http://git-wip-us.apache.org/repos/asf/spark/blob/d181c2a1/sql/core/src/test/scala/org/apache/spark/sql/SQLQuerySuite.scala ---------------------------------------------------------------------- diff --git a/sql/core/src/test/scala/org/apache/spark/sql/SQLQuerySuite.scala b/sql/core/src/test/scala/org/apache/spark/sql/SQLQuerySuite.scala index 54fabc5..03b44ca 100644 --- a/sql/core/src/test/scala/org/apache/spark/sql/SQLQuerySuite.scala +++ b/sql/core/src/test/scala/org/apache/spark/sql/SQLQuerySuite.scala @@ -46,7 +46,7 @@ class SQLQuerySuite extends QueryTest with BeforeAndAfterAll { test("SPARK-4625 support SORT BY in SimpleSQLParser & DSL") { checkAnswer( sql("SELECT a FROM testData2 SORT BY a"), - Seq(1, 1, 2 ,2 ,3 ,3).map(Seq(_)) + Seq(1, 1, 2 ,2 ,3 ,3).map(Row(_)) ) } @@ -70,13 +70,13 @@ class SQLQuerySuite extends QueryTest with BeforeAndAfterAll { test("SPARK-3176 Added Parser of SQL ABS()") { checkAnswer( sql("SELECT ABS(-1.3)"), - 1.3) + Row(1.3)) checkAnswer( sql("SELECT ABS(0.0)"), - 0.0) + Row(0.0)) checkAnswer( sql("SELECT ABS(2.5)"), - 2.5) + Row(2.5)) } test("aggregation with codegen") { @@ -89,13 +89,13 @@ class SQLQuerySuite extends QueryTest with BeforeAndAfterAll { test("SPARK-3176 Added Parser of SQL LAST()") { checkAnswer( sql("SELECT LAST(n) FROM lowerCaseData"), - 4) + Row(4)) } test("SPARK-2041 column name equals tablename") { checkAnswer( sql("SELECT tableName FROM tableName"), - "test") + Row("test")) } test("SQRT") { @@ -115,40 +115,40 @@ class SQLQuerySuite extends QueryTest with BeforeAndAfterAll { test("SPARK-2407 Added Parser of SQL SUBSTR()") { checkAnswer( sql("SELECT substr(tableName, 1, 2) FROM tableName"), - "te") + Row("te")) checkAnswer( sql("SELECT substr(tableName, 3) FROM tableName"), - "st") + Row("st")) checkAnswer( sql("SELECT substring(tableName, 1, 2) FROM tableName"), - "te") + Row("te")) checkAnswer( sql("SELECT substring(tableName, 3) FROM tableName"), - "st") + Row("st")) } test("SPARK-3173 Timestamp support in the parser") { checkAnswer(sql( "SELECT time FROM timestamps WHERE time=CAST('1970-01-01 00:00:00.001' AS TIMESTAMP)"), - Seq(Seq(java.sql.Timestamp.valueOf("1970-01-01 00:00:00.001")))) + Row(java.sql.Timestamp.valueOf("1970-01-01 00:00:00.001"))) checkAnswer(sql( "SELECT time FROM timestamps WHERE time='1970-01-01 00:00:00.001'"), - Seq(Seq(java.sql.Timestamp.valueOf("1970-01-01 00:00:00.001")))) + Row(java.sql.Timestamp.valueOf("1970-01-01 00:00:00.001"))) checkAnswer(sql( "SELECT time FROM timestamps WHERE '1970-01-01 00:00:00.001'=time"), - Seq(Seq(java.sql.Timestamp.valueOf("1970-01-01 00:00:00.001")))) + Row(java.sql.Timestamp.valueOf("1970-01-01 00:00:00.001"))) checkAnswer(sql( """SELECT time FROM timestamps WHERE time<'1970-01-01 00:00:00.003' AND time>'1970-01-01 00:00:00.001'"""), - Seq(Seq(java.sql.Timestamp.valueOf("1970-01-01 00:00:00.002")))) + Row(java.sql.Timestamp.valueOf("1970-01-01 00:00:00.002"))) checkAnswer(sql( "SELECT time FROM timestamps WHERE time IN ('1970-01-01 00:00:00.001','1970-01-01 00:00:00.002')"), - Seq(Seq(java.sql.Timestamp.valueOf("1970-01-01 00:00:00.001")), - Seq(java.sql.Timestamp.valueOf("1970-01-01 00:00:00.002")))) + Seq(Row(java.sql.Timestamp.valueOf("1970-01-01 00:00:00.001")), + Row(java.sql.Timestamp.valueOf("1970-01-01 00:00:00.002")))) checkAnswer(sql( "SELECT time FROM timestamps WHERE time='123'"), @@ -158,13 +158,13 @@ class SQLQuerySuite extends QueryTest with BeforeAndAfterAll { test("index into array") { checkAnswer( sql("SELECT data, data[0], data[0] + data[1], data[0 + 1] FROM arrayData"), - arrayData.map(d => (d.data, d.data(0), d.data(0) + d.data(1), d.data(1))).collect().toSeq) + arrayData.map(d => Row(d.data, d.data(0), d.data(0) + d.data(1), d.data(1))).collect()) } test("left semi greater than predicate") { checkAnswer( sql("SELECT * FROM testData2 x LEFT SEMI JOIN testData2 y ON x.a >= y.a + 2"), - Seq((3,1), (3,2)) + Seq(Row(3,1), Row(3,2)) ) } @@ -173,7 +173,7 @@ class SQLQuerySuite extends QueryTest with BeforeAndAfterAll { sql( "SELECT nestedData, nestedData[0][0], nestedData[0][0] + nestedData[0][1] FROM arrayData"), arrayData.map(d => - (d.nestedData, + Row(d.nestedData, d.nestedData(0)(0), d.nestedData(0)(0) + d.nestedData(0)(1))).collect().toSeq) } @@ -181,13 +181,13 @@ class SQLQuerySuite extends QueryTest with BeforeAndAfterAll { test("agg") { checkAnswer( sql("SELECT a, SUM(b) FROM testData2 GROUP BY a"), - Seq((1,3),(2,3),(3,3))) + Seq(Row(1,3), Row(2,3), Row(3,3))) } test("aggregates with nulls") { checkAnswer( sql("SELECT MIN(a), MAX(a), AVG(a), SUM(a), COUNT(a) FROM nullInts"), - (1, 3, 2, 6, 3) :: Nil + Row(1, 3, 2, 6, 3) ) } @@ -200,29 +200,29 @@ class SQLQuerySuite extends QueryTest with BeforeAndAfterAll { test("simple select") { checkAnswer( sql("SELECT value FROM testData WHERE key = 1"), - Seq(Seq("1"))) + Row("1")) } def sortTest() = { checkAnswer( sql("SELECT * FROM testData2 ORDER BY a ASC, b ASC"), - Seq((1,1), (1,2), (2,1), (2,2), (3,1), (3,2))) + Seq(Row(1,1), Row(1,2), Row(2,1), Row(2,2), Row(3,1), Row(3,2))) checkAnswer( sql("SELECT * FROM testData2 ORDER BY a ASC, b DESC"), - Seq((1,2), (1,1), (2,2), (2,1), (3,2), (3,1))) + Seq(Row(1,2), Row(1,1), Row(2,2), Row(2,1), Row(3,2), Row(3,1))) checkAnswer( sql("SELECT * FROM testData2 ORDER BY a DESC, b DESC"), - Seq((3,2), (3,1), (2,2), (2,1), (1,2), (1,1))) + Seq(Row(3,2), Row(3,1), Row(2,2), Row(2,1), Row(1,2), Row(1,1))) checkAnswer( sql("SELECT * FROM testData2 ORDER BY a DESC, b ASC"), - Seq((3,1), (3,2), (2,1), (2,2), (1,1), (1,2))) + Seq(Row(3,1), Row(3,2), Row(2,1), Row(2,2), Row(1,1), Row(1,2))) checkAnswer( sql("SELECT b FROM binaryData ORDER BY a ASC"), - (1 to 5).map(Row(_)).toSeq) + (1 to 5).map(Row(_))) checkAnswer( sql("SELECT b FROM binaryData ORDER BY a DESC"), @@ -230,19 +230,19 @@ class SQLQuerySuite extends QueryTest with BeforeAndAfterAll { checkAnswer( sql("SELECT * FROM arrayData ORDER BY data[0] ASC"), - arrayData.collect().sortBy(_.data(0)).toSeq) + arrayData.collect().sortBy(_.data(0)).map(Row.fromTuple).toSeq) checkAnswer( sql("SELECT * FROM arrayData ORDER BY data[0] DESC"), - arrayData.collect().sortBy(_.data(0)).reverse.toSeq) + arrayData.collect().sortBy(_.data(0)).reverse.map(Row.fromTuple).toSeq) checkAnswer( sql("SELECT * FROM mapData ORDER BY data[1] ASC"), - mapData.collect().sortBy(_.data(1)).toSeq) + mapData.collect().sortBy(_.data(1)).map(Row.fromTuple).toSeq) checkAnswer( sql("SELECT * FROM mapData ORDER BY data[1] DESC"), - mapData.collect().sortBy(_.data(1)).reverse.toSeq) + mapData.collect().sortBy(_.data(1)).reverse.map(Row.fromTuple).toSeq) } test("sorting") { @@ -266,94 +266,94 @@ class SQLQuerySuite extends QueryTest with BeforeAndAfterAll { checkAnswer( sql("SELECT * FROM arrayData LIMIT 1"), - arrayData.collect().take(1).toSeq) + arrayData.collect().take(1).map(Row.fromTuple).toSeq) checkAnswer( sql("SELECT * FROM mapData LIMIT 1"), - mapData.collect().take(1).toSeq) + mapData.collect().take(1).map(Row.fromTuple).toSeq) } test("from follow multiple brackets") { checkAnswer(sql( "select key from ((select * from testData limit 1) union all (select * from testData limit 1)) x limit 1"), - 1 + Row(1) ) checkAnswer(sql( "select key from (select * from testData) x limit 1"), - 1 + Row(1) ) checkAnswer(sql( "select key from (select * from testData limit 1 union all select * from testData limit 1) x limit 1"), - 1 + Row(1) ) } test("average") { checkAnswer( sql("SELECT AVG(a) FROM testData2"), - 2.0) + Row(2.0)) } test("average overflow") { checkAnswer( sql("SELECT AVG(a),b FROM largeAndSmallInts group by b"), - Seq((2147483645.0,1),(2.0,2))) + Seq(Row(2147483645.0,1), Row(2.0,2))) } test("count") { checkAnswer( sql("SELECT COUNT(*) FROM testData2"), - testData2.count()) + Row(testData2.count())) } test("count distinct") { checkAnswer( sql("SELECT COUNT(DISTINCT b) FROM testData2"), - 2) + Row(2)) } test("approximate count distinct") { checkAnswer( sql("SELECT APPROXIMATE COUNT(DISTINCT a) FROM testData2"), - 3) + Row(3)) } test("approximate count distinct with user provided standard deviation") { checkAnswer( sql("SELECT APPROXIMATE(0.04) COUNT(DISTINCT a) FROM testData2"), - 3) + Row(3)) } test("null count") { checkAnswer( sql("SELECT a, COUNT(b) FROM testData3 GROUP BY a"), - Seq((1, 0), (2, 1))) + Seq(Row(1, 0), Row(2, 1))) checkAnswer( sql("SELECT COUNT(a), COUNT(b), COUNT(1), COUNT(DISTINCT a), COUNT(DISTINCT b) FROM testData3"), - (2, 1, 2, 2, 1) :: Nil) + Row(2, 1, 2, 2, 1)) } test("inner join where, one match per row") { checkAnswer( sql("SELECT * FROM upperCaseData JOIN lowerCaseData WHERE n = N"), Seq( - (1, "A", 1, "a"), - (2, "B", 2, "b"), - (3, "C", 3, "c"), - (4, "D", 4, "d"))) + Row(1, "A", 1, "a"), + Row(2, "B", 2, "b"), + Row(3, "C", 3, "c"), + Row(4, "D", 4, "d"))) } test("inner join ON, one match per row") { checkAnswer( sql("SELECT * FROM upperCaseData JOIN lowerCaseData ON n = N"), Seq( - (1, "A", 1, "a"), - (2, "B", 2, "b"), - (3, "C", 3, "c"), - (4, "D", 4, "d"))) + Row(1, "A", 1, "a"), + Row(2, "B", 2, "b"), + Row(3, "C", 3, "c"), + Row(4, "D", 4, "d"))) } test("inner join, where, multiple matches") { @@ -363,10 +363,10 @@ class SQLQuerySuite extends QueryTest with BeforeAndAfterAll { | (SELECT * FROM testData2 WHERE a = 1) x JOIN | (SELECT * FROM testData2 WHERE a = 1) y |WHERE x.a = y.a""".stripMargin), - (1,1,1,1) :: - (1,1,1,2) :: - (1,2,1,1) :: - (1,2,1,2) :: Nil) + Row(1,1,1,1) :: + Row(1,1,1,2) :: + Row(1,2,1,1) :: + Row(1,2,1,2) :: Nil) } test("inner join, no matches") { @@ -397,38 +397,38 @@ class SQLQuerySuite extends QueryTest with BeforeAndAfterAll { | SELECT * FROM testData) y |WHERE x.key = y.key""".stripMargin), testData.flatMap( - row => Seq.fill(16)((row ++ row).toSeq)).collect().toSeq) + row => Seq.fill(16)(Row.merge(row, row))).collect().toSeq) } ignore("cartesian