[
https://issues.apache.org/jira/browse/SPARK-27265?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16800389#comment-16800389
]
Michael Reynolds commented on SPARK-27265:
------------------------------------------
PR is here https://github.com/apache/spark/pull/24199
> Provide convenience accessors for values by Column Name
> -------------------------------------------------------
>
> Key: SPARK-27265
> URL: https://issues.apache.org/jira/browse/SPARK-27265
> Project: Spark
> Issue Type: Improvement
> Components: SQL
> Affects Versions: 2.4.0
> Reporter: Michael Reynolds
> Priority: Minor
>
> There are convenient methods for accessing SQL primitive values by column
> index such as *getDouble*, *getBoolean*, *getTimestamp*, etc.
>
> However, when doing pivot operations sometimes it is easier to code it
> against column names instead of column indexes. When doing this, there is no
> way to conveniently access the values which leads to really annoying
> boilerplate code such as:
> {code:scala}
> desc.flatMap(row => {
> val metric = row.getAs[String](columns.head)
> columns.tail.map(columnName => (metric, columnName,
> row.getAs[String].(columnName).toDouble))
> }).toDF("metric", "field", "value")
> {code}
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscr...@spark.apache.org
For additional commands, e-mail: issues-h...@spark.apache.org
