[
https://issues.apache.org/jira/browse/SPARK-4551?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14632997#comment-14632997
]
Reynold Xin commented on SPARK-4551:
------------------------------------
[~huangjs] while this might be useful in a small number of cases, I think it
complicates the design more and could also lead to more confusing behavior for
users (imagine somebody asking on the mailing list "why are you changing my
column names?").
Unless there is a compelling enough use case for it, it's best not to do this
one.
(I'm going to close it for now. We can always reopen if new evidence/argument
surface.)
> Allow auto-conversion of field names of case class from camelCase to
> lower_case convention
> ------------------------------------------------------------------------------------------
>
> Key: SPARK-4551
> URL: https://issues.apache.org/jira/browse/SPARK-4551
> Project: Spark
> Issue Type: Improvement
> Components: SQL
> Affects Versions: 1.3.0
> Reporter: Jianshi Huang
> Priority: Minor
>
> In SQL, lower_case is the naming convention for column names.
> We want to keep this convention as much as possible. So when converting a
> RDD[case class] to SchemaRDD, let's make it an option to convert the filed
> names from camelCase to lower_case (e.g. unitAmount => unit_amount)
> My proposal:
> 1) Utility function for case conversion (e.g.
> https://gist.github.com/sidharthkuruvila/3154845)'
> 2) Add option convertToLowerCase to registerTempTable
> def registerTempTable(tableName: String, convertToLowerCase = false)
> Jianshi
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
