[SPARK-28276][SQL][PYTHON][TEST] Convert and port 'cross-join.sql' into UDF test base

[viirya]

What changes were proposed in this pull request?

This PR adds some tests converted from cross-join.sql' to test UDFs.

Diff comparing to 'cross-join.sql'

diff --git a/sql/core/src/test/resources/sql-tests/results/cross-join.sql.out b/sql/core/src/test/resources/sql-tests/results/udf/udf-cross-join.sql.out
index 3833c42bdf..11c1e01d54 100644
--- a/sql/core/src/test/resources/sql-tests/results/cross-join.sql.out
+++ b/sql/core/src/test/resources/sql-tests/results/udf/udf-cross-join.sql.out
@@ -43,7 +43,7 @@ two   2       two     22
 
 
 -- !query 3
-SELECT * FROM nt1 cross join nt2 where nt1.k = nt2.k
+SELECT * FROM nt1 cross join nt2 where udf(nt1.k) = udf(nt2.k)
 -- !query 3 schema
 struct<k:string,v1:int,k:string,v2:int>
 -- !query 3 output
@@ -53,7 +53,7 @@ two   2       two     22
 
 
 -- !query 4
-SELECT * FROM nt1 cross join nt2 on (nt1.k = nt2.k)
+SELECT * FROM nt1 cross join nt2 on (udf(nt1.k) = udf(nt2.k))
 -- !query 4 schema
 struct<k:string,v1:int,k:string,v2:int>
 -- !query 4 output
@@ -63,7 +63,7 @@ two   2       two     22
 
 
 -- !query 5
-SELECT * FROM nt1 cross join nt2 where nt1.v1 = 1 and nt2.v2 = 22
+SELECT * FROM nt1 cross join nt2 where udf(nt1.v1) = "1" and udf(nt2.v2) = "22"
 -- !query 5 schema
 struct<k:string,v1:int,k:string,v2:int>
 -- !query 5 output
@@ -71,12 +71,12 @@ one 1       two     22
 
 
-- !query 6
-SELECT a.key, b.key FROM
-(SELECT k key FROM nt1 WHERE v1 < 2) a
+SELECT udf(a.key), udf(b.key) FROM
+(SELECT udf(k) key FROM nt1 WHERE v1 < 2) a
 CROSS JOIN
-(SELECT k key FROM nt2 WHERE v2 = 22) b
+(SELECT udf(k) key FROM nt2 WHERE v2 = 22) b
 -- !query 6 schema
-struct<key:string,key:string>
+struct<udf(key):string,udf(key):string>
 -- !query 6 output
 one    two
 
@@ -114,23 +114,29 @@ struct<>
 
 
-- !query 11
-select * from ((A join B on (a = b)) cross join C) join D on (a = d)
+select * from ((A join B on (udf(a) = udf(b))) cross join C) join D on (udf(a) = udf(d))
 -- !query 11 schema
-struct<a:string,va:int,b:string,vb:int,c:string,vc:int,d:string,vd:int>
+struct<>
 -- !query 11 output
-one    1       one     1       one     1       one     1
-one    1       one     1       three   3       one     1
-one    1       one     1       two     2       one     1
-three  3       three   3       one     1       three   3
-three  3       three   3       three   3       three   3
-three  3       three   3       two     2       three   3
-two    2       two     2       one     1       two     2
-two    2       two     2       three   3       two     2
-two    2       two     2       two     2       two     2
+org.apache.spark.sql.AnalysisException
+Detected implicit cartesian product for INNER join between logical plans
+Filter (udf(a#x) = udf(b#x))
++- Join Inner
+   :- Project [k#x AS a#x, v1#x AS va#x]
+   :  +- LocalRelation [k#x, v1#x]
+   +- Project [k#x AS b#x, v1#x AS vb#x]
+      +- LocalRelation [k#x, v1#x]
+and
+Project [k#x AS d#x, v1#x AS vd#x]
++- LocalRelation [k#x, v1#x]
+Join condition is missing or trivial.
+Either: use the CROSS JOIN syntax to allow cartesian products between these
+relations, or: enable implicit cartesian products by setting the configuration
+variable spark.sql.crossJoin.enabled=true;
 
 
 -- !query 12
-SELECT * FROM nt1 CROSS JOIN nt2 ON (nt1.k > nt2.k)
+SELECT * FROM nt1 CROSS JOIN nt2 ON (udf(nt1.k) > udf(nt2.k))
 -- !query 12 schema
 struct<k:string,v1:int,k:string,v2:int>
 -- !query 12 output

How was this patch tested?

Added test.

[SparkQA]

Test build #107719 has finished for PR 25168 at commit af80ac7.

• This patch fails due to an unknown error code, -9.
• This patch merges cleanly.
• This patch adds no public classes.

[viirya]

retest this please

[HyukjinKwon]

Looks good .. but let's get #25130 first .. just in case that causes some changes in those PRs.

[SparkQA]

Test build #107724 has finished for PR 25168 at commit af80ac7.

• This patch fails Spark unit tests.
• This patch merges cleanly.
• This patch adds no public classes.

[SparkQA]

Test build #107727 has finished for PR 25168 at commit ac20743.

• This patch passes all tests.
• This patch merges cleanly.
• This patch adds no public classes.

[HyukjinKwon]

LGTM if the tests pass

[HyukjinKwon]

retest this please

[HyukjinKwon]

BTW, @viirya, please feel free to review those PRs when you have some times since you know those codes pretty well as well.

[viirya]

I synced this with master.

[viirya]

@HyukjinKwon yeah, thanks. I will take some times to review those PRs.

[SparkQA]

Test build #107820 has finished for PR 25168 at commit ac20743.

• This patch fails Spark unit tests.
• This patch merges cleanly.
• This patch adds no public classes.

[SparkQA]

Test build #107828 has finished for PR 25168 at commit 762ffb6.

• This patch fails due to an unknown error code, -9.
• This patch merges cleanly.
• This patch adds the following public classes (experimental):
• case class Division(conf: SQLConf) extends TypeCoercionRule

[SparkQA]

Test build #107829 has finished for PR 25168 at commit 0dbc985.

• This patch passes all tests.
• This patch merges cleanly.
• This patch adds no public classes.

[HyukjinKwon]

Merged to master.