[
https://issues.apache.org/jira/browse/SPARK-50133?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Xinrong Meng updated SPARK-50133:
---------------------------------
Description:
We want to implement a function df.table_argument() (the name can be discussed
further), which returns a Column object. This Column can then be passed as
input to call a User-Defined Table Function (UDTF). For example:
{{@udtf(returnType="a: int")}}
{{class TestUDTF:}}
{{ def eval(self, row: Row):}}
{{ if row[0] > 5:}}
{{ yield row[0]}}
{{TestUDTF(df.table_agument())}}
The df.table_argument() function might also take additional parameters, such as:
* partition_by: Specifies how the data should be partitioned for processing.
* order_by: Defines the order within each partition.
> Support conversion to table argument
> ------------------------------------
>
> Key: SPARK-50133
> URL: https://issues.apache.org/jira/browse/SPARK-50133
> Project: Spark
> Issue Type: Sub-task
> Components: PySpark, SQL
> Affects Versions: 4.0.0
> Reporter: Takuya Ueshin
> Priority: Major
>
> We want to implement a function df.table_argument() (the name can be
> discussed further), which returns a Column object. This Column can then be
> passed as input to call a User-Defined Table Function (UDTF). For example:
>
> {{@udtf(returnType="a: int")}}
> {{class TestUDTF:}}
> {{ def eval(self, row: Row):}}
> {{ if row[0] > 5:}}
> {{ yield row[0]}}
> {{TestUDTF(df.table_agument())}}
>
> The df.table_argument() function might also take additional parameters, such
> as:
> * partition_by: Specifies how the data should be partitioned for processing.
> * order_by: Defines the order within each partition.
>
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
