Github user ajantha-bhat commented on the issue:
https://github.com/apache/carbondata/pull/2289
@jackylk : ObjectSizeCalculator is called during ReverseDictonaryCache init
time [please refer the issue callstack in jira] but reverse dictionary cache no
need to initialize for sdk flows . Because it is needed only for dictionary
encoding. But sdk don't support dictionary encoding.
So, I have moved the initialize of Reverse dictionary cache inside
dictionary encoding check. so sdk flow this will not be called. so, that spark
class no need to find. hence we dont get this issue exception
---
