[SPARK-19602][SQL][TESTS] Add tests for qualified column names

sql/core/src/test/resources/sql-tests/inputs/columnresolution-negative.sql

@@ -0,0 +1,36 @@
+-- Negative testcases for column resolution
+CREATE DATABASE mydb1;
+USE mydb1;
+CREATE TABLE t1 USING parquet AS SELECT 1 AS i1;
+
+CREATE DATABASE mydb2;
+USE mydb2;
+CREATE TABLE t1 USING parquet AS SELECT 20 AS i1;
+
+-- Negative tests: column resolution scenarios with ambiguous cases in join queries
+SET spark.sql.crossJoin.enabled = true;
+USE mydb1;
+SELECT i1 FROM t1, mydb1.t1;
+SELECT t1.i1 FROM t1, mydb1.t1;
+SELECT mydb1.t1.i1 FROM t1, mydb1.t1;
+SELECT i1 FROM t1, mydb2.t1;
+SELECT t1.i1 FROM t1, mydb2.t1;
+USE mydb2;
+SELECT i1 FROM t1, mydb1.t1;
+SELECT t1.i1 FROM t1, mydb1.t1;
+SELECT i1 FROM t1, mydb2.t1;
+SELECT t1.i1 FROM t1, mydb2.t1;
+SELECT db1.t1.i1 FROM t1, mydb2.t1;
+SET spark.sql.crossJoin.enabled = false;
+
+-- Negative tests
+USE mydb1;
+SELECT mydb1.t1 FROM t1;
+SELECT t1.x.y.* FROM t1;
+SELECT t1 FROM mydb1.t1;
+USE mydb2;
+SELECT mydb1.t1.i1 FROM t1;
+
+-- reset
+DROP DATABASE mydb1 CASCADE;
+DROP DATABASE mydb2 CASCADE;

sql/core/src/test/resources/sql-tests/inputs/columnresolution-views.sql

@@ -0,0 +1,25 @@
+-- Tests for qualified column names for the view code-path
+-- Test scenario with Temporary view
+CREATE OR REPLACE TEMPORARY VIEW view1 AS SELECT 2 AS i1;
+SELECT view1.* FROM view1;
+SELECT * FROM view1;
+SELECT view1.i1 FROM view1;
+SELECT i1 FROM view1;
+SELECT a.i1 FROM view1 AS a;
+SELECT i1 FROM view1 AS a;
+-- cleanup
+DROP VIEW view1;
+
+-- Test scenario with Global Temp view
+CREATE OR REPLACE GLOBAL TEMPORARY VIEW view1 as SELECT 1 as i1;
+SELECT * FROM global_temp.view1;
+-- TODO: Support this scenario
+SELECT global_temp.view1.* FROM global_temp.view1;
+SELECT i1 FROM global_temp.view1;
+-- TODO: Support this scenario
+SELECT global_temp.view1.i1 FROM global_temp.view1;
+SELECT view1.i1 FROM global_temp.view1;
+SELECT a.i1 FROM global_temp.view1 AS a;
+SELECT i1 FROM global_temp.view1 AS a;
+-- cleanup
+DROP VIEW global_temp.view1;

