[ https://issues.apache.org/jira/browse/CALCITE-5547?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17694180#comment-17694180 ]
Julian Hyde commented on CALCITE-5547: -------------------------------------- Here's a more straightforward use case: {noformat} > !connect jdbc:calcite:model=core/src/test/resources/hsqldb-model.json sa sa > select 2 as two, * from emp natural join dept; +---------+-----+-------+--------+-----------+------+------------+---------+---------+------------+ | COMM | TWO | EMPNO | ENAME | JOB | MGR | HIREDATE | SAL | DEPTNO0 | DNAME | +---------+-----+-------+--------+-----------+------+------------+---------+---------+------------+ | 20.00 | 2 | 7369 | SMITH | CLERK | 7902 | 1980-12-17 | 800.00 | 20 | RESEARCH | | 300.00 | 2 | 7499 | ALLEN | SALESMAN | 7698 | 1981-02-20 | 1600.00 | 30 | SALES | | 500.00 | 2 | 7521 | WARD | SALESMAN | 7698 | 1981-02-22 | 1250.00 | 30 | SALES | | 20.00 | 2 | 7566 | JONES | MANAGER | 7839 | 1981-02-04 | 2975.00 | 20 | RESEARCH | | 1400.00 | 2 | 7654 | MARTIN | SALESMAN | 7698 | 1981-09-28 | 1250.00 | 30 | SALES | | 30.00 | 2 | 7698 | BLAKE | MANAGER | 7839 | 1981-01-05 | 2850.00 | 30 | SALES | | 10.00 | 2 | 7782 | CLARK | MANAGER | 7839 | 1981-06-09 | 2450.00 | 10 | ACCOUNTING | | 20.00 | 2 | 7788 | SCOTT | ANALYST | 7566 | 1987-04-19 | 3000.00 | 20 | RESEARCH | | 10.00 | 2 | 7839 | KING | PRESIDENT | null | 1981-11-17 | 5000.00 | 10 | ACCOUNTING | | 0.00 | 2 | 7844 | TURNER | SALESMAN | 7698 | 1981-09-08 | 1500.00 | 30 | SALES | | 20.00 | 2 | 7876 | ADAMS | CLERK | 7788 | 1987-05-23 | 1100.00 | 20 | RESEARCH | | 30.00 | 2 | 7900 | JAMES | CLERK | 7698 | 1981-12-03 | 950.00 | 30 | SALES | | 20.00 | 2 | 7902 | FORD | ANALYST | 7566 | 1981-12-03 | 3000.00 | 20 | RESEARCH | | 10.00 | 2 | 7934 | MILLER | CLERK | 7782 | 1982-01-23 | 1300.00 | 10 | ACCOUNTING | +---------+-----+-------+--------+-----------+------+------------+---------+---------+------------+ 14 rows selected (0.082 seconds) > select *, 2 as two from emp natural join dept; +--------+-------+--------+-----------+------+------------+---------+---------+------------+----------+-----+ | DEPTNO | EMPNO | ENAME | JOB | MGR | HIREDATE | SAL | COMM | DNAME | LOC | TWO | +--------+-------+--------+-----------+------+------------+---------+---------+------------+----------+-----+ | 20 | 7369 | SMITH | CLERK | 7902 | 1980-12-17 | 800.00 | null | RESEARCH | DALLAS | 2 | | 30 | 7499 | ALLEN | SALESMAN | 7698 | 1981-02-20 | 1600.00 | 300.00 | SALES | CHICAGO | 2 | | 30 | 7521 | WARD | SALESMAN | 7698 | 1981-02-22 | 1250.00 | 500.00 | SALES | CHICAGO | 2 | | 20 | 7566 | JONES | MANAGER | 7839 | 1981-02-04 | 2975.00 | null | RESEARCH | DALLAS | 2 | | 30 | 7654 | MARTIN | SALESMAN | 7698 | 1981-09-28 | 1250.00 | 1400.00 | SALES | CHICAGO | 2 | | 30 | 7698 | BLAKE | MANAGER | 7839 | 1981-01-05 | 2850.00 | null | SALES | CHICAGO | 2 | | 10 | 7782 | CLARK | MANAGER | 7839 | 1981-06-09 | 2450.00 | null | ACCOUNTING | NEW YORK | 2 | | 20 | 7788 | SCOTT | ANALYST | 7566 | 1987-04-19 | 3000.00 | null | RESEARCH | DALLAS | 2 | | 10 | 7839 | KING | PRESIDENT | null | 1981-11-17 | 5000.00 | null | ACCOUNTING | NEW YORK | 2 | | 30 | 7844 | TURNER | SALESMAN | 7698 | 1981-09-08 | 1500.00 | 0.00 | SALES | CHICAGO | 2 | | 20 | 7876 | ADAMS | CLERK | 7788 | 1987-05-23 | 1100.00 | null | RESEARCH | DALLAS | 2 | | 30 | 7900 | JAMES | CLERK | 7698 | 1981-12-03 | 950.00 | null | SALES | CHICAGO | 2 | | 20 | 7902 | FORD | ANALYST | 7566 | 1981-12-03 | 3000.00 | null | RESEARCH | DALLAS | 2 | | 10 | 7934 | MILLER | CLERK | 7782 | 1982-01-23 | 1300.00 | null | ACCOUNTING | NEW YORK | 2 | +--------+-------+--------+-----------+------+------------+---------+---------+------------+----------+-----+ {noformat} The second query has the correct number of columns. In the first query, the names and orderings of columns are messed up. I suspect that the expansion of "*" has clashed with the machinery trying to move the DEPTNO column to the front and make it unique. > Join using returns incorrect column names > ----------------------------------------- > > Key: CALCITE-5547 > URL: https://issues.apache.org/jira/browse/CALCITE-5547 > Project: Calcite > Issue Type: Bug > Components: core > Affects Versions: 1.33.0 > Reporter: Sean Broeder > Priority: Major > > The query SELECT '' as \"xxx\", * > FROM (values (1, 4, 'one')) as t1(i ,j ,t ) > LEFT JOIN (values (1, -1)) as t2(i ,k ) > USING (i) > returns columns (xxx, i, j, i0), > but should return columns (xxx, i, j, t, k) -- This message was sent by Atlassian Jira (v8.20.10#820010)