[ https://issues.apache.org/jira/browse/CALCITE-5486?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17681485#comment-17681485 ]
Julian Hyde commented on CALCITE-5486: -------------------------------------- [~clesaec], SPARK-32726 seems to be about WHERE. HAVING is very a different case. Calcite doesn't support aliases in HAVING by default, but it can be enabled via a flag, and when it does, it should apply to sub-queries as well as top-level queries. > Subquery not support HAVING alias in condition > ---------------------------------------------- > > Key: CALCITE-5486 > URL: https://issues.apache.org/jira/browse/CALCITE-5486 > Project: Calcite > Issue Type: Improvement > Components: core > Reporter: Jiajun Xie > Priority: Major > > Spark can allow HAVING alias in condition. > We also should allow HAVING alias in condition, here is the simple case. > {code:java} > sql("select * from emp where sal >\n" > + " (select avg(sal) as s" > + " from emp having ^s^ > 0" > + " )") > .withConformance(SqlConformanceEnum.LENIENT) > .fails("Column 'S' not found in any table"); // should be ok, but now is > fails{code} -- This message was sent by Atlassian Jira (v8.20.10#820010)