sql/core/src/test/resources/sql-tests/inputs/columnresolution.sql

@@ -0,0 +1,88 @@
+-- Tests covering different scenarios with qualified column names
+-- Scenario: column resolution scenarios with datasource table
+CREATE DATABASE mydb1;
+USE mydb1;
+CREATE TABLE t1 USING parquet AS SELECT 1 AS i1;
+
+CREATE DATABASE mydb2;
+USE mydb2;
+CREATE TABLE t1 USING parquet AS SELECT 20 AS i1;
+
+USE mydb1;
+SELECT i1 FROM t1;
+SELECT i1 FROM mydb1.t1;
+SELECT t1.i1 FROM t1;
+SELECT t1.i1 FROM mydb1.t1;
+
+-- TODO: Support this scenario
+SELECT mydb1.t1.i1 FROM t1;
+-- TODO: Support this scenario
+SELECT mydb1.t1.i1 FROM mydb1.t1;
+
+USE mydb2;
+SELECT i1 FROM t1;
+SELECT i1 FROM mydb1.t1;
+SELECT t1.i1 FROM t1;
+SELECT t1.i1 FROM mydb1.t1;
+-- TODO: Support this scenario
+SELECT mydb1.t1.i1 FROM mydb1.t1;
+
+-- Scenario: resolve fully qualified table name in star expansion
+USE mydb1;
+SELECT t1.* FROM t1;
+SELECT mydb1.t1.* FROM mydb1.t1;
+SELECT t1.* FROM mydb1.t1;
+USE mydb2;
+SELECT t1.* FROM t1;
+-- TODO: Support this scenario
+SELECT mydb1.t1.* FROM mydb1.t1;
+SELECT t1.* FROM mydb1.t1;
+SELECT a.* FROM mydb1.t1 AS a;
+
+-- Scenario: resolve in case of subquery
+
+USE mydb1;
+CREATE TABLE t3 USING parquet AS SELECT * FROM VALUES (4,1), (3,1) AS t3(c1, c2);
+CREATE TABLE t4 USING parquet AS SELECT * FROM VALUES (4,1), (2,1) AS t4(c2, c3);
+
+SELECT * FROM t3 WHERE c1 IN (SELECT c2 FROM t4 WHERE t4.c3 = t3.c2);
+
+-- TODO: Support this scenario
+SELECT * FROM mydb1.t3 WHERE c1 IN
+  (SELECT mydb1.t4.c2 FROM mydb1.t4 WHERE mydb1.t4.c3 = mydb1.t3.c2);
+
+-- Scenario: column resolution scenarios in join queries
+SET spark.sql.crossJoin.enabled = true;
+
+-- TODO: Support this scenario
+SELECT mydb1.t1.i1 FROM t1, mydb2.t1;
+
+-- TODO: Support this scenario
+SELECT mydb1.t1.i1 FROM mydb1.t1, mydb2.t1;
+
+USE mydb2;
+-- TODO: Support this scenario
+SELECT mydb1.t1.i1 FROM t1, mydb1.t1;
+SET spark.sql.crossJoin.enabled = false;
+
+-- Scenario: Table with struct column
+USE mydb1;
+CREATE TABLE t5(i1 INT, t5 STRUCT<i1:INT, i2:INT>) USING parquet;
+INSERT INTO t5 VALUES(1, (2, 3));
+SELECT t5.i1 FROM t5;
+SELECT t5.t5.i1 FROM t5;
+SELECT t5.t5.i1 FROM mydb1.t5;
+SELECT t5.i1 FROM mydb1.t5;
+SELECT t5.* FROM mydb1.t5;
+SELECT t5.t5.* FROM mydb1.t5;
+-- TODO: Support this scenario
+SELECT mydb1.t5.t5.i1 FROM mydb1.t5;
+-- TODO: Support this scenario
+SELECT mydb1.t5.t5.i2 FROM mydb1.t5;
+-- TODO: Support this scenario
+SELECT mydb1.t5.* FROM mydb1.t5;
+
+-- Cleanup and Reset
+USE default;
+DROP DATABASE mydb1 CASCADE;
+DROP DATABASE mydb2 CASCADE;

