[
https://issues.apache.org/jira/browse/CALCITE-6161?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17794898#comment-17794898
]
Mingcan Wang commented on CALCITE-6161:
---------------------------------------
[~julianhyde] Yes, you're right. I just forgot the case of unsolved function
compared to resolved function. They can just compare names.But wouldn’t that be
a bit subjective?
I actually saw the comment above my change. I may not have fully understood
the meaning of unresovled at that time.But just comparing names does lead to
incorrectness in some cases.
I agree It should check the operator itself. I'll think about it again
> The equalsDeep of sqlCall should compare sqlOperator's sqlKind
> --------------------------------------------------------------
>
> Key: CALCITE-6161
> URL: https://issues.apache.org/jira/browse/CALCITE-6161
> Project: Calcite
> Issue Type: Bug
> Components: core
> Affects Versions: 1.36.0
> Reporter: Mingcan Wang
> Assignee: Mingcan Wang
> Priority: Minor
> Labels: pull-request-available
> Fix For: 1.37.0
>
>
> Here is the situation:
> when I create two SqlBasicCalls, one uses UNARY_MINUS operator and the other
> uses MINUS operator. But their operandList is the same. However, The result
> after calling their equalsDeep() method is true.
> I found that the sqlKind of many operators is reused, and the name is not
> guaranteed to be unique. Only the combination of kind and name can determine
> the unique operator.
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
