[SPARK-23905][SQL] Add UDF weekday

Author: yucai

sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/analysis/FunctionRegistry.scala

@@ -395,6 +395,7 @@ object FunctionRegistry {
     expression[TruncTimestamp]("date_trunc"),
     expression[UnixTimestamp]("unix_timestamp"),
     expression[DayOfWeek]("dayofweek"),
+    expression[WeekDay]("weekday"),
     expression[WeekOfYear]("weekofyear"),
     expression[Year]("year"),
     expression[TimeWindow]("window"),

sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/expressions/datetimeExpressions.scala

@@ -426,15 +426,7 @@ case class DayOfMonth(child: Expression) extends UnaryExpression with ImplicitCa
   """,
   since = "2.3.0")
 // scalastyle:on line.size.limit
-case class DayOfWeek(child: Expression) extends UnaryExpression with ImplicitCastInputTypes {
-
-  override def inputTypes: Seq[AbstractDataType] = Seq(DateType)
-
-  override def dataType: DataType = IntegerType
-
-  @transient private lazy val c = {
-    Calendar.getInstance(DateTimeUtils.getTimeZone("UTC"))
-  }
+case class DayOfWeek(child: Expression) extends DayWeek {
 
   override protected def nullSafeEval(date: Any): Any = {
     c.setTimeInMillis(date.asInstanceOf[Int] * 1000L * 3600L * 24L)
@@ -456,6 +448,49 @@ case class DayOfWeek(child: Expression) extends UnaryExpression with ImplicitCas
   }
 }
 
+// scalastyle:off line.size.limit
+@ExpressionDescription(
+  usage = "_FUNC_(date) - Returns the day of the week for date/timestamp (0 = Monday, 1 = Tuesday, ..., 6 = Sunday).",
+  examples = """
+    Examples:
+      > SELECT _FUNC_('2009-07-30');
+       3
+  """,
+  since = "2.4.0")
+// scalastyle:on line.size.limit
+case class WeekDay(child: Expression) extends DayWeek {
+
+  override protected def nullSafeEval(date: Any): Any = {
+    c.setTimeInMillis(date.asInstanceOf[Int] * 1000L * 3600L * 24L)
+    (c.get(Calendar.DAY_OF_WEEK) + 5 ) % 7
+  }
+
+  override protected def doGenCode(ctx: CodegenContext, ev: ExprCode): ExprCode = {
+    nullSafeCodeGen(ctx, ev, time => {
+      val cal = classOf[Calendar].getName
+      val dtu = DateTimeUtils.getClass.getName.stripSuffix("$")
+      val c = "calWeekDay"
+      ctx.addImmutableStateIfNotExists(cal, c,
+        v => s"""$v = $cal.getInstance($dtu.getTimeZone("UTC"));""")
+      s"""
+        $c.setTimeInMillis($time * 1000L * 3600L * 24L);
+        ${ev.value} = ($c.get($cal.DAY_OF_WEEK) + 5) % 7;
+      """
+    })
+  }
+}
+
+abstract class DayWeek extends UnaryExpression with ImplicitCastInputTypes {
+
+  override def inputTypes: Seq[AbstractDataType] = Seq(DateType)
+
+  override def dataType: DataType = IntegerType
+
+  @transient protected lazy val c: Calendar = {
+    Calendar.getInstance(DateTimeUtils.getTimeZone("UTC"))
+  }
+}
+
 // scalastyle:off line.size.limit
 @ExpressionDescription(
   usage = "_FUNC_(date) - Returns the week of the year of the given date. A week is considered to start on a Monday and week 1 is the first week with >3 days.",

sql/catalyst/src/test/scala/org/apache/spark/sql/catalyst/expressions/DateExpressionsSuite.scala

@@ -211,6 +211,17 @@ class DateExpressionsSuite extends SparkFunSuite with ExpressionEvalHelper {
     checkConsistencyBetweenInterpretedAndCodegen(DayOfWeek, DateType)
   }
 
+  test("WeekDay") {
+    checkEvaluation(WeekDay(Literal.create(null, DateType)), null)
+    checkEvaluation(WeekDay(Literal(d)), 2)
+    checkEvaluation(WeekDay(Cast(Literal(sdfDate.format(d)), DateType, gmtId)), 2)
+    checkEvaluation(WeekDay(Cast(Literal(ts), DateType, gmtId)), 4)
+    checkEvaluation(WeekDay(Cast(Literal("2011-05-06"), DateType, gmtId)), 4)
+    checkEvaluation(WeekDay(Literal(new Date(sdf.parse("2017-05-27 13:10:15").getTime))), 5)
+    checkEvaluation(WeekDay(Literal(new Date(sdf.parse("1582-10-15 13:10:15").getTime))), 4)
+    checkConsistencyBetweenInterpretedAndCodegen(WeekDay, DateType)
+  }
+
   test("WeekOfYear") {
     checkEvaluation(WeekOfYear(Literal.create(null, DateType)), null)
     checkEvaluation(WeekOfYear(Literal(d)), 15)

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

@@ -25,3 +25,5 @@ create temporary view ttf2 as select * from values
 select current_date = current_date(), current_timestamp = current_timestamp(), a, b from ttf2;
 
 select a, b from ttf2 order by a, current_date;
+
+select weekday('2007-02-03'), weekday('2009-07-30'), weekday('2017-05-27'), weekday(null), weekday('1582-10-15 13:10:15');

sql/core/src/test/resources/sql-tests/results/datetime.sql.out

@@ -1,5 +1,5 @@
 -- Automatically generated by SQLQueryTestSuite
--- Number of queries: 9
+-- Number of queries: 10
 
 
 -- !query 0
@@ -81,3 +81,10 @@ struct<a:int,b:int>
 -- !query 8 output
 1	2
 2	3
+
+-- !query 9
+select weekday('2007-02-03'), weekday('2009-07-30'), weekday('2017-05-27'), weekday(null), weekday('1582-10-15 13:10:15')
+-- !query 3 schema
+struct<weekday(CAST(2007-02-03 AS DATE)):int,weekday(CAST(2009-07-30 AS DATE)):int,weekday(CAST(2017-05-27 AS DATE)):int,weekday(CAST(NULL AS DATE)):int,weekday(CAST(1582-10-15 13:10:15 AS DATE)):int>
+-- !query 3 output
+5	3	5	NULL	4