[
https://issues.apache.org/jira/browse/CALCITE-3658?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17530643#comment-17530643
]
Jiajun Xie edited comment on CALCITE-3658 at 5/2/22 8:46 AM:
-------------------------------------------------------------
This solution does not solve all the problems, such as the following two queries
{code:java}
update emp a set ename = (select ^ename^ from empdefaults b where
a.empno=b.empno)
// sourceExpressionList is $0 that is EMPNO
LogicalTableModify(table=[[CATALOG, SALES, EMP]], operation=[UPDATE],
updateColumnList=[[ENAME]], sourceExpressionList=[[^$0^]], flattened=[true])
LogicalProject(EMPNO=[$0], ENAME=[$1], JOB=[$2], MGR=[$3], HIREDATE=[$4],
SAL=[$5], COMM=[$6], DEPTNO=[$7], SLACKER=[$8], EXPR$0=[$9])
LogicalProject(EMPNO=[$0], ENAME=[$1], JOB=[$2], MGR=[$3], HIREDATE=[$4],
SAL=[$5], COMM=[$6], DEPTNO=[$7], SLACKER=[$8], ENAME9=[$10])
LogicalJoin(condition=[=($0, $9)], joinType=[left])
LogicalTableScan(table=[[CATALOG, SALES, EMP]])
LogicalTableScan(table=[[CATALOG, SALES, EMPDEFAULTS]]){code}
{code:java}
update emp a set job = (select ^job^ from empdefaults b where a.empno=b.empno)
// sourceExpressionList is $0 that is EMPNO
LogicalTableModify(table=[[CATALOG, SALES, EMP]], operation=[UPDATE],
updateColumnList=[[JOB]], sourceExpressionList=[[$0]], flattened=[true])
LogicalProject(EMPNO=[$0], ENAME=[$1], JOB=[$2], MGR=[$3], HIREDATE=[$4],
SAL=[$5], COMM=[$6], DEPTNO=[$7], SLACKER=[$8], EXPR$0=[$9])
LogicalProject(EMPNO=[$0], ENAME=[$1], JOB=[$2], MGR=[$3], HIREDATE=[$4],
SAL=[$5], COMM=[$6], DEPTNO=[$7], SLACKER=[$8], JOB9=[$11])
LogicalJoin(condition=[=($0, $9)], joinType=[left])
LogicalTableScan(table=[[CATALOG, SALES, EMP]])
LogicalTableScan(table=[[CATALOG, SALES, EMPDEFAULTS]]){code}
When the field is not in the updated target table, the sourceExpressionList
always is $0. I will fix it in
https://issues.apache.org/jira/browse/CALCITE-5122
was (Author: jiajunbernoulli):
This solution does not solve all the problems, such as the following two queries
{code:java}
update emp a set ename = (select ^ename^ from empdefaults b where
a.empno=b.empno)
// sourceExpressionList is $0 that is EMPNO
LogicalTableModify(table=[[CATALOG, SALES, EMP]], operation=[UPDATE],
updateColumnList=[[ENAME]], sourceExpressionList=[[^$0^]], flattened=[true])
LogicalProject(EMPNO=[$0], ENAME=[$1], JOB=[$2], MGR=[$3], HIREDATE=[$4],
SAL=[$5], COMM=[$6], DEPTNO=[$7], SLACKER=[$8], EXPR$0=[$9])
LogicalProject(EMPNO=[$0], ENAME=[$1], JOB=[$2], MGR=[$3], HIREDATE=[$4],
SAL=[$5], COMM=[$6], DEPTNO=[$7], SLACKER=[$8], ENAME9=[$10])
LogicalJoin(condition=[=($0, $9)], joinType=[left])
LogicalTableScan(table=[[CATALOG, SALES, EMP]])
LogicalTableScan(table=[[CATALOG, SALES, EMPDEFAULTS]]){code}
{code:java}
update emp a set job = (select ^job^ from empdefaults b where a.empno=b.empno)
{code}
> TableModify of Update contains correlated variable by mistake.
> --------------------------------------------------------------
>
> Key: CALCITE-3658
> URL: https://issues.apache.org/jira/browse/CALCITE-3658
> Project: Calcite
> Issue Type: Improvement
> Reporter: Jin Xing
> Priority: Major
> Labels: pull-request-available
> Fix For: 1.22.0
>
> Time Spent: 0.5h
> Remaining Estimate: 0h
>
> UPDATE clause like below
> {code:java}
> update emp set empno = empno + 1
> {code}
> will be converted to
> {code:java}
> LogicalTableModify(table=[[CATALOG, SALES, EMP]], operation=[UPDATE],
> updateColumnList=[[EMPNO]], sourceExpressionList=[[+($cor0.EMPNO, 1)]],
> flattened=[true])
> LogicalProject(EMPNO=[$0], ENAME=[$1], JOB=[$2], MGR=[$3], HIREDATE=[$4],
> SAL=[$5], COMM=[$6], DEPTNO=[$7], SLACKER=[$8], EXPR$0=[+($0, 1)])
> LogicalTableScan(table=[[CATALOG, SALES, EMP]])
> {code}
> However the correlated variables $cor0.EMPNO should not exist in
> *sourceExpressionList*.
> It also brings trouble when convert TableModify back to Sql string.
--
This message was sent by Atlassian Jira
(v8.20.7#820007)