product join") { checkAnswer( testData3.join(testData3), - (1, null, 1, null) :: - (1, null, 2, 2) :: - (2, 2, 1, null) :: - (2, 2, 2, 2) :: Nil) + Row(1, null, 1, null) :: + Row(1, null, 2, 2) :: + Row(2, 2, 1, null) :: + Row(2, 2, 2, 2) :: Nil) } test("left outer join") { checkAnswer( sql("SELECT * FROM upperCaseData LEFT OUTER JOIN lowerCaseData ON n = N"), - (1, "A", 1, "a") :: - (2, "B", 2, "b") :: - (3, "C", 3, "c") :: - (4, "D", 4, "d") :: - (5, "E", null, null) :: - (6, "F", null, null) :: Nil) + Row(1, "A", 1, "a") :: + Row(2, "B", 2, "b") :: + Row(3, "C", 3, "c") :: + Row(4, "D", 4, "d") :: + Row(5, "E", null, null) :: + Row(6, "F", null, null) :: Nil) } test("right outer join") { checkAnswer( sql("SELECT * FROM lowerCaseData RIGHT OUTER JOIN upperCaseData ON n = N"), - (1, "a", 1, "A") :: - (2, "b", 2, "B") :: - (3, "c", 3, "C") :: - (4, "d", 4, "D") :: - (null, null, 5, "E") :: - (null, null, 6, "F") :: Nil) + Row(1, "a", 1, "A") :: + Row(2, "b", 2, "B") :: + Row(3, "c", 3, "C") :: + Row(4, "d", 4, "D") :: + Row(null, null, 5, "E") :: + Row(null, null, 6, "F") :: Nil) } test("full outer join") { @@ -440,12 +440,12 @@ class SQLQuerySuite extends QueryTest with BeforeAndAfterAll { | (SELECT * FROM upperCaseData WHERE N >= 3) rightTable | ON leftTable.N = rightTable.N """.stripMargin), - (1, "A", null, null) :: - (2, "B", null, null) :: - (3, "C", 3, "C") :: - (4, "D", 4, "D") :: - (null, null, 5, "E") :: - (null, null, 6, "F") :: Nil) + Row(1, "A", null, null) :: + Row(2, "B", null, null) :: + Row(3, "C", 3, "C") :: + Row (4, "D", 4, "D") :: + Row(null, null, 5, "E") :: + Row(null, null, 6, "F") :: Nil) } test("SPARK-3349 partitioning after limit") { @@ -457,12 +457,12 @@ class SQLQuerySuite extends QueryTest with BeforeAndAfterAll { .registerTempTable("subset2") checkAnswer( sql("SELECT * FROM lowerCaseData INNER JOIN subset1 ON subset1.n = lowerCaseData.n"), - (3, "c", 3) :: - (4, "d", 4) :: Nil) + Row(3, "c", 3) :: + Row(4, "d", 4) :: Nil) checkAnswer( sql("SELECT * FROM lowerCaseData INNER JOIN subset2 ON subset2.n = lowerCaseData.n"), - (1, "a", 1) :: - (2, "b", 2) :: Nil) + Row(1, "a", 1) :: + Row(2, "b", 2) :: Nil) } test("mixed-case keywords") { @@ -474,28 +474,28 @@ class SQLQuerySuite extends QueryTest with BeforeAndAfterAll { | (sElEcT * FROM upperCaseData whERe N >= 3) rightTable | oN leftTable.N = rightTable.N """.stripMargin), - (1, "A", null, null) :: - (2, "B", null, null) :: - (3, "C", 3, "C") :: - (4, "D", 4, "D") :: - (null, null, 5, "E") :: - (null, null, 6, "F") :: Nil) + Row(1, "A", null, null) :: + Row(2, "B", null, null) :: + Row(3, "C", 3, "C") :: + Row(4, "D", 4, "D") :: + Row(null, null, 5, "E") :: + Row(null, null, 6, "F") :: Nil) } test("select with table name as qualifier") { checkAnswer( sql("SELECT testData.value FROM testData WHERE testData.key = 1"), - Seq(Seq("1"))) + Row("1")) } test("inner join ON with table name as qualifier") { checkAnswer( sql("SELECT * FROM upperCaseData JOIN lowerCaseData ON lowerCaseData.n = upperCaseData.N"), Seq( - (1, "A", 1, "a"), - (2, "B", 2, "b"), - (3, "C", 3, "c"), - (4, "D", 4, "d"))) + Row(1, "A", 1, "a"), + Row(2, "B", 2, "b"), + Row(3, "C", 3, "c"), + Row(4, "D", 4, "d"))) } test("qualified select with inner join ON with table name as qualifier") { @@ -503,72 +503,72 @@ class SQLQuerySuite extends QueryTest with BeforeAndAfterAll { sql("SELECT upperCaseData.N, upperCaseData.L FROM upperCaseData JOIN lowerCaseData " + "ON lowerCaseData.n = upperCaseData.N"), Seq( - (1, "A"), - (2, "B"), - (3, "C"), - (4, "D"))) + Row(1, "A"), + Row(2, "B"), + Row(3, "C"), + Row(4, "D"))) } test("system function upper()") { checkAnswer( sql("SELECT n,UPPER(l) FROM lowerCaseData"), Seq( - (1, "A"), - (2, "B"), - (3, "C"), - (4, "D"))) + Row(1, "A"), + Row(2, "B"), + Row(3, "C"), + Row(4, "D"))) checkAnswer( sql("SELECT n, UPPER(s) FROM nullStrings"), Seq( - (1, "ABC"), - (2, "ABC"), - (3, null))) + Row(1, "ABC"), + Row(2, "ABC"), + Row(3, null))) } test("system function lower()") { checkAnswer( sql("SELECT N,LOWER(L) FROM upperCaseData"), Seq( - (1, "a"), - (2, "b"), - (3, "c"), - (4, "d"), - (5, "e"), - (6, "f"))) + Row(1, "a"), + Row(2, "b"), + Row(3, "c"), + Row(4, "d"), + Row(5, "e"), + Row(6, "f"))) checkAnswer( sql("SELECT n, LOWER(s) FROM nullStrings"), Seq( - (1, "abc"), - (2, "abc"), - (3, null))) + Row(1, "abc"), + Row(2, "abc"), + Row(3, null))) } test("UNION") { checkAnswer( sql("SELECT * FROM lowerCaseData UNION SELECT * FROM upperCaseData"), - (1, "A") :: (1, "a") :: (2, "B") :: (2, "b") :: (3, "C") :: (3, "c") :: - (4, "D") :: (4, "d") :: (5, "E") :: (6, "F") :: Nil) + Row(1, "A") :: Row(1, "a") :: Row(2, "B") :: Row(2, "b") :: Row(3, "C") :: Row(3, "c") :: + Row(4, "D") :: Row(4, "d") :: Row(5, "E") :: Row(6, "F") :: Nil) checkAnswer( sql("SELECT * FROM lowerCaseData UNION SELECT * FROM lowerCaseData"), - (1, "a") :: (2, "b") :: (3, "c") :: (4, "d") :: Nil) + Row(1, "a") :: Row(2, "b") :: Row(3, "c") :: Row(4, "d") :: Nil) checkAnswer( sql("SELECT * FROM lowerCaseData UNION ALL SELECT * FROM lowerCaseData"), - (1, "a") :: (1, "a") :: (2, "b") :: (2, "b") :: (3, "c") :: (3, "c") :: - (4, "d") :: (4, "d") :: Nil) + Row(1, "a") :: Row(1, "a") :: Row(2, "b") :: Row(2, "b") :: Row(3, "c") :: Row(3, "c") :: + Row(4, "d") :: Row(4, "d") :: Nil) } test("UNION with column mismatches") { // Column name mismatches are allowed. checkAnswer( sql("SELECT n,l FROM lowerCaseData UNION SELECT N as x1, L as x2 FROM upperCaseData"), - (1, "A") :: (1, "a") :: (2, "B") :: (2, "b") :: (3, "C") :: (3, "c") :: - (4, "D") :: (4, "d") :: (5, "E") :: (6, "F") :: Nil) + Row(1, "A") :: Row(1, "a") :: Row(2, "B") :: Row(2, "b") :: Row(3, "C") :: Row(3, "c") :: + Row(4, "D") :: Row(4, "d") :: Row(5, "E") :: Row(6, "F") :: Nil) // Column type mismatches are not allowed, forcing a type coercion. checkAnswer( sql("SELECT n FROM lowerCaseData UNION SELECT L FROM upperCaseData"), - ("1" :: "2" :: "3" :: "4" :: "A" :: "B" :: "C" :: "D" :: "E" :: "F" :: Nil).map(Tuple1(_))) + ("1" :: "2" :: "3" :: "4" :: "A" :: "B" :: "C" :: "D" :: "E" :: "F" :: Nil).map(Row(_))) // Column type mismatches where a coercion is not possible, in this case between integer // and array types, trigger a TreeNodeException. intercept[TreeNodeException[_]] { @@ -579,10 +579,10 @@ class SQLQuerySuite extends QueryTest with BeforeAndAfterAll { test("EXCEPT") { checkAnswer( sql("SELECT * FROM lowerCaseData EXCEPT SELECT * FROM upperCaseData"), - (1, "a") :: - (2, "b") :: - (3, "c") :: - (4, "d") :: Nil) + Row(1, "a") :: + Row(2, "b") :: + Row(3, "c") :: + Row(4, "d") :: Nil) checkAnswer( sql("SELECT * FROM lowerCaseData EXCEPT SELECT * FROM lowerCaseData"), Nil) checkAnswer( @@ -592,10 +592,10 @@ class SQLQuerySuite extends QueryTest with BeforeAndAfterAll { test("INTERSECT") { checkAnswer( sql("SELECT * FROM lowerCaseData INTERSECT SELECT * FROM lowerCaseData"), - (1, "a") :: - (2, "b") :: - (3, "c") :: - (4, "d") :: Nil) + Row(1, "a") :: + Row(2, "b") :: + Row(3, "c") :: + Row(4, "d") :: Nil) checkAnswer( sql("SELECT * FROM lowerCaseData INTERSECT SELECT * FROM upperCaseData"), Nil) } @@ -613,25 +613,25 @@ class SQLQuerySuite extends QueryTest with BeforeAndAfterAll { sql(s"SET $testKey=$testVal") checkAnswer( sql("SET"), - Seq(Seq(s"$testKey=$testVal")) + Row(s"$testKey=$testVal") ) sql(s"SET ${testKey + testKey}=${testVal + testVal}") checkAnswer( sql("set"), Seq( - Seq(s"$testKey=$testVal"), - Seq(s"${testKey + testKey}=${testVal + testVal}")) + Row(s"$testKey=$testVal"), + Row(s"${testKey + testKey}=${testVal + testVal}")) ) // "set key" checkAnswer( sql(s"SET $testKey"), - Seq(Seq(s"$testKey=$testVal")) + Row(s"$testKey=$testVal") ) checkAnswer( sql(s"SET $nonexistentKey"), - Seq(Seq(s"$nonexistentKey=<undefined>")) + Row(s"$nonexistentKey=<undefined>") ) conf.clear() } @@ -655,17 +655,17 @@ class SQLQuerySuite extends QueryTest with BeforeAndAfterAll { schemaRDD1.registerTempTable("applySchema1") checkAnswer( sql("SELECT * FROM applySchema1"), - (1, "A1", true, null) :: - (2, "B2", false, null) :: - (3, "C3", true, null) :: - (4, "D4", true, 2147483644) :: Nil) + Row(1, "A1", true, null) :: + Row(2, "B2", false, null) :: + Row(3, "C3", true, null) :: + Row(4, "D4", true, 2147483644) :: Nil) checkAnswer( sql("SELECT f1, f4 FROM applySchema1"), - (1, null) :: - (2, null) :: - (3, null) :: - (4, 2147483644) :: Nil) + Row(1, null) :: + Row(2, null) :: + Row(3, null) :: + Row(4, 2147483644) :: Nil) val schema2 = StructType( StructField("f1", StructType( @@ -685,17 +685,17 @@ class SQLQuerySuite extends QueryTest with BeforeAndAfterAll { schemaRDD2.registerTempTable("applySchema2") checkAnswer( sql("SELECT * FROM applySchema2"), - (Seq(1, true), Map("A1" -> null)) :: - (Seq(2, false), Map("B2" -> null)) :: - (Seq(3, true), Map("C3" -> null)) :: - (Seq(4, true), Map("D4" -> 2147483644)) :: Nil) + Row(Row(1, true), Map("A1" -> null)) :: + Row(Row(2, false), Map("B2" -> null)) :: + Row(Row(3, true), Map("C3" -> null)) :: + Row(Row(4, true), Map("D4" -> 2147483644)) :: Nil) checkAnswer( sql("SELECT f1.f11, f2['D4'] FROM applySchema2"), - (1, null) :: - (2, null) :: - (3, null) :: - (4, 2147483644) :: Nil) + Row(1, null) :: + Row(2, null) :: + Row(3, null) :: + Row(4, 2147483644) :: Nil) // The value of a MapType column can be a mutable map. val rowRDD3 = unparsedStrings.map { r => @@ -711,26 +711,26 @@ class SQLQuerySuite extends QueryTest with BeforeAndAfterAll { checkAnswer( sql("SELECT f1.f11, f2['D4'] FROM applySchema3"), - (1, null) :: - (2, null) :: - (3, null) :: - (4, 2147483644) :: Nil) + Row(1, null) :: + Row(2, null) :: + Row(3, null) :: + Row(4, 2147483644) :: Nil) } test("SPARK-3423 BETWEEN") { checkAnswer( sql("SELECT key, value FROM testData WHERE key BETWEEN 5 and 