pvary commented on a change in pull request #1430:
URL: https://github.com/apache/iceberg/pull/1430#discussion_r485539679
##########
File path: api/src/main/java/org/apache/iceberg/Metrics.java
##########
@@ -120,4 +124,83 @@ public Long recordCount() {
public Map<Integer, ByteBuffer> upperBounds() {
return upperBounds;
}
+
+ /**
+ * Implemented the method to enable serialization of ByteBuffers.
+ * @param out The stream where to write
+ * @throws IOException On serialization error
+ */
+ private void writeObject(ObjectOutputStream out) throws IOException {
+ out.writeObject(rowCount);
+ out.writeObject(columnSizes);
+ out.writeObject(valueCounts);
+ out.writeObject(nullValueCounts);
+
+ writeByteBufferMap(out, lowerBounds);
+ writeByteBufferMap(out, upperBounds);
+ }
+
+ private static void writeByteBufferMap(ObjectOutputStream out, Map<Integer,
ByteBuffer> byteBufferMap)
+ throws IOException {
+ if (byteBufferMap == null) {
+ out.writeInt(-1);
+
+ } else {
+ // Write the size
+ out.writeInt(byteBufferMap.size());
+
+ for (Map.Entry<Integer, ByteBuffer> entry : byteBufferMap.entrySet()) {
+ // Write the key
+ out.writeObject(entry.getKey());
+
+ // Write the value
+ if (entry.getValue() != null) {
+ out.writeObject(entry.getValue().array());
Review comment:
@RussellSpitzer: Fixed with your suggested changes. I have chickened out
from using different serialization method for different buffers because I think
the added complexity might not worth the gained performance.
Could you please recheck the change?
Thanks,
Peter
----------------------------------------------------------------
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]
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
