An example that calls JOIN::make_simple_join() twice:

create table t5 (a int , b int) ;

create table t6 like t5 ;

create table t7 like t5 ;

insert into t5 values (0, 100), (1, 2), (1, 3), (2, 2), (2, 7),

                      (2, -1), (3, 10) ;

insert into t6 values (0, 0), (1, 1), (2, 1), (3, 1), (4, 1) ;

insert into t7 values (3, 3), (2, 2), (1, 1) ;

select count(distinct t5.b) as sum from t5, t6

                     where t5.a=t6.a and t6.b > 0 and t5.a <= 5

                     group by t5.a order by sum limit 1

This executes as follows:

1. create a tmp. table $TMP1 (without any keys)
2. run the join, put its output into $TMP1
3. create a temporary table $TMP2 (without any keys again)
4. use filesort() sort data from $TMP1 (produces an ordered stream)
5. read data from filesort result, do group-by-on-the-fly
and write into $TMP2.
// make_simple_join call.
6. use filesort(), read from $TMP2 and sort
7. read from filesort results and write to output.

MDEV-6995 has info about research done for EXPLAIN=JSON on this topic

Automated message:
----------------------------
Since this issue has not been updated since 6 weeks, it's time to move it back to Stalled.