
+          }
+        }
+      }
+    }
+  }
+}
+drop table t1;
+set @@in_subquery_conversion_threshold= default;
diff --git a/mysql-test/t/opt_tvc.test b/mysql-test/t/opt_tvc.test
new file mode 100644
index 00000000000..1529672fc69
--- /dev/null
+++ b/mysql-test/t/opt_tvc.test
@@ -0,0 +1,124 @@
+create table t1 (a int, b int);
+
+insert into t1 
+values (1,2), (4,6), (9,7),
+       (1,1), (2,5), (7,8);
+
+create table t2 (a int, b int, c int);
+
+insert into t2 
+values (1,2,3), (5,1,2), (4,3,7), 
+       (8,9,0), (10,7,1), (5,5,1);
+
+--echo # optimization is not used
+
+let $query= select * from t1 where a in (1,2);
+eval $query;
+eval explain $query;
+eval explain format=json $query; 
+
+--echo # set minimum number of values in VALUEs list when optimization works to 2
+
+set @@in_subquery_conversion_threshold= 2;
+
+--echo # single IN-predicate in WHERE-part
+
+let $query= select * from t1 where a in (1,2);
+
+let $optimized_query= 
+select * from t1 
+where a in 
+      (
+        select * 
+        from (values (1),(2)) as new_tvc
+      );
+
+eval $query;
+eval $optimized_query;
+eval explain $query;
+eval explain $optimized_query;
+eval explain format=json $query; 
+
+--echo # AND-condition with IN-predicates in WHERE-part
+
+let $query= 
+select * from t1 
+where a in (1,2) and 
+      b in (1,5);
+
+let $optimized_query= 
+select * from t1 
+where a in 
+(
+  select * 
+  from (values (1),(2)) as new_tvc
+) 
+and b in
+(
+  select * 
+  from (values (1),(5)) as new_tvc
+);
+      
+eval $query;
+eval $optimized_query;
+eval explain $query;
+eval explain $optimized_query;
+eval explain format=json $query; 
+
+--echo # OR-condition with IN-predicates in WHERE-part
+
+let $query= 
+select * from t1 
+where a in (1,2) or 
+      b in (4,5);
+
+let $optimized_query= 
+select * from t1 
+where a in 
+(
+  select * 
+  from (values (1),(2)) as new_tvc
+) 
+or b in
+(
+  select * 
+  from (values (4),(5)) as new_tvc
+);
+      
+eval $query;
+eval $optimized_query;
+eval explain $query;
+eval explain $optimized_query;
+eval explain format=json $query; 
+
+--echo # subquery with IN-predicate
+
+let $query= 
+select * from t1 
+where a in 
+  (
+   select a 
+   from t2 where b in (3,4)
+  )
+;
+
+let $optimized_query= 
+select * from t1 
+where a in (
+  select a from t2 
+  where b in 
+  (
+    select * 
+    from (values (3),(4)) as new_tvc)
+  )
+;
+      
+eval $query;
+eval $optimized_query;
+eval explain $query;
+eval explain $optimized_query;
+eval explain format=json $query; 
+
+drop table t1;
+set @@in_subquery_conversion_threshold= default;
+