pavibhai commented on a change in pull request #635:
URL: https://github.com/apache/orc/pull/635#discussion_r578825345
##########
File path: java/core/src/java/org/apache/orc/impl/RecordReaderImpl.java
##########
@@ -83,17 +85,24 @@
private final boolean[] fileIncluded;
private final long rowIndexStride;
private long rowInStripe = 0;
+ private long followRowInStripe = 0;
private int currentStripe = -1;
private long rowBaseInStripe = 0;
private long rowCountInStripe = 0;
private final BatchReader reader;
private final OrcIndex indexes;
+ // identifies the columns requiring row indexes
+ private final boolean[] rowIndexCols;
private final SargApplier sargApp;
// an array about which row groups aren't skipped
private boolean[] includedRowGroups = null;
private final DataReader dataReader;
private final int maxDiskRangeChunkLimit;
private final StripePlanner planner;
+ // identifies the type of read: FULL (no filters), LEAD (filters are present)
+ private final ReadLevel readLevel;
Review comment:
FOLLOW is never set at the RecordReader level. It will only be FULL or
LEAD. If this is LEAD then as part of read FOLLOW is implicitly used.
I can clarify further in the comment if that helps.
##########
File path: java/core/src/java/org/apache/orc/impl/BufferChunkList.java
##########
@@ -22,6 +22,14 @@
* Builds a list of buffer chunks
*/
public class BufferChunkList {
+ public BufferChunk getHead() {
Review comment:
Removed
##########
File path: java/core/src/java/org/apache/orc/impl/reader/StripePlanner.java
##########
@@ -363,11 +396,26 @@ private void addChunk(BufferChunkList list,
StreamInformation stream,
stream.firstChunk = chunk;
}
list.add(chunk);
+ stream.lastChunk = chunk;
offset += thisLen;
length -= thisLen;
}
}
+ private BufferChunkList addFollowChunk(long offset, long length) {
Review comment:
Removed
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
[email protected]
