Commit
This commit does not belong to any branch on this repository, and may belong to a fork outside of the repository.
[SPARK-28284][SQL][PYTHON][TESTS] Convert and port 'join-empty-relati…
…on.sql' into UDF test base ## What changes were proposed in this pull request? This PR adds some tests converted from `join-empty-relation.sql` to test UDFs. Please see contribution guide of this umbrella ticket - [SPARK-27921](https://issues.apache.org/jira/browse/SPARK-27921). <details><summary>Diff comparing to 'join-empty-relation.sql'</summary> <p> ```diff diff --git a/sql/core/src/test/resources/sql-tests/results/join-empty-relation.sql.out b/sql/core/src/test/resources/sql-tests/results/udf/udf-join-empty-relation.sql.out index 857073a827..e79d01fb14 100644 --- a/sql/core/src/test/resources/sql-tests/results/join-empty-relation.sql.out +++ b/sql/core/src/test/resources/sql-tests/results/udf/udf-join-empty-relation.sql.out -27,111 +27,111 struct<> -- !query 3 -SELECT * FROM t1 INNER JOIN empty_table +SELECT udf(t1.a), udf(empty_table.a) FROM t1 INNER JOIN empty_table ON (udf(t1.a) = udf(udf(empty_table.a))) -- !query 3 schema -struct<a:int,a:int> +struct<CAST(udf(cast(a as string)) AS INT):int,CAST(udf(cast(a as string)) AS INT):int> -- !query 3 output -- !query 4 -SELECT * FROM t1 CROSS JOIN empty_table +SELECT udf(t1.a), udf(udf(empty_table.a)) FROM t1 CROSS JOIN empty_table ON (udf(udf(t1.a)) = udf(empty_table.a)) -- !query 4 schema -struct<a:int,a:int> +struct<CAST(udf(cast(a as string)) AS INT):int,CAST(udf(cast(cast(udf(cast(a as string)) as int) as string)) AS INT):int> -- !query 4 output -- !query 5 -SELECT * FROM t1 LEFT OUTER JOIN empty_table +SELECT udf(udf(t1.a)), empty_table.a FROM t1 LEFT OUTER JOIN empty_table ON (udf(t1.a) = udf(empty_table.a)) -- !query 5 schema -struct<a:int,a:int> +struct<CAST(udf(cast(cast(udf(cast(a as string)) as int) as string)) AS INT):int,a:int> -- !query 5 output 1 NULL -- !query 6 -SELECT * FROM t1 RIGHT OUTER JOIN empty_table +SELECT udf(t1.a), udf(empty_table.a) FROM t1 RIGHT OUTER JOIN empty_table ON (udf(t1.a) = udf(empty_table.a)) -- !query 6 schema -struct<a:int,a:int> +struct<CAST(udf(cast(a as string)) AS INT):int,CAST(udf(cast(a as string)) AS INT):int> -- !query 6 output -- !query 7 -SELECT * FROM t1 FULL OUTER JOIN empty_table +SELECT udf(t1.a), empty_table.a FROM t1 FULL OUTER JOIN empty_table ON (udf(t1.a) = udf(empty_table.a)) -- !query 7 schema -struct<a:int,a:int> +struct<CAST(udf(cast(a as string)) AS INT):int,a:int> -- !query 7 output 1 NULL -- !query 8 -SELECT * FROM t1 LEFT SEMI JOIN empty_table +SELECT udf(udf(t1.a)) FROM t1 LEFT SEMI JOIN empty_table ON (udf(t1.a) = udf(udf(empty_table.a))) -- !query 8 schema -struct<a:int> +struct<CAST(udf(cast(cast(udf(cast(a as string)) as int) as string)) AS INT):int> -- !query 8 output -- !query 9 -SELECT * FROM t1 LEFT ANTI JOIN empty_table +SELECT udf(t1.a) FROM t1 LEFT ANTI JOIN empty_table ON (udf(t1.a) = udf(empty_table.a)) -- !query 9 schema -struct<a:int> +struct<CAST(udf(cast(a as string)) AS INT):int> -- !query 9 output 1 -- !query 10 -SELECT * FROM empty_table INNER JOIN t1 +SELECT udf(empty_table.a), udf(t1.a) FROM empty_table INNER JOIN t1 ON (udf(udf(empty_table.a)) = udf(t1.a)) -- !query 10 schema -struct<a:int,a:int> +struct<CAST(udf(cast(a as string)) AS INT):int,CAST(udf(cast(a as string)) AS INT):int> -- !query 10 output -- !query 11 -SELECT * FROM empty_table CROSS JOIN t1 +SELECT udf(empty_table.a), udf(udf(t1.a)) FROM empty_table CROSS JOIN t1 ON (udf(empty_table.a) = udf(udf(t1.a))) -- !query 11 schema -struct<a:int,a:int> +struct<CAST(udf(cast(a as string)) AS INT):int,CAST(udf(cast(cast(udf(cast(a as string)) as int) as string)) AS INT):int> -- !query 11 output -- !query 12 -SELECT * FROM empty_table LEFT OUTER JOIN t1 +SELECT udf(udf(empty_table.a)), udf(t1.a) FROM empty_table LEFT OUTER JOIN t1 ON (udf(empty_table.a) = udf(t1.a)) -- !query 12 schema -struct<a:int,a:int> +struct<CAST(udf(cast(cast(udf(cast(a as string)) as int) as string)) AS INT):int,CAST(udf(cast(a as string)) AS INT):int> -- !query 12 output -- !query 13 -SELECT * FROM empty_table RIGHT OUTER JOIN t1 +SELECT empty_table.a, udf(t1.a) FROM empty_table RIGHT OUTER JOIN t1 ON (udf(empty_table.a) = udf(t1.a)) -- !query 13 schema -struct<a:int,a:int> +struct<a:int,CAST(udf(cast(a as string)) AS INT):int> -- !query 13 output NULL 1 -- !query 14 -SELECT * FROM empty_table FULL OUTER JOIN t1 +SELECT empty_table.a, udf(udf(t1.a)) FROM empty_table FULL OUTER JOIN t1 ON (udf(empty_table.a) = udf(t1.a)) -- !query 14 schema -struct<a:int,a:int> +struct<a:int,CAST(udf(cast(cast(udf(cast(a as string)) as int) as string)) AS INT):int> -- !query 14 output NULL 1 -- !query 15 -SELECT * FROM empty_table LEFT SEMI JOIN t1 +SELECT udf(udf(empty_table.a)) FROM empty_table LEFT SEMI JOIN t1 ON (udf(empty_table.a) = udf(udf(t1.a))) -- !query 15 schema -struct<a:int> +struct<CAST(udf(cast(cast(udf(cast(a as string)) as int) as string)) AS INT):int> -- !query 15 output -- !query 16 -SELECT * FROM empty_table LEFT ANTI JOIN t1 +SELECT empty_table.a FROM empty_table LEFT ANTI JOIN t1 ON (udf(empty_table.a) = udf(t1.a)) -- !query 16 schema struct<a:int> -- !query 16 output -139,56 +139,56 struct<a:int> -- !query 17 -SELECT * FROM empty_table INNER JOIN empty_table +SELECT udf(empty_table.a) FROM empty_table INNER JOIN empty_table AS empty_table2 ON (udf(empty_table.a) = udf(udf(empty_table2.a))) -- !query 17 schema -struct<a:int,a:int> +struct<CAST(udf(cast(a as string)) AS INT):int> -- !query 17 output -- !query 18 -SELECT * FROM empty_table CROSS JOIN empty_table +SELECT udf(udf(empty_table.a)) FROM empty_table CROSS JOIN empty_table AS empty_table2 ON (udf(udf(empty_table.a)) = udf(empty_table2.a)) -- !query 18 schema -struct<a:int,a:int> +struct<CAST(udf(cast(cast(udf(cast(a as string)) as int) as string)) AS INT):int> -- !query 18 output -- !query 19 -SELECT * FROM empty_table LEFT OUTER JOIN empty_table +SELECT udf(empty_table.a) FROM empty_table LEFT OUTER JOIN empty_table AS empty_table2 ON (udf(empty_table.a) = udf(empty_table2.a)) -- !query 19 schema -struct<a:int,a:int> +struct<CAST(udf(cast(a as string)) AS INT):int> -- !query 19 output -- !query 20 -SELECT * FROM empty_table RIGHT OUTER JOIN empty_table +SELECT udf(udf(empty_table.a)) FROM empty_table RIGHT OUTER JOIN empty_table AS empty_table2 ON (udf(empty_table.a) = udf(udf(empty_table2.a))) -- !query 20 schema -struct<a:int,a:int> +struct<CAST(udf(cast(cast(udf(cast(a as string)) as int) as string)) AS INT):int> -- !query 20 output -- !query 21 -SELECT * FROM empty_table FULL OUTER JOIN empty_table +SELECT udf(empty_table.a) FROM empty_table FULL OUTER JOIN empty_table AS empty_table2 ON (udf(empty_table.a) = udf(empty_table2.a)) -- !query 21 schema -struct<a:int,a:int> +struct<CAST(udf(cast(a as string)) AS INT):int> -- !query 21 output -- !query 22 -SELECT * FROM empty_table LEFT SEMI JOIN empty_table +SELECT udf(udf(empty_table.a)) FROM empty_table LEFT SEMI JOIN empty_table AS empty_table2 ON (udf(empty_table.a) = udf(empty_table2.a)) -- !query 22 schema -struct<a:int> +struct<CAST(udf(cast(cast(udf(cast(a as string)) as int) as string)) AS INT):int> -- !query 22 output -- !query 23 -SELECT * FROM empty_table LEFT ANTI JOIN empty_table +SELECT udf(empty_table.a) FROM empty_table LEFT ANTI JOIN empty_table AS empty_table2 ON (udf(empty_table.a) = udf(empty_table2.a)) -- !query 23 schema -struct<a:int> +struct<CAST(udf(cast(a as string)) AS INT):int> -- !query 23 output ``` </p> </details> ## How was this patch tested? Tested as guided in [SPARK-27921](https://issues.apache.org/jira/browse/SPARK-27921). Closes apache#25127 from imback82/join-empty-relation-sql. Authored-by: Terry Kim <[email protected]> Signed-off-by: HyukjinKwon <[email protected]>
- Loading branch information
1 parent
52ddf03
commit 453cbf3