[GitHub] spark pull request: [SPARK-9301] [SQL] Add collect_set aggregate f...

[justinuang <gi...@git.apache.org>]

Github user justinuang commented on the pull request:

    https://github.com/apache/spark/pull/8592#issuecomment-152189895
  
    @rxin I would really like to see this merged in as well. I agree with @nburoojy that collect_as_list and collect_as_set will always have issues if the size of one list/set gets too big, but that doesn't mean that this isn't useful. I didn't quite follow the other parts about the API refactor, but if that isn't a huge issue, it would be nice to merge something like this in soon.
    
    If people are careful, this can be extremely useful, especially when it's impossible to write a UDAF that has a mergeValue and mergeCombiner function, or when you just want to restructure how the table is layed-out. In addition, until pyspark gets UDAFs, this will be a good substitute for most cases. Right now, to get around this in pyspark, I'm using hive's collect_list, but it's annoying because I have to register a temp table and use a SQL query instead of the dataframe API.


---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at infrastructure@apache.org or file a JIRA ticket
with INFRA.
---

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