编码如下：

org.apache.flink.api.java.hadoop.mapreduce.HadoopInputFormat<LongWritable, Text> hadoopFile = org.apache.flink.hadoopcompatibility.HadoopInputs.readHadoopFile(
            new org.apache.hadoop.mapreduce.lib.input.TextInputFormat(), // extends org.apache.hadoop.mapreduce.lib.input.FileInputFormat
            org.apache.hadoop.io.LongWritable.class,
            org.apache.hadoop.io.Text.class,
            inputPath);

    DataSet<Tuple2<LongWritable, Text>> input = env.createInput(hadoopFile);

除此之外，maven pom还需要这种依赖：

<dependency>
        <groupId>org.apache.flink</groupId>
        <artifactId>flink-hadoop-compatibility_2.10</artifactId>
        <version>${flink.version}</version>
 </dependency>

下面是一个如何在flink中使用hadoopinputformat的简短示例：

final ExecutionEnvironment env = ExecutionEnvironment.getExecutionEnvironment();

final String inputPath = ...
final HadoopInputFormat<LongWritable, Text> hadoopFile = HadoopInputs.readHadoopFile(
        new TextInputFormat(),
        LongWritable.class,
        Text.class,
        inputPath);
DataSet<Tuple2<LongWritable, Text>> input = env.createInput(hadoopFile);