lin-zhao commented on issue #17947:
URL: https://github.com/apache/pulsar/issues/17947#issuecomment-1282990768

   > For the message with protobuf schema, maybe we are impossible to 
deserialize by the AUTO_CONSUME schema. We have added protobuf_native 
https://docs.google.com/document/d/1XR_MNOuSXyig-CKsdVhr6IXvFwziBRdSoS3oEUiLFe8/edit
   > 
   > And for the encrypted message will not provide human-readable format data.
   > 
   > Currently, the peek message API is based on the entry-level. It can be a 
list of messages. We need to split the batch message and then apply the schema 
decoding.
   > 
   > How does the message with key-value schema react to the human-readable 
format data? I think it's better to have some examples in the proposal to show 
the expected result of primitive schema, struct schema, and key-value schema. 
The PulsarSQL also has similar logic for this part 
https://github.com/apache/pulsar/tree/master/pulsar-sql/presto-pulsar/src/main/java/org/apache/pulsar/sql/presto
   
   For batched messages, I tested and it appeared to work with the existing 
peek endpoint. I will double check but if it doesn't work for batched messages 
as stands today, we should return the right error code.
   
   For encrypted messages, by definition the peek API shouldn't be able to read 
the data. We should return the right error message and HTTP code.
   
    For other types of schemas that this approach does not address, how about 
we address it for the several schema types that works and address the others in 
future phases? 


-- 
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]

Reply via email to