NicoK commented on a change in pull request #6705: [FLINK-10356][network] add
sanity checks to SpillingAdaptiveSpanningRecordDeserializer
URL: https://github.com/apache/flink/pull/6705#discussion_r252359412
##########
File path:
flink-runtime/src/main/java/org/apache/flink/runtime/io/network/api/serialization/SpillingAdaptiveSpanningRecordDeserializer.java
##########
@@ -575,6 +615,18 @@ private void
moveRemainderToNonSpanningDeserializer(NonSpanningWrapper deseriali
}
}
+ private int getRemainingBytes() {
+ if (this.spillFileReader == null) {
+ return this.serializationReadBuffer.available();
+ } else {
+ try {
+ return this.spillFileReader.available();
+ } catch (IOException ignored) {
Review comment:
Hmm - it may look dangerous but honestly, if retrieving the number of
remaining bytes gets to some error, it will still be an error when trying to
continue retrieving actual data (if we even do so). Since this is only used for
better error reporting, I didn't want to add another source of errors.
I could maybe rename the method to state that it tolerates failures so that
it is not confusing for future users, e.g. `getRemainingBytesOrZero()`. How
about that instead?
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on 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
