Can't find column when uses the clustered index

[wjhuang2016]

Bug Report

Please answer these questions before submitting your issue. Thanks!

1. Minimal reproduce step (Required)

CREATE TABLE `tbl_1` (
  `col_5` bigint(20) unsigned DEFAULT NULL,
  `col_6` double NOT NULL,
  `col_7` tinyint(4) NOT NULL DEFAULT '-47',
  `col_8` smallint(5) unsigned NOT NULL,
  `col_9` date DEFAULT NULL,
  PRIMARY KEY (`col_8`,`col_6`,`col_7`) /*T![clustered_index] CLUSTERED */,
  KEY `idx_5` (`col_9`,`col_8`),
  KEY `idx_6` (`col_6`)
) ENGINE=InnoDB DEFAULT CHARSET=utf8mb4 COLLATE=utf8mb4_bin
PARTITION BY RANGE ( `col_7` ) (
  PARTITION `p0` VALUES LESS THAN (-87),
  PARTITION `p1` VALUES LESS THAN (-44),
  PARTITION `p2` VALUES LESS THAN (46),
  PARTITION `p3` VALUES LESS THAN (118),
  PARTITION `p4` VALUES LESS THAN (127)
);
desc select  /*+ use_index_merge( tbl_1 ) */ col_5 from tbl_1 where tbl_1.col_6 >= 9637.317973853435 or tbl_1.col_9 in ( '1975-07-15' , '1999-11-30' ) and tbl_1.col_8 = 6073 and tbl_1.col_5 in ( 6390338814472142299 ) or tbl_1.col_8 <= 17132 or tbl_1.col_6 <> 4755.671685443747 or tbl_1.col_8 <= 39946 and tbl_1.col_6 in ( 6283.76957320476 , 6046.095273053335 , 9637.317973853435 , 7405.304278203997 , 3924.166154298398 , 7651.113042850394 , 4618.630644850198 , 5969.08193059866 , 6283.76957320476 , 6283.76957320476 ) and tbl_1.col_6 in ( 7150.103012479594 , 8579.144890325371 , 9495.762329500238 ) and tbl_1.col_9 <> '1999-11-30';

2. What did you expect to see? (Required)

No error

3. What did you see instead (Required)

ERROR 1105 (HY000): Can't find column clustered_index_go_c_1617801763.tbl_1.col_7 in schema Column: [clustered_index_go_c_1617801763.tbl_1.col_5,clustered_index_go_c_1617801763.tbl_1.col_6,clustered_index_go_c_1617801763.tbl_1.col_8,clustered_index_go_c_1617801763.tbl_1.col_9] Unique key: []

4. What is your TiDB version? (Required)

master

[tangenta]

Minimal reproducible case:

drop table if exists t;
create table t (a int, b int, c int, primary key (a, b) clustered, key idx_c(c));
select /*+ use_index_merge(t) */ c from t where c > 10 or a < 1;

This is introduced in #22940 (support pruning common handle columns).

[eurekaka]

Another reproduction:

MySQL [test]> create table t (
    ->   c1 float, c2 int, c3 int,
    ->   primary key (c1) /*T![clustered_index] CLUSTERED */,
    ->   key idx_1 (c2), key idx_2 (c3)
    -> );
Query OK, 0 rows affected (0.02 sec)

MySQL [test]> desc select /*+ use_index_merge(t) */ c3 from t where c3 = 1 or c2 = 1;
ERROR 1105 (HY000): Can't find column test.t.c1 in schema Column: [test.t.c2,test.t.c3] Unique key: []

[ti-srebot]

Bug

Note: Make Sure that 'component', and 'severity' labels are added
Example for how to fill out the template: #20100

5. Affected versions

[v5.0.0:v5.0.0]

6. Fixed versions

v5.0.1