jecsand838 commented on code in PR #8349:
URL: https://github.com/apache/arrow-rs/pull/8349#discussion_r2371137299


##########
arrow-avro/src/reader/record.rs:
##########
@@ -991,10 +1096,327 @@ impl Decoder {
                     .map_err(|e| ArrowError::ParseError(e.to_string()))?;
                 Arc::new(vals)
             }
+            Self::Union(u) => u.flush(nulls)?,
         })
     }
 }
 
+#[derive(Debug)]
+struct DispatchLookupTable {
+    to_reader: Box<[i16]>,
+    promotion: Box<[Promotion]>,

Review Comment:
   I think there's definitely a few candidates in the Decoder. I'd just want to 
be thorough and not overuse boxed slices. To your point, it doesn't seem to be 
a super common approach. I did plan to come back through the `RecordDecoder` 
and benchmark optimizations/improvements like this in a dedicated PR.
   
   The places I applied the boxed slices in this PR just seemed to be the most 
obvious ones and I was sort of experimenting with it a bit lol.



-- 
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