[SPARK-5578][SQL][DataFrame] Provide a convenient way for Scala users to use UDFs
A more convenient way to define user-defined functions. Author: Reynold Xin <r...@databricks.com> Closes #4345 from rxin/defineUDF and squashes the following commits: 639c0f8 [Reynold Xin] udf tests. 0a0b339 [Reynold Xin] defineUDF -> udf. b452b8d [Reynold Xin] Fix UDF registration. d2e42c3 [Reynold Xin] SQLContext.udf.register() returns a UserDefinedFunction also. 4333605 [Reynold Xin] [SQL][DataFrame] defineUDF. Project: http://git-wip-us.apache.org/repos/asf/spark/repo Commit: http://git-wip-us.apache.org/repos/asf/spark/commit/1077f2e1 Tree: http://git-wip-us.apache.org/repos/asf/spark/tree/1077f2e1 Diff: http://git-wip-us.apache.org/repos/asf/spark/diff/1077f2e1 Branch: refs/heads/master Commit: 1077f2e1def6266aee6ad6f0640a8f46cd273e21 Parents: e380d2d Author: Reynold Xin <r...@databricks.com> Authored: Tue Feb 3 20:07:46 2015 -0800 Committer: Reynold Xin <r...@databricks.com> Committed: Tue Feb 3 20:07:46 2015 -0800 ---------------------------------------------------------------------- .../ml/classification/LogisticRegression.scala | 10 +- .../spark/ml/feature/StandardScaler.scala | 6 +- .../apache/spark/ml/recommendation/ALS.scala | 14 +- .../scala/org/apache/spark/sql/DataFrame.scala | 4 +- .../org/apache/spark/sql/DataFrameImpl.scala | 2 +- .../main/scala/org/apache/spark/sql/Dsl.scala | 194 +++-- .../org/apache/spark/sql/UDFRegistration.scala | 571 ++++++++++++++ .../org/apache/spark/sql/UdfRegistration.scala | 774 ------------------- .../apache/spark/sql/UserDefinedFunction.scala | 39 + .../org/apache/spark/sql/DataFrameSuite.scala | 6 +- .../org/apache/spark/sql/SQLQuerySuite.scala | 6 +- .../apache/spark/sql/UserDefinedTypeSuite.scala | 5 +- 12 files changed, 731 insertions(+), 900 deletions(-) ---------------------------------------------------------------------- http://git-wip-us.apache.org/repos/asf/spark/blob/1077f2e1/mllib/src/main/scala/org/apache/spark/ml/classification/LogisticRegression.scala ---------------------------------------------------------------------- diff --git a/mllib/src/main/scala/org/apache/spark/ml/classification/LogisticRegression.scala b/mllib/src/main/scala/org/apache/spark/ml/classification/LogisticRegression.scala index 18be35a..df90078 100644 --- a/mllib/src/main/scala/org/apache/spark/ml/classification/LogisticRegression.scala +++ b/mllib/src/main/scala/org/apache/spark/ml/classification/LogisticRegression.scala @@ -132,14 +132,14 @@ class LogisticRegressionModel private[ml] ( override def transform(dataset: DataFrame, paramMap: ParamMap): DataFrame = { transformSchema(dataset.schema, paramMap, logging = true) val map = this.paramMap ++ paramMap - val scoreFunction: Vector => Double = (v) => { + val scoreFunction = udf((v: Vector) => { val margin = BLAS.dot(v, weights) 1.0 / (1.0 + math.exp(-margin)) - } + } : Double) val t = map(threshold) - val predictFunction: Double => Double = (score) => { if (score > t) 1.0 else 0.0 } + val predictFunction = udf((score: Double) => { if (score > t) 1.0 else 0.0 } : Double) dataset - .select($"*", callUDF(scoreFunction, col(map(featuresCol))).as(map(scoreCol))) - .select($"*", callUDF(predictFunction, col(map(scoreCol))).as(map(predictionCol))) + .select($"*", scoreFunction(col(map(featuresCol))).as(map(scoreCol))) + .select($"*", predictFunction(col(map(scoreCol))).as(map(predictionCol))) } } http://git-wip-us.apache.org/repos/asf/spark/blob/1077f2e1/mllib/src/main/scala/org/apache/spark/ml/feature/StandardScaler.scala ---------------------------------------------------------------------- diff --git a/mllib/src/main/scala/org/apache/spark/ml/feature/StandardScaler.scala b/mllib/src/main/scala/org/apache/spark/ml/feature/StandardScaler.scala index 01a4f5e..4745a7a 100644 --- a/mllib/src/main/scala/org/apache/spark/ml/feature/StandardScaler.scala +++ b/mllib/src/main/scala/org/apache/spark/ml/feature/StandardScaler.scala @@ -81,10 +81,8 @@ class StandardScalerModel private[ml] ( override def transform(dataset: DataFrame, paramMap: ParamMap): DataFrame = { transformSchema(dataset.schema, paramMap, logging = true) val map = this.paramMap ++ paramMap - val scale: (Vector) => Vector = (v) => { - scaler.transform(v) - } - dataset.select($"*", callUDF(scale, col(map(inputCol))).as(map(outputCol))) + val scale = udf((v: Vector) => { scaler.transform(v) } : Vector) + dataset.select($"*", scale(col(map(inputCol))).as(map(outputCol))) } private[ml] override def transformSchema(schema: StructType, paramMap: ParamMap): StructType = { http://git-wip-us.apache.org/repos/asf/spark/blob/1077f2e1/mllib/src/main/scala/org/apache/spark/ml/recommendation/ALS.scala ---------------------------------------------------------------------- diff --git a/mllib/src/main/scala/org/apache/spark/ml/recommendation/ALS.scala b/mllib/src/main/scala/org/apache/spark/ml/recommendation/ALS.scala index 511cb2f..c7bec7a 100644 --- a/mllib/src/main/scala/org/apache/spark/ml/recommendation/ALS.scala +++ b/mllib/src/main/scala/org/apache/spark/ml/recommendation/ALS.scala @@ -126,22 +126,20 @@ class ALSModel private[ml] ( val map = this.paramMap ++ paramMap val users = userFactors.toDataFrame("id", "features") val items = itemFactors.toDataFrame("id", "features") - val predict: (Seq[Float], Seq[Float]) => Float = (userFeatures, itemFeatures) => { + + // Register a UDF for DataFrame, and then + // create a new column named map(predictionCol) by running the predict UDF. + val predict = udf((userFeatures: Seq[Float], itemFeatures: Seq[Float]) => { if (userFeatures != null && itemFeatures != null) { blas.sdot(k, userFeatures.toArray, 1, itemFeatures.toArray, 1) } else { Float.NaN } - } - val inputColumns = dataset.schema.fieldNames - val prediction = callUDF(predict, users("features"), items("features")).as(map(predictionCol)) - val outputColumns = inputColumns.map(f => dataset(f)) :+ prediction + } : Float) dataset .join(users, dataset(map(userCol)) === users("id"), "left") .join(items, dataset(map(itemCol)) === items("id"), "left") - .select(outputColumns: _*) - // TODO: Just use a dataset("*") - // .select(dataset("*"), prediction) + .select(dataset("*"), predict(users("features"), items("features")).as(map(predictionCol))) } override private[ml] def transformSchema(schema: StructType, paramMap: ParamMap): StructType = { http://git-wip-us.apache.org/repos/asf/spark/blob/1077f2e1/sql/core/src/main/scala/org/apache/spark/sql/DataFrame.scala ---------------------------------------------------------------------- diff --git a/sql/core/src/main/scala/org/apache/spark/sql/DataFrame.scala b/sql/core/src/main/scala/org/apache/spark/sql/DataFrame.scala index f3bc07a..732b685 100644 --- a/sql/core/src/main/scala/org/apache/spark/sql/DataFrame.scala +++ b/sql/core/src/main/scala/org/apache/spark/sql/DataFrame.scala @@ -54,10 +54,10 @@ private[sql] object DataFrame { * }}} * * Note that the [[Column]] type can also be manipulated through its various functions. - * {{ + * {{{ * // The following creates a new column that increases everybody's age by 10. * people("age") + 10 // in Scala - * }} + * }}} * * A more concrete example: * {{{ http://git-wip-us.apache.org/repos/asf/spark/blob/1077f2e1/sql/core/src/main/scala/org/apache/spark/sql/DataFrameImpl.scala ---------------------------------------------------------------------- diff --git a/sql/core/src/main/scala/org/apache/spark/sql/DataFrameImpl.scala b/sql/core/src/main/scala/org/apache/spark/sql/DataFrameImpl.scala index 0b0623d..a52bfa5 100644 --- a/sql/core/src/main/scala/org/apache/spark/sql/DataFrameImpl.scala +++ b/sql/core/src/main/scala/org/apache/spark/sql/DataFrameImpl.scala @@ -329,7 +329,7 @@ private[sql] class DataFrameImpl protected[sql]( override def save(path: String): Unit = { val dataSourceName = sqlContext.conf.defaultDataSourceName - save(dataSourceName, ("path" -> path)) + save(dataSourceName, "path" -> path) } override def save( http://git-wip-us.apache.org/repos/asf/spark/blob/1077f2e1/sql/core/src/main/scala/org/apache/spark/sql/Dsl.scala ---------------------------------------------------------------------- diff --git a/sql/core/src/main/scala/org/apache/spark/sql/Dsl.scala b/sql/core/src/main/scala/org/apache/spark/sql/Dsl.scala index 71365c7..8cf59f0 100644 --- a/sql/core/src/main/scala/org/apache/spark/sql/Dsl.scala +++ b/sql/core/src/main/scala/org/apache/spark/sql/Dsl.scala @@ -186,15 +186,13 @@ object Dsl { (0 to 22).map { x => val types = (1 to x).foldRight("RT")((i, s) => {s"A$i, $s"}) val typeTags = (1 to x).map(i => s"A$i: TypeTag").foldLeft("RT: TypeTag")(_ + ", " + _) - val args = (1 to x).map(i => s"arg$i: Column").mkString(", ") - val argsInUdf = (1 to x).map(i => s"arg$i.expr").mkString(", ") println(s""" /** - * Call a Scala function of ${x} arguments as user-defined function (UDF), and automatically - * infer the data types based on the function's signature. + * Defines a user-defined function of ${x} arguments as user-defined function (UDF). + * The data types are automatically inferred based on the function's signature. */ - def callUDF[$typeTags](f: Function$x[$types]${if (args.length > 0) ", " + args else ""}): Column = { - ScalaUdf(f, ScalaReflection.schemaFor(typeTag[RT]).dataType, Seq($argsInUdf)) + def udf[$typeTags](f: Function$x[$types]): UserDefinedFunction = { + UserDefinedFunction(f, ScalaReflection.schemaFor(typeTag[RT]).dataType) }""") } @@ -214,187 +212,187 @@ object Dsl { } */ /** - * Call a Scala function of 0 arguments as user-defined function (UDF), and automatically - * infer the data types based on the function's signature. + * Defines a user-defined function of 0 arguments as user-defined function (UDF). + * The data types are automatically inferred based on the function's signature. */ - def callUDF[RT: TypeTag](f: Function0[RT]): Column = { - ScalaUdf(f, ScalaReflection.schemaFor(typeTag[RT]).dataType, Seq()) + def udf[RT: TypeTag](f: Function0[RT]): UserDefinedFunction = { + UserDefinedFunction(f, ScalaReflection.schemaFor(typeTag[RT]).dataType) } /** - * Call a Scala function of 1 arguments as user-defined function (UDF), and automatically - * infer the data types based on the function's signature. + * Defines a user-defined function of 1 arguments as user-defined function (UDF). + * The data types are automatically inferred based on the function's signature. */ - def callUDF[RT: TypeTag, A1: TypeTag](f: Function1[A1, RT], arg1: Column): Column = { - ScalaUdf(f, ScalaReflection.schemaFor(typeTag[RT]).dataType, Seq(arg1.expr)) + def udf[RT: TypeTag, A1: TypeTag](f: Function1[A1, RT]): UserDefinedFunction = { + UserDefinedFunction(f, ScalaReflection.schemaFor(typeTag[RT]).dataType) } /** - * Call a Scala function of 2 arguments as user-defined function (UDF), and automatically - * infer the data types based on the function's signature. + * Defines a user-defined function of 2 arguments as user-defined function (UDF). + * The data types are automatically inferred based on the function's signature. */ - def callUDF[RT: TypeTag, A1: TypeTag, A2: TypeTag](f: Function2[A1, A2, RT], arg1: Column, arg2: Column): Column = { - ScalaUdf(f, ScalaReflection.schemaFor(typeTag[RT]).dataType, Seq(arg1.expr, arg2.expr)) + def udf[RT: TypeTag, A1: TypeTag, A2: TypeTag](f: Function2[A1, A2, RT]): UserDefinedFunction = { + UserDefinedFunction(f, ScalaReflection.schemaFor(typeTag[RT]).dataType) } /** - * Call a Scala function of 3 arguments as user-defined function (UDF), and automatically - * infer the data types based on the function's signature. + * Defines a user-defined function of 3 arguments as user-defined function (UDF). + * The data types are automatically inferred based on the function's signature. */ - def callUDF[RT: TypeTag, A1: TypeTag, A2: TypeTag, A3: TypeTag](f: Function3[A1, A2, A3, RT], arg1: Column, arg2: Column, arg3: Column): Column = { - ScalaUdf(f, ScalaReflection.schemaFor(typeTag[RT]).dataType, Seq(arg1.expr, arg2.expr, arg3.expr)) + def udf[RT: TypeTag, A1: TypeTag, A2: TypeTag, A3: TypeTag](f: Function3[A1, A2, A3, RT]): UserDefinedFunction = { + UserDefinedFunction(f, ScalaReflection.schemaFor(typeTag[RT]).dataType) } /** - * Call a Scala function of 4 arguments as user-defined function (UDF), and automatically - * infer the data types based on the function's signature. + * Defines a user-defined function of 4 arguments as user-defined function (UDF). + * The data types are automatically inferred based on the function's signature. */ - def callUDF[RT: TypeTag, A1: TypeTag, A2: TypeTag, A3: TypeTag, A4: TypeTag](f: Function4[A1, A2, A3, A4, RT], arg1: Column, arg2: Column, arg3: Column, arg4: Column): Column = { - ScalaUdf(f, ScalaReflection.schemaFor(typeTag[RT]).dataType, Seq(arg1.expr, arg2.expr, arg3.expr, arg4.expr)) + def udf[RT: TypeTag, A1: TypeTag, A2: TypeTag, A3: TypeTag, A4: TypeTag](f: Function4[A1, A2, A3, A4, RT]): UserDefinedFunction = { + UserDefinedFunction(f, ScalaReflection.schemaFor(typeTag[RT]).dataType) } /** - * Call a Scala function of 5 arguments as user-defined function (UDF), and automatically - * infer the data types based on the function's signature. + * Defines a user-defined function of 5 arguments as user-defined function (UDF). + * The data types are automatically inferred based on the function's signature. */ - def callUDF[RT: TypeTag, A1: TypeTag, A2: TypeTag, A3: TypeTag, A4: TypeTag, A5: TypeTag](f: Function5[A1, A2, A3, A4, A5, RT], arg1: Column, arg2: Column, arg3: Column, arg4: Column, arg5: Column): Column = { - ScalaUdf(f, ScalaReflection.schemaFor(typeTag[RT]).dataType, Seq(arg1.expr, arg2.expr, arg3.expr, arg4.expr, arg5.expr)) + def udf[RT: TypeTag, A1: TypeTag, A2: TypeTag, A3: TypeTag, A4: TypeTag, A5: TypeTag](f: Function5[A1, A2, A3, A4, A5, RT]): UserDefinedFunction = { + UserDefinedFunction(f, ScalaReflection.schemaFor(typeTag[RT]).dataType) } /** - * Call a Scala function of 6 arguments as user-defined function (UDF), and automatically - * infer the data types based on the function's signature. + * Defines a user-defined function of 6 arguments as user-defined function (UDF). + * The data types are automatically inferred based on the function's signature. */ - def callUDF[RT: TypeTag, A1: TypeTag, A2: TypeTag, A3: TypeTag, A4: TypeTag, A5: TypeTag, A6: TypeTag](f: Function6[A1, A2, A3, A4, A5, A6, RT], arg1: Column, arg2: Column, arg3: Column, arg4: Column, arg5: Column, arg6: Column): Column = { - ScalaUdf(f, ScalaReflection.schemaFor(typeTag[RT]).dataType, Seq(arg1.expr, arg2.expr, arg3.expr, arg4.expr, arg5.expr, arg6.expr)) + def udf[RT: TypeTag, A1: TypeTag, A2: TypeTag, A3: TypeTag, A4: TypeTag, A5: TypeTag, A6: TypeTag](f: Function6[A1, A2, A3, A4, A5, A6, RT]): UserDefinedFunction = { + UserDefinedFunction(f, ScalaReflection.schemaFor(typeTag[RT]).dataType) } /** - * Call a Scala function of 7 arguments as user-defined function (UDF), and automatically - * infer the data types based on the function's signature. + * Defines a user-defined function of 7 arguments as user-defined function (UDF). + * The data types are automatically inferred based on the function's signature. */ - def callUDF[RT: TypeTag, A1: TypeTag, A2: TypeTag, A3: TypeTag, A4: TypeTag, A5: TypeTag, A6: TypeTag, A7: TypeTag](f: Function7[A1, A2, A3, A4, A5, A6, A7, RT], arg1: Column, arg2: Column, arg3: Column, arg4: Column, arg5: Column, arg6: Column, arg7: Column): Column = { - ScalaUdf(f, ScalaReflection.schemaFor(typeTag[RT]).dataType, Seq(arg1.expr, arg2.expr, arg3.expr, arg4.expr, arg5.expr, arg6.expr, arg7.expr)) + def udf[RT: TypeTag, A1: TypeTag, A2: TypeTag, A3: TypeTag, A4: TypeTag, A5: TypeTag, A6: TypeTag, A7: TypeTag](f: Function7[A1, A2, A3, A4, A5, A6, A7, RT]): UserDefinedFunction = { + UserDefinedFunction(f, ScalaReflection.schemaFor(typeTag[RT]).dataType) } /** - * Call a Scala function of 8 arguments as user-defined function (UDF), and automatically - * infer the data types based on the function's signature. + * Defines a user-defined function of 8 arguments as user-defined function (UDF). + * The data types are automatically inferred based on the function's signature. */ - def callUDF[RT: TypeTag, A1: TypeTag, A2: TypeTag, A3: TypeTag, A4: TypeTag, A5: TypeTag, A6: TypeTag, A7: TypeTag, A8: TypeTag](f: Function8[A1, A2, A3, A4, A5, A6, A7, A8, RT], arg1: Column, arg2: Column, arg3: Column, arg4: Column, arg5: Column, arg6: Column, arg7: Column, arg8: Column): Column = { - ScalaUdf(f, ScalaReflection.schemaFor(typeTag[RT]).dataType, Seq(arg1.expr, arg2.expr, arg3.expr, arg4.expr, arg5.expr, arg6.expr, arg7.expr, arg8.expr)) + def udf[RT: TypeTag, A1: TypeTag, A2: TypeTag, A3: TypeTag, A4: TypeTag, A5: TypeTag, A6: TypeTag, A7: TypeTag, A8: TypeTag](f: Function8[A1, A2, A3, A4, A5, A6, A7, A8, RT]): UserDefinedFunction = { + UserDefinedFunction(f, ScalaReflection.schemaFor(typeTag[RT]).dataType) } /** - * Call a Scala function of 9 arguments as user-defined function (UDF), and automatically - * infer the data types based on the function's signature. + * Defines a user-defined function of 9 arguments as user-defined function (UDF). + * The data types are automatically inferred based on the function's signature. */ - def callUDF[RT: TypeTag, A1: TypeTag, A2: TypeTag, A3: TypeTag, A4: TypeTag, A5: TypeTag, A6: TypeTag, A7: TypeTag, A8: TypeTag, A9: TypeTag](f: Function9[A1, A2, A3, A4, A5, A6, A7, A8, A9, RT], arg1: Column, arg2: Column, arg3: Column, arg4: Column, arg5: Column, arg6: Column, arg7: Column, arg8: Column, arg9: Column): Column = { - ScalaUdf(f, ScalaReflection.schemaFor(typeTag[RT]).dataType, Seq(arg1.expr, arg2.expr, arg3.expr, arg4.expr, arg5.expr, arg6.expr, arg7.expr, arg8.expr, arg9.expr)) + def udf[RT: TypeTag, A1: TypeTag, A2: TypeTag, A3: TypeTag, A4: TypeTag, A5: TypeTag, A6: TypeTag, A7: TypeTag, A8: TypeTag, A9: TypeTag](f: Function9[A1, A2, A3, A4, A5, A6, A7, A8, A9, RT]): UserDefinedFunction = { + UserDefinedFunction(f, ScalaReflection.schemaFor(typeTag[RT]).dataType) } /** - * Call a Scala function of 10 arguments as user-defined function (UDF), and automatically - * infer the data types based on the function's signature. + * Defines a user-defined function of 10 arguments as user-defined function (UDF). + * The data types are automatically inferred based on the function's signature. */ - def callUDF[RT: TypeTag, A1: TypeTag, A2: TypeTag, A3: TypeTag, A4: TypeTag, A5: TypeTag, A6: TypeTag, A7: TypeTag, A8: TypeTag, A9: TypeTag, A10: TypeTag](f: Function10[A1, A2, A3, A4, A5, A6, A7, A8, A9, A10, RT], arg1: Column, arg2: Column, arg3: Column, arg4: Column, arg5: Column, arg6: Column, arg7: Column, arg8: Column, arg9: Column, arg10: Column): Column = { - ScalaUdf(f, ScalaReflection.schemaFor(typeTag[RT]).dataType, Seq(arg1.expr, arg2.expr, arg3.expr, arg4.expr, arg5.expr, arg6.expr, arg7.expr, arg8.expr, arg9.expr, arg10.expr)) + def udf[RT: TypeTag, A1: TypeTag, A2: TypeTag, A3: TypeTag, A4: TypeTag, A5: TypeTag, A6: TypeTag, A7: TypeTag, A8: TypeTag, A9: TypeTag, A10: TypeTag](f: Function10[A1, A2, A3, A4, A5, A6, A7, A8, A9, A10, RT]): UserDefinedFunction = { + UserDefinedFunction(f, ScalaReflection.schemaFor(typeTag[RT]).dataType) } /** - * Call a Scala function of 11 arguments as user-defined function (UDF), and automatically - * infer the data types based on the function's signature. + * Defines a user-defined function of 11 arguments as user-defined function (UDF). + * The data types are automatically inferred based on the function's signature. */ - def callUDF[RT: TypeTag, A1: TypeTag, A2: TypeTag, A3: TypeTag, A4: TypeTag, A5: TypeTag, A6: TypeTag, A7: TypeTag, A8: TypeTag, A9: TypeTag, A10: TypeTag, A11: TypeTag](f: Function11[A1, A2, A3, A4, A5, A6, A7, A8, A9, A10, A11, RT], arg1: Column, arg2: Column, arg3: Column, arg4: Column, arg5: Column, arg6: Column, arg7: Column, arg8: Column, arg9: Column, arg10: Column, arg11: Column): Column = { - ScalaUdf(f, ScalaReflection.schemaFor(typeTag[RT]).dataType, Seq(arg1.expr, arg2.expr, arg3.expr, arg4.expr, arg5.expr, arg6.expr, arg7.expr, arg8.expr, arg9.expr, arg10.expr, arg11.expr)) + def udf[RT: TypeTag, A1: TypeTag, A2: TypeTag, A3: TypeTag, A4: TypeTag, A5: TypeTag, A6: TypeTag, A7: TypeTag, A8: TypeTag, A9: TypeTag, A10: TypeTag, A11: TypeTag](f: Function11[A1, A2, A3, A4, A5, A6, A7, A8, A9, A10, A11, RT]): UserDefinedFunction = { + UserDefinedFunction(f, ScalaReflection.schemaFor(typeTag[RT]).dataType) } /** - * Call a Scala function of 12 arguments as user-defined function (UDF), and automatically - * infer the data types based on the function's signature. + * Defines a user-defined function of 12 arguments as user-defined function (UDF). + * The data types are automatically inferred based on the function's signature. */ - def callUDF[RT: TypeTag, A1: TypeTag, A2: TypeTag, A3: TypeTag, A4: TypeTag, A5: TypeTag, A6: TypeTag, A7: TypeTag, A8: TypeTag, A9: TypeTag, A10: TypeTag, A11: TypeTag, A12: TypeTag](f: Function12[A1, A2, A3, A4, A5, A6, A7, A8, A9, A10, A11, A12, RT], arg1: Column, arg2: Column, arg3: Column, arg4: Column, arg5: Column, arg6: Column, arg7: Column, arg8: Column, arg9: Column, arg10: Column, arg11: Column, arg12: Column): Column = { - ScalaUdf(f, ScalaReflection.schemaFor(typeTag[RT]).dataType, Seq(arg1.expr, arg2.expr, arg3.expr, arg4.expr, arg5.expr, arg6.expr, arg7.expr, arg8.expr, arg9.expr, arg10.expr, arg11.expr, arg12.expr)) + def udf[RT: TypeTag, A1: TypeTag, A2: TypeTag, A3: TypeTag, A4: TypeTag, A5: TypeTag, A6: TypeTag, A7: TypeTag, A8: TypeTag, A9: TypeTag, A10: TypeTag, A11: TypeTag, A12: TypeTag](f: Function12[A1, A2, A3, A4, A5, A6, A7, A8, A9, A10, A11, A12, RT]): UserDefinedFunction = { + UserDefinedFunction(f, ScalaReflection.schemaFor(typeTag[RT]).dataType) } /** - * Call a Scala function of 13 arguments as user-defined function (UDF), and automatically - * infer the data types based on the function's signature. + * Defines a user-defined function of 13 arguments as user-defined function (UDF). + * The data types are automatically inferred based on the function's signature. */ - def callUDF[RT: TypeTag, A1: TypeTag, A2: TypeTag, A3: TypeTag, A4: TypeTag, A5: TypeTag, A6: TypeTag, A7: TypeTag, A8: TypeTag, A9: TypeTag, A10: TypeTag, A11: TypeTag, A12: TypeTag, A13: TypeTag](f: Function13[A1, A2, A3, A4, A5, A6, A7, A8, A9, A10, A11, A12, A13, RT], arg1: Column, arg2: Column, arg3: Column, arg4: Column, arg5: Column, arg6: Column, arg7: Column, arg8: Column, arg9: Column, arg10: Column, arg11: Column, arg12: Column, arg13: Column): Column = { - ScalaUdf(f, ScalaReflection.schemaFor(typeTag[RT]).dataType, Seq(arg1.expr, arg2.expr, arg3.expr, arg4.expr, arg5.expr, arg6.expr, arg7.expr, arg8.expr, arg9.expr, arg10.expr, arg11.expr, arg12.expr, arg13.expr)) + def udf[RT: TypeTag, A1: TypeTag, A2: TypeTag, A3: TypeTag, A4: TypeTag, A5: TypeTag, A6: TypeTag, A7: TypeTag, A8: TypeTag, A9: TypeTag, A10: TypeTag, A11: TypeTag, A12: TypeTag, A13: TypeTag](f: Function13[A1, A2, A3, A4, A5, A6, A7, A8, A9, A10, A11, A12, A13, RT]): UserDefinedFunction = { + UserDefinedFunction(f, ScalaReflection.schemaFor(typeTag[RT]).dataType) } /** - * Call a Scala function of 14 arguments as user-defined function (UDF), and automatically - * infer the data types based on the function's signature. + * Defines a user-defined function of 14 arguments as user-defined function (UDF). + * The data types are automatically inferred based on the function's signature. */ - def callUDF[RT: TypeTag, A1: TypeTag, A2: TypeTag, A3: TypeTag, A4: TypeTag, A5: TypeTag, A6: TypeTag, A7: TypeTag, A8: TypeTag, A9: TypeTag, A10: TypeTag, A11: TypeTag, A12: TypeTag, A13: TypeTag, A14: TypeTag](f: Function14[A1, A2, A3, A4, A5, A6, A7, A8, A9, A10, A11, A12, A13, A14, RT], arg1: Column, arg2: Column, arg3: Column, arg4: Column, arg5: Column, arg6: Column, arg7: Column, arg8: Column, arg9: Column, arg10: Column, arg11: Column, arg12: Column, arg13: Column, arg14: Column): Column = { - ScalaUdf(f, ScalaReflection.schemaFor(typeTag[RT]).dataType, Seq(arg1.expr, arg2.expr, arg3.expr, arg4.expr, arg5.expr, arg6.expr, arg7.expr, arg8.expr, arg9.expr, arg10.expr, arg11.expr, arg12.expr, arg13.expr, arg14.expr)) + def udf[RT: TypeTag, A1: TypeTag, A2: TypeTag, A3: TypeTag, A4: TypeTag, A5: TypeTag, A6: TypeTag, A7: TypeTag, A8: TypeTag, A9: TypeTag, A10: TypeTag, A11: TypeTag, A12: TypeTag, A13: TypeTag, A14: TypeTag](f: Function14[A1, A2, A3, A4, A5, A6, A7, A8, A9, A10, A11, A12, A13, A14, RT]): UserDefinedFunction = { + UserDefinedFunction(f, ScalaReflection.schemaFor(typeTag[RT]).dataType) } /** - * Call a Scala function of 15 arguments as user-defined function (UDF), and automatically - * infer the data types based on the function's signature. + * Defines a user-defined function of 15 arguments as user-defined function (UDF). + * The data types are automatically inferred based on the function's signature. */ - def callUDF[RT: TypeTag, A1: TypeTag, A2: TypeTag, A3: TypeTag, A4: TypeTag, A5: TypeTag, A6: TypeTag, A7: TypeTag, A8: TypeTag, A9: TypeTag, A10: TypeTag, A11: TypeTag, A12: TypeTag, A13: TypeTag, A14: TypeTag, A15: TypeTag](f: Function15[A1, A2, A3, A4, A5, A6, A7, A8, A9, A10, A11, A12, A13, A14, A15, RT], arg1: Column, arg2: Column, arg3: Column, arg4: