[jira] [Updated] (SPARK-9488) pyspark.sql.types.Row very slow when using named arguments

["Alexis Benoist (JIRA)" <ji...@apache.org>]

     [ https://issues.apache.org/jira/browse/SPARK-9488?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Alexis Benoist updated SPARK-9488:
----------------------------------
    Summary: pyspark.sql.types.Row very slow when using named arguments  (was: pyspark.sql.types.Row very slow when used named arguments)

> pyspark.sql.types.Row very slow when using named arguments
> ----------------------------------------------------------
>
>                 Key: SPARK-9488
>                 URL: https://issues.apache.org/jira/browse/SPARK-9488
>             Project: Spark
>          Issue Type: Bug
>          Components: PySpark
>    Affects Versions: 1.4.0
>         Environment: 
>            Reporter: Alexis Benoist
>              Labels: performance
>
> We can see that the implementation of the Row is accessing items in O(n).
> https://github.com/apache/spark/blob/master/python/pyspark/sql/types.py#L1217
> We could use an OrderedDict instead of a tuple to make the access time in O(1). Can the keys be of an unhashable type?
> I'm ok to do the edit.
> Cheers,
> Alexis.



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

---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscribe@spark.apache.org
For additional commands, e-mail: issues-help@spark.apache.org