This is to work around what is a limitation in the JVM's implementation of a hash map entry spliterator which results from the stream grouping collector. As such, it would be very reasonable for a future JVM implementation to change its grouping collector to support a more robust spliterator on the grouped results, or for some other grouping collector implementation to come along that does the same thing (or a different map constructor to be used in the grouping collector definition). As such, this is just a temporary workaround until a better group collector solution can be found.
[ Full content available at: https://github.com/apache/incubator-druid/pull/5913 ] This message was relayed via gitbox.apache.org for [email protected]
