Github user kumarvishal09 commented on a diff in the pull request:
https://github.com/apache/carbondata/pull/1830#discussion_r162298090
--- Diff:
integration/spark2/src/main/scala/org/apache/spark/sql/execution/strategy/CarbonLateDecodeStrategy.scala
---
@@ -612,6 +612,10 @@ private[sql] class CarbonLateDecodeStrategy extends
SparkStrategy {
CastExpressionOptimization.checkIfCastCanBeRemove(c)
case StartsWith(a: Attribute, Literal(v, t)) =>
Some(sources.StringStartsWith(a.name, v.toString))
+ case c@EndsWith(a: Attribute, Literal(v, t)) =>
--- End diff --
I think we need to handle Like case also when filter is [column like '%_a']
expression will be like expression so we need to pushdown the same
---
