lucasbru commented on code in PR #22514:
URL: https://github.com/apache/kafka/pull/22514#discussion_r3380763743
##########
clients/src/main/java/org/apache/kafka/common/serialization/ListDeserializer.java:
##########
@@ -153,8 +153,8 @@ private List<Integer> deserializeNullIndexList(final
DataInputStream dis, final
if (nullIndexListSize < 0) {
throw new SerializationException("Corrupted byte[]. The number of
null list entries cannot be negative.");
}
- if (nullIndexListSize > length) {
- throw new SerializationException("Corrupted byte[]. The number of
null list entries cannot be larger than overall number of bytes.");
+ if (nullIndexListSize > length / primitiveSize) {
Review Comment:
I think `primitiveSize` is the wrong divisor here. The null-index list holds
indices, each written as a 4-byte `readInt()`.
##########
streams/src/main/java/org/apache/kafka/streams/processor/internals/assignment/AssignmentInfo.java:
##########
@@ -389,7 +389,7 @@ private static void decodeActiveTasks(final AssignmentInfo
assignmentInfo,
final DataInputStream in,
final int length) throws IOException
{
final int count = in.readInt();
- if (count < 0 || count > length) {
+ if (count < 0 || count > length / (2 * Integer.BYTES)) { // task-id is
<subtopologyId[INTEGER]><partition[INTEGER]>
Review Comment:
`decodePartitionsByHost`, the non-dictionary `readTopicPartitions`, and
`decodeTopicIndexAndGet` are still left at the loose `> length`. Not wrong,
just inconsistent if the intent is to bound allocation everywhere.
##########
clients/src/test/java/org/apache/kafka/common/serialization/ListDeserializerTest.java:
##########
@@ -380,14 +380,14 @@ public void shouldThrowOnTooLargeNullEntryLength() {
(byte) 0x00, (byte) 0x00, (byte) 0x00, (byte) 0xFF // encodes
number of null entries == 255
};
- final ListDeserializer<String> testDeserializer = new
ListDeserializer<>(ArrayList.class, new StringDeserializer());
+ final ListDeserializer<Integer> testDeserializer = new
ListDeserializer<>(ArrayList.class, new IntegerDeserializer());
Review Comment:
Did you switch here to integer to make the regression above pass?
--
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.
To unsubscribe, e-mail: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
