[
https://issues.apache.org/jira/browse/CALCITE-987?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15024652#comment-15024652
]
Pengcheng Xiong commented on CALCITE-987:
-----------------------------------------
[~jcamachorodriguez], thanks for your comments. i partially agree with your
opinion. But please also note that we had a similar discussion before "whether
to use 0 or 1 for explain when a hive operator outputs 0 rows". We followed
what Postgres does, i.e., validate it to 1. So, if this "getRowCount" is used
externally, e.g., in explain, i would prefer to validate it to 1. Else if it is
used internally, e.g., in pushing limit through union/join, i would prefer not
to validate it. Thus, i would like to have two functions, i.e., what i did in
the patch. Let's see [~julianhyde]'s opinion on this. :)
> Push limit 0 will result in an infinite loop
> --------------------------------------------
>
> Key: CALCITE-987
> URL: https://issues.apache.org/jira/browse/CALCITE-987
> Project: Calcite
> Issue Type: Bug
> Reporter: Pengcheng Xiong
> Assignee: Pengcheng Xiong
> Attachments: CALCITE-987.01.patch
>
>
> We use "checkInputForCollationAndLimit" in RelMdUtil.java to check the input
> #rows. However, it calls RelMetadataQuery.getRowCount which will validate the
> #rows. The validation will change #row=0 to #row=1. This will result in an
> infinite loop to push limit. The affected rules include
> SortUnionTransposeRule and any Sort***TransposeRules that call
> checkInputForCollationAndLimit.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
