- release_tracks: [ALPHA, BETA, GA]

  help_text:
    brief: Detect and transcribe text from PDF files stored in Google Cloud Storage.
    description: |
      Detect and transcribe text from PDF files stored in Google Cloud Storage.

      The Vision API accepts PDF files up to 2000 pages.
      Larger files will return an error.
    examples: |
      To detect text for input PDF file 'gs://my_bucket/input_file' and store output in 'gs://my_bucket/out_put_prefix':

        $ {command} gs://my_bucket/input_file
        gs://my_bucket/out_put_prefix

  request:
    collection: vision.files
    method: asyncBatchAnnotate
    api_version: v1
    static_fields:
      requests.features.type: DOCUMENT_TEXT_DETECTION
      requests.inputConfig.mimeType: 'application/pdf'

  response:
    error:
      field: error
      code: code
      message: message

  arguments:
    params:
    - api_field: requests.inputConfig.gcsSource
      arg_name: input_file
      is_positional: true
      help_text: |
        Google Cloud Storage location to read the input from. It must be in
        Google Cloud Storage format (gs://bucket/object)
      processor: googlecloudsdk.command_lib.ml.vision.util:GetGcsSourceFromPath
    - api_field: requests.outputConfig.gcsDestination
      arg_name: output_path
      is_positional: true
      help_text: |
        Google Cloud Storage location to store the output file. It must be in
        Google Cloud Storage format (gs://bucket/object)
      processor: googlecloudsdk.command_lib.ml.vision.util:GetGcsDestinationFromPath
    - api_field: requests.outputConfig.batchSize
      arg_name: batch-size
      help_text: |
        Maximum number of response protos to put into each output JSON file on
        Google Cloud Storage.
        The valid range is [1, 100]. If not specified, the default value is 20.
    - _COMMON_: args.model_field


  output:
    format: json