correlated subquery in the SELECT clause will raise an error

[tankilo]

• dble version：
• preconditions ：
  no
• configs：

schema.xml

<table name="TK_TEST" dataNode="dn1,dn2" rule="sharding-by-mod2" primaryKey="SUBS_ID"/>

rule.xml

server.xml

• steps：
  step1. create table

CREATE TABLE `TK_TEST` (
  `SUBS_ID` bigint(19) NOT NULL,
  `SENDING_DATE` datetime DEFAULT NULL,
  PRIMARY KEY (SUBS_ID)
);

step2. query

SELECT COUNT(*) AS TEMP2,
       (SELECT COUNT(*)
          FROM TK_TEST S
         WHERE MONTH(s.SENDING_DATE) = MONTH(t.SENDING_DATE)
       ) AS TEMP1
  FROM TK_TEST t

• expect result：
  1.
• real result：

mysql> SELECT COUNT(*) AS TEMP2,
    -> (SELECT COUNT(*)
    -> FROM TK_TEST S
    -> WHERE MONTH(s.SENDING_DATE) = MONTH(t.SENDING_DATE)
    -> ) AS TEMP1
    -> FROM TK_TEST t;
ERROR 4004 (42S22): column t.SENDING_DATE not found

• supplements：
  1.

/label ~BUG

[tankilo]

It seems that Correlated SubQuery is expected to be unsupported.

Code only deal with scenes of WHERE clause and HAVING clause, but missing SELECT clause.

[tankilo]

I may post an PR to fix it.