Differences between exists replacement in SQL

When using exists in SQL to replace in queries, note that when using exists, you must associate the primary query with the subquery. Otherwise, the query will not receive the following results:
Statement 1 use in to query:
Select realname from users where users. userid in
(Select gallery. galleries. creatorid from gallery. galleries group by gallery. galleries. creatorid
Having count (gallery. galleries. creatorid)> 1) Order by userid
Statement 2 Use exists to query:
Select realname from users where exists
(Select gallery. galleries. creatorid from gallery. galleries group by gallery. galleries. creatorid
Having count (gallery. galleries. creatorid)> 1) Order by userid
At first glance, there is no error, but Statement 2 forgets the association between the primary key of the primary query and the subquery, so that all the content of the primary query can be found out.
Statement 2 should be written correctly:
Select * From gallery. galleries as a where exists (select a. creatorid from gallery. galleries as B
Where a. creatorid = B. creatorid group by B. creatorid having count (B. creatorid)> 1) Order by creatorid
Note: exists is a Boolean operation. As long as the word query result set has a data result, the where condition is true, so the second query statement is equivalent
Select realname from users where 1 = 1 order by userid (if a piece of data is found in the subquery)