[
https://issues.apache.org/jira/browse/SPARK-22478?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16275985#comment-16275985
]
Sean Owen commented on SPARK-22478:
-----------------------------------
Yeah the issue is that there are a hundred useful functions you could build
into Spark. Which ones do you choose? The ones that are there exist for Hive
and SQL compatibility.
One point of Spark is mixing SQL and code more easily. UDFs are really easy,
and language functions are often actually harder to use. So this isn't often
helping users over just writing simple UDFs that do exactly what the user
wants.
> Spark - Truncate date by Day / Hour
> ------------------------------------
>
> Key: SPARK-22478
> URL: https://issues.apache.org/jira/browse/SPARK-22478
> Project: Spark
> Issue Type: New Feature
> Components: Spark Core
> Affects Versions: 2.2.0
> Reporter: david hodeffi
> Priority: Trivial
>
> Currently it is possible to truncate date/timestamp only by month or year,
> There is a missing functionality to truncate by hour/day.
> A usecase that requires this functionality is: Aggregating financial
> transaction of a party in order to create daily/hour summary.
> Those summaries are required for machine learning algorithms.
--
This message was sent by Atlassian JIRA
(v6.4.14#64029)
---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscr...@spark.apache.org
For additional commands, e-mail: issues-h...@spark.apache.org
