name: Automl export data to gcs
description: |
  Exports dataset data to GCS.
inputs:
- name: dataset_path
  type: String
- name: gcs_output_uri_prefix
  optional: true
  type: String
- name: timeout
  optional: true
  type: Float
- default: '{}'
  name: metadata
  optional: true
  type: JsonObject
outputs:
- name: gcs_output_uri_prefix
  type: String
implementation:
  container:
    image: python:3.7
    command:
    - python3
    - -u
    - -c
    - |
      from typing import NamedTuple

      def automl_export_data_to_gcs(
          dataset_path: str,
          gcs_output_uri_prefix: str = None,
          #retry=None, #=google.api_core.gapic_v1.method.DEFAULT,
          timeout: float = None, #=google.api_core.gapic_v1.method.DEFAULT,
          metadata: dict = {},
      ) -> NamedTuple('Outputs', [('gcs_output_uri_prefix', str)]):
          """Exports dataset data to GCS."""
          import sys
          import subprocess
          subprocess.run([sys.executable, "-m", "pip", "install", "google-cloud-automl==0.4.0", "--quiet", "--no-warn-script-location"], env={"PIP_DISABLE_PIP_VERSION_CHECK": "1"}, check=True)

          import google
          from google.cloud import automl
          client = automl.AutoMlClient()

          output_config = {"gcs_destination": {"output_uri_prefix": gcs_output_uri_prefix}}

          response = client.export_data(
              name=dataset_path,
              output_config=output_config,
              #retry=retry or google.api_core.gapic_v1.method.DEFAULT
              timeout=timeout or google.api_core.gapic_v1.method.DEFAULT,
              metadata=metadata,
          )
          print('Operation started:')
          print(response.operation)
          result = response.result()
          metadata = response.metadata
          print('Operation finished:')
          print(metadata)
          return (gcs_output_uri_prefix, )

      import json
      import argparse
      _parser = argparse.ArgumentParser(prog='Automl export data to gcs', description='Exports dataset data to GCS.\n')
      _parser.add_argument("--dataset-path", dest="dataset_path", type=str, required=True, default=argparse.SUPPRESS)
      _parser.add_argument("--gcs-output-uri-prefix", dest="gcs_output_uri_prefix", type=str, required=False, default=argparse.SUPPRESS)
      _parser.add_argument("--timeout", dest="timeout", type=float, required=False, default=argparse.SUPPRESS)
      _parser.add_argument("--metadata", dest="metadata", type=json.loads, required=False, default=argparse.SUPPRESS)
      _parser.add_argument("----output-paths", dest="_output_paths", type=str, nargs=1)
      _parsed_args = vars(_parser.parse_args())
      _output_files = _parsed_args.pop("_output_paths", [])

      _outputs = automl_export_data_to_gcs(**_parsed_args)

      if not hasattr(_outputs, '__getitem__') or isinstance(_outputs, str):
          _outputs = [_outputs]

      _output_serializers = [
          str
      ]

      import os
      for idx, output_file in enumerate(_output_files):
          try:
              os.makedirs(os.path.dirname(output_file))
          except OSError:
              pass
          with open(output_file, 'w') as f:
              f.write(_output_serializers[idx](_outputs[idx]))
    args:
    - --dataset-path
    - inputValue: dataset_path
    - if:
        cond:
          isPresent: gcs_output_uri_prefix
        then:
        - --gcs-output-uri-prefix
        - inputValue: gcs_output_uri_prefix
    - if:
        cond:
          isPresent: timeout
        then:
        - --timeout
        - inputValue: timeout
    - if:
        cond:
          isPresent: metadata
        then:
        - --metadata
        - inputValue: metadata
    - '----output-paths'
    - outputPath: gcs_output_uri_prefix