spark-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Nicholas Chammas (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (SPARK-7548) Add explode expression
Date Tue, 12 May 2015 00:12:00 GMT

    [ https://issues.apache.org/jira/browse/SPARK-7548?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14538937#comment-14538937
] 

Nicholas Chammas commented on SPARK-7548:
-----------------------------------------

To provide a motivating example for the record, consider a DataFrame that looks like this:

{code}
>>> a = {
    'test_name': 'abracadabra',
    'results': [{'time': 14.7}, {'time': 22.3}]
}
>>> df = sqlContext.jsonRDD(sc.parallelize([json.dumps(a)]))
>>> df.printSchema()
root
 |-- results: array (nullable = true)
 |    |-- element: struct (containsNull = true)
 |    |    |-- time: double (nullable = true)
 |-- test_name: string (nullable = true)

>>> print df.select('results.time').collect()
[Row(time=[14.7, 22.3])]
{code}

It is currently not possible to aggregate over the nested {{time}} field as follows:

{code}
df.groupBy('test_name').avg('results.time')
{code}

An alternative to supporting this kind of aggregation would be to offer some way to "promote"
the nested column to a top-level column. Hence, {{explode()}}.

> Add explode expression
> ----------------------
>
>                 Key: SPARK-7548
>                 URL: https://issues.apache.org/jira/browse/SPARK-7548
>             Project: Spark
>          Issue Type: Sub-task
>          Components: SQL
>            Reporter: Reynold Xin
>            Assignee: Michael Armbrust
>            Priority: Blocker
>




--
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


Mime
View raw message