diff options
author | Reynold Xin <rxin@databricks.com> | 2015-04-23 21:21:03 -0700 |
---|---|---|
committer | Reynold Xin <rxin@databricks.com> | 2015-04-23 21:21:03 -0700 |
commit | d3a302defc45768492dec9da4c40d78d28997a65 (patch) | |
tree | 52111ca80f4fae52a9406d67bde2f482176a9a1c /sql | |
parent | 67bccbda1e3ed7db2753daa7e6ae8b1441356177 (diff) | |
download | spark-d3a302defc45768492dec9da4c40d78d28997a65.tar.gz spark-d3a302defc45768492dec9da4c40d78d28997a65.tar.bz2 spark-d3a302defc45768492dec9da4c40d78d28997a65.zip |
[SQL] Fixed expression data type matching.
Also took the chance to improve documentation for various types.
Author: Reynold Xin <rxin@databricks.com>
Closes #5675 from rxin/data-type-matching-expr and squashes the following commits:
0f31856 [Reynold Xin] One more function documentation.
27c1973 [Reynold Xin] Added more documentation.
336a36d [Reynold Xin] [SQL] Fixed expression data type matching.
Diffstat (limited to 'sql')
-rw-r--r-- | sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/expressions/codegen/CodeGenerator.scala | 2 | ||||
-rw-r--r-- | sql/catalyst/src/main/scala/org/apache/spark/sql/types/DataType.scala | 50 |
2 files changed, 42 insertions, 10 deletions
diff --git a/sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/expressions/codegen/CodeGenerator.scala b/sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/expressions/codegen/CodeGenerator.scala index cbe5203473..dbc92fb93e 100644 --- a/sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/expressions/codegen/CodeGenerator.scala +++ b/sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/expressions/codegen/CodeGenerator.scala @@ -279,7 +279,7 @@ abstract class CodeGenerator[InType <: AnyRef, OutType <: AnyRef] extends Loggin org.apache.spark.sql.types.UTF8String(${eval.primitiveTerm}.toString) """.children - case EqualTo(e1: BinaryType, e2: BinaryType) => + case EqualTo(e1 @ BinaryType(), e2 @ BinaryType()) => (e1, e2).evaluateAs (BooleanType) { case (eval1, eval2) => q""" diff --git a/sql/catalyst/src/main/scala/org/apache/spark/sql/types/DataType.scala b/sql/catalyst/src/main/scala/org/apache/spark/sql/types/DataType.scala index e6bfcd9adf..06bff7d70e 100644 --- a/sql/catalyst/src/main/scala/org/apache/spark/sql/types/DataType.scala +++ b/sql/catalyst/src/main/scala/org/apache/spark/sql/types/DataType.scala @@ -40,32 +40,46 @@ import org.apache.spark.util.Utils */ @DeveloperApi abstract class DataType { - /** Matches any expression that evaluates to this DataType */ - def unapply(a: Expression): Boolean = a match { + /** + * Enables matching against NumericType for expressions: + * {{{ + * case Cast(child @ BinaryType(), StringType) => + * ... + * }}} + */ + private[sql] def unapply(a: Expression): Boolean = a match { case e: Expression if e.dataType == this => true case _ => false } - /** The default size of a value of this data type. */ + /** + * The default size of a value of this data type, used internally for size estimation. + */ def defaultSize: Int + /** Name of the type used in JSON serialization. */ def typeName: String = this.getClass.getSimpleName.stripSuffix("$").dropRight(4).toLowerCase private[sql] def jsonValue: JValue = typeName + /** The compact JSON representation of this data type. */ def json: String = compact(render(jsonValue)) + /** The pretty (i.e. indented) JSON representation of this data type. */ def prettyJson: String = pretty(render(jsonValue)) + /** Readable string representation for the type. */ def simpleString: String = typeName - /** Check if `this` and `other` are the same data type when ignoring nullability - * (`StructField.nullable`, `ArrayType.containsNull`, and `MapType.valueContainsNull`). + /** + * Check if `this` and `other` are the same data type when ignoring nullability + * (`StructField.nullable`, `ArrayType.containsNull`, and `MapType.valueContainsNull`). */ private[spark] def sameType(other: DataType): Boolean = DataType.equalsIgnoreNullability(this, other) - /** Returns the same data type but set all nullability fields are true + /** + * Returns the same data type but set all nullability fields are true * (`StructField.nullable`, `ArrayType.containsNull`, and `MapType.valueContainsNull`). */ private[spark] def asNullable: DataType @@ -104,12 +118,25 @@ abstract class NumericType extends AtomicType { private[sql] object NumericType { + /** + * Enables matching against NumericType for expressions: + * {{{ + * case Cast(child @ NumericType(), StringType) => + * ... + * }}} + */ def unapply(e: Expression): Boolean = e.dataType.isInstanceOf[NumericType] } -/** Matcher for any expressions that evaluate to [[IntegralType]]s */ private[sql] object IntegralType { + /** + * Enables matching against IntegralType for expressions: + * {{{ + * case Cast(child @ IntegralType(), StringType) => + * ... + * }}} + */ def unapply(a: Expression): Boolean = a match { case e: Expression if e.dataType.isInstanceOf[IntegralType] => true case _ => false @@ -122,9 +149,14 @@ private[sql] abstract class IntegralType extends NumericType { } - -/** Matcher for any expressions that evaluate to [[FractionalType]]s */ private[sql] object FractionalType { + /** + * Enables matching against FractionalType for expressions: + * {{{ + * case Cast(child @ FractionalType(), StringType) => + * ... + * }}} + */ def unapply(a: Expression): Boolean = a match { case e: Expression if e.dataType.isInstanceOf[FractionalType] => true case _ => false |