如何从GCP存储桶读取Apache Beam中的多个文件

时间:2019-11-08 10:11:36

标签: python python-3.x apache-beam dataflow apache-beam-io

我正在尝试使用Apache Beam在GCP中的多个文件上读取并应用一些子设置。我准备了两个仅对一个文件有效的管道,但是在多个文件上尝试它们时失败。除此之外,如果可能的话,我会很方便地将我的管道合并成一个管道,或者有一种方法可以对它们进行编排,以便它们按顺序工作。现在,管道可以在本地运行,但最终目标是使用Dataflow运行它们。

我是textio.ReadFromText和textio.ReadAllFromText,但是在有多个文件的情况下我都无法正常工作。

You can convert with this 

public static String saveHtml(Activity activity, String html) {
            String filePath = "";
            String fileName = "";
            try {
                if (Build.VERSION.SDK_INT >= Build.VERSION_CODES.M
                        && 
          activity.checkSelfPermission(Manifest.permission.WRITE_EXTERNAL_STORAGE)
                        != PackageManager.PERMISSION_GRANTED) {
                    ActivityCompat.requestPermissions(activity, new String[] 
          {Manifest.permission.WRITE_EXTERNAL_STORAGE}, ARE_Toolbar.REQ_VIDEO);
                    return "";
                }

                filePath = Environment.getExternalStorageDirectory() + 
                File.separator + "ARE" + File.separator;
                File dir = new File(filePath);
                if (!dir.exists()) {
                    dir.mkdir();
                }

                DateFormat dateFormat = new SimpleDateFormat("yyyy-MM- 
                dd_hh_mm_ss");
                String time = dateFormat.format(new Date());
                fileName = time.concat(".html");

                File file = new File(filePath + fileName);
                if (!file.exists()) {
                    boolean isCreated = file.createNewFile();

                    if (!isCreated) {
                        com.chinalwb.are.Util.toast(activity, "Cannot create file 
                    at: " + filePath);
                        return "";
                    }
                }

                FileWriter fileWriter = new FileWriter(file);
                fileWriter.write(html);
                fileWriter.close();

                com.chinalwb.are.Util.toast(activity, fileName + " has been saved 
                at " + filePath);
            } catch (IOException e) {
                e.printStackTrace();
                com.chinalwb.are.Util.toast(activity, "Run into error: " + 
            e.getMessage());
            }
            return filePath + fileName;
        }

    **Call this method as -** 

    saveHtml(this, html);

这两个管道对于单个文件来说效果很好,但是我有数百个相同格式的文件,并且希望利用并行计算的优势。

是否有一种方法可以使该管道对同一目录下的多个文件起作用?

是否可以在单个管道中执行此操作,而不是创建两个不同的管道? (将文件从存储桶中写入工作节点并不方便。)

非常感谢!

1 个答案:

答案 0 :(得分:0)

我解决了如何使它适用于多个文件的问题,但是却无法使其在单个管道中运行。我先使用循环,然后再使用beam.Flatten选项。

这是我的解决方法:

file_list = ["gs://my_bucket/file*.txt.gz"]
res_list = ["/home/subject_test_{}-00000-of-00001.json".format(i) for i in range(len(file_list))]

with beam.Pipeline(options=PipelineOptions()) as p:
    for i,file in enumerate(file_list):
       (p 
        | "Read Text {}".format(i) >> beam.io.textio.ReadFromText(file, skip_header_lines = 0)
        | "Write TExt {}".format(i) >> beam.io.WriteToText("/home/subject_test_{}".format(i),
                   file_name_suffix=".json", num_shards=1 , append_trailing_newlines = True))

pcols = []
with beam.Pipeline(options=PipelineOptions()) as p:
   for i,res in enumerate(res_list):
         pcol = (p   | 'read_data_{}'.format(i) >> beam.Create([res])
            | "toJson_{}".format(i) >> beam.Map(toJson)
            | "takeItems_{}".format(i) >> beam.FlatMap(lambda line: line["Items"])
            | "takeSubjects_{}".format(i) >> beam.FlatMap(lambda line: line['data']['subjects']))
        pcols.append(pcol)
   out = (pcols
    | beam.Flatten()
    | beam.combiners.Count.PerElement()
    | beam.io.WriteToText("/home/items",
                   file_name_suffix=".txt", num_shards=1 , append_trailing_newlines = True))