[ https://issues.apache.org/jira/browse/CALCITE-2440?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16585420#comment-16585420 ]
Shuyi Chen commented on CALCITE-2440: ------------------------------------- [~vvysotskyi], thanks a lot. There are 52 rules together, it's hard to identify the set. After running the volcanoplanner, the relsubset.best is null and bestCost is {inf}. Do you have any idea on what might be the cause so I can narrow down the search space? > SQL with UNNEST fail in volcano planner in some cases > ----------------------------------------------------- > > Key: CALCITE-2440 > URL: https://issues.apache.org/jira/browse/CALCITE-2440 > Project: Calcite > Issue Type: Bug > Components: core > Affects Versions: 1.17.0 > Reporter: Shuyi Chen > Assignee: Julian Hyde > Priority: Major > > After [CALCITE-2291|https://issues.apache.org/jira/browse/CALCITE-2291], > specifically, [this > line|https://github.com/apache/calcite/blob/7088dc7261d294b7c6d5c9f4463435e95f0b9c15/core/src/main/java/org/apache/calcite/sql2rel/SqlToRelConverter.java#L1952] > causes volcano planner to fail to find the best plan for query > {code:java} > SELECT a, s FROM T, UNNEST(T.c) as A (s) > {code} > . I'll try to write a unittest in Calcite to reproduce. [~cshi] [~vvysotskyi] > [~HanumathRao], can you provide some context? Thanks a lot. -- This message was sent by Atlassian JIRA (v7.6.3#76005)