JingsongLi commented on a change in pull request #10170: [FLINK-14722][hadoop]
Optimize mapred.HadoopInputSplit to not serialize conf when split is not
configurable
URL: https://github.com/apache/flink/pull/10170#discussion_r349457137
##########
File path:
flink-connectors/flink-hadoop-compatibility/src/main/java/org/apache/flink/api/java/hadoop/mapred/wrapper/HadoopInputSplit.java
##########
@@ -70,8 +77,8 @@ public HadoopInputSplit(int splitNumber,
org.apache.hadoop.mapred.InputSplit hIn
public String[] getHostnames() {
try {
return this.hadoopInputSplit.getLocations();
- }
- catch (IOException e) {
+ } catch (IOException e) {
+ e.printStackTrace();
Review comment:
Sorry, it is just for debug. I was thinking about why to return 'new string
[0]' here. Later, I found that this is a convention of `InputSplit`. They all
do this. Maybe they think this is just an optimization.
----------------------------------------------------------------
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]
With regards,
Apache Git Services