7"), - Seq((5, "5"), (6, "6"), (7, "7")) + Seq(Row(5, "5"), Row(6, "6"), Row(7, "7")) ) checkAnswer( sql("SELECT key, value FROM testData WHERE key BETWEEN 7 and 7"), - Seq((7, "7")) + Row(7, "7") ) checkAnswer( sql("SELECT key, value FROM testData WHERE key BETWEEN 9 and 7"), - Seq() + Nil ) } @@ -738,7 +738,7 @@ class SQLQuerySuite extends QueryTest with BeforeAndAfterAll { // TODO Ensure true/false string letter casing is consistent with Hive in all cases. checkAnswer( sql("SELECT CAST(TRUE AS STRING), CAST(FALSE AS STRING) FROM testData LIMIT 1"), - ("true", "false") :: Nil) + Row("true", "false")) } test("metadata is propagated correctly") { @@ -768,17 +768,20 @@ class SQLQuerySuite extends QueryTest with BeforeAndAfterAll { test("SPARK-3371 Renaming a function expression with group by gives error") { udf.register("len", (s: String) => s.length) checkAnswer( - sql("SELECT len(value) as temp FROM testData WHERE key = 1 group by len(value)"), 1) + sql("SELECT len(value) as temp FROM testData WHERE key = 1 group by len(value)"), + Row(1)) } test("SPARK-3813 CASE a WHEN b THEN c [WHEN d THEN e]* [ELSE f] END") { checkAnswer( - sql("SELECT CASE key WHEN 1 THEN 1 ELSE 0 END FROM testData WHERE key = 1 group by key"), 1) + sql("SELECT CASE key WHEN 1 THEN 1 ELSE 0 END FROM testData WHERE key = 1 group by key"), + Row(1)) } test("SPARK-3813 CASE WHEN a THEN b [WHEN c THEN d]* [ELSE e] END") { checkAnswer( - sql("SELECT CASE WHEN key = 1 THEN 1 ELSE 2 END FROM testData WHERE key = 1 group by key"), 1) + sql("SELECT CASE WHEN key = 1 THEN 1 ELSE 2 END FROM testData WHERE key = 1 group by key"), + Row(1)) } test("throw errors for non-aggregate attributes with aggregation") { @@ -808,130 +811,131 @@ class SQLQuerySuite extends QueryTest with BeforeAndAfterAll { test("Test to check we can use Long.MinValue") { checkAnswer( - sql(s"SELECT ${Long.MinValue} FROM testData ORDER BY key LIMIT 1"), Long.MinValue + sql(s"SELECT ${Long.MinValue} FROM testData ORDER BY key LIMIT 1"), Row(Long.MinValue) ) checkAnswer( - sql(s"SELECT key FROM testData WHERE key > ${Long.MinValue}"), (1 to 100).map(Row(_)).toSeq + sql(s"SELECT key FROM testData WHERE key > ${Long.MinValue}"), + (1 to 100).map(Row(_)).toSeq ) } test("Floating point number format") { checkAnswer( - sql("SELECT 0.3"), 0.3 + sql("SELECT 0.3"), Row(0.3) ) checkAnswer( - sql("SELECT -0.8"), -0.8 + sql("SELECT -0.8"), Row(-0.8) ) checkAnswer( - sql("SELECT .5"), 0.5 + sql("SELECT .5"), Row(0.5) ) checkAnswer( - sql("SELECT -.18"), -0.18 + sql("SELECT -.18"), Row(-0.18) ) } test("Auto cast integer type") { checkAnswer( - sql(s"SELECT ${Int.MaxValue + 1L}"), Int.MaxValue + 1L + sql(s"SELECT ${Int.MaxValue + 1L}"), Row(Int.MaxValue + 1L) ) checkAnswer( - sql(s"SELECT ${Int.MinValue - 1L}"), Int.MinValue - 1L + sql(s"SELECT ${Int.MinValue - 1L}"), Row(Int.MinValue - 1L) ) checkAnswer( - sql("SELECT 9223372036854775808"), new java.math.BigDecimal("9223372036854775808") + sql("SELECT 9223372036854775808"), Row(new java.math.BigDecimal("9223372036854775808")) ) checkAnswer( - sql("SELECT -9223372036854775809"), new java.math.BigDecimal("-9223372036854775809") + sql("SELECT -9223372036854775809"), Row(new java.math.BigDecimal("-9223372036854775809")) ) } test("Test to check we can apply sign to expression") { checkAnswer( - sql("SELECT -100"), -100 + sql("SELECT -100"), Row(-100) ) checkAnswer( - sql("SELECT +230"), 230 + sql("SELECT +230"), Row(230) ) checkAnswer( - sql("SELECT -5.2"), -5.2 + sql("SELECT -5.2"), Row(-5.2) ) checkAnswer( - sql("SELECT +6.8"), 6.8 + sql("SELECT +6.8"), Row(6.8) ) checkAnswer( - sql("SELECT -key FROM testData WHERE key = 2"), -2 + sql("SELECT -key FROM testData WHERE key = 2"), Row(-2) ) checkAnswer( - sql("SELECT +key FROM testData WHERE key = 3"), 3 + sql("SELECT +key FROM testData WHERE key = 3"), Row(3) ) checkAnswer( - sql("SELECT -(key + 1) FROM testData WHERE key = 1"), -2 + sql("SELECT -(key + 1) FROM testData WHERE key = 1"), Row(-2) ) checkAnswer( - sql("SELECT - key + 1 FROM testData WHERE key = 10"), -9 + sql("SELECT - key + 1 FROM testData WHERE key = 10"), Row(-9) ) checkAnswer( - sql("SELECT +(key + 5) FROM testData WHERE key = 5"), 10 + sql("SELECT +(key + 5) FROM testData WHERE key = 5"), Row(10) ) checkAnswer( - sql("SELECT -MAX(key) FROM testData"), -100 + sql("SELECT -MAX(key) FROM testData"), Row(-100) ) checkAnswer( - sql("SELECT +MAX(key) FROM testData"), 100 + sql("SELECT +MAX(key) FROM testData"), Row(100) ) checkAnswer( - sql("SELECT - (-10)"), 10 + sql("SELECT - (-10)"), Row(10) ) checkAnswer( - sql("SELECT + (-key) FROM testData WHERE key = 32"), -32 + sql("SELECT + (-key) FROM testData WHERE key = 32"), Row(-32) ) checkAnswer( - sql("SELECT - (+Max(key)) FROM testData"), -100 + sql("SELECT - (+Max(key)) FROM testData"), Row(-100) ) checkAnswer( - sql("SELECT - - 3"), 3 + sql("SELECT - - 3"), Row(3) ) checkAnswer( - sql("SELECT - + 20"), -20 + sql("SELECT - + 20"), Row(-20) ) checkAnswer( - sql("SELEcT - + 45"), -45 + sql("SELEcT - + 45"), Row(-45) ) checkAnswer( - sql("SELECT + + 100"), 100 + sql("SELECT + + 100"), Row(100) ) checkAnswer( - sql("SELECT - - Max(key) FROM testData"), 100 + sql("SELECT - - Max(key) FROM testData"), Row(100) ) checkAnswer( - sql("SELECT + - key FROM testData WHERE key = 33"), -33 + sql("SELECT + - key FROM testData WHERE key = 33"), Row(-33) ) } @@ -943,7 +947,7 @@ class SQLQuerySuite extends QueryTest with BeforeAndAfterAll { |JOIN testData b ON a.key = b.key |JOIN testData c ON a.key = c.key """.stripMargin), - (1 to 100).map(i => Seq(i, i, i))) + (1 to 100).map(i => Row(i, i, i))) } test("SPARK-3483 Special chars in column names") { @@ -953,19 +957,19 @@ class SQLQuerySuite extends QueryTest with BeforeAndAfterAll { } test("SPARK-3814 Support Bitwise & operator") { - checkAnswer(sql("SELECT key&1 FROM testData WHERE key = 1 "), 1) + checkAnswer(sql("SELECT key&1 FROM testData WHERE key = 1 "), Row(1)) } test("SPARK-3814 Support Bitwise | operator") { - checkAnswer(sql("SELECT key|0 FROM testData WHERE key = 1 "), 1) + checkAnswer(sql("SELECT key|0 FROM testData WHERE key = 1 "), Row(1)) } test("SPARK-3814 Support Bitwise ^ operator") { - checkAnswer(sql("SELECT key^0 FROM testData WHERE key = 1 "), 1) + checkAnswer(sql("SELECT key^0 FROM testData WHERE key = 1 "), Row(1)) } test("SPARK-3814 Support Bitwise ~ operator") { - checkAnswer(sql("SELECT ~key FROM testData WHERE key = 1 "), -2) + checkAnswer(sql("SELECT ~key FROM testData WHERE key = 1 "), Row(-2)) } test("SPARK-4120 Join of multiple tables does not work in SparkSQL") { @@ -975,40 +979,40 @@ class SQLQuerySuite extends QueryTest with BeforeAndAfterAll { |FROM testData a,testData b,testData c |where a.key = b.key and a.key = c.key """.stripMargin), - (1 to 100).map(i => Seq(i, i, i))) + (1 to 100).map(i => Row(i, i, i))) } test("SPARK-4154 Query does not work if it has 'not between' in Spark SQL and HQL") { checkAnswer(sql("SELECT key FROM testData WHERE key not between 0 and 10 order by key"), - (11 to 100).map(i => Seq(i))) + (11 to 100).map(i => Row(i))) } test("SPARK-4207 Query which has syntax like 'not like' is not working in Spark SQL") { checkAnswer(sql("SELECT key FROM testData WHERE value not like '100%' order by key"), - (1 to 99).map(i => Seq(i))) + (1 to 99).map(i => Row(i))) } test("SPARK-4322 Grouping field with struct field as sub expression") { jsonRDD(sparkContext.makeRDD("""{"a": {"b": [{"c": 1}]}}""" :: Nil)).registerTempTable("data") - checkAnswer(sql("SELECT a.b[0].c FROM data GROUP BY a.b[0].c"), 1) + checkAnswer(sql("SELECT a.b[0].c FROM data GROUP BY a.b[0].c"), Row(1)) dropTempTable("data") jsonRDD(sparkContext.makeRDD("""{"a": {"b": 1}}""" :: Nil)).registerTempTable("data") - checkAnswer(sql("SELECT a.b + 1 FROM data GROUP BY a.b + 1"), 2) + checkAnswer(sql("SELECT a.b + 1 FROM data GROUP BY a.b + 1"), Row(2)) dropTempTable("data") } test("SPARK-4432 Fix attribute reference resolution error when using ORDER BY") { checkAnswer( sql("SELECT a + b FROM testData2 ORDER BY a"), - Seq(2, 3, 3 ,4 ,4 ,5).map(Seq(_)) + Seq(2, 3, 3 ,4 ,4 ,5).map(Row(_)) ) } test("oder by asc by default when not specify ascending and descending") { checkAnswer( sql("SELECT a, b FROM testData2 ORDER BY a desc, b"), - Seq((3, 1), (3, 2), (2, 1), (2,2), (1, 1), (1, 2)) + Seq(Row(3, 1), Row(3, 2), Row(2, 1), Row(2,2), Row(1, 1), Row(1, 2)) ) } @@ -1021,13 +1025,13 @@ class SQLQuerySuite extends QueryTest with BeforeAndAfterAll { rdd2.registerTempTable("nulldata2") checkAnswer(sql("SELECT nulldata1.key FROM nulldata1 join " + "nulldata2 on nulldata1.value <=> nulldata2.value"), - (1 to 2).map(i => Seq(i))) + (1 to 2).map(i => Row(i))) } test("Multi-column COUNT(DISTINCT ...)") { val data = TestData(1,"val_1") :: TestData(2,"val_2") :: Nil val rdd = sparkContext.parallelize((0 to 1).map(i => data(i))) rdd.registerTempTable("distinctData") - checkAnswer(sql("SELECT COUNT(DISTINCT key,value) FROM distinctData"), 2) + checkAnswer(sql("SELECT COUNT(DISTINCT key,value) FROM distinctData"), Row(2)) } }