Column, arg5: Column, arg6: Column, arg7: Column, arg8: Column, arg9: Column, arg10: Column, arg11: Column, arg12: Column, arg13: Column, arg14: Column, arg15: Column): Column = { - ScalaUdf(f, ScalaReflection.schemaFor(typeTag[RT]).dataType, Seq(arg1.expr, arg2.expr, arg3.expr, arg4.expr, arg5.expr, arg6.expr, arg7.expr, arg8.expr, arg9.expr, arg10.expr, arg11.expr, arg12.expr, arg13.expr, arg14.expr, arg15.expr)) + def udf[RT: TypeTag, A1: TypeTag, A2: TypeTag, A3: TypeTag, A4: TypeTag, A5: TypeTag, A6: TypeTag, A7: TypeTag, A8: TypeTag, A9: TypeTag, A10: TypeTag, A11: TypeTag, A12: TypeTag, A13: TypeTag, A14: TypeTag, A15: TypeTag](f: Function15[A1, A2, A3, A4, A5, A6, A7, A8, A9, A10, A11, A12, A13, A14, A15, RT]): UserDefinedFunction = { + UserDefinedFunction(f, ScalaReflection.schemaFor(typeTag[RT]).dataType) } /** - * Call a Scala function of 16 arguments as user-defined function (UDF), and automatically - * infer the data types based on the function's signature. + * Defines a user-defined function of 16 arguments as user-defined function (UDF). + * The data types are automatically inferred based on the function's signature. */ - def callUDF[RT: TypeTag, A1: TypeTag, A2: TypeTag, A3: TypeTag, A4: TypeTag, A5: TypeTag, A6: TypeTag, A7: TypeTag, A8: TypeTag, A9: TypeTag, A10: TypeTag, A11: TypeTag, A12: TypeTag, A13: TypeTag, A14: TypeTag, A15: TypeTag, A16: TypeTag](f: Function16[A1, A2, A3, A4, A5, A6, A7, A8, A9, A10, A11, A12, A13, A14, A15, A16, RT], arg1: Column, arg2: Column, arg3: Column, arg4: Column, arg5: Column, arg6: Column, arg7: Column, arg8: Column, arg9: Column, arg10: Column, arg11: Column, arg12: Column, arg13: Column, arg14: Column, arg15: Column, arg16: Column): Column = { - ScalaUdf(f, ScalaReflection.schemaFor(typeTag[RT]).dataType, Seq(arg1.expr, arg2.expr, arg3.expr, arg4.expr, arg5.expr, arg6.expr, arg7.expr, arg8.expr, arg9.expr, arg10.expr, arg11.expr, arg12.expr, arg13.expr, arg14.expr, arg15.expr, arg16.expr)) + def udf[RT: TypeTag, A1: TypeTag, A2: TypeTag, A3: TypeTag, A4: TypeTag, A5: TypeTag, A6: TypeTag, A7: TypeTag, A8: TypeTag, A9: TypeTag, A10: TypeTag, A11: TypeTag, A12: TypeTag, A13: TypeTag, A14: TypeTag, A15: TypeTag, A16: TypeTag](f: Function16[A1, A2, A3, A4, A5, A6, A7, A8, A9, A10, A11, A12, A13, A14, A15, A16, RT]): UserDefinedFunction = { + UserDefinedFunction(f, ScalaReflection.schemaFor(typeTag[RT]).dataType) } /** - * Call a Scala function of 17 arguments as user-defined function (UDF), and automatically - * infer the data types based on the function's signature. + * Defines a user-defined function of 17 arguments as user-defined function (UDF). + * The data types are automatically inferred based on the function's signature. */ - def callUDF[RT: TypeTag, A1: TypeTag, A2: TypeTag, A3: TypeTag, A4: TypeTag, A5: TypeTag, A6: TypeTag, A7: TypeTag, A8: TypeTag, A9: TypeTag, A10: TypeTag, A11: TypeTag, A12: TypeTag, A13: TypeTag, A14: TypeTag, A15: TypeTag, A16: TypeTag, A17: TypeTag](f: Function17[A1, A2, A3, A4, A5, A6, A7, A8, A9, A10, A11, A12, A13, A14, A15, A16, A17, RT], arg1: Column, arg2: Column, arg3: Column, arg4: Column, arg5: Column, arg6: Column, arg7: Column, arg8: Column, arg9: Column, arg10: Column, arg11: Column, arg12: Column, arg13: Column, arg14: Column, arg15: Column, arg16: Column, arg17: Column): Column = { - ScalaUdf(f, ScalaReflection.schemaFor(typeTag[RT]).dataType, Seq(arg1.expr, arg2.expr, arg3.expr, arg4.expr, arg5.expr, arg6.expr, arg7.expr, arg8.expr, arg9.expr, arg10.expr, arg11.expr, arg12.expr, arg13.expr, arg14.expr, arg15.expr, arg16.expr, arg17.expr)) + def udf[RT: TypeTag, A1: TypeTag, A2: TypeTag, A3: TypeTag, A4: TypeTag, A5: TypeTag, A6: TypeTag, A7: TypeTag, A8: TypeTag, A9: TypeTag, A10: TypeTag, A11: TypeTag, A12: TypeTag, A13: TypeTag, A14: TypeTag, A15: TypeTag, A16: TypeTag, A17: TypeTag](f: Function17[A1, A2, A3, A4, A5, A6, A7, A8, A9, A10, A11, A12, A13, A14, A15, A16, A17, RT]): UserDefinedFunction = { + UserDefinedFunction(f, ScalaReflection.schemaFor(typeTag[RT]).dataType) } /** - * Call a Scala function of 18 arguments as user-defined function (UDF), and automatically - * infer the data types based on the function's signature. + * Defines a user-defined function of 18 arguments as user-defined function (UDF). + * The data types are automatically inferred based on the function's signature. */ - def callUDF[RT: TypeTag, A1: TypeTag, A2: TypeTag, A3: TypeTag, A4: TypeTag, A5: TypeTag, A6: TypeTag, A7: TypeTag, A8: TypeTag, A9: TypeTag, A10: TypeTag, A11: TypeTag, A12: TypeTag, A13: TypeTag, A14: TypeTag, A15: TypeTag, A16: TypeTag, A17: TypeTag, A18: TypeTag](f: Function18[A1, A2, A3, A4, A5, A6, A7, A8, A9, A10, A11, A12, A13, A14, A15, A16, A17, A18, RT], arg1: Column, arg2: Column, arg3: Column, arg4: Column, arg5: Column, arg6: Column, arg7: Column, arg8: Column, arg9: Column, arg10: Column, arg11: Column, arg12: Column, arg13: Column, arg14: Column, arg15: Column, arg16: Column, arg17: Column, arg18: Column): Column = { - ScalaUdf(f, ScalaReflection.schemaFor(typeTag[RT]).dataType, Seq(arg1.expr, arg2.expr, arg3.expr, arg4.expr, arg5.expr, arg6.expr, arg7.expr, arg8.expr, arg9.expr, arg10.expr, arg11.expr, arg12.expr, arg13.expr, arg14.expr, arg15.expr, arg16.expr, arg17.expr, arg18.expr)) + def udf[RT: TypeTag, A1: TypeTag, A2: TypeTag, A3: TypeTag, A4: TypeTag, A5: TypeTag, A6: TypeTag, A7: TypeTag, A8: TypeTag, A9: TypeTag, A10: TypeTag, A11: TypeTag, A12: TypeTag, A13: TypeTag, A14: TypeTag, A15: TypeTag, A16: TypeTag, A17: TypeTag, A18: TypeTag](f: Function18[A1, A2, A3, A4, A5, A6, A7, A8, A9, A10, A11, A12, A13, A14, A15, A16, A17, A18, RT]): UserDefinedFunction = { + UserDefinedFunction(f, ScalaReflection.schemaFor(typeTag[RT]).dataType) } /** - * Call a Scala function of 19 arguments as user-defined function (UDF), and automatically - * infer the data types based on the function's signature. + * Defines a user-defined function of 19 arguments as user-defined function (UDF). + * The data types are automatically inferred based on the function's signature. */ - def callUDF[RT: TypeTag, A1: TypeTag, A2: TypeTag, A3: TypeTag, A4: TypeTag, A5: TypeTag, A6: TypeTag, A7: TypeTag, A8: TypeTag, A9: TypeTag, A10: TypeTag, A11: TypeTag, A12: TypeTag, A13: TypeTag, A14: TypeTag, A15: TypeTag, A16: TypeTag, A17: TypeTag, A18: TypeTag, A19: TypeTag](f: Function19[A1, A2, A3, A4, A5, A6, A7, A8, A9, A10, A11, A12, A13, A14, A15, A16, A17, A18, A19, RT], arg1: Column, arg2: Column, arg3: Column, arg4: Column, arg5: Column, arg6: Column, arg7: Column, arg8: Column, arg9: Column, arg10: Column, arg11: Column, arg12: Column, arg13: Column, arg14: Column, arg15: Column, arg16: Column, arg17: Column, arg18: Column, arg19: Column): Column = { - ScalaUdf(f, ScalaReflection.schemaFor(typeTag[RT]).dataType, Seq(arg1.expr, arg2.expr, arg3.expr, arg4.expr, arg5.expr, arg6.expr, arg7.expr, arg8.expr, arg9.expr, arg10.expr, arg11.expr, arg12.expr, arg13.expr, arg14.expr, arg15.expr, arg16.expr, arg17.expr, arg18.expr, arg19.expr)) + def udf[RT: TypeTag, A1: TypeTag, A2: TypeTag, A3: TypeTag, A4: TypeTag, A5: TypeTag, A6: TypeTag, A7: TypeTag, A8: TypeTag, A9: TypeTag, A10: TypeTag, A11: TypeTag, A12: TypeTag, A13: TypeTag, A14: TypeTag, A15: TypeTag, A16: TypeTag, A17: TypeTag, A18: TypeTag, A19: TypeTag](f: Function19[A1, A2, A3, A4, A5, A6, A7, A8, A9, A10, A11, A12, A13, A14, A15, A16, A17, A18, A19, RT]): UserDefinedFunction = { + UserDefinedFunction(f, ScalaReflection.schemaFor(typeTag[RT]).dataType) } /** - * Call a Scala function of 20 arguments as user-defined function (UDF), and automatically - * infer the data types based on the function's signature. + * Defines a user-defined function of 20 arguments as user-defined function (UDF). + * The data types are automatically inferred based on the function's signature. */ - def callUDF[RT: TypeTag, A1: TypeTag, A2: TypeTag, A3: TypeTag, A4: TypeTag, A5: TypeTag, A6: TypeTag, A7: TypeTag, A8: TypeTag, A9: TypeTag, A10: TypeTag, A11: TypeTag, A12: TypeTag, A13: TypeTag, A14: TypeTag, A15: TypeTag, A16: TypeTag, A17: TypeTag, A18: TypeTag, A19: TypeTag, A20: TypeTag](f: Function20[A1, A2, A3, A4, A5, A6, A7, A8, A9, A10, A11, A12, A13, A14, A15, A16, A17, A18, A19, A20, RT], arg1: Column, arg2: Column, arg3: Column, arg4: Column, arg5: Column, arg6: Column, arg7: Column, arg8: Column, arg9: Column, arg10: Column, arg11: Column, arg12: