[ 
https://issues.apache.org/jira/browse/PHOENIX-2827?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15537155#comment-15537155
 ] 

ASF GitHub Bot commented on PHOENIX-2827:
-----------------------------------------

Github user lomoree commented on a diff in the pull request:

    https://github.com/apache/phoenix/pull/213#discussion_r81422631
  
    --- Diff: 
phoenix-core/src/main/java/org/apache/phoenix/calcite/rules/PhoenixConverterRules.java
 ---
    @@ -232,8 +232,7 @@ public RelNode convert(RelNode rel) {
             private static Predicate<LogicalSort> HAS_FETCH = new 
Predicate<LogicalSort>() {
                 @Override
                 public boolean apply(LogicalSort input) {
    -                return input.offset == null 
    -                        && input.fetch != null;
    +                return input.fetch != null;
    --- End diff --
    
    With respect to my comment below, if there is a limit, then we apply the 
rule, otherwise we do not (regardless of offset).
     I may have the wrong idea here, but based on my understanding of offset, 
all of the test cases pass as expected. Are the test cases I have laid out in 
CalciteIT what you would expect?


> Support OFFSET in Calcite-Phoenix
> ---------------------------------
>
>                 Key: PHOENIX-2827
>                 URL: https://issues.apache.org/jira/browse/PHOENIX-2827
>             Project: Phoenix
>          Issue Type: Task
>            Reporter: Maryann Xue
>            Assignee: Eric Lomore
>              Labels: calcite
>




--
This message was sent by Atlassian JIRA
(v6.3.4#6332)

Reply via email to