http://git-wip-us.apache.org/repos/asf/spark/blob/d181c2a1/sql/core/src/test/scala/org/apache/spark/sql/ScalaReflectionRelationSuite.scala ---------------------------------------------------------------------- diff --git a/sql/core/src/test/scala/org/apache/spark/sql/ScalaReflectionRelationSuite.scala b/sql/core/src/test/scala/org/apache/spark/sql/ScalaReflectionRelationSuite.scala index ee381da..a015884 100644 --- a/sql/core/src/test/scala/org/apache/spark/sql/ScalaReflectionRelationSuite.scala +++ b/sql/core/src/test/scala/org/apache/spark/sql/ScalaReflectionRelationSuite.scala @@ -82,7 +82,7 @@ class ScalaReflectionRelationSuite extends FunSuite { rdd.registerTempTable("reflectData") assert(sql("SELECT * FROM reflectData").collect().head === - Seq("a", 1, 1L, 1.toFloat, 1.toDouble, 1.toShort, 1.toByte, true, + Row("a", 1, 1L, 1.toFloat, 1.toDouble, 1.toShort, 1.toByte, true, new java.math.BigDecimal(1), new Date(12345), new Timestamp(12345), Seq(1,2,3))) } @@ -91,7 +91,7 @@ class ScalaReflectionRelationSuite extends FunSuite { val rdd = sparkContext.parallelize(data :: Nil) rdd.registerTempTable("reflectNullData") - assert(sql("SELECT * FROM reflectNullData").collect().head === Seq.fill(7)(null)) + assert(sql("SELECT * FROM reflectNullData").collect().head === Row.fromSeq(Seq.fill(7)(null))) } test("query case class RDD with Nones") { @@ -99,7 +99,7 @@ class ScalaReflectionRelationSuite extends FunSuite { val rdd = sparkContext.parallelize(data :: Nil) rdd.registerTempTable("reflectOptionalData") - assert(sql("SELECT * FROM reflectOptionalData").collect().head === Seq.fill(7)(null)) + assert(sql("SELECT * FROM reflectOptionalData").collect().head === Row.fromSeq(Seq.fill(7)(null))) } // Equality is broken for Arrays, so we test that separately. http://git-wip-us.apache.org/repos/asf/spark/blob/d181c2a1/sql/core/src/test/scala/org/apache/spark/sql/columnar/ColumnStatsSuite.scala ---------------------------------------------------------------------- diff --git a/sql/core/src/test/scala/org/apache/spark/sql/columnar/ColumnStatsSuite.scala b/sql/core/src/test/scala/org/apache/spark/sql/columnar/ColumnStatsSuite.scala index 9be0b38..be2b34d 100644 --- a/sql/core/src/test/scala/org/apache/spark/sql/columnar/ColumnStatsSuite.scala +++ b/sql/core/src/test/scala/org/apache/spark/sql/columnar/ColumnStatsSuite.scala @@ -42,8 +42,8 @@ class ColumnStatsSuite extends FunSuite { test(s"$columnStatsName: empty") { val columnStats = columnStatsClass.newInstance() - columnStats.collectedStatistics.zip(initialStatistics).foreach { case (actual, expected) => - assert(actual === expected) + columnStats.collectedStatistics.toSeq.zip(initialStatistics.toSeq).foreach { + case (actual, expected) => assert(actual === expected) } } @@ -54,7 +54,7 @@ class ColumnStatsSuite extends FunSuite { val rows = Seq.fill(10)(makeRandomRow(columnType)) ++ Seq.fill(10)(makeNullRow(1)) rows.foreach(columnStats.gatherStats(_, 0)) - val values = rows.take(10).map(_.head.asInstanceOf[T#JvmType]) + val values = rows.take(10).map(_(0).asInstanceOf[T#JvmType]) val ordering = columnType.dataType.ordering.asInstanceOf[Ordering[T#JvmType]] val stats = columnStats.collectedStatistics http://git-wip-us.apache.org/repos/asf/spark/blob/d181c2a1/sql/core/src/test/scala/org/apache/spark/sql/columnar/InMemoryColumnarQuerySuite.scala ---------------------------------------------------------------------- diff --git a/sql/core/src/test/scala/org/apache/spark/sql/columnar/InMemoryColumnarQuerySuite.scala b/sql/core/src/test/scala/org/apache/spark/sql/columnar/InMemoryColumnarQuerySuite.scala index d94729b..e61f3c3 100644 --- a/sql/core/src/test/scala/org/apache/spark/sql/columnar/InMemoryColumnarQuerySuite.scala +++ b/sql/core/src/test/scala/org/apache/spark/sql/columnar/InMemoryColumnarQuerySuite.scala @@ -49,7 +49,7 @@ class InMemoryColumnarQuerySuite extends QueryTest { checkAnswer(scan, testData.collect().map { case Row(key: Int, value: String) => value -> key - }.toSeq) + }.map(Row.fromTuple)) } test("SPARK-1436 regression: in-memory columns must be able to be accessed multiple times") { @@ -63,49 +63,49 @@ class InMemoryColumnarQuerySuite extends QueryTest { test("SPARK-1678 regression: compression must not lose repeated values") { checkAnswer( sql("SELECT * FROM repeatedData"), - repeatedData.collect().toSeq) + repeatedData.collect().toSeq.map(Row.fromTuple)) cacheTable("repeatedData") checkAnswer( sql("SELECT * FROM repeatedData"), - repeatedData.collect().toSeq) + repeatedData.collect().toSeq.map(Row.fromTuple)) } test("with null values") { checkAnswer( sql("SELECT * FROM nullableRepeatedData"), - nullableRepeatedData.collect().toSeq) + nullableRepeatedData.collect().toSeq.map(Row.fromTuple)) cacheTable("nullableRepeatedData") checkAnswer( sql("SELECT * FROM nullableRepeatedData"), - nullableRepeatedData.collect().toSeq) + nullableRepeatedData.collect().toSeq.map(Row.fromTuple)) } test("SPARK-2729 regression: timestamp data type") { checkAnswer( sql("SELECT time FROM timestamps"), - timestamps.collect().toSeq) + timestamps.collect().toSeq.map(Row.fromTuple)) cacheTable("timestamps") checkAnswer( sql("SELECT time FROM timestamps"), - timestamps.collect().toSeq) + timestamps.collect().toSeq.map(Row.fromTuple)) } test("SPARK-3320 regression: batched column buffer building should work with empty partitions") { checkAnswer( sql("SELECT * FROM withEmptyParts"), - withEmptyParts.collect().toSeq) + withEmptyParts.collect().toSeq.map(Row.fromTuple)) cacheTable("withEmptyParts") checkAnswer( sql("SELECT * FROM withEmptyParts"), - withEmptyParts.collect().toSeq) + withEmptyParts.collect().toSeq.map(Row.fromTuple)) } test("SPARK-4182 Caching complex types") { http://git-wip-us.apache.org/repos/asf/spark/blob/d181c2a1/sql/core/src/test/scala/org/apache/spark/sql/columnar/PartitionBatchPruningSuite.scala ---------------------------------------------------------------------- diff --git a/sql/core/src/test/scala/org/apache/spark/sql/columnar/PartitionBatchPruningSuite.scala b/sql/core/src/test/scala/org/apache/spark/sql/columnar/PartitionBatchPruningSuite.scala index 592cafb..c3a3f8d 100644 --- a/sql/core/src/test/scala/org/apache/spark/sql/columnar/PartitionBatchPruningSuite.scala +++ b/sql/core/src/test/scala/org/apache/spark/sql/columnar/PartitionBatchPruningSuite.scala @@ -108,7 +108,7 @@ class PartitionBatchPruningSuite extends FunSuite with BeforeAndAfterAll with Be val queryExecution = schemaRdd.queryExecution assertResult(expectedQueryResult.toArray, s"Wrong query result: $queryExecution") { - schemaRdd.collect().map(_.head).toArray + schemaRdd.collect().map(_(0)).toArray } val (readPartitions, readBatches) = schemaRdd.queryExecution.executedPlan.collect { http://git-wip-us.apache.org/repos/asf/spark/blob/d181c2a1/sql/core/src/test/scala/org/apache/spark/sql/columnar/compression/BooleanBitSetSuite.scala ---------------------------------------------------------------------- diff --git a/sql/core/src/test/scala/org/apache/spark/sql/columnar/compression/BooleanBitSetSuite.scala b/sql/core/src/test/scala/org/apache/spark/sql/columnar/compression/BooleanBitSetSuite.scala index d9e488e..8b518f0 100644 --- a/sql/core/src/test/scala/org/apache/spark/sql/columnar/compression/BooleanBitSetSuite.scala +++ b/sql/core/src/test/scala/org/apache/spark/sql/columnar/compression/BooleanBitSetSuite.scala @@ -34,7 +34,7 @@ class BooleanBitSetSuite extends FunSuite { val builder = TestCompressibleColumnBuilder(new NoopColumnStats, BOOLEAN, BooleanBitSet) val rows = Seq.fill[Row](count)(makeRandomRow(BOOLEAN)) - val values = rows.map(_.head) + val values = rows.map(_(0)) rows.foreach(builder.appendFrom(_, 0)) val buffer = builder.build() http://git-wip-us.apache.org/repos/asf/spark/blob/d181c2a1/sql/core/src/test/scala/org/apache/spark/sql/execution/TgfSuite.scala ---------------------------------------------------------------------- diff --git a/sql/core/src/test/scala/org/apache/spark/sql/execution/TgfSuite.scala b/sql/core/src/test/scala/org/apache/spark/sql/execution/TgfSuite.scala index 2cab5e0..272c0d4 100644 --- a/sql/core/src/test/scala/org/apache/spark/sql/execution/TgfSuite.scala +++ b/sql/core/src/test/scala/org/apache/spark/sql/execution/TgfSuite.scala @@ -59,7 +59,7 @@ class TgfSuite extends QueryTest { checkAnswer( inputData.generate(ExampleTGF()), Seq( - "michael is 29 years old" :: Nil, - "Next year, michael will be 30 years old" :: Nil)) + Row("michael is 29 years old"), + Row("Next year, michael will be 30 years old"))) } } http://git-wip-us.apache.org/repos/asf/spark/blob/d181c2a1/sql/core/src/test/scala/org/apache/spark/sql/json/JsonSuite.scala ---------------------------------------------------------------------- diff --git a/sql/core/src/test/scala/org/apache/spark/sql/json/JsonSuite.scala b/sql/core/src/test/scala/org/apache/spark/sql/json/JsonSuite.scala index 2bc9aed..94d14ac 100644 --- a/sql/core/src/test/scala/org/apache/spark/sql/json/JsonSuite.scala +++ b/sql/core/src/test/scala/org/apache/spark/sql/json/JsonSuite.scala @@ -229,13 +229,13 @@ class JsonSuite extends QueryTest { checkAnswer( sql("select * from jsonTable"), - (new java.math.BigDecimal("92233720368547758070"), - true, - 1.7976931348623157E308, - 10, - 21474836470L, - null, - "this is a simple string.") :: Nil + Row(new java.math.BigDecimal("92233720368547758070"), + true, + 1.7976931348623157E308, + 10, + 21474836470L, + null, + "this is a simple string.") ) } @@ -271,48 +271,49 @@ class JsonSuite extends QueryTest { // Access elements of a primitive array. checkAnswer( sql("select arrayOfString[0], arrayOfString[1], arrayOfString[2] from jsonTable"), - ("str1", "str2", null) :: Nil + Row("str1", "str2", null) ) // Access an array of null values. checkAnswer( sql("select arrayOfNull from jsonTable"), - Seq(Seq(null, null, null, null)) :: Nil + Row(Seq(null, null, null, null)) ) // Access elements of a BigInteger array (we use DecimalType internally). checkAnswer( sql("select arrayOfBigInteger[0], arrayOfBigInteger[1], arrayOfBigInteger[2] from jsonTable"), - (new java.math.BigDecimal("922337203685477580700"), - new java.math.BigDecimal("-922337203685477580800"), null) :: Nil + Row(new java.math.BigDecimal("922337203685477580700"), + new java.math.BigDecimal("-922337203685477580800"), null) ) // Access elements of an array of arrays. checkAnswer( sql("select arrayOfArray1[0], arrayOfArray1[1] from jsonTable"), - (Seq("1", "2", "3"), Seq("str1", "str2")) :: Nil + Row(Seq("1", "2", "3"), Seq("str1", "str2")) ) // Access elements of an array of arrays. checkAnswer( sql("select