Github user BJangir commented on a diff in the pull request:
https://github.com/apache/carbondata/pull/2897#discussion_r243812841
--- Diff:
hadoop/src/main/java/org/apache/carbondata/hadoop/util/CarbonVectorizedRecordReader.java
---
@@ -68,6 +67,10 @@
private AbstractDetailQueryResultIterator iterator;
private QueryModel queryModel;
+ //This holds mapping of fetch index with respect to project col index.
+ // it is used when same col is used in projection many times.So need to
fetch only that col.
+ private List<Integer> projectionMapping = new ArrayList<>();
--- End diff --
During initBatch only it is done(not frequent lookup/creation ).
---
