[SPARK-10245] [SQL] Fix decimal literals with precision < scale

In BigDecimal or java.math.BigDecimal, the precision could be smaller than scale, for example, BigDecimal("0.001") has precision = 1 and scale = 3. But DecimalType require that the precision should be larger than scale, so we should use the maximum of precision and scale when inferring the schema from decimal literal.

Author: Davies Liu <[email protected]>

Closes #8428 from davies/smaller_decimal.

Author: Davies Liu

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

@@ -36,9 +36,10 @@ object Literal {
     case s: Short => Literal(s, ShortType)
     case s: String => Literal(UTF8String.fromString(s), StringType)
     case b: Boolean => Literal(b, BooleanType)
-    case d: BigDecimal => Literal(Decimal(d), DecimalType(d.precision, d.scale))
-    case d: java.math.BigDecimal => Literal(Decimal(d), DecimalType(d.precision(), d.scale()))
-    case d: Decimal => Literal(d, DecimalType(d.precision, d.scale))
+    case d: BigDecimal => Literal(Decimal(d), DecimalType(Math.max(d.precision, d.scale), d.scale))
+    case d: java.math.BigDecimal =>
+      Literal(Decimal(d), DecimalType(Math.max(d.precision, d.scale), d.scale()))
+    case d: Decimal => Literal(d, DecimalType(Math.max(d.precision, d.scale), d.scale))
     case t: Timestamp => Literal(DateTimeUtils.fromJavaTimestamp(t), TimestampType)
     case d: Date => Literal(DateTimeUtils.fromJavaDate(d), DateType)
     case a: Array[Byte] => Literal(a, BinaryType)

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

@@ -83,12 +83,14 @@ class LiteralExpressionSuite extends SparkFunSuite with ExpressionEvalHelper {
   }
 
   test("decimal") {
-    List(0.0, 1.2, 1.1111, 5).foreach { d =>
+    List(-0.0001, 0.0, 0.001, 1.2, 1.1111, 5).foreach { d =>
       checkEvaluation(Literal(Decimal(d)), Decimal(d))
       checkEvaluation(Literal(Decimal(d.toInt)), Decimal(d.toInt))
       checkEvaluation(Literal(Decimal(d.toLong)), Decimal(d.toLong))
-      checkEvaluation(Literal(Decimal((d * 1000L).toLong, 10, 1)),
-        Decimal((d * 1000L).toLong, 10, 1))
+      checkEvaluation(Literal(Decimal((d * 1000L).toLong, 10, 3)),
+        Decimal((d * 1000L).toLong, 10, 3))
+      checkEvaluation(Literal(BigDecimal(d.toString)), Decimal(d))
+      checkEvaluation(Literal(new java.math.BigDecimal(d.toString)), Decimal(d))
     }
   }
 

sql/core/src/test/scala/org/apache/spark/sql/SQLQuerySuite.scala

@@ -1627,6 +1627,16 @@ class SQLQuerySuite extends QueryTest with SharedSQLContext {
       Row(null))
   }
 
+  test("precision smaller than scale") {
+    checkAnswer(sql("select 10.00"), Row(BigDecimal("10.00")))
+    checkAnswer(sql("select 1.00"), Row(BigDecimal("1.00")))
+    checkAnswer(sql("select 0.10"), Row(BigDecimal("0.10")))
+    checkAnswer(sql("select 0.01"), Row(BigDecimal("0.01")))
+    checkAnswer(sql("select 0.001"), Row(BigDecimal("0.001")))
+    checkAnswer(sql("select -0.01"), Row(BigDecimal("-0.01")))
+    checkAnswer(sql("select -0.001"), Row(BigDecimal("-0.001")))
+  }
+
   test("external sorting updates peak execution memory") {
     withSQLConf((SQLConf.EXTERNAL_SORT.key, "true")) {
       val sc = sqlContext.sparkContext