tustvold commented on issue #3949:
URL: https://github.com/apache/arrow-rs/issues/3949#issuecomment-1484077983
The major reason for wanting to remove the old decoder, aside from its
incredibly underwhelming performance, is that the code is very hard to maintain
and reason about, in particular the way it handles nested types is very
convoluted, and is likely also incorrect.
That being said I wonder if we can accommodate your use-case in a different
manner, you mention "using serde's deserialization code", does this mean that
the types are known statically ahead of time? I ask as providing a nicer story
for creating `RecordBatch` from statically typed rows is something I have been
playing around with, would this work for your use-case?
--
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]
