如何使用Flink dataset API对源代码中读取的总记录进行计数

<dependency>
    <groupId>org.apache.flink</groupId>
    <artifactId>flink-java</artifactId>
    <version>${flink.version}</version>
</dependency>

ExecutionEnvironment env = ExecutionEnvironment.getExecutionEnvironment();

DataSet<String> input = env.readTextFile("path/to/source/code");

DataSet<Tuple2<String, Integer>> counts = input
    .flatMap((String line, Collector<Tuple2<String, Integer>> out) -> {
        // 按照空格分割每行记录为单词
        String[] words = line.split(" ");
        for (String word : words) {
            out.collect(new Tuple2<>(word, 1));
        }
    })
    .groupBy(0)  // 按照单词进行分组
    .sum(1);     // 对单词进行计数求和

counts.writeAsText("path/to/output");

env.execute("Word Count");