etseidl commented on code in PR #6987:
URL: https://github.com/apache/arrow-rs/pull/6987#discussion_r1918983332
##########
parquet/src/arrow/array_reader/primitive_array.rs:
##########
@@ -270,6 +270,59 @@ where
Arc::new(array) as ArrayRef
}
+ ArrowType::Dictionary(_, value_type) => match value_type.as_ref() {
+ ArrowType::Decimal128(p, s) => {
+ let array = match array.data_type() {
+ ArrowType::Int32 => array
+ .as_any()
+ .downcast_ref::<Int32Array>()
+ .unwrap()
+ .unary(|i| i as i128)
+ as Decimal128Array,
+ ArrowType::Int64 => array
+ .as_any()
+ .downcast_ref::<Int64Array>()
+ .unwrap()
+ .unary(|i| i as i128)
+ as Decimal128Array,
+ _ => {
+ return Err(arrow_err!(
+ "Cannot convert {:?} to decimal dictionary",
+ array.data_type()
+ ));
+ }
+ }
+ .with_precision_and_scale(*p, *s)?;
+
+ arrow_cast::cast(&array, target_type)?
Review Comment:
Taking a second look, I'm curious why the cast here rather than creating the
`ArrayRef` directly as is done above? Will this have any performance
implications?
--
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]
