转换JavaRDD< Tuple2< Object，long []>进入Spark Dataset< Row>在Java中 [英] Convert a JavaRDD<Tuple2<Object, long[]>> into a Spark Dataset<Row> in Java

问题描述

在Java(不是Scala！)中，Spark 3.0.1具有

In Java (not Scala!) Spark 3.0.1 have a JavaRDD instance object neighborIdsRDD which its type is JavaRDD<Tuple2<Object, long[]>>.

与JavaRDD生成相关的部分代码如下:

Part of my code related to the generation of the JavaRDD is the following:

GraphOps<String, String> graphOps = new GraphOps<>(graph, stringTag, stringTag);
JavaRDD<Tuple2<Object, long[]>> neighborIdsRDD = graphOps.collectNeighborIds(EdgeDirection.Either()).toJavaRDD();

我必须使用 toJavaRDD()获得JavaRDD，因为 collectNeighborIds 返回 org.apache.spark.graphx.VertexRDD< long []>.对象( VertexRDD文档).

I have had to get a JavaRDD using toJavaRDD() because collectNeighborIds returns a org.apache.spark.graphx.VertexRDD<long[]> object (VertexRDD doc).

但是，在我的其余应用程序中，我需要有一个 Dataset< Row> 由 collectNeighborIds 对象构建.

However, in the rest of my application I need to have a Spark Dataset<Row> built from the collectNeighborIds object.

获得 JavaRDD< Tuple2< Object，long []>> 被转换为我根据注释调整了代码:

I adjusted the code basing from comments:

        GraphOps<String, String> graphOps = new GraphOps<>(graph, stringTag, stringTag);
        JavaRDD<Tuple2<Object, long[]>> neighborIdsRDD = graphOps.collectNeighborIds(EdgeDirection.Either()).toJavaRDD();
        System.out.println("VertexRDD neighborIdsRDD is:");
        for (int i = 0; i < neighborIdsRDD.collect().size(); i++) {
            System.out.println(
                    ((Tuple2<Object, long[]>) neighborIdsRDD.collect().get(i))._1() + " -- " +
                            Arrays.toString(((Tuple2<Object, long[]>) neighborIdsRDD.collect().get(i))._2())
            );
        }

        Dataset<Row> dr = spark_session.createDataFrame(neighborIdsRDD.rdd(), Tuple2.class);
        System.out.println("converted Dataset<Row> is:");
        dr.show();

但是我得到一个空的数据集，如下所示:

but I get an empty Dataset as follows:

VertexRDD neighborIdsRDD is:
4 -- [3]
1 -- [2, 3]
5 -- [3, 2]
2 -- [1, 3, 5]
3 -- [1, 2, 5, 4]
converted Dataset<Row> is:
++
||
++
||
||
||
||
||
++

推荐答案

我也遇到了同样的情况，但是幸运的是，我找到了一种解决方法，可以找回 Dataframe .

I was in your same situation, but fortunately I found a solution to get back a Dataframe.

在步骤 [1] ， [2] 和 [3] 处注释解决方案代码.

Solution code is commented at steps [1], [2] and [3].

GraphOps<String, String> graphOps = new GraphOps<>(graph, stringTag, stringTag);
System.out.println("VertexRDD neighborIdsRDD is:");
JavaRDD<Tuple2<Object, long[]>> neighborIdsRDD = graphOps.collectNeighborIds(EdgeDirection.Either()).toJavaRDD();
for (int i = 0; i < neighborIdsRDD.collect().size(); i++) {
    System.out.println(
            ((Tuple2<Object, long[]>) neighborIdsRDD.collect().get(i))._1() + " -- " +
                    Arrays.toString(((Tuple2<Object, long[]>) neighborIdsRDD.collect().get(i))._2())
    );
}

// [1] Define encoding schema
StructType graphStruct =  new StructType(new StructField[]{
        new StructField("father", DataTypes.LongType, false, Metadata.empty()),
        new StructField("children", DataTypes.createArrayType(DataTypes.LongType), false, Metadata.empty()),
});

// [2] Build a JavaRDD<Row> from a JavaRDD<Tuple2<Object,long[]>>
JavaRDD<Row> dr = neighborIdsRDD.map(tupla -> RowFactory.create(tupla._1(), tupla._2()));
        
// [3] Finally build the reqired Dataframe<Row>
Dataset<Row> dsr = spark_session.createDataFrame(dr.rdd(), graphStruct);

System.out.println("DATASET IS:");
dsr.show();

打印输出:

VertexRDD neighborIdsRDD is:
4 -- [3]
1 -- [2, 3]
5 -- [3, 2]
2 -- [1, 3, 5]
3 -- [1, 2, 5, 4]
DATASET IS:
+------+------------+
|father|    children|
+------+------------+
|     4|         [3]|
|     1|      [2, 3]|
|     5|      [3, 2]|
|     2|   [1, 3, 5]|
|     3|[1, 2, 5, 4]|
+------+------------+

这篇关于转换JavaRDD&lt; Tuple2&lt; Object，long []&gt;进入Spark Dataset&lt; Row&gt;在Java中的文章就介绍到这了，希望我们推荐的答案对大家有所帮助，也希望大家多多支持IT屋！