alamb commented on code in PR #9497:
URL: https://github.com/apache/arrow-rs/pull/9497#discussion_r2955801123
##########
arrow-json/src/reader/list_array.rs:
##########
@@ -93,15 +93,14 @@ impl<O: OffsetSizeTrait> ArrayDecoder for
ListArrayDecoder<O> {
let child_data = self.decoder.decode(tape, &child_pos)?;
let nulls = nulls.as_mut().map(|x| NullBuffer::new(x.finish()));
+ let values = make_array(child_data);
+ let field = match &self.data_type {
+ DataType::List(f) | DataType::LargeList(f) => f.clone(),
+ _ => unreachable!(),
+ };
+ let offsets =
OffsetBuffer::<O>::new(ScalarBuffer::from(offsets.finish()));
Review Comment:
I think this is mostly responsible for the different in performance --
perhaps we can revert
https://github.com/apache/arrow-rs/pull/9497/commits/bfb0ee51ef1a01db3fd26ad4d4ff80b115510208
and try the `new_unchecked` 🤔
--
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]
