[
https://issues.apache.org/jira/browse/CALCITE-5798?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17739765#comment-17739765
]
Runkang He edited comment on CALCITE-5798 at 7/4/23 1:32 AM:
-------------------------------------------------------------
[~nobigo] I agree with you. One thing to add is that the PR extends support for
'(x < y) IS NOT TRUE', when x and y are nullable. Before this PR, this is not
supported. And '(1 < x) IS NOT TRUE' is a special case of '(x < y) IS NOT
TRUE', so maybe we should name the title to "Improve simplification of '(x < y)
IS NOT TRUE' when x and y are not nullable".
was (Author: JIRAUSER280488):
[~nobigo] I agree with you. One thing to add is that the PR extends support for
'(x < y) is not true', when x and y are nullable. Before this PR, this is not
supported. And '(1 < x) is not true' is a special case of '(x < y) is not
true', so maybe we should name the title to "Improve simplification of '(x < y)
IS NOT TRUE' when x and y are not nullable".
> Improve simplification of '(x < y) IS NOT TRUE' and '(x < y) IS TRUE' when x
> and y are not nullable
> ---------------------------------------------------------------------------------------------------
>
> Key: CALCITE-5798
> URL: https://issues.apache.org/jira/browse/CALCITE-5798
> Project: Calcite
> Issue Type: Improvement
> Components: core
> Affects Versions: 1.34.0
> Reporter: Runkang He
> Assignee: Runkang He
> Priority: Major
> Labels: pull-request-available
>
> Currently '(1 < x) IS NOT TRUE' when pulled up predicates contain 'x IS NOT
> NULL' is not simplified, and keeps it as it is.
>
> It can and should be simplified to '1 >= x', which is more concise and useful
> for further simplification, such as CALCITE-5780
>
> Note that this Jira is splited from CALCITE-5780, which makes this part of
> codes more clear, and follows the principle that a Jira only discusses one
> issue.
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
