piccolbo/dplyr.spark.hive

case support

Closed this issue · 1 comments

I initially worked under the assumption backend was case insensitive, which should be the default for a hivecontext, but is configurable. I made the decision to lowercase names that the user enters. Then I found the csv parser creates colnames with uppercases, breaking tests. Found related bug and added report

https://issues.apache.org/jira/browse/SPARK-10754

Right now one users are advised to use only lowercase.

to be continued on rzilla fork