arrayOfArray2[0], arrayOfArray2[1] from jsonTable"), - (Seq(1.0, 2.0, 3.0), Seq(1.1, 2.1, 3.1)) :: Nil + Row(Seq(1.0, 2.0, 3.0), Seq(1.1, 2.1, 3.1)) ) // Access elements of an array inside a filed with the type of ArrayType(ArrayType). checkAnswer( sql("select arrayOfArray1[1][1], arrayOfArray2[1][1] from jsonTable"), - ("str2", 2.1) :: Nil + Row("str2", 2.1) ) // Access elements of an array of structs. checkAnswer( sql("select arrayOfStruct[0], arrayOfStruct[1], arrayOfStruct[2], arrayOfStruct[3] " + "from jsonTable"), - (true :: "str1" :: null :: Nil, - false :: null :: null :: Nil, - null :: null :: null :: Nil, - null) :: Nil + Row( + Row(true, "str1", null), + Row(false, null, null), + Row(null, null, null), + null) ) // Access a struct and fields inside of it. @@ -327,13 +328,13 @@ class JsonSuite extends QueryTest { // Access an array field of a struct. checkAnswer( sql("select structWithArrayFields.field1, structWithArrayFields.field2 from jsonTable"), - (Seq(4, 5, 6), Seq("str1", "str2")) :: Nil + Row(Seq(4, 5, 6), Seq("str1", "str2")) ) // Access elements of an array field of a struct. checkAnswer( sql("select structWithArrayFields.field1[1], structWithArrayFields.field2[3] from jsonTable"), - (5, null) :: Nil + Row(5, null) ) } @@ -344,14 +345,14 @@ class JsonSuite extends QueryTest { // Right now, "field1" and "field2" are treated as aliases. We should fix it. checkAnswer( sql("select arrayOfStruct[0].field1, arrayOfStruct[0].field2 from jsonTable"), - (true, "str1") :: Nil + Row(true, "str1") ) // Right now, the analyzer cannot resolve arrayOfStruct.field1 and arrayOfStruct.field2. // Getting all values of a specific field from an array of structs. checkAnswer( sql("select arrayOfStruct.field1, arrayOfStruct.field2 from jsonTable"), - (Seq(true, false), Seq("str1", null)) :: Nil + Row(Seq(true, false), Seq("str1", null)) ) } @@ -372,57 +373,57 @@ class JsonSuite extends QueryTest { checkAnswer( sql("select * from jsonTable"), - ("true", 11L, null, 1.1, "13.1", "str1") :: - ("12", null, new java.math.BigDecimal("21474836470.9"), null, null, "true") :: - ("false", 21474836470L, new java.math.BigDecimal("92233720368547758070"), 100, "str1", "false") :: - (null, 21474836570L, new java.math.BigDecimal("1.1"), 21474836470L, "92233720368547758070", null) :: Nil + Row("true", 11L, null, 1.1, "13.1", "str1") :: + Row("12", null, new java.math.BigDecimal("21474836470.9"), null, null, "true") :: + Row("false", 21474836470L, new java.math.BigDecimal("92233720368547758070"), 100, "str1", "false") :: + Row(null, 21474836570L, new java.math.BigDecimal("1.1"), 21474836470L, "92233720368547758070", null) :: Nil ) // Number and Boolean conflict: resolve the type as number in this query. checkAnswer( sql("select num_bool - 10 from jsonTable where num_bool > 11"), - 2 + Row(2) ) // Widening to LongType checkAnswer( sql("select num_num_1 - 100 from jsonTable where num_num_1 > 11"), - Seq(21474836370L) :: Seq(21474836470L) :: Nil + Row(21474836370L) :: Row(21474836470L) :: Nil ) checkAnswer( sql("select num_num_1 - 100 from jsonTable where num_num_1 > 10"), - Seq(-89) :: Seq(21474836370L) :: Seq(21474836470L) :: Nil + Row(-89) :: Row(21474836370L) :: Row(21474836470L) :: Nil ) // Widening to DecimalType checkAnswer( sql("select num_num_2 + 1.2 from jsonTable where num_num_2 > 1.1"), - Seq(new java.math.BigDecimal("21474836472.1")) :: Seq(new java.math.BigDecimal("92233720368547758071.2")) :: Nil + Row(new java.math.BigDecimal("21474836472.1")) :: Row(new java.math.BigDecimal("92233720368547758071.2")) :: Nil ) // Widening to DoubleType checkAnswer( sql("select num_num_3 + 1.2 from jsonTable where num_num_3 > 1.1"), - Seq(101.2) :: Seq(21474836471.2) :: Nil + Row(101.2) :: Row(21474836471.2) :: Nil ) // Number and String conflict: resolve the type as number in this query. checkAnswer( sql("select num_str + 1.2 from jsonTable where num_str > 14"), - 92233720368547758071.2 + Row(92233720368547758071.2) ) // Number and String conflict: resolve the type as number in this query. checkAnswer( sql("select num_str + 1.2 from jsonTable where num_str > 92233720368547758060"), - new java.math.BigDecimal("92233720368547758061.2").doubleValue + Row(new java.math.BigDecimal("92233720368547758061.2").doubleValue) ) // String and Boolean conflict: resolve the type as string. checkAnswer( sql("select * from jsonTable where str_bool = 'str1'"), - ("true", 11L, null, 1.1, "13.1", "str1") :: Nil + Row("true", 11L, null, 1.1, "13.1", "str1") ) } @@ -434,24 +435,24 @@ class JsonSuite extends QueryTest { // Number and Boolean conflict: resolve the type as boolean in this query. checkAnswer( sql("select num_bool from jsonTable where NOT num_bool"), - false + Row(false) ) checkAnswer( sql("select str_bool from jsonTable where NOT str_bool"), - false + Row(false) ) // Right now, the analyzer does not know that num_bool should be treated as a boolean. // Number and Boolean conflict: resolve the type as boolean in this query. checkAnswer( sql("select num_bool from jsonTable where num_bool"), - true + Row(true) ) checkAnswer( sql("select str_bool from jsonTable where str_bool"), - false + Row(false) ) // The plan of the following DSL is @@ -464,7 +465,7 @@ class JsonSuite extends QueryTest { jsonSchemaRDD. where('num_str > BigDecimal("92233720368547758060")). select('num_str + 1.2 as Symbol("num")), - new java.math.BigDecimal("92233720368547758061.2") + Row(new java.math.BigDecimal("92233720368547758061.2")) ) // The following test will fail. The type of num_str is StringType. @@ -475,7 +476,7 @@ class JsonSuite extends QueryTest { // Number and String conflict: resolve the type as number in this query. checkAnswer( sql("select num_str + 1.2 from jsonTable where num_str > 13"), - Seq(14.3) :: Seq(92233720368547758071.2) :: Nil + Row(14.3) :: Row(92233720368547758071.2) :: Nil ) } @@ -496,10 +497,10 @@ class JsonSuite extends QueryTest { checkAnswer( sql("select * from jsonTable"), - (Seq(), "11", "[1,2,3]", Seq(null), "[]") :: - (null, """{"field":false}""", null, null, "{}") :: - (Seq(4, 5, 6), null, "str", Seq(null), "[7,8,9]") :: - (Seq(7), "{}","[str1,str2,33]", Seq("str"), """{"field":true}""") :: Nil + Row(Seq(), "11", "[1,2,3]", Row(null), "[]") :: + Row(null, """{"field":false}""", null, null, "{}") :: + Row(Seq(4, 5, 6), null, "str", Row(null), "[7,8,9]") :: + Row(Seq(7), "{}","[str1,str2,33]", Row("str"), """{"field":true}""") :: Nil ) } @@ -518,16 +519,16 @@ class JsonSuite extends QueryTest { checkAnswer( sql("select * from jsonTable"), - Seq(Seq("1", "1.1", "true", null, "[]", "{}", "[2,3,4]", - """{"field":"str"}"""), Seq(Seq(214748364700L), Seq(1)), null) :: - Seq(null, null, Seq("""{"field":"str"}""", """{"field":1}""")) :: - Seq(null, null, Seq("1", "2", "3")) :: Nil + Row(Seq("1", "1.1", "true", null, "[]", "{}", "[2,3,4]", + """{"field":"str"}"""), Seq(Row(214748364700L), Row(1)), null) :: + Row(null, null, Seq("""{"field":"str"}""", """{"field":1}""")) :: + Row(null, null, Seq("1", "2", "3")) :: Nil ) // Treat an element as a number. checkAnswer( sql("select array1[0] + 1 from jsonTable where array1 is not null"), - 2 + Row(2) ) } @@ -568,13 +569,13 @@ class JsonSuite extends QueryTest { checkAnswer( sql("select * from jsonTable"), - (new java.math.BigDecimal("92233720368547758070"), + Row(new java.math.BigDecimal("92233720368547758070"), true, 1.7976931348623157E308, 10, 21474836470L, null, - "this is a simple string.") :: Nil + "this is a simple string.") ) } @@ -594,13 +595,13 @@ class JsonSuite extends QueryTest { checkAnswer( sql("select * from jsonTableSQL"), - (new java.math.BigDecimal("92233720368547758070"), + Row(new java.math.BigDecimal("92233720368547758070"), true, 1.7976931348623157E308, 10, 21474836470L, null, - "this is a simple string.") :: Nil + "this is a simple string.") ) } @@ -626,13 +627,13 @@ class JsonSuite extends QueryTest { checkAnswer( sql("select * from jsonTable1"), - (new java.math.BigDecimal("92233720368547758070"), + Row(new java.math.BigDecimal("92233720368547758070"), true, 1.7976931348623157E308, 10, 21474836470L, null, - "this is a simple string.") :: Nil + "this is a simple string.") ) val jsonSchemaRDD2 = jsonRDD(primitiveFieldAndType, schema) @@ -643,13 +644,13 @@ class JsonSuite extends QueryTest { checkAnswer( sql("select * from jsonTable2"), - (new java.math.BigDecimal("92233720368547758070"), + Row(new java.math.BigDecimal("92233720368547758070"), true, 1.7976931348623157E308, 10, 21474836470L, null, - "this is a simple string.") :: Nil + "this is a simple string.") ) } @@ -659,7 +660,7 @@ class JsonSuite extends QueryTest { checkAnswer( sql("select arrayOfStruct[0].field1, arrayOfStruct[0].field2 from jsonTable"), - (true, "str1") :: Nil + Row(true, "str1") ) checkAnswer( sql( @@ -667,7 +668,7 @@ class JsonSuite extends QueryTest { |select complexArrayOfStruct[0].field1[1].inner2[0], complexArrayOfStruct[1].field2[0][1] |from jsonTable """.stripMargin), - ("str2", 6) :: Nil + Row("str2", 6) ) } @@ -681,7 +682,7 @@ class JsonSuite extends QueryTest { |select arrayOfArray1[0][0][0], arrayOfArray1[1][0][1], arrayOfArray1[1][1][0] |from jsonTable """.stripMargin), - (5, 7, 8) :: Nil + Row(5, 7, 8) ) checkAnswer( sql( @@ -690,7 +691,7 @@ class JsonSuite extends QueryTest { |arrayOfArray2[1][1][1].inner2[0], arrayOfArray2[2][0][0].inner3[0][0].inner4 |from jsonTable """.stripMargin), - ("str1", Nil, "str4", 2) :: Nil + Row("str1", Nil, "str4", 2) ) } @@ -704,10 +705,10 @@ class JsonSuite extends QueryTest { |select a, b, c |from jsonTable """.stripMargin), - ("str_a_1", null, null) :: - ("str_a_2", null, null) :: - (null, "str_b_3", null) :: - ("str_a_4", "str_b_4", "str_c_4") :: Nil + Row("str_a_1", null, null) :: + Row("str_a_2", null, null) :: + Row(null, "str_b_3", null) :: + Row("str_a_4", "str_b_4", "str_c_4") :: Nil ) } @@ -734,12 +735,12 @@ class JsonSuite extends QueryTest { |SELECT a, b, c, _unparsed |FROM jsonTable """.stripMargin), - (null, null, null, "{") :: - (null, null, null, "") :: - (null, null, null, """{"a":1, b:2}""") :: - (null, null, null, """{"a":{, b:3}""") :: - ("str_a_4", "str_b_4", "str_c_4", null) :: - (null, null, null, "]") :: Nil + Row(null, null, null, "{") :: + Row(null, null, null, "") :: + Row(null, null, null, """{"a":1, b:2}""") :: + Row(null, null, null, """{"a":{, b:3}""") :: + Row("str_a_4", "str_b_4", "str_c_4", null) :: + Row(null, null, null, "]") :: Nil ) checkAnswer( @@ -749,7 +750,7 @@ class JsonSuite extends QueryTest { |FROM jsonTable |WHERE _unparsed IS NULL """.stripMargin), - ("str_a_4", "str_b_4", "str_c_4") :: Nil + Row("str_a_4", "str_b_4", "str_c_4") ) checkAnswer( @@ -759,11 +760,11 @@ class JsonSuite extends QueryTest { |FROM jsonTable |WHERE _unparsed IS NOT NULL """.stripMargin), - Seq("{") :: - Seq("") :: - Seq("""{"a":1, b:2}""") :: - Seq("""{"a":{, b:3}""") :: - Seq("]") :: Nil + Row("{") :: + Row("") :: + Row("""{"a":1, b:2}""") :: + Row("""{"a":{, b:3}""") :: + Row("]") :: Nil ) TestSQLContext.setConf(SQLConf.COLUMN_NAME_OF_CORRUPT_RECORD, oldColumnNameOfCorruptRecord) @@ -793,10 +794,10 @@ class JsonSuite extends QueryTest { |SELECT field1, field2, field3, field4 |FROM jsonTable """.stripMargin), - Seq(Seq(Seq(null), Seq(Seq(Seq("Test")))), null, null, null) :: - Seq(null, Seq(null, Seq(Seq(1))), null, null) :: - Seq(null, null, Seq(Seq(null), Seq(Seq("2"))), null) :: - Seq(null, null, null, Seq(Seq(null, Seq(1, 2, 3)))) :: Nil + Row(Seq(Seq(null), Seq(Seq(Seq("Test")))), null, null, null) :: + Row(null, Seq(null, Seq(Row(1))), null, null) :: + Row(null, null, Seq(Seq(null), Seq(Row("2"))), null) :: + Row(null, null, null, Seq(Seq(null, Seq(1, 2, 3)))) :: Nil ) } @@ -851,12 +852,12 @@ class JsonSuite extends QueryTest { primTable.registerTempTable("primativeTable") checkAnswer( sql("select * from primativeTable"), - (new java.math.BigDecimal("92233720368547758070"), + Row(new java.math.BigDecimal("92233720368547758070"), true, 1.7976931348623157E308, 10, 21474836470L, - "this is a simple string.") :: Nil + "this is a simple string.") ) val complexJsonSchemaRDD = jsonRDD(complexFieldAndType1) @@ -865,38 +866,38 @@ class JsonSuite extends QueryTest { // Access elements of a primitive array. checkAnswer( sql("select arrayOfString[0], arrayOfString[1], arrayOfString[2] from complexTable"), - ("str1", "str2", null) :: Nil + Row("str1", "str2", null) ) // Access an array of null values. checkAnswer( sql("select arrayOfNull from complexTable"), - Seq(Seq(null, null, null, null)) :: Nil + Row(Seq(null, null, null, null)) ) // Access elements of a BigInteger array (we use DecimalType internally). checkAnswer( sql("select arrayOfBigInteger[0], arrayOfBigInteger[1], arrayOfBigInteger[2] from complexTable"), - (new java.math.BigDecimal("922337203685477580700"), - new java.math.BigDecimal("-922337203685477580800"), null) :: Nil + Row(new java.math.BigDecimal("922337203685477580700"), + new java.math.BigDecimal("-922337203685477580800"), null) ) // Access elements of an array of arrays. checkAnswer( sql("select arrayOfArray1[0], arrayOfArray1[1] from complexTable"), - (Seq("1", "2", "3"), Seq("str1", "str2")) :: Nil + Row(Seq("1", "2", "3"), Seq("str1", "str2")) ) // Access elements of an array of arrays. checkAnswer( sql("select arrayOfArray2[0], arrayOfArray2[1] from complexTable"), - (Seq(1.0, 2.0, 3.0), Seq(1.1, 2.1, 3.1)) :: Nil + Row(Seq(1.0, 2.0, 3.0), Seq(1.1, 2.1, 3.1)) ) // Access elements of an array inside a filed with the type of ArrayType(ArrayType). checkAnswer( sql("select arrayOfArray1[1][1], arrayOfArray2[1][1] from complexTable"), - ("str2", 2.1) :: Nil + Row("str2", 2.1) ) // Access a struct and fields inside of it. @@ -911,13 +912,13 @@ class JsonSuite extends QueryTest { // Access an array field of a struct. checkAnswer( sql("select structWithArrayFields.field1, structWithArrayFields.field2 from complexTable"), - (Seq(4, 5, 6), Seq("str1", "str2")) :: Nil + Row(Seq(4, 5, 6), Seq("str1", "str2")) ) // Access elements of an array field of a struct. checkAnswer( sql("select structWithArrayFields.field1[1], structWithArrayFields.field2[3] from complexTable"), - (5, null) :: Nil + Row(5, null) ) } http://git-wip-us.apache.org/repos/asf/spark/blob/d181c2a1/sql/core/src/test/scala/org/apache/spark/sql/parquet/ParquetFilterSuite.scala ---------------------------------------------------------------------- diff --git a/sql/core/src/test/scala/org/apache/spark/sql/parquet/ParquetFilterSuite.scala b/sql/core/src/test/scala/org/apache/spark/sql/parquet/ParquetFilterSuite.scala index 4c3a045..4ad8c47 100644 --- a/sql/core/src/test/scala/org/apache/spark/sql/parquet/ParquetFilterSuite.scala +++ b/sql/core/src/test/scala/org/apache/spark/sql/parquet/ParquetFilterSuite.scala @@ -46,7 +46,7 @@ class ParquetFilterSuite extends QueryTest with ParquetTest { predicate: Predicate, filterClass: Class[_ <: FilterPredicate], checker: (SchemaRDD, Any) => Unit, - expectedResult: => Any): Unit = { + expectedResult: Any): Unit = { withSQLConf(SQLConf.PARQUET_FILTER_PUSHDOWN_ENABLED -> "true") { val query = rdd.select(output.map(_.attr): _*).where(predicate) @@ -65,11 +65,20 @@ class ParquetFilterSuite extends QueryTest with ParquetTest { } } - private def checkFilterPushdown + private def checkFilterPushdown1 (rdd: SchemaRDD, output: Symbol*) (predicate: Predicate, filterClass: Class[_ <: FilterPredicate]) - (expectedResult: => Any): Unit = { - checkFilterPushdown(rdd, output, predicate, filterClass, checkAnswer _, expectedResult) + (expectedResult: => Seq[Row]): Unit = { + checkFilterPushdown(rdd, output, predicate, filterClass, + (query, expected) => checkAnswer(query, expected.asInstanceOf[Seq[Row]]), expectedResult) + } + + private def checkFilterPushdown + (rdd: SchemaRDD, output: Symbol*) + (predicate: Predicate, filterClass: Class[_ <: FilterPredicate]) + (expectedResult: Int): Unit = { + checkFilterPushdown(rdd, output, predicate, filterClass, + (query, expected) => checkAnswer(query, expected.asInstanceOf[Seq[Row]]), Seq(Row(expectedResult))) } def checkBinaryFilterPushdown @@ -89,27 +98,25 @@ class ParquetFilterSuite extends QueryTest with ParquetTest { test("filter pushdown - boolean") { withParquetRDD((true :: false :: Nil).map(b => Tuple1.apply(Option(b)))) { rdd => - checkFilterPushdown(rdd, '_1)('_1.isNull, classOf[Eq[java.lang.Boolean]])(Seq.empty[Row]) - checkFilterPushdown(rdd, '_1)('_1.isNotNull, classOf[NotEq[java.lang.Boolean]]) { + checkFilterPushdown1(rdd, '_1)('_1.isNull, classOf[Eq[java.lang.Boolean]])(Seq.empty[Row]) + checkFilterPushdown1(rdd, '_1)('_1.isNotNull, classOf[NotEq[java.lang.Boolean]]) { Seq(Row(true), Row(false)) } - checkFilterPushdown(rdd, '_1)('_1 === true, classOf[Eq[java.lang.Boolean]])(true) - checkFilterPushdown(rdd, '_1)('_1 !== true, classOf[Operators.NotEq[java.lang.Boolean]]) { - false - } + checkFilterPushdown1(rdd, '_1)('_1 === true, classOf[Eq[java.lang.Boolean]])(Seq(Row(true))) + checkFilterPushdown1(rdd, '_1)('_1 !== true, classOf[Operators.NotEq[java.lang.Boolean]])(Seq(Row(false))) } } test("filter pushdown - integer") { withParquetRDD((1 to 4).map(i => Tuple1(Option(i)))) { rdd => - checkFilterPushdown(rdd, '_1)('_1.isNull, classOf[Eq[Integer]])(Seq.empty[Row]) - checkFilterPushdown(rdd, '_1)('_1.isNotNull, classOf[NotEq[Integer]]) { + checkFilterPushdown1(rdd, '_1)('_1.isNull, classOf[Eq[Integer]])(Seq.empty[Row]) + checkFilterPushdown1(rdd, '_1)('_1.isNotNull, classOf[NotEq[Integer]]) { (1 to 4).map(Row.apply(_)) } checkFilterPushdown(rdd, '_1)('_1 === 1, classOf[Eq[Integer]])(1) - checkFilterPushdown(rdd, '_1)('_1 !== 1, classOf[Operators.NotEq[Integer]]) { + checkFilterPushdown1(rdd, '_1)('_1 !== 1, classOf[Operators.NotEq[Integer]]) { (2 to 4).map(Row.apply(_)) } @@ -126,7 +133,7 @@ class ParquetFilterSuite extends QueryTest with ParquetTest { checkFilterPushdown(rdd, '_1)(!('_1 < 4), classOf[Operators.GtEq[Integer]])(4) checkFilterPushdown(rdd, '_1)('_1 > 2 && '_1 < 4, classOf[Operators.And])(3) - checkFilterPushdown(rdd, '_1)('_1 < 2 || '_1 > 3, classOf[Operators.Or]) { + checkFilterPushdown1(rdd, '_1)('_1 < 2 || '_1 > 3, classOf[Operators.Or]) { Seq(Row(1), Row(4)) } } @@ -134,13 +141,13 @@ class ParquetFilterSuite extends QueryTest with ParquetTest { test("filter pushdown - long") { withParquetRDD((1 to 4).map(i => Tuple1(Option(i.toLong)))) { rdd => - checkFilterPushdown(rdd, '_1)('_1.isNull, classOf[Eq[java.lang.Long]])(Seq.empty[Row]) - checkFilterPushdown(rdd, '_1)('_1.isNotNull, classOf[NotEq[java.lang.Long]]) { + checkFilterPushdown1(rdd, '_1)('_1.isNull, classOf[Eq[java.lang.Long]])(Seq.empty[Row]) + checkFilterPushdown1(rdd, '_1)('_1.isNotNull, classOf[NotEq[java.lang.Long]]) { (1 to 4).map(Row.apply(_)) } checkFilterPushdown(rdd, '_1)('_1 === 1, classOf[Eq[java.lang.Long]])(1) - checkFilterPushdown(rdd, '_1)('_1 !== 1, classOf[Operators.NotEq[java.lang.Long]]) { + checkFilterPushdown1(rdd, '_1)('_1 !== 1, classOf[Operators.NotEq[java.lang.Long]]) { (2 to 4).map(Row.apply(_)) } @@ -157,7 +164,7 @@ class ParquetFilterSuite extends QueryTest with ParquetTest { checkFilterPushdown(rdd, '_1)(!('_1 < 4), classOf[Operators.GtEq[java.lang.Long]])(4) checkFilterPushdown(rdd, '_1)('_1 > 2 && '_1 < 4, classOf[Operators.And])(3) - checkFilterPushdown(rdd, '_1)('_1 < 2 || '_1 > 3, classOf[Operators.Or]) { + checkFilterPushdown1(rdd, '_1)('_1 < 2 || '_1 > 3, classOf[Operators.Or]) { Seq(Row(1), Row(4)) } } @@ -165,13 +172,13 @@ class ParquetFilterSuite extends QueryTest with ParquetTest { test("filter pushdown - float") { withParquetRDD((1 to 4).map(i => Tuple1(Option(i.toFloat)))) { rdd => - checkFilterPushdown(rdd, '_1)('_1.isNull, classOf[Eq[java.lang.Float]])(Seq.empty[Row]) - checkFilterPushdown(rdd, '_1)('_1.isNotNull, classOf[NotEq[java.lang.Float]]) { + checkFilterPushdown1(rdd, '_1)('_1.isNull, classOf[Eq[java.lang.Float]])(Seq.empty[Row]) + checkFilterPushdown1(rdd, '_1)('_1.isNotNull, classOf[NotEq[java.lang.Float]]) { (1 to 4).map(Row.apply(_)) } checkFilterPushdown(rdd, '_1)('_1 === 1, classOf[Eq[java.lang.Float]])(1) - checkFilterPushdown(rdd, '_1)('_1 !