[
https://issues.apache.org/jira/browse/CALCITE-5376?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17635458#comment-17635458
]
Sophie edited comment on CALCITE-5376 at 11/17/22 5:14 PM:
-----------------------------------------------------------
Thanks for this.
Just to make sure I understand, would the idea be to do something similar to
defining the function to be something along the lines of this, given that it
would be an aggregate function?
{code:java}
@LibraryOperator(libraries = {SPARK})
public static final SqlAggFunction FIRST =
new SqlFirstLastValueAggFunction(SqlKind.FIRST_VALUE);{code}
But probably not with SqlFirstLastValueAggFunction, but the equivalent
And the validator change might be around SqlValidatorImpl.validateCall where
there's a check with SqlFunctionCategory.MATCH_RECOGNIZE?
Is there a better place for testing this than BabelQuidemTest?
was (Author: JIRAUSER298340):
Thanks for this.
Just to make sure I understand, would the idea be to do something similar to
defining the function to be something along the lines of this, given that it
would be an aggregate function?
{code:java}
@LibraryOperator(libraries = {SPARK})
public static final SqlAggFunction FIRST =
new SparkSqlFirstLastValueAggFunction(SqlKind.FIRST_VALUE);{code}
And the validator change might be around SqlValidatorImpl.validateCall where
there's a check with SqlFunctionCategory.MATCH_RECOGNIZE?
Is there a better place for testing this than BabelQuidemTest?
> Support Spark’s FIRST and LAST functions
> ----------------------------------------
>
> Key: CALCITE-5376
> URL: https://issues.apache.org/jira/browse/CALCITE-5376
> Project: Calcite
> Issue Type: Task
> Reporter: Sophie
> Priority: Major
>
> Following on from the mailing list thread -
> [https://lists.apache.org/thread/3rk8zo5r38c6gpj7j0oy0453cxwlmnm7]
>
> Spark has the following function definition for FIRST and LAST
> [https://spark.apache.org/docs/3.0.1/api/sql/index.html#first]
> [https://spark.apache.org/docs/3.0.1/api/sql/index.html#last]
>
> At the moment trying to use these functions with a query like
> {code:java}
> select first(age) from customers{code}
> gives an error along the lines of
> {code:java}
> Function 'FIRST(`customers`.`age`, 0)' can only be used in
> MATCH_RECOGNIZE{code}
>
> In terms of testing, I can reproduce the error using BabelQuidemTest and
> setting up a case for spark.
>
> Given that Spark’s first function is very similar to the FIRST_VALUE
> function, I think SqlFirstLastValueAggFunction is probably a useful starting
> point, but some pointers about where to look would be helpful. Particularly
> around potential parser changes.
>
> Thanks in advance
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