Column, arg13: Column, arg14: Column, arg15: Column, arg16: Column, arg17: Column, arg18: Column, arg19: Column, arg20: Column): Column = { - ScalaUdf(f, ScalaReflection.schemaFor(typeTag[RT]).dataType, Seq(arg1.expr, arg2.expr, arg3.expr, arg4.expr, arg5.expr, arg6.expr, arg7.expr, arg8.expr, arg9.expr, arg10.expr, arg11.expr, arg12.expr, arg13.expr, arg14.expr, arg15.expr, arg16.expr, arg17.expr, arg18.expr, arg19.expr, arg20.expr)) + def udf[RT: TypeTag, A1: TypeTag, A2: TypeTag, A3: TypeTag, A4: TypeTag, A5: TypeTag, A6: TypeTag, A7: TypeTag, A8: TypeTag, A9: TypeTag, A10: TypeTag, A11: TypeTag, A12: TypeTag, A13: TypeTag, A14: TypeTag, A15: TypeTag, A16: TypeTag, A17: TypeTag, A18: TypeTag, A19: TypeTag, A20: TypeTag](f: Function20[A1, A2, A3, A4, A5, A6, A7, A8, A9, A10, A11, A12, A13, A14, A15, A16, A17, A18, A19, A20, RT]): UserDefinedFunction = { + UserDefinedFunction(f, ScalaReflection.schemaFor(typeTag[RT]).dataType) } /** - * Call a Scala function of 21 arguments as user-defined function (UDF), and automatically - * infer the data types based on the function's signature. + * Defines a user-defined function of 21 arguments as user-defined function (UDF). + * The data types are automatically inferred based on the function's signature. */ - def callUDF[RT: TypeTag, A1: TypeTag, A2: TypeTag, A3: TypeTag, A4: TypeTag, A5: TypeTag, A6: TypeTag, A7: TypeTag, A8: TypeTag, A9: TypeTag, A10: TypeTag, A11: TypeTag, A12: TypeTag, A13: TypeTag, A14: TypeTag, A15: TypeTag, A16: TypeTag, A17: TypeTag, A18: TypeTag, A19: TypeTag, A20: TypeTag, A21: TypeTag](f: Function21[A1, A2, A3, A4, A5, A6, A7, A8, A9, A10, A11, A12, A13, A14, A15, A16, A17, A18, A19, A20, A21, RT], arg1: Column, arg2: Column, arg3: Column, arg4: Column, arg5: Column, arg6: Column, arg7: Column, arg8: Column, arg9: Column, arg10: Column, arg11: Column, arg12: Column, arg13: Column, arg14: Column, arg15: Column, arg16: Column, arg17: Column, arg18: Column, arg19: Column, arg20: Column, arg21: Column): Column = { - ScalaUdf(f, ScalaReflection.schemaFor(typeTag[RT]).dataType, Seq(arg1.expr, arg2.expr, arg3.expr, arg4.expr, arg5.expr, arg6.expr, arg7.expr, arg8.expr, arg9.expr, arg10.expr, arg11.expr, arg12.expr, arg13.expr, arg14.expr, arg15.expr, arg16.expr, arg17.expr, arg18.expr, arg19.expr, arg20.expr, arg21.expr)) + def udf[RT: TypeTag, A1: TypeTag, A2: TypeTag, A3: TypeTag, A4: TypeTag, A5: TypeTag, A6: TypeTag, A7: TypeTag, A8: TypeTag, A9: TypeTag, A10: TypeTag, A11: TypeTag, A12: TypeTag, A13: TypeTag, A14: TypeTag, A15: TypeTag, A16: TypeTag, A17: TypeTag, A18: TypeTag, A19: TypeTag, A20: TypeTag, A21: TypeTag](f: Function21[A1, A2, A3, A4, A5, A6, A7, A8, A9, A10, A11, A12, A13, A14, A15, A16, A17, A18, A19, A20, A21, RT]): UserDefinedFunction = { + UserDefinedFunction(f, ScalaReflection.schemaFor(typeTag[RT]).dataType) } /** - * Call a Scala function of 22 arguments as user-defined function (UDF), and automatically - * infer the data types based on the function's signature. + * Defines a user-defined function of 22 arguments as user-defined function (UDF). + * The data types are automatically inferred based on the function's signature. */ - def callUDF[RT: TypeTag, A1: TypeTag, A2: TypeTag, A3: TypeTag, A4: TypeTag, A5: TypeTag, A6: TypeTag, A7: TypeTag, A8: TypeTag, A9: TypeTag, A10: TypeTag, A11: TypeTag, A12: TypeTag, A13: TypeTag, A14: TypeTag, A15: TypeTag, A16: TypeTag, A17: TypeTag, A18: TypeTag, A19: TypeTag, A20: TypeTag, A21: TypeTag, A22: TypeTag](f: Function22[A1, A2, A3, A4, A5, A6, A7, A8, A9, A10, A11, A12, A13, A14, A15, A16, A17, A18, A19, A20, A21, A22, RT], arg1: Column, arg2: Column, arg3: Column, arg4: Column, arg5: Column, arg6: Column, arg7: Column, arg8: Column, arg9: Column, arg10: Column, arg11: Column, arg12: Column, arg13: Column, arg14: Column, arg15: Column, arg16: Column, arg17: Column, arg18: Column, arg19: Column, arg20: Column, arg21: Column, arg22: Column): Column = { - ScalaUdf(f, ScalaReflection.schemaFor(typeTag[RT]).dataType, Seq(arg1.expr, arg2.expr, arg3.expr, arg4.expr, arg5.expr, arg6.expr, arg7.expr, arg8.expr, arg9.expr, arg10.expr, arg11.expr, arg12.expr, arg13.expr, arg14.expr, arg15.expr, arg16.expr, arg17.expr, arg18.expr, arg19.expr, arg20.expr, arg21.expr, arg22.expr)) + def udf[RT: TypeTag, A1: TypeTag, A2: TypeTag, A3: TypeTag, A4: TypeTag, A5: TypeTag, A6: TypeTag, A7: TypeTag, A8: TypeTag, A9: TypeTag, A10: TypeTag, A11: TypeTag, A12: TypeTag, A13: TypeTag, A14: TypeTag, A15: TypeTag, A16: TypeTag, A17: TypeTag, A18: TypeTag, A19: TypeTag, A20: TypeTag, A21: TypeTag, A22: TypeTag](f: Function22[A1, A2, A3, A4, A5, A6, A7, A8, A9, A10, A11, A12, A13, A14, A15, A16, A17, A18, A19, A20, A21, A22, RT]): UserDefinedFunction = { + UserDefinedFunction(f, ScalaReflection.schemaFor(typeTag[RT]).dataType) } ////////////////////////////////////////////////////////////////////////////////////////////////// http://git-wip-us.apache.org/repos/asf/spark/blob/1077f2e1/sql/core/src/main/scala/org/apache/spark/sql/UDFRegistration.scala ---------------------------------------------------------------------- diff --git a/sql/core/src/main/scala/org/apache/spark/sql/UDFRegistration.scala b/sql/core/src/main/scala/org/apache/spark/sql/UDFRegistration.scala new file mode 100644 index 0000000..d8b0a3b --- /dev/null +++ b/sql/core/src/main/scala/org/apache/spark/sql/UDFRegistration.scala @@ -0,0 +1,571 @@ +/* + * Licensed to the Apache Software Foundation (ASF) under one or more + * contributor license agreements. See the NOTICE file distributed with + * this work for additional information regarding copyright ownership. + * The ASF licenses this file to You under the Apache License, Version 2.0 + * (the "License"); you may not use this file except in compliance with + * the License. You may obtain a copy of the License at + * + * http://www.apache.org/licenses/LICENSE-2.0 + * + * Unless required by applicable law or agreed to in writing, software + * distributed under the License is distributed on an "AS IS" BASIS, + * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. + * See the License for the specific language governing permissions and + * limitations under the License. + */ + +package org.apache.spark.sql + +import java.util.{List => JList, Map => JMap} + +import scala.reflect.runtime.universe.TypeTag + +import org.apache.spark.{Accumulator, Logging} +import org.apache.spark.api.python.PythonBroadcast +import org.apache.spark.broadcast.Broadcast +import org.apache.spark.sql.api.java._ +import org.apache.spark.sql.catalyst.ScalaReflection +import org.apache.spark.sql.catalyst.expressions.{Expression, ScalaUdf} +import org.apache.spark.sql.execution.PythonUDF +import org.apache.spark.sql.types.DataType + + +/** + * Functions for registering user-defined functions. + */ +class UDFRegistration(sqlContext: SQLContext) extends Logging { + + private val functionRegistry = sqlContext.functionRegistry + + protected[sql] def registerPython( + name: String, + command: Array[Byte], + envVars: JMap[String, String], + pythonIncludes: JList[String], + pythonExec: String, + broadcastVars: JList[Broadcast[PythonBroadcast]], + accumulator: Accumulator[JList[Array[Byte]]], + stringDataType: String): Unit = { + log.debug( + s""" + | Registering new PythonUDF: + | name: $name + | command: ${command.toSeq} + | envVars: $envVars + | pythonIncludes: $pythonIncludes + | pythonExec: $pythonExec + | dataType: $stringDataType + """.stripMargin) + + + val dataType = sqlContext.parseDataType(stringDataType) + + def builder(e: Seq[Expression]) = + PythonUDF( + name, + command, + envVars, + pythonIncludes, + pythonExec, + broadcastVars, + accumulator, + dataType, + e) + + functionRegistry.registerFunction(name, builder) + } + + // scalastyle:off + + /* register 0-22 were generated by this script + + (0 to 22).map { x => + val types = (1 to x).foldRight("RT")((i, s) => {s"A$i, $s"}) + val typeTags = (1 to x).map(i => s"A${i}: TypeTag").foldLeft("RT: TypeTag")(_ + ", " + _) + println(s""" + /** + * Register a Scala closure of ${x} arguments as user-defined function (UDF). + * @tparam RT return type of UDF. + */ + def register[$typeTags](name: String, func: Function$x[$types]): UserDefinedFunction = { + val dataType = ScalaReflection.schemaFor[RT].dataType + def builder(e: Seq[Expression]) = ScalaUdf(func, dataType, e) + functionRegistry.registerFunction(name, builder) + UserDefinedFunction(func, dataType) + }""") + } + + (1 