Github user manishgupta88 commented on a diff in the pull request:
https://github.com/apache/carbondata/pull/2644#discussion_r214311329
--- Diff:
hadoop/src/main/java/org/apache/carbondata/hadoop/api/CarbonTableInputFormat.java
---
@@ -342,60 +341,52 @@ public void refreshSegmentCacheIfRequired(JobContext
job, CarbonTable carbonTabl
/**
* use file list in .carbonindex file to get the split of streaming.
*/
- public List<InputSplit> getSplitsOfStreaming(JobContext job,
AbsoluteTableIdentifier identifier,
- List<Segment> streamSegments) throws IOException {
+ public List<InputSplit> getSplitsOfStreaming(JobContext job,
List<Segment> streamSegments,
+ CarbonTable carbonTable, FilterResolverIntf filterResolverIntf)
throws IOException {
--- End diff --
You can write an overloaded method for getSplitsOfStreaming. One which
accepts 3 parameters and one with 4 parameters.
1. getSplitsOfStreaming(JobContext job, AbsoluteTableIdentifier
identifier,List<Segment> streamSegments)
-- From this method you can the other method and pass null as the 4th
argument. This will avoid passing null at all places above.
2. getSplitsOfStreaming(JobContext job, List<Segment> streamSegments,
CarbonTable carbonTable, FilterResolverIntf filterResolverIntf)
---