== 1, classOf[Operators.NotEq[java.lang.Float]]) { + checkFilterPushdown1(rdd, '_1)('_1 !== 1, classOf[Operators.NotEq[java.lang.Float]]) { (2 to 4).map(Row.apply(_)) } @@ -188,7 +195,7 @@ class ParquetFilterSuite extends QueryTest with ParquetTest { checkFilterPushdown(rdd, '_1)(!('_1 < 4), classOf[Operators.GtEq[java.lang.Float]])(4) checkFilterPushdown(rdd, '_1)('_1 > 2 && '_1 < 4, classOf[Operators.And])(3) - checkFilterPushdown(rdd, '_1)('_1 < 2 || '_1 > 3, classOf[Operators.Or]) { + checkFilterPushdown1(rdd, '_1)('_1 < 2 || '_1 > 3, classOf[Operators.Or]) { Seq(Row(1), Row(4)) } } @@ -196,13 +203,13 @@ class ParquetFilterSuite extends QueryTest with ParquetTest { test("filter pushdown - double") { withParquetRDD((1 to 4).map(i => Tuple1(Option(i.toDouble)))) { rdd => - checkFilterPushdown(rdd, '_1)('_1.isNull, classOf[Eq[java.lang.Double]])(Seq.empty[Row]) - checkFilterPushdown(rdd, '_1)('_1.isNotNull, classOf[NotEq[java.lang.Double]]) { + checkFilterPushdown1(rdd, '_1)('_1.isNull, classOf[Eq[java.lang.Double]])(Seq.empty[Row]) + checkFilterPushdown1(rdd, '_1)('_1.isNotNull, classOf[NotEq[java.lang.Double]]) { (1 to 4).map(Row.apply(_)) } checkFilterPushdown(rdd, '_1)('_1 === 1, classOf[Eq[java.lang.Double]])(1) - checkFilterPushdown(rdd, '_1)('_1 !== 1, classOf[Operators.NotEq[java.lang.Double]]) { + checkFilterPushdown1(rdd, '_1)('_1 !== 1, classOf[Operators.NotEq[java.lang.Double]]) { (2 to 4).map(Row.apply(_)) } @@ -219,7 +226,7 @@ class ParquetFilterSuite extends QueryTest with ParquetTest { checkFilterPushdown(rdd, '_1)(!('_1 < 4), classOf[Operators.GtEq[java.lang.Double]])(4) checkFilterPushdown(rdd, '_1)('_1 > 2 && '_1 < 4, classOf[Operators.And])(3) - checkFilterPushdown(rdd, '_1)('_1 < 2 || '_1 > 3, classOf[Operators.Or]) { + checkFilterPushdown1(rdd, '_1)('_1 < 2 || '_1 > 3, classOf[Operators.Or]) { Seq(Row(1), Row(4)) } } @@ -227,30 +234,30 @@ class ParquetFilterSuite extends QueryTest with ParquetTest { test("filter pushdown - string") { withParquetRDD((1 to 4).map(i => Tuple1(i.toString))) { rdd => - checkFilterPushdown(rdd, '_1)('_1.isNull, classOf[Eq[java.lang.String]])(Seq.empty[Row]) - checkFilterPushdown(rdd, '_1)('_1.isNotNull, classOf[NotEq[java.lang.String]]) { + checkFilterPushdown1(rdd, '_1)('_1.isNull, classOf[Eq[java.lang.String]])(Seq.empty[Row]) + checkFilterPushdown1(rdd, '_1)('_1.isNotNull, classOf[NotEq[java.lang.String]]) { (1 to 4).map(i => Row.apply(i.toString)) } - checkFilterPushdown(rdd, '_1)('_1 === "1", classOf[Eq[String]])("1") - checkFilterPushdown(rdd, '_1)('_1 !== "1", classOf[Operators.NotEq[String]]) { + checkFilterPushdown1(rdd, '_1)('_1 === "1", classOf[Eq[String]])(Seq(Row("1"))) + checkFilterPushdown1(rdd, '_1)('_1 !== "1", classOf[Operators.NotEq[String]]) { (2 to 4).map(i => Row.apply(i.toString)) } - checkFilterPushdown(rdd, '_1)('_1 < "2", classOf[Lt [java.lang.String]])("1") - checkFilterPushdown(rdd, '_1)('_1 > "3", classOf[Gt [java.lang.String]])("4") - checkFilterPushdown(rdd, '_1)('_1 <= "1", classOf[LtEq[java.lang.String]])("1") - checkFilterPushdown(rdd, '_1)('_1 >= "4", classOf[GtEq[java.lang.String]])("4") + checkFilterPushdown1(rdd, '_1)('_1 < "2", classOf[Lt [java.lang.String]])(Seq(Row("1"))) + checkFilterPushdown1(rdd, '_1)('_1 > "3", classOf[Gt [java.lang.String]])(Seq(Row("4"))) + checkFilterPushdown1(rdd, '_1)('_1 <= "1", classOf[LtEq[java.lang.String]])(Seq(Row("1"))) + checkFilterPushdown1(rdd, '_1)('_1 >= "4", classOf[GtEq[java.lang.String]])(Seq(Row("4"))) - checkFilterPushdown(rdd, '_1)(Literal("1") === '_1, classOf[Eq [java.lang.String]])("1") - checkFilterPushdown(rdd, '_1)(Literal("2") > '_1, classOf[Lt [java.lang.String]])("1") - checkFilterPushdown(rdd, '_1)(Literal("3") < '_1, classOf[Gt [java.lang.String]])("4") - checkFilterPushdown(rdd, '_1)(Literal("1") >= '_1, classOf[LtEq[java.lang.String]])("1") - checkFilterPushdown(rdd, '_1)(Literal("4") <= '_1, classOf[GtEq[java.lang.String]])("4") + checkFilterPushdown1(rdd, '_1)(Literal("1") === '_1, classOf[Eq [java.lang.String]])(Seq(Row("1"))) + checkFilterPushdown1(rdd, '_1)(Literal("2") > '_1, classOf[Lt [java.lang.String]])(Seq(Row("1"))) + checkFilterPushdown1(rdd, '_1)(Literal("3") < '_1, classOf[Gt [java.lang.String]])(Seq(Row("4"))) + checkFilterPushdown1(rdd, '_1)(Literal("1") >= '_1, classOf[LtEq[java.lang.String]])(Seq(Row("1"))) + checkFilterPushdown1(rdd, '_1)(Literal("4") <= '_1, classOf[GtEq[java.lang.String]])(Seq(Row("4"))) - checkFilterPushdown(rdd, '_1)(!('_1 < "4"), classOf[Operators.GtEq[java.lang.String]])("4") - checkFilterPushdown(rdd, '_1)('_1 > "2" && '_1 < "4", classOf[Operators.And])("3") - checkFilterPushdown(rdd, '_1)('_1 < "2" || '_1 > "3", classOf[Operators.Or]) { + checkFilterPushdown1(rdd, '_1)(!('_1 < "4"), classOf[Operators.GtEq[java.lang.String]])(Seq(Row("4"))) + checkFilterPushdown1(rdd, '_1)('_1 > "2" && '_1 < "4", classOf[Operators.And])(Seq(Row("3"))) + checkFilterPushdown1(rdd, '_1)('_1 < "2" || '_1 > "3", classOf[Operators.Or]) { Seq(Row("1"), Row("4")) } } http://git-wip-us.apache.org/repos/asf/spark/blob/d181c2a1/sql/core/src/test/scala/org/apache/spark/sql/parquet/ParquetIOSuite.scala ---------------------------------------------------------------------- diff --git a/sql/core/src/test/scala/org/apache/spark/sql/parquet/ParquetIOSuite.scala b/sql/core/src/test/scala/org/apache/spark/sql/parquet/ParquetIOSuite.scala index 973819a..a57e4e8 100644 --- a/sql/core/src/test/scala/org/apache/spark/sql/parquet/ParquetIOSuite.scala +++ b/sql/core/src/test/scala/org/apache/spark/sql/parquet/ParquetIOSuite.scala @@ -68,8 +68,8 @@ class ParquetIOSuite extends QueryTest with ParquetTest { /** * Writes `data` to a Parquet file, reads it back and check file contents. */ - protected def checkParquetFile[T <: Product: ClassTag: TypeTag](data: Seq[T]): Unit = { - withParquetRDD(data)(checkAnswer(_, data)) + protected def checkParquetFile[T <: Product : ClassTag: TypeTag](data: Seq[T]): Unit = { + withParquetRDD(data)(r => checkAnswer(r, data.map(Row.fromTuple))) } test("basic data types (without binary)") { @@ -143,7 +143,7 @@ class ParquetIOSuite extends QueryTest with ParquetTest { withParquetRDD(data) { rdd => // Structs are converted to `Row`s checkAnswer(rdd, data.map { case Tuple1(struct) => - Tuple1(Row(struct.productIterator.toSeq: _*)) + Row(Row(struct.productIterator.toSeq: _*)) }) } } @@ -153,7 +153,7 @@ class ParquetIOSuite extends QueryTest with ParquetTest { withParquetRDD(data) { rdd => // Structs are converted to `Row`s checkAnswer(rdd, data.map { case Tuple1(struct) => - Tuple1(Row(struct.productIterator.toSeq: _*)) + Row(Row(struct.productIterator.toSeq: _*)) }) } } @@ -162,7 +162,7 @@ class ParquetIOSuite extends QueryTest with ParquetTest { val data = (1 to 4).map(i => Tuple1(Map(i -> (i, s"val_$i")))) withParquetRDD(data) { rdd => checkAnswer(rdd, data.map { case Tuple1(m) => - Tuple1(m.mapValues(struct => Row(struct.productIterator.toSeq: _*))) + Row(m.mapValues(struct => Row(struct.productIterator.toSeq: _*))) }) } } @@ -261,7 +261,7 @@ class ParquetIOSuite extends QueryTest with ParquetTest { val path = new Path(dir.toURI.toString, "part-r-0.parquet") makeRawParquetFile(path) checkAnswer(parquetFile(path.toString), (0 until 10).map { i => - (i % 2 == 0, i, i.toLong, i.toFloat, i.toDouble) + Row(i % 2 == 0, i, i.toLong, i.toFloat, i.toDouble) }) } } http://git-wip-us.apache.org/repos/asf/spark/blob/d181c2a1/sql/core/src/test/scala/org/apache/spark/sql/parquet/ParquetQuerySuite.scala ---------------------------------------------------------------------- diff --git a/sql/core/src/test/scala/org/apache/spark/sql/parquet/ParquetQuerySuite.scala b/sql/core/src/test/scala/org/apache/spark/sql/parquet/ParquetQuerySuite.scala index 3a073a6..2c5345b 100644 --- a/sql/core/src/test/scala/org/apache/spark/sql/parquet/ParquetQuerySuite.scala +++ b/sql/core/src/test/scala/org/apache/spark/sql/parquet/ParquetQuerySuite.scala @@ -28,7 +28,7 @@ import parquet.hadoop.util.ContextUtil import parquet.io.api.Binary import org.apache.spark.sql._ -import org.apache.spark.sql.catalyst.expressions._ +import org.apache.spark.sql.catalyst.expressions.