to 22).foreach { i => + val extTypeArgs = (1 to i).map(_ => "_").mkString(", ") + val anyTypeArgs = (1 to i).map(_ => "Any").mkString(", ") + val anyCast = s".asInstanceOf[UDF$i[$anyTypeArgs, Any]]" + val anyParams = (1 to i).map(_ => "_: Any").mkString(", ") + println(s""" + |/** + | * Register a user-defined function with ${i} arguments. + | */ + |def register(name: String, f: UDF$i[$extTypeArgs, _], returnType: DataType) = { + | functionRegistry.registerFunction( + | name, + | (e: Seq[Expression]) => ScalaUdf(f$anyCast.call($anyParams), returnType, e)) + |}""".stripMargin) + } + */ + + /** + * Register a Scala closure of 0 arguments as user-defined function (UDF). + * @tparam RT return type of UDF. + */ + def register[RT: TypeTag](name: String, func: Function0[RT]): UserDefinedFunction = { + val dataType = ScalaReflection.schemaFor[RT].dataType + def builder(e: Seq[Expression]) = ScalaUdf(func, dataType, e) + functionRegistry.registerFunction(name, builder) + UserDefinedFunction(func, dataType) + } + + /** + * Register a Scala closure of 1 arguments as user-defined function (UDF). + * @tparam RT return type of UDF. + */ + def register[RT: TypeTag, A1: TypeTag](name: String, func: Function1[A1, RT]): UserDefinedFunction = { + val dataType = ScalaReflection.schemaFor[RT].dataType + def builder(e: Seq[Expression]) = ScalaUdf(func, dataType, e) + functionRegistry.registerFunction(name, builder) + UserDefinedFunction(func, dataType) + } + + /** + * Register a Scala closure of 2 arguments as user-defined function (UDF). + * @tparam RT return type of UDF. + */ + def register[RT: TypeTag, A1: TypeTag, A2: TypeTag](name: String, func: Function2[A1, A2, RT]): UserDefinedFunction = { + val dataType = ScalaReflection.schemaFor[RT].dataType + def builder(e: Seq[Expression]) = ScalaUdf(func, dataType, e) + functionRegistry.registerFunction(name, builder) + UserDefinedFunction(func, dataType) + } + + /** + * Register a Scala closure of 3 arguments as user-defined function (UDF). + * @tparam RT return type of UDF. + */ + def register[RT: TypeTag, A1: TypeTag, A2: TypeTag, A3: TypeTag](name: String, func: Function3[A1, A2, A3, RT]): UserDefinedFunction = { + val dataType = ScalaReflection.schemaFor[RT].dataType + def builder(e: Seq[Expression]) = ScalaUdf(func, dataType, e) + functionRegistry.registerFunction(name, builder) + UserDefinedFunction(func, dataType) + } + + /** + * Register a Scala closure of 4 arguments as user-defined function (UDF). + * @tparam RT return type of UDF. + */ + def register[RT: TypeTag, A1: TypeTag, A2: TypeTag, A3: TypeTag, A4: TypeTag](name: String, func: Function4[A1, A2, A3, A4, RT]): UserDefinedFunction = { + val dataType = ScalaReflection.schemaFor[RT].dataType + def builder(e: Seq[Expression]) = ScalaUdf(func, dataType, e) + functionRegistry.registerFunction(name, builder) + UserDefinedFunction(func, dataType) + } + + /** + * Register a Scala closure of 5 arguments as user-defined function (UDF). + * @tparam RT return type of UDF. + */ + def register[RT: TypeTag, A1: TypeTag, A2: TypeTag, A3: TypeTag, A4: TypeTag, A5: TypeTag](name: String, func: Function5[A1, A2, A3, A4, A5, RT]): UserDefinedFunction = { + val dataType = ScalaReflection.schemaFor[RT].dataType + def builder(e: Seq[Expression]) = ScalaUdf(func, dataType, e) + functionRegistry.registerFunction(name, builder) + UserDefinedFunction(func, dataType) + } + + /** + * Register a Scala closure of 6 arguments as user-defined function (UDF). + * @tparam RT return type of UDF. + */ + def register[RT: TypeTag, A1: TypeTag, A2: TypeTag, A3: TypeTag, A4: TypeTag, A5: TypeTag, A6: TypeTag](name: String, func: Function6[A1, A2, A3, A4, A5, A6, RT]): UserDefinedFunction = { + val dataType = ScalaReflection.schemaFor[RT].dataType + def builder(e: Seq[Expression]) = ScalaUdf(func, dataType, e) + functionRegistry.registerFunction(name, builder) + UserDefinedFunction(func, dataType) + } + + /** + * Register a Scala closure of 7 arguments as user-defined function (UDF). + * @tparam RT return type of UDF. + */ + def register[RT: TypeTag, A1: TypeTag, A2: TypeTag, A3: TypeTag, A4: TypeTag, A5: TypeTag, A6: TypeTag, A7: TypeTag](name: String, func: Function7[A1, A2, A3, A4, A5, A6, A7, RT]): UserDefinedFunction = { + val dataType = ScalaReflection.schemaFor[RT].dataType + def builder(e: Seq[Expression]) = ScalaUdf(func, dataType, e) + functionRegistry.registerFunction(name, builder) + UserDefinedFunction(func, dataType) + } + + /** + * Register a Scala closure of 8 arguments as user-defined function (UDF). + * @tparam RT return type of UDF. + */ + def register[RT: TypeTag, A1: TypeTag, A2: TypeTag, A3: TypeTag, A4: TypeTag, A5: TypeTag, A6: TypeTag, A7: TypeTag, A8: TypeTag](name: String, func: Function8[A1, A2, A3, A4, A5, A6, A7, A8, RT]): UserDefinedFunction = { + val dataType = ScalaReflection.schemaFor[RT].dataType + def builder(e: Seq[Expression]) = ScalaUdf(func, dataType, e) + functionRegistry.registerFunction(name, builder) + UserDefinedFunction(func, dataType) + } + + /** + * Register a Scala closure of 9 arguments as user-defined function (UDF). + * @tparam RT return type of UDF. + */ + def register[RT: TypeTag, A1: TypeTag, A2: TypeTag, A3: TypeTag, A4: TypeTag, A5: TypeTag, A6: TypeTag, A7: TypeTag, A8: TypeTag, A9: TypeTag](name: String, func: Function9[A1, A2, A3, A4, A5, A6, A7, A8, A9, RT]): UserDefinedFunction = { + val dataType = ScalaReflection.schemaFor[RT].dataType + def builder(e: Seq[Expression]) = ScalaUdf(func, dataType, e) + functionRegistry.registerFunction(name, builder) + UserDefinedFunction(func, dataType) + } + + /** + * Register a Scala closure of 10 arguments as user-defined function (UDF). + * @tparam RT return type of UDF. + */ + def register[RT: TypeTag, A1: TypeTag, A2: TypeTag, A3: TypeTag, A4: TypeTag, A5: TypeTag, A6: TypeTag, A7: TypeTag, A8: TypeTag, A9: TypeTag, A10: TypeTag](name: String, func: Function10[A1, A2, A3, A4, A5, A6, A7, A8, A9, A10, RT]): UserDefinedFunction = { + val dataType = ScalaReflection.schemaFor[RT].dataType + def builder(e: Seq[Expression]) = ScalaUdf(func, dataType, e) + functionRegistry.registerFunction(name, builder) + UserDefinedFunction(func, dataType) + } + + /** + * Register a Scala closure of 11 arguments as user-defined function (UDF). + * @tparam RT return type of UDF. + */ + def register[RT: TypeTag, A1: TypeTag, A2: TypeTag, A3: TypeTag, A4: TypeTag, A5: TypeTag, A6: TypeTag, A7: TypeTag, A8: TypeTag, A9: TypeTag, A10: TypeTag, A11: TypeTag](name: String, func: Function11[A1, A2, A3, A4, A5, A6, A7, A8, A9, A10, A11, RT]): UserDefinedFunction = { + val dataType = ScalaReflection.schemaFor[RT].dataType + def builder(e: Seq[Expression]) = ScalaUdf(func, dataType, e) + functionRegistry.registerFunction(name, builder) + UserDefinedFunction(func, dataType) + } + + /** + * Register a Scala closure of 12 arguments as user-defined function (UDF). + * @tparam RT return type of UDF. + */ + def register[RT: TypeTag, A1: TypeTag, A2: TypeTag, A3: TypeTag, A4: TypeTag, A5: TypeTag, A6: TypeTag, A7: TypeTag, A8: TypeTag, A9: TypeTag, A10: TypeTag, A11: TypeTag, A12: TypeTag](name: String, func: Function12[A1, A2, A3, A4, A5, A6, A7, A8, A9, A10, A11, A12, RT]): UserDefinedFunction = { + val dataType = ScalaReflection.schemaFor[RT].dataType + def builder(e: Seq[Expression]) = ScalaUdf(func, dataType, e) + functionRegistry.registerFunction(name, builder) + UserDefinedFunction(func, dataType) + } + + /** + * Register a Scala closure of 13 arguments as user-defined function (UDF). + * @tparam RT return type of UDF. + */ + def register[RT: TypeTag, A1: TypeTag, A2: TypeTag, A3: TypeTag, A4: TypeTag, A5: TypeTag, A6: TypeTag, A7: TypeTag, A8: TypeTag, A9: TypeTag, A10: TypeTag, A11: TypeTag, A12: TypeTag, A13: TypeTag](name: String, func: Function13[A1, A2, A3, A4, A5, A6, A7, A8, A9, A10, A11, A12, A13, RT]): UserDefinedFunction = { + val dataType = ScalaReflection.schemaFor[RT].dataType + def builder(e: Seq[Expression]) = ScalaUdf(func, dataType, e) + functionRegistry.registerFunction(name, builder) + UserDefinedFunction(func, dataType) + } + + /** + * Register a Scala closure of 14 arguments as user-defined function (UDF). + * @tparam RT return type of UDF. + */ + def register[RT: TypeTag, A1: TypeTag, A2: TypeTag, A3: TypeTag, A4: TypeTag, A5: TypeTag, A6: TypeTag, A7: TypeTag, A8: TypeTag, A9: TypeTag, A10: TypeTag, A11: TypeTag, A12: TypeTag, A13: TypeTag, A14: TypeTag](name: String, func: Function14[A1, A2, A3, A4, A5, A6, A7, A8, A9, A10, A11, A12, A13, A14, RT]): UserDefinedFunction = { + val dataType = ScalaReflection.schemaFor[RT].dataType + def builder(e: Seq[Expression]) = ScalaUdf(func, dataType, e) + functionRegistry.registerFunction(name, builder) + UserDefinedFunction(func, dataType) + } + + /** + * Register a Scala closure of 15 arguments as user-defined function (UDF). + * @tparam RT return type of UDF. + */ + def register[RT: TypeTag, A1: TypeTag, A2: TypeTag, A3: TypeTag, A4: TypeTag, A5: TypeTag, A6: TypeTag, A7: TypeTag, A8: TypeTag, A9: TypeTag, A10: TypeTag, A11: TypeTag, A12: TypeTag, A13: TypeTag, A14: TypeTag, A15: TypeTag](name: String, func: Function15[A1, A2, A3, A4, A5, A6, A7, A8, A9, A10, A11, A12, A13, A14, A15, RT]): UserDefinedFunction = { + val dataType = ScalaReflection.schemaFor[RT].dataType + def builder(e: Seq[Expression]) = ScalaUdf(func, dataType, e) + functionRegistry.registerFunction(name, builder) + UserDefinedFunction(func, dataType) + } + + /** + * Register a Scala closure of 16 arguments as user-defined function (UDF). + * @tparam RT return type of UDF. + */ + def register[RT: TypeTag, A1: TypeTag, A2: TypeTag, A3: TypeTag, A4: TypeTag, A5: TypeTag, A6: TypeTag, A7: TypeTag, A8: TypeTag, A9: TypeTag, A10: TypeTag, A11: TypeTag, A12: TypeTag, A13: TypeTag, A14: TypeTag, A15: TypeTag, A16: TypeTag](name: String, func: Function16[A1, A2, A3, A4, A5, A6, A7, A8, A9, A10, A11, A12, A13, A14, A15, A16, RT]): UserDefinedFunction = { + val dataType = ScalaReflection.schemaFor[RT].dataType + def builder(e: Seq[Expression]) = ScalaUdf(func, dataType, e) + functionRegistry.registerFunction(name, builder) + UserDefinedFunction(func, dataType) + } + + /** + * Register a Scala closure of 17 arguments as user-defined function (UDF). + * @tparam RT return type of UDF. + */ + def register[RT: TypeTag, A1: TypeTag, A2: TypeTag, A3: TypeTag, A4: TypeTag, A5: TypeTag, A6: TypeTag, A7: TypeTag, A8: TypeTag, A9: TypeTag, A10: TypeTag, A11: TypeTag, A12: TypeTag, A13: TypeTag, A14: TypeTag, A15: TypeTag, A16: TypeTag, A17: TypeTag](name: String, func: Function17[A1, A2, A3, A4, A5, A6, A7, A8, A9, A10, A11, A12, A13, A14, A15, A16, A17, RT]): UserDefinedFunction = { + val dataType = ScalaReflection.schemaFor[RT].dataType + def builder(e: Seq[Expression]) = ScalaUdf(func, dataType, e) + functionRegistry.registerFunction(name, builder) + UserDefinedFunction(func, dataType) + } + + /** + * Register a Scala closure of 18 arguments as user-defined function (UDF). + * @tparam RT return type of UDF. + */ + def register[RT: TypeTag, A1: TypeTag, A2: TypeTag, A3: TypeTag, A4: TypeTag, A5: TypeTag, A6: TypeTag, A7: TypeTag, A8: TypeTag, A9: TypeTag, A10: TypeTag, A11: TypeTag, A12: TypeTag, A13: TypeTag, A14: TypeTag, A15: TypeTag, A16: TypeTag, A17: TypeTag, A18: TypeTag](name: String, func: Function18[A1, A2, A3, A4, A5, A6, A7, A8, A9, A10, A11, A12, A13, A14, A15, A16, A17, A18, RT]): UserDefinedFunction = { + val dataType = ScalaReflection.schemaFor[RT].dataType + def builder(e: Seq[Expression]) = ScalaUdf(func, dataType, e) + functionRegistry.registerFunction(name, builder) + UserDefinedFunction(func, dataType) + } + + /** + * Register a Scala closure of 19 arguments as user-defined function (UDF). + * @tparam RT return type of UDF. + */ + def register[RT: TypeTag, A1: TypeTag, A2: TypeTag, A3: TypeTag, A4: TypeTag, A5: TypeTag, A6: TypeTag, A7: TypeTag, A8: TypeTag, A9: TypeTag, A10: TypeTag, A11: TypeTag, A12: TypeTag, A13: TypeTag, A14: TypeTag, A15: TypeTag, A16: TypeTag, A17: TypeTag, A18: TypeTag, A19: TypeTag](name: String, func: Function19[A1, A2, A3, A4, A5, A6, A7, A8, A9, A10, A11, A12, A13, A14, A15, A16, A17, A18, A19, RT]): UserDefinedFunction = { + val dataType = ScalaReflection.schemaFor[RT].dataType + def builder(e: Seq[Expression]) = ScalaUdf(func, dataType, e) + functionRegistry.registerFunction(name, builder) + UserDefinedFunction(func, dataType) + } + + /** + * Register a Scala closure of 20 arguments as user-defined function (UDF). + * @tparam RT return type of UDF. + */ + def register[RT: TypeTag, A1: TypeTag, A2: TypeTag, A3: TypeTag, A4: TypeTag, A5: TypeTag, A6: TypeTag, A7: TypeTag, A8: TypeTag, A9: TypeTag, A10: TypeTag, A11: TypeTag, A12: TypeTag, A13: TypeTag, A14: TypeTag, A15: TypeTag, A16: TypeTag, A17: TypeTag, A18: TypeTag, A19: TypeTag, A20: TypeTag](name: String, func: Function20[A1, A2, A3, A4, A5, A6, A7, A8, A9, A10, A11, A12, A13, A14, A15, A16, A17, A18, A19, A20, RT]): UserDefinedFunction = { + val dataType = ScalaReflection.schemaFor[RT].dataType + def builder(e: Seq[Expression]) = ScalaUdf(func, dataType, e) + functionRegistry.registerFunction(name, builder) + UserDefinedFunction(func, dataType) + } + + /** + * Register a Scala closure of 21 arguments as user-defined function (UDF). + * @tparam RT return type of UDF. + */ + def register[RT: TypeTag, A1: TypeTag, A2: TypeTag, A3: TypeTag, A4: TypeTag, A5: TypeTag, A6: TypeTag, A7: TypeTag, A8: TypeTag, A9: TypeTag, A10: TypeTag, A11: TypeTag, A12: TypeTag, A13: TypeTag, A14: TypeTag, A15: TypeTag, A16: TypeTag, A17: TypeTag, A18: TypeTag, A19: TypeTag, A20: TypeTag, A21: TypeTag](name: String, func: Function21[A1, A2, A3, A4, A5, A6, A7, A8, A9, A10, A11, A12, A13, A14, A15, A16, A17, A18, A19, A20, A21, RT]): UserDefinedFunction = { + val dataType = ScalaReflection.schemaFor[RT].dataType + def builder(e: Seq[Expression]) = ScalaUdf(func, dataType, e) + functionRegistry.registerFunction(name, builder) + UserDefinedFunction(func, dataType) + } + + /** + * Register a Scala closure of 22 arguments as user-defined function (UDF). + * @tparam RT return type of UDF. + */ + def register[RT: TypeTag, A1: TypeTag, A2: TypeTag, A3: TypeTag, A4: TypeTag, A5: TypeTag, A6: TypeTag, A7: TypeTag, A8: TypeTag, A9: TypeTag, A10: TypeTag, A11: TypeTag, A12: TypeTag, A13: TypeTag, A14: TypeTag, A15: TypeTag, A16: TypeTag, A17: TypeTag, A18: TypeTag, A19: TypeTag, A20: TypeTag, A21: TypeTag, A22: TypeTag](name: String, func: Function22[A1, A2, A3, A4, A5, A6, A7, A8, A9, A10, A11, A12, A13, A14, A15, A16, A17, A18, A19, A20, A21, A22, RT]): UserDefinedFunction = { + val dataType = ScalaReflection.schemaFor[RT].dataType + def builder(e: Seq[Expression]) = ScalaUdf(func, dataType, e) + functionRegistry.registerFunction(name, builder) + UserDefinedFunction(func, dataType) + } + + ////////////////////////////////////////////////////////////////////////////////////////////// + ////////////////////////////////////////////////////////////////////////////////////////////// + + /** + * Register a user-defined function with 1 arguments. + */ + def register(name: String, f: UDF1[_, _], returnType: DataType) = { + functionRegistry.registerFunction( + name, + (e: Seq[Expression]) => ScalaUdf(f.asInstanceOf[UDF1[Any, Any]].call(_: Any), returnType, e)) + } + + /** + * Register a user-defined function with 2 arguments. + */ + def register(name: String, f: UDF2[_, _, _], returnType: DataType) = { + functionRegistry.registerFunction( + name, + (e: Seq[Expression]) => ScalaUdf(f.asInstanceOf[UDF2[Any, Any, Any]].call(_: Any, _: Any), returnType, e)) + } + + /** + * Register a user-defined function with 3 arguments. + */ + def register(name: String, f: UDF3[_, _, _, _], returnType: DataType) = { + functionRegistry.registerFunction( + name, + (e: Seq[Expression]) => ScalaUdf(f.asInstanceOf[UDF3[Any, Any, Any, Any]].call(_: Any, _: Any, _: Any), returnType, e)) + } + + /** + * Register a user-defined function with 4 arguments. + */ + def register(name: String, f: UDF4[_, _, _, _, _], returnType: DataType) = { + functionRegistry.registerFunction( + name, + (e: Seq[Expression]) => ScalaUdf(f.asInstanceOf[UDF4[Any, Any, Any, Any, Any]].call(_: Any, _: Any, _: Any, _: Any), returnType, e)) + } + + /** + * Register a user-defined function with 5 arguments. + */ + def register(name: String, f: UDF5[_, _, _, _, _, _], returnType: DataType) = { + functionRegistry.registerFunction( + name, + (e: Seq[Expression]) => ScalaUdf(f.asInstanceOf[UDF5[Any, Any, Any, Any, Any, Any]].call(_: Any, _: Any, _: Any, _: Any, _: Any), returnType, e)) + } + + /** + * Register a user-defined function with 6 arguments. + */ + def register(name: String, f: UDF6[_, _, _, _, _, _, _], returnType: DataType) = { + functionRegistry.registerFunction( + name, + (e: Seq[Expression]) => ScalaUdf(f.asInstanceOf[UDF6[Any, Any, Any, Any, Any, Any, Any]].call(_: Any, _: Any, _: Any, _: Any, _: Any, _: Any), returnType, e)) + } + + /** + * Register a user-defined function with 7 arguments. + */ + def register(name: String, f: UDF7[_, _, _, _, _, _, _, _], returnType: DataType) = { + functionRegistry.registerFunction( + name, + (e: Seq[Expression]) => ScalaUdf(f.asInstanceOf[UDF7[Any, Any, Any, Any, Any, Any, Any, Any]].call(_: Any, _: Any, _: Any, _: Any, _: Any, _: Any, _: Any), returnType, e)) + } + + /** + * Register a user-defined function with 8 arguments. + */ + def register(name: String, f: UDF8[_, _, _, _, _, _, _, _, _], returnType: DataType) = { + functionRegistry.registerFunction( + name, + (e: Seq[Expression]) => ScalaUdf(f.asInstanceOf[UDF8[Any, Any, Any, Any, Any, Any, Any, Any, Any]].call(_: Any, _: Any, _: Any, _: Any, _: Any, _: Any, _: Any, _: Any), returnType, e)) + } + + /** + * Register a user-defined function with 9 arguments. + */ + def register(name: String, f: UDF9[_, _, _, _, _, _, _, _, _, _], returnType: DataType) = { + functionRegistry.registerFunction( + name, + (e: Seq[Expression]) => ScalaUdf(f.asInstanceOf[UDF9[Any, Any, Any, Any, Any, Any, Any, Any, Any, Any]].call(_: Any, _: Any, _: Any, _: Any, _: Any, _: Any, _: Any, _: Any, _: Any), returnType, e)) + } + + /** + * Register a user-defined function with 10 arguments. + */ + def register(name: String, f: UDF10[_, _, _, _, _, _, _, _, _, _, _], returnType: DataType) = { + functionRegistry.registerFunction( + name, + (e: Seq[Expression]) => ScalaUdf(f.asInstanceOf[UDF10[Any, Any, Any, Any, Any, Any, Any, Any, Any, Any, Any]].call(_: Any, _: Any, _: Any, _: Any, _: Any, _: Any, _: Any, _: Any, _: Any, _: Any), returnType, e)) + } + + /** + * Register a user-defined function with 11 arguments. + */ + def register(name: String, f: UDF11[_, _, _, _, _, _, _, _, _, _, _, _], returnType: DataType) = { + functionRegistry.registerFunction( + name, + (e: Seq[Expression]) => ScalaUdf(f.asInstanceOf[UDF11[Any, Any, Any, Any, Any, Any, Any, Any, Any, Any, Any, Any]].call(_: Any, _: Any, _: Any, _: Any, _: Any, _: Any, _: Any, _: Any, _: Any, _: Any, _: Any), returnType, e)) + } + + /** + * Register a user-defined function with 12 arguments. + */ + def register(name: String, f: UDF12[_, _, _, _, _, _, _, _, _, _, _, _, _], returnType: DataType) = { + functionRegistry.registerFunction( + name, + (e: Seq[Expression]) => ScalaUdf(f.asInstanceOf[UDF12[Any, Any, Any, Any, Any, Any, Any, Any, Any, Any, Any, Any, Any]].call(_: Any, _: Any, _: Any, _: Any, _: Any, _: Any, _: Any, _: Any, _: Any, _: Any, _: Any, _: Any), returnType, e)) + } + + /** + * Register a user-defined function with 13 arguments. + */ + def register(name: String, f: UDF13[_, _, _, _, _, _, _, _, _, _, _, _, _, _], returnType: DataType) = { + functionRegistry.registerFunction( + name, + (e: Seq[Expression]) => ScalaUdf(f.asInstanceOf[UDF13[Any, Any, Any, Any, Any, Any, Any, Any, Any, Any, Any, Any, Any, Any]].call(_: Any, _: Any, _: Any, _: Any, _: Any, _: Any, _: Any, _: Any, _: Any, _: Any, _: Any, _: Any, _: Any), returnType, e)) + } + + /** + * Register a user-defined function with 14 arguments. + */ + def register(name: String, f: UDF14[_, _, _, _, _, _, _, _, _, _, _, _, _, _, _], returnType: DataType) = { + functionRegistry.registerFunction( + name, + (e: Seq[Expression]) => ScalaUdf(f.asInstanceOf[UDF14[Any, Any, Any, Any, Any, Any, Any, Any, Any, Any, Any, Any, Any, Any, Any]].call(_: Any, _: Any, _: Any, _: Any, _: Any, _: Any, _: Any, _: Any, _: Any, _: Any, _: Any, _: Any, _: Any, _: Any), returnType, e)) + } + + /** + * Register a user-defined function with 15 arguments. + */ + def register(name: String, f: UDF15[_, _, _, _, _, _, _, _, _, _, _, _, _, _, _, _], returnType: DataType) = { + functionRegistry.registerFunction( + name, + (e: Seq[Expression]) => ScalaUdf(f.asInstanceOf[UDF15[Any, Any, Any, Any, Any, Any, Any, Any, Any, Any, Any, Any, Any, Any, Any, Any]].call(_: Any, _: Any, _: Any, _: Any, _: Any, _: Any, _: Any, _: Any, _: Any, _: Any, _: Any, _: Any, _: Any, _: Any, _: Any), returnType, e)) + } + + /** + * Register a user-defined function with 16 arguments. + */ + def register(name: String, f: UDF16[_, _, _, _, _, _, _, _, _, _, _, _, _, _, _, _, _], returnType: DataType) = { + functionRegistry.registerFunction( + name, + (e: Seq[Expression]) => ScalaUdf(f.asInstanceOf[UDF16[Any, Any, Any, Any, Any, Any, Any, Any, Any, Any, Any, Any, Any, Any, Any, Any, Any]].call(_: Any, _: Any, _: Any, _: Any, _: Any, _: Any, _: Any, _: Any, _: Any, _: Any, _: Any, _: Any, _: Any, _: Any, _: Any, _: Any), returnType, e)) + } + + /** + * Register a user-defined function with 17 arguments. + */ + def register(name: String, f: UDF17[_, _, _, _, _, _, _, _, _, _, _, _, _, _, _, _, _, _], returnType: DataType) = { + functionRegistry.registerFunction( + name, + (e: Seq[Expression]) => ScalaUdf(f.asInstanceOf[UDF17[Any, Any, Any, Any, Any, Any, Any, Any, Any, Any, Any, Any, Any, Any, Any, Any, Any, Any]].call(_: Any, _: Any, _: Any, _: Any, _: Any, _: Any, _: Any, _: Any, _: Any, _: Any, _: Any, _: Any, _: Any, _: Any, _: Any, _: Any, _: Any), returnType, e)) + } + + /** + * Register a user-defined function with 18 arguments. + */ + def register(name: String, f: UDF18[_, _, _, _, _, _, _, _, _, _, _, _, _, _, _, _, _, _, _], returnType: DataType) = { + functionRegistry.registerFunction( + name, + (e: Seq[Expression]) => ScalaUdf(f.asInstanceOf[UDF18[Any, Any, Any, Any, Any, Any, Any, Any, Any, Any, Any, Any, Any, Any, Any, Any, Any, Any, Any]].call(_: Any, _: Any, _: Any, _: Any, _: Any, _: Any, _: Any, _: Any, _: Any, _: Any, _: Any, _: Any, _: Any, _: Any, _: Any, _: Any, _: Any, _: Any), returnType, e)) + } + + /** + * Register a user-defined function with 19 arguments. + */ + def register(name: String, f: UDF19[_, _, _, _, _, _, _, _, _, _, _, _, _, _, _, _, _, _, _, _], returnType: DataType) = { + functionRegistry.registerFunction( + name, + (e: Seq[Expression]) => ScalaUdf(f.asInstanceOf[UDF19[Any, Any, Any, Any, Any, Any, Any, Any, Any, Any, Any, Any, Any, Any, Any, Any, Any, Any, Any, Any]].call(_: Any, _: Any, _: Any, _: Any, _: Any, _: Any, _: Any, _: Any, _: Any, _: Any, _: Any, _: Any, _: Any, _: Any, _: Any, _: Any, _: Any, _: Any, _: Any), returnType, e)) + } + + /** + * Register a user-defined function with 20 arguments. + */ + def register(name: String, f: UDF20[_, _, _, _, _, _, _, _, _, _, _, _, _, _, _, _, _, _, _, _, _], returnType: DataType) = { + functionRegistry.registerFunction( + name, + (e: Seq[Expression]) => ScalaUdf(f.asInstanceOf[UDF20[Any, Any, Any, Any, Any, Any, Any, Any, Any, Any, Any, Any, Any, Any, Any, Any, Any, Any, Any, Any, Any]].call(_: Any, _: Any, _: Any, _: Any, _: Any, _: Any, _: Any, _: Any, _: Any, _: Any, _: Any, _: Any, _: Any, _: Any, _: Any, _: Any, _: Any, _: Any, _: Any, _: Any), returnType, e)) + } + + /** + * Register a user-defined function with 21 arguments. + */ + def register(name: String, f: UDF21[_, _, _, _, _, _, _, _, _, _, _, _, _, _, _, _, _, _, _, _, _, _], returnType: DataType) = { + functionRegistry.registerFunction( + name, + (e: Seq[Expression]) => ScalaUdf(f.asInstanceOf[UDF21[Any, Any, Any, Any, Any, Any, Any, Any, Any, Any, Any, Any, Any, Any, Any, Any, Any, Any, Any, Any, Any, Any]].call(_: Any, _: Any, _: Any, _: Any, _: Any, _: Any, _: Any, _: Any, _: Any, _: Any, _: Any, _: Any, _: Any, _: Any, _: Any, _: Any, _: Any, _: Any, _: Any, _: Any, _: Any), returnType, e)) + } + + /** + * Register a user-defined function with 22 arguments. + */ + def register(name: String, f: UDF22[_, _, _, _, _, _, _, _, _, _, _, _, _, _, _, _, _, _, _, _, _, _, _], returnType: DataType) = { + functionRegistry.registerFunction( + name, + (e: Seq[Expression]) => ScalaUdf(f.asInstanceOf[UDF22[Any, Any, Any, Any, Any, Any, Any, Any, Any, Any, Any, Any, Any, Any, Any, Any, Any, Any, Any, Any, Any, Any, Any]].call(_: Any, _: Any, _: Any, _: Any, _: Any, _: Any, _: Any, _: Any, _: Any, _: Any, _: Any, _: Any, _: Any, _: Any, _: Any, _: Any, _: Any, _: Any, _: Any, _: Any, _: Any, _: Any), returnType, e)) + } + + // scalastyle:on +} --------------------------------------------------------------------- To unsubscribe, e-mail: commits-unsubscr...@spark.apache.org For additional commands, e-mail: commits-h...@spark.apache.org
