Github user NicoK commented on a diff in the pull request:
https://github.com/apache/flink/pull/5105#discussion_r156413073
--- Diff:
flink-runtime/src/main/java/org/apache/flink/runtime/io/network/api/serialization/SpanningRecordSerializer.java
---
@@ -169,33 +151,29 @@ public Buffer getCurrentBuffer() {
if (targetBuffer == null) {
return null;
}
-
- targetBuffer.setSize(position);
- return targetBuffer;
+ Buffer result = targetBuffer.build();
+ targetBuffer = null;
+ return result;
}
@Override
public void clearCurrentBuffer() {
targetBuffer = null;
- position = 0;
- limit = 0;
}
@Override
public void clear() {
targetBuffer = null;
- position = 0;
- limit = 0;
- // ensure clear state with hasRemaining false (for correct
setNextBuffer logic)
+ // ensure clear state with hasRemaining false (for correct
setNextBufferBuilder logic)
dataBuffer.position(dataBuffer.limit());
lengthBuffer.position(4);
}
@Override
public boolean hasData() {
// either data in current target buffer or intermediate buffers
- return position > 0 || (lengthBuffer.hasRemaining() ||
dataBuffer.hasRemaining());
+ return (targetBuffer != null && !targetBuffer.isEmpty()) ||
(lengthBuffer.hasRemaining() || dataBuffer.hasRemaining());
--- End diff --
it has been there before, but the second set of parentheses is actually not
needed...
`return (targetBuffer != null && !targetBuffer.isEmpty()) ||
lengthBuffer.hasRemaining() || dataBuffer.hasRemaining();`
---
