scovich commented on code in PR #8349:
URL: https://github.com/apache/arrow-rs/pull/8349#discussion_r2368634667
##########
arrow-avro/src/reader/record.rs:
##########
@@ -486,6 +734,70 @@ impl Decoder {
Self::Decimal256(_, _, _, builder) =>
builder.append_value(i256::ZERO),
Self::Enum(indices, _, _) => indices.push(0),
Self::Duration(builder) => builder.append_null(),
+ Self::Union(fields, type_ids, offsets, encodings, encoding_counts,
None) => {
+ let mut chosen = None;
+ for (i, ch) in encodings.iter().enumerate() {
+ if matches!(ch, Decoder::Null(_)) {
+ chosen = Some(i);
+ break;
+ }
+ }
+ let idx = chosen.unwrap_or(0);
+ let type_id = fields
+ .iter()
+ .nth(idx)
+ .map(|(type_id, _)| type_id)
+ .unwrap_or_else(|| i8::try_from(idx).unwrap_or(0));
+ type_ids.push(type_id);
+ offsets.push(encoding_counts[idx]);
+ encodings[idx].append_null();
+ encoding_counts[idx] += 1;
+ }
+ Self::Union(
+ fields,
+ type_ids,
+ offsets,
+ encodings,
+ encoding_counts,
+ Some(union_resolution),
+ ) => match &mut union_resolution.kind {
+ UnionResolvedKind::Both { .. } => {
+ let mut chosen = None;
+ for (i, ch) in encodings.iter().enumerate() {
+ if matches!(ch, Decoder::Null(_)) {
+ chosen = Some(i);
+ break;
+ }
+ }
+ let idx = chosen.unwrap_or(0);
+ let type_id = fields
+ .iter()
+ .nth(idx)
+ .map(|(type_id, _)| type_id)
+ .unwrap_or_else(|| i8::try_from(idx).unwrap_or(0));
Review Comment:
I agree we shouldn't conflate Avro branch index with Arrow union type
code... but (how) did all this shake out in the latest revisions? I'm not
confident I understand this bit.
--
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: github-unsubscr...@arrow.apache.org
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org
