gty404 commented on code in PR #1054:
URL: https://github.com/apache/iceberg-cpp/pull/1054#discussion_r4384896339
##########
src/iceberg/manifest/manifest_reader.cc:
##########
@@ -318,6 +323,7 @@ Result<std::vector<ManifestFile>>
ParseManifestList(ArrowSchema* arrow_schema,
ICEBERG_RETURN_UNEXPECTED(AssertViewTypeAndChildren(
&view, ArrowType::NANOARROW_TYPE_STRUCT, schema.fields().size(),
"manifest_list"));
+ ICEBERG_RETURN_UNEXPECTED(AssertChildrenLength(&view, "manifest_list"));
Review Comment:
Checked this — I don't think it's a gap. nanoarrow's
ArrowArrayViewValidate(FULL) already rejects a struct child shorter than its
parent, recursively, for every STRUCT node including map_view and elem_view
(array.c:1134-1145), so "shorter" is already caught. And for the "longer" case
this PR fixes elsewhere: ParseIntMapField/ParsePartitionFieldSummaryList index
key_view/value_view/contains_null etc. by an offset taken from the outer
list/map's own offsets buffer, not by the child's own length, so a longer child
can't drive those loops out of bounds the way it does in ParseField-style
parsers.
--
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]
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
