我有 apache 光束管道，我使用 pubsub 從輸入文件中獲取一些文本，然后我正在做一些轉(zhuǎn)換，我正在獲取句子和分?jǐn)?shù)，但我的作者過度寫入結(jié)果而不是附加，我想知道有沒有為beam.filesystems 添加模塊？from __future__ import absolute_importimport argparseimport loggingfrom datetime import datetimefrom past.builtins import unicodeimport jsonfrom google.cloud import languagefrom google.cloud.language import enumsfrom google.cloud.language import typesimport apache_beam as beamimport apache_beam.transforms.window as windowfrom apache_beam.io.filesystems import FileSystemsfrom apache_beam.io.gcp.pubsub import WriteToPubSubfrom apache_beam.examples.wordcount import WordExtractingDoFnfrom apache_beam.options.pipeline_options import PipelineOptionsfrom apache_beam.options.pipeline_options import SetupOptionsfrom apache_beam.options.pipeline_options import StandardOptionsfrom apache_beam.io.textio import ReadFromText, WriteToTextdef run(argv=None):  """Build and run the pipeline."""  parser = argparse.ArgumentParser()  parser.add_argument(      '--output',         dest='output',        required=True,        help='GCS destination folder to save the images to (example: gs://BUCKET_NAME/path')  group = parser.add_mutually_exclusive_group(required=True)  group.add_argument(      '--input_topic',      help=('Input PubSub topic of the form '            '"projects<project name>/subscriptions/<topic name>".'))  group.add_argument(      '--input_subscription',      help=('Input PubSub subscription of the form '            '"projects<project name>/subscriptions/<subsciption name>."'))  known_args, pipeline_args = parser.parse_known_args(argv)  # We use the save_main_session option because one or more DoFn's in this  # workflow rely on global context (e.g., a module imported at module level).  pipeline_options = PipelineOptions(pipeline_args)  pipeline_options.view_as(SetupOptions).save_main_session = True  pipeline_options.view_as(StandardOptions).streaming = True  p = beam.Pipeline(options=pipeline_options)我只是明白這個：<sentence n> <score>我需要一些小的修復(fù)，我被卡住了請幫助我。