Omega359 commented on code in PR #15242: URL: https://github.com/apache/datafusion/pull/15242#discussion_r2001009802
########## datafusion/sqllogictest/test_files/union_by_name.slt: ########## @@ -287,3 +287,137 @@ SELECT '0' as c UNION ALL BY NAME SELECT 0 as c; ---- 0 0 + +# Regression tests for https://github.com/apache/datafusion/issues/15236 +# Ensure that the correct output is produced even if the width of an input node's +# schema is the same as the resulting schema width after the union is applied. + +statement ok +create table t3 (x varchar(255), y varchar(255), z varchar(255)); + +statement ok +create table t4 (x varchar(255), y varchar(255), z varchar(255)); + +statement ok +insert into t3 values ('a', 'b', 'c'); + +statement ok +insert into t4 values ('a', 'b', 'c'); + +query TTTT rowsort +select t3.x, t3.y, t3.z from t3 union by name select t3.z, t3.y, t3.x, 'd' as zz from t3; +---- +a b c NULL +a b c d + +query TTTT rowsort +select t3.x, t3.y, t3.z from t3 union by name select t4.z, t4.y, t4.x, 'd' as zz from t4; +---- +a b c NULL +a b c d + +query TTT rowsort +select x, y, z from t3 union all by name select z, y, x from t3; +---- +a b c +a b c + +query TTT rowsort +select x, y, z from t3 union all by name select z, y, x from t4; +---- +a b c +a b c + +query TTT +select x, y, z from t3 union all by name select z, y, x from t4 order by x; +---- +a b c +a b c + + +# FIXME: The following should pass without error, but currently it is failing +# due to differing record batch schemas when the SLT runner collects results in +# normalize::convert_batches. +# +# More context can be found here: https://github.com/apache/datafusion/pull/15242#issuecomment-2730402547 +query error +select x, y, z from t3 union all by name select z, y, x, 'd' as zz from t3; +---- +DataFusion error: Internal error: Schema mismatch. Previously had +Schema { + fields: [ + Field { + name: "x", + data_type: Utf8, + nullable: true, + dict_id: 0, + dict_is_ordered: false, + metadata: {}, + }, + Field { + name: "y", + data_type: Utf8, + nullable: true, + dict_id: 0, + dict_is_ordered: false, + metadata: {}, + }, + Field { + name: "z", + data_type: Utf8, + nullable: true, + dict_id: 0, + dict_is_ordered: false, + metadata: {}, + }, + Field { + name: "zz", + data_type: Utf8, + nullable: false, + dict_id: 0, + dict_is_ordered: false, + metadata: {}, + }, + ], + metadata: {}, +} + +Got: +Schema { + fields: [ + Field { + name: "x", + data_type: Utf8, + nullable: true, + dict_id: 0, + dict_is_ordered: false, + metadata: {}, + }, + Field { + name: "y", + data_type: Utf8, + nullable: true, + dict_id: 0, + dict_is_ordered: false, + metadata: {}, + }, + Field { + name: "z", + data_type: Utf8, + nullable: true, + dict_id: 0, + dict_is_ordered: false, + metadata: {}, + }, + Field { + name: "zz", + data_type: Utf8, + nullable: true, Review Comment: Interesting that the nullable changed from false to true here. -- 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...@datafusion.apache.org For queries about this service, please contact Infrastructure at: us...@infra.apache.org --------------------------------------------------------------------- To unsubscribe, e-mail: github-unsubscr...@datafusion.apache.org For additional commands, e-mail: github-h...@datafusion.apache.org