{Row => _, _} import org.apache.spark.sql.catalyst.util.getTempFilePath import org.apache.spark.sql.test.TestSQLContext import org.apache.spark.sql.test.TestSQLContext._ @@ -191,8 +191,8 @@ class ParquetQuerySuite extends QueryTest with FunSuiteLike with BeforeAndAfterA parquetFile(path).registerTempTable("tmp") checkAnswer( sql("SELECT key, value FROM tmp WHERE value = 'val_5' OR value = 'val_7'"), - (5, "val_5") :: - (7, "val_7") :: Nil) + Row(5, "val_5") :: + Row(7, "val_7") :: Nil) Utils.deleteRecursively(file) @@ -207,8 +207,8 @@ class ParquetQuerySuite extends QueryTest with FunSuiteLike with BeforeAndAfterA parquetFile(path).registerTempTable("tmp") checkAnswer( sql("SELECT key, value FROM tmp WHERE value = 'val_5' OR value = 'val_7'"), - (5, "val_5") :: - (7, "val_7") :: Nil) + Row(5, "val_5") :: + Row(7, "val_7") :: Nil) Utils.deleteRecursively(file) @@ -223,8 +223,8 @@ class ParquetQuerySuite extends QueryTest with FunSuiteLike with BeforeAndAfterA parquetFile(path).registerTempTable("tmp") checkAnswer( sql("SELECT key, value FROM tmp WHERE value = 'val_5' OR value = 'val_7'"), - (5, "val_5") :: - (7, "val_7") :: Nil) + Row(5, "val_5") :: + Row(7, "val_7") :: Nil) Utils.deleteRecursively(file) @@ -239,8 +239,8 @@ class ParquetQuerySuite extends QueryTest with FunSuiteLike with BeforeAndAfterA parquetFile(path).registerTempTable("tmp") checkAnswer( sql("SELECT key, value FROM tmp WHERE value = 'val_5' OR value = 'val_7'"), - (5, "val_5") :: - (7, "val_7") :: Nil) + Row(5, "val_5") :: + Row(7, "val_7") :: Nil) Utils.deleteRecursively(file) @@ -255,8 +255,8 @@ class ParquetQuerySuite extends QueryTest with FunSuiteLike with BeforeAndAfterA parquetFile(path).registerTempTable("tmp") checkAnswer( sql("SELECT key, value FROM tmp WHERE value = 'val_5' OR value = 'val_7'"), - (5, "val_5") :: - (7, "val_7") :: Nil) + Row(5, "val_5") :: + Row(7, "val_7") :: Nil) Utils.deleteRecursively(file) @@ -303,7 +303,7 @@ class ParquetQuerySuite extends QueryTest with FunSuiteLike with BeforeAndAfterA assert(result.size === 9, "self-join result has incorrect size") assert(result(0).size === 12, "result row has incorrect size") result.zipWithIndex.foreach { - case (row, index) => row.zipWithIndex.foreach { + case (row, index) => row.toSeq.zipWithIndex.foreach { case (field, column) => assert(field != null, s"self-join contains null value in row $index field $column") } } @@ -423,7 +423,7 @@ class ParquetQuerySuite extends QueryTest with FunSuiteLike with BeforeAndAfterA val readFile = parquetFile(path) val rdd_saved = readFile.collect() - assert(rdd_saved(0) === Seq.fill(5)(null)) + assert(rdd_saved(0) === Row(null, null, null, null, null)) Utils.deleteRecursively(file) assert(true) } @@ -438,7 +438,7 @@ class ParquetQuerySuite extends QueryTest with FunSuiteLike with BeforeAndAfterA val readFile = parquetFile(path) val rdd_saved = readFile.collect() - assert(rdd_saved(0) === Seq.fill(5)(null)) + assert(rdd_saved(0) === Row(null, null, null, null, null)) Utils.deleteRecursively(file) assert(true) } http://git-wip-us.apache.org/repos/asf/spark/blob/d181c2a1/sql/core/src/test/scala/org/apache/spark/sql/parquet/ParquetQuerySuite2.scala ---------------------------------------------------------------------- diff --git a/sql/core/src/test/scala/org/apache/spark/sql/parquet/ParquetQuerySuite2.scala b/sql/core/src/test/scala/org/apache/spark/sql/parquet/ParquetQuerySuite2.scala index 4c081fb..7b3f8c2 100644 --- a/sql/core/src/test/scala/org/apache/spark/sql/parquet/ParquetQuerySuite2.scala +++ b/sql/core/src/test/scala/org/apache/spark/sql/parquet/ParquetQuerySuite2.scala @@ -38,7 +38,7 @@ class ParquetQuerySuite2 extends QueryTest with ParquetTest { val data = (0 until 10).map(i => (i, i.toString)) withParquetTable(data, "t") { sql("INSERT INTO t SELECT * FROM t") - checkAnswer(table("t"), data ++ data) + checkAnswer(table("t"), (data ++ data).map(Row.fromTuple)) } } http://git-wip-us.apache.org/repos/asf/spark/blob/d181c2a1/sql/core/src/test/scala/org/apache/spark/sql/sources/TableScanSuite.scala ---------------------------------------------------------------------- diff --git a/sql/core/src/test/scala/org/apache/spark/sql/sources/TableScanSuite.scala b/sql/core/src/test/scala/org/apache/spark/sql/sources/TableScanSuite.scala index 264f6d9..b1e0919 100644 --- a/sql/core/src/test/scala/org/apache/spark/sql/sources/TableScanSuite.scala +++ b/sql/core/src/test/scala/org/apache/spark/sql/sources/TableScanSuite.scala @@ -244,7 +244,7 @@ class TableScanSuite extends DataSourceTest { sqlTest( "SELECT count(*) FROM tableWithSchema", - 10) + Seq(Row(10))) sqlTest( "SELECT `string$%Field` FROM tableWithSchema", @@ -260,7 +260,7 @@ class TableScanSuite extends DataSourceTest { sqlTest( "SELECT structFieldSimple.key, arrayFieldSimple[1] FROM tableWithSchema a where int_Field=1", - Seq(Seq(1, 2))) + Seq(Row(1, 2))) sqlTest( "SELECT structFieldComplex.Value.`value_(2)` FROM tableWithSchema", http://git-wip-us.apache.org/repos/asf/spark/blob/d181c2a1/sql/hive/src/main/scala/org/apache/spark/sql/hive/HiveContext.scala ---------------------------------------------------------------------- diff --git a/sql/hive/src/main/scala/org/apache/spark/sql/hive/HiveContext.scala b/sql/hive/src/main/scala/org/apache/spark/sql/hive/HiveContext.scala index 10833c1..3e26fe3 100644 --- a/sql/hive/src/main/scala/org/apache/spark/sql/hive/HiveContext.scala +++ b/sql/hive/src/main/scala/org/apache/spark/sql/hive/HiveContext.scala @@ -368,10 +368,10 @@ class HiveContext(sc: SparkContext) extends SQLContext(sc) { .mkString("\t") } case command: ExecutedCommand => - command.executeCollect().map(_.head.toString) + command.executeCollect().map(_(0).toString) case other => - val result: Seq[Seq[Any]] = other.executeCollect().toSeq + val result: Seq[Seq[Any]] = other.executeCollect().map(_.toSeq).toSeq // We need the types so we can output struct field names val types = analyzed.output.map(_.dataType) // Reformat to match hive tab delimited output. @@ -395,7 +395,7 @@ private object HiveContext { protected[sql] def toHiveString(a: (Any, DataType)): String = a match { case (struct: Row, StructType(fields)) => - struct.zip(fields).map { + struct.toSeq.zip(fields).map { case (v, t) => s""""${t.name}":${toHiveStructString(v, t.dataType)}""" }.mkString("{", ",", "}") case (seq: Seq[_], ArrayType(typ, _)) => @@ -418,7 +418,7 @@ private object HiveContext { /** Hive outputs fields of structs slightly differently than top level attributes. */ protected def toHiveStructString(a: (Any, DataType)): String = a match { case (struct: Row, StructType(fields)) => - struct.zip(fields).map { + struct.toSeq.zip(fields).map { case (v, t) => s""""${t.name}":${toHiveStructString(v, t.dataType)}""" }.mkString("{", ",", "}") case (seq: Seq[_], ArrayType(typ, _)) => http://git-wip-us.apache.org/repos/asf/spark/blob/d181c2a1/sql/hive/src/main/scala/org/apache/spark/sql/hive/HiveInspectors.scala ---------------------------------------------------------------------- diff --git a/sql/hive/src/main/scala/org/apache/spark/sql/hive/HiveInspectors.scala b/sql/hive/src/main/scala/org/apache/spark/sql/hive/HiveInspectors.scala index eeabfdd..82dba99 100644 --- a/sql/hive/src/main/scala/org/apache/spark/sql/hive/HiveInspectors.scala +++ b/sql/hive/src/main/scala/org/apache/spark/sql/hive/HiveInspectors.scala @@ -348,7 +348,7 @@ private[hive] trait HiveInspectors { (o: Any) => { if (o != null) { val struct = soi.create() - (soi.getAllStructFieldRefs, wrappers, o.asInstanceOf[Row]).zipped.foreach { + (soi.getAllStructFieldRefs, wrappers, o.asInstanceOf[Row].toSeq).zipped.foreach { (field, wrapper, data) => soi.setStructFieldData(struct, field, wrapper(data)) } struct @@ -432,7 +432,7 @@ private[hive] trait HiveInspectors { } case x: SettableStructObjectInspector => val fieldRefs = x.getAllStructFieldRefs - val row = a.asInstanceOf[Seq[_]] + val row = a.asInstanceOf[Row] // 1. create the pojo (most likely) object val result = x.create() var i = 0 @@ -448,7 +448,7 @@ private[hive] trait HiveInspectors { result case x: StructObjectInspector => val fieldRefs = x.getAllStructFieldRefs - val row = a.asInstanceOf[Seq[_]] + val row = a.asInstanceOf[Row] val result = new java.util.ArrayList[AnyRef](fieldRefs.length) var i = 0 while (i < fieldRefs.length) { @@ -475,7 +475,7 @@ private[hive] trait HiveInspectors { } def wrap( - row: Seq[Any], + row: Row, inspectors: Seq[ObjectInspector], cache: Array[AnyRef]): Array[AnyRef] = { var i = 0 @@ -486,6 +486,18 @@ private[hive] trait HiveInspectors { cache } + def wrap( + row: Seq[Any], + inspectors: Seq[ObjectInspector], + cache: Array[AnyRef]): Array[AnyRef] = { + var i = 0 + while (i < inspectors.length) { + cache(i) = wrap(row(i), inspectors(i)) + i += 1 + } + cache + } + /** * @param dataType Catalyst data type * @return Hive java object inspector (recursively), not the Writable ObjectInspector http://git-wip-us.apache.org/repos/asf/spark/blob/d181c2a1/sql/hive/src/main/scala/org/apache/spark/sql/hive/hiveUdfs.scala ---------------------------------------------------------------------- diff --git a/sql/hive/src/main/scala/org/apache/spark/sql/hive/hiveUdfs.scala b/sql/hive/src/main/scala/org/apache/spark/sql/hive/hiveUdfs.scala index d898b87..76d2140 100644 --- a/sql/hive/src/main/scala/org/apache/spark/sql/hive/hiveUdfs.scala +++ b/sql/hive/src/main/scala/org/apache/spark/sql/hive/hiveUdfs.scala @@ -360,7 +360,7 @@ private[hive] case class HiveUdafFunction( protected lazy val cached = new Array[AnyRef](exprs.length) def update(input: Row): Unit = { - val inputs = inputProjection(input).asInstanceOf[Seq[AnyRef]].toArray + val inputs = inputProjection(input) function.iterate(buffer, wrap(inputs, inspectors, cached)) } } http://git-wip-us.apache.org/repos/asf/spark/blob/d181c2a1/sql/hive/src/main/scala/org/apache/spark/sql/hive/hiveWriterContainers.scala ---------------------------------------------------------------------- diff --git a/sql/hive/src/main/scala/org/apache/spark/sql/hive/hiveWriterContainers.scala b/sql/hive/src/main/scala/org/apache/spark/sql/hive/hiveWriterContainers.scala index cc8bb3e..aae175e 100644 --- a/sql/hive/src/main/scala/org/apache/spark/sql/hive/hiveWriterContainers.scala +++ b/sql/hive/src/main/scala/org/apache/spark/sql/hive/hiveWriterContainers.scala @@ -209,7 +209,7 @@ private[spark] class SparkHiveDynamicPartitionWriterContainer( override def getLocalFileWriter(row: Row): FileSinkOperator.RecordWriter = { val dynamicPartPath = dynamicPartColNames - .zip(row.takeRight(dynamicPartColNames.length)) + .zip(row.toSeq.takeRight(dynamicPartColNames.length)) .map { case (col, rawVal) => val string = if (rawVal == null) null else String.valueOf(rawVal) s"/$col=${if (string == null || string.isEmpty) defaultPartName else string}" --------------------------------------------------------------------- To unsubscribe, e-mail: commits-unsubscr...@spark.apache.org For additional commands, e-mail: commits-h...@spark.apache.org