sql/core/src/test/resources/sql-tests/results/columnresolution-negative.sql.out

@@ -0,0 +1,240 @@
+-- Automatically generated by SQLQueryTestSuite
+-- Number of queries: 28
+
+
+-- !query 0
+CREATE DATABASE mydb1
+-- !query 0 schema
+struct<>
+-- !query 0 output
+
+
+
+-- !query 1
+USE mydb1
+-- !query 1 schema
+struct<>
+-- !query 1 output
+
+
+
+-- !query 2
+CREATE TABLE t1 USING parquet AS SELECT 1 AS i1
+-- !query 2 schema
+struct<>
+-- !query 2 output
+
+
+
+-- !query 3
+CREATE DATABASE mydb2
+-- !query 3 schema
+struct<>
+-- !query 3 output
+
+
+
+-- !query 4
+USE mydb2
+-- !query 4 schema
+struct<>
+-- !query 4 output
+
+
+
+-- !query 5
+CREATE TABLE t1 USING parquet AS SELECT 20 AS i1
+-- !query 5 schema
+struct<>
+-- !query 5 output
+
+
+
+-- !query 6
+SET spark.sql.crossJoin.enabled = true
+-- !query 6 schema
+struct<key:string,value:string>
+-- !query 6 output
+spark.sql.crossJoin.enabled	true
+
+
+-- !query 7
+USE mydb1
+-- !query 7 schema
+struct<>
+-- !query 7 output
+
+
+
+-- !query 8
+SELECT i1 FROM t1, mydb1.t1
+-- !query 8 schema
+struct<>
+-- !query 8 output
+org.apache.spark.sql.AnalysisException
+Reference 'i1' is ambiguous, could be: i1#x, i1#x.; line 1 pos 7
+
+
+-- !query 9
+SELECT t1.i1 FROM t1, mydb1.t1
+-- !query 9 schema
+struct<>
+-- !query 9 output
+org.apache.spark.sql.AnalysisException
+Reference 't1.i1' is ambiguous, could be: i1#x, i1#x.; line 1 pos 7
+
+
+-- !query 10
+SELECT mydb1.t1.i1 FROM t1, mydb1.t1
+-- !query 10 schema
+struct<>
+-- !query 10 output
+org.apache.spark.sql.AnalysisException
+cannot resolve '`mydb1.t1.i1`' given input columns: [i1, i1]; line 1 pos 7
+
+
+-- !query 11
+SELECT i1 FROM t1, mydb2.t1
+-- !query 11 schema
+struct<>
+-- !query 11 output
+org.apache.spark.sql.AnalysisException
+Reference 'i1' is ambiguous, could be: i1#x, i1#x.; line 1 pos 7
+
+
+-- !query 12
+SELECT t1.i1 FROM t1, mydb2.t1
+-- !query 12 schema
+struct<>
+-- !query 12 output
+org.apache.spark.sql.AnalysisException
+Reference 't1.i1' is ambiguous, could be: i1#x, i1#x.; line 1 pos 7
+
+
+-- !query 13
+USE mydb2
+-- !query 13 schema
+struct<>
+-- !query 13 output
+
+
+
+-- !query 14
+SELECT i1 FROM t1, mydb1.t1
+-- !query 14 schema
+struct<>
+-- !query 14 output
+org.apache.spark.sql.AnalysisException
+Reference 'i1' is ambiguous, could be: i1#x, i1#x.; line 1 pos 7
+
+
+-- !query 15
+SELECT t1.i1 FROM t1, mydb1.t1
+-- !query 15 schema
+struct<>
+-- !query 15 output
+org.apache.spark.sql.AnalysisException
+Reference 't1.i1' is ambiguous, could be: i1#x, i1#x.; line 1 pos 7
+
+
+-- !query 16
+SELECT i1 FROM t1, mydb2.t1
+-- !query 16 schema
+struct<>
+-- !query 16 output
+org.apache.spark.sql.AnalysisException
+Reference 'i1' is ambiguous, could be: i1#x, i1#x.; line 1 pos 7
+
+
+-- !query 17
+SELECT t1.i1 FROM t1, mydb2.t1
+-- !query 17 schema
+struct<>
+-- !query 17 output
+org.apache.spark.sql.AnalysisException
+Reference 't1.i1' is ambiguous, could be: i1#x, i1#x.; line 1 pos 7
+
+
+-- !query 18
+SELECT db1.t1.i1 FROM t1, mydb2.t1
+-- !query 18 schema
+struct<>
+-- !query 18 output
+org.apache.spark.sql.AnalysisException
+cannot resolve '`db1.t1.i1`' given input columns: [i1, i1]; line 1 pos 7
+
+
+-- !query 19
+SET spark.sql.crossJoin.enabled = false
+-- !query 19 schema
+struct<key:string,value:string>
+-- !query 19 output
+spark.sql.crossJoin.enabled	false
+
+
+-- !query 20
+USE mydb1
+-- !query 20 schema
+struct<>
+-- !query 20 output
+
+
+
+-- !query 21
+SELECT mydb1.t1 FROM t1
+-- !query 21 schema
+struct<>
+-- !query 21 output
+org.apache.spark.sql.AnalysisException
+cannot resolve '`mydb1.t1`' given input columns: [i1]; line 1 pos 7
+
+
+-- !query 22
+SELECT t1.x.y.* FROM t1
+-- !query 22 schema
+struct<>
+-- !query 22 output
+org.apache.spark.sql.AnalysisException
+cannot resolve 't1.x.y.*' give input columns 'i1';
+
+
+-- !query 23
+SELECT t1 FROM mydb1.t1
+-- !query 23 schema
+struct<>
+-- !query 23 output
+org.apache.spark.sql.AnalysisException
+cannot resolve '`t1`' given input columns: [i1]; line 1 pos 7
+
+
+-- !query 24
+USE mydb2
+-- !query 24 schema
+struct<>
+-- !query 24 output
+
+
+
+-- !query 25
+SELECT mydb1.t1.i1 FROM t1
+-- !query 25 schema
+struct<>
+-- !query 25 output
+org.apache.spark.sql.AnalysisException
+cannot resolve '`mydb1.t1.i1`' given input columns: [i1]; line 1 pos 7
+
+
+-- !query 26
+DROP DATABASE mydb1 CASCADE
+-- !query 26 schema
+struct<>
+-- !query 26 output
+
+
+
+-- !query 27
+DROP DATABASE mydb2 CASCADE
+-- !query 27 schema
+struct<>
+-- !query 27 output
+