BewareMyPower commented on code in PR #16196:
URL: https://github.com/apache/pulsar/pull/16196#discussion_r906124600
##########
pulsar-client/src/main/java/org/apache/pulsar/client/impl/ProducerImpl.java:
##########
@@ -488,6 +488,10 @@ public void sendAsync(Message<?> message, SendCallback
callback) {
return;
}
+ // Update the message metadata before computing the payload chunk size
to avoid a large message cannot be split
+ // into chunks.
+ final long sequenceId = updateMessageMetadata(msgMetadata,
uncompressedSize);
Review Comment:
Yes, we can skip `isMessageSizeExceeded` if client chunking is enabled. But
I think the existing changes in this PR make the computation of the payload
chunk size **more accurate**.
Regarding to the metadata update after the computation,
> msgMetadata.setUuid(...) and msgMetadata.setChunkId().....
These metadata fields are updated only for chunks:
```protobuf
optional string uuid = 26;
optional int32 num_chunks_from_msg = 27;
optional int32 total_chunk_msg_size = 28;
optional int32 chunk_id = 29;
```
In this case, `isMessageSizeExceeded` is already skipped.
And yes, `ProducerImpl#encryptMessage` could update the metadata as well. I
think it's better to skip the message size check when encryption is enabled,
because message encryption could also increase the size of the payload buffer.
--
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]
