Bright Data Marketplace Dataset Snapshot and Deliver

arazzo: 1.0.1
info:
  title: Bright Data Marketplace Dataset Snapshot and Deliver
  summary: Inspect a marketplace dataset's metadata, read a snapshot, and deliver it to cloud.
  description: >-
    A marketplace consumption pattern. The workflow reads a dataset's metadata
    to confirm its shape, retrieves a snapshot of its rows, and schedules
    delivery of that snapshot to a cloud destination. Every step spells out its
    request inline so the flow can be read and executed without opening the
    underlying OpenAPI description.
  version: 1.0.0
sourceDescriptions:
- name: datasetsApi
  url: ../openapi/bright-data-datasets-api-openapi.yml
  type: openapi
- name: deliveryApi
  url: ../openapi/bright-data-delivery-api-openapi.yml
  type: openapi
- name: snapshotsApi
  url: ../openapi/bright-data-snapshots-api-openapi.yml
  type: openapi
workflows:
- workflowId: inspect-snapshot-and-deliver
  summary: Read dataset metadata, fetch a snapshot, then deliver it to cloud storage.
  description: >-
    Reads a marketplace dataset's metadata, retrieves a snapshot of its rows,
    and schedules delivery of the snapshot to a configured cloud destination.
  inputs:
    type: object
    required:
    - apiToken
    - datasetId
    - snapshotId
    - destinationType
    - bucket
    properties:
      apiToken:
        type: string
        description: Bright Data API token used as a Bearer credential.
      datasetId:
        type: string
        description: Marketplace dataset identifier to inspect.
      snapshotId:
        type: string
        description: Snapshot identifier to read and deliver.
      destinationType:
        type: string
        description: Delivery destination type (s3, azure, gcs, snowflake, webhook).
      bucket:
        type: string
        description: Destination bucket or container name.
      credentials:
        type: object
        description: Destination credentials object.
      format:
        type: string
        description: Delivery format (json, ndjson, csv, parquet).
  steps:
  - stepId: readMetadata
    description: >-
      Read the dataset metadata to confirm its shape before consuming a
      snapshot.
    operationId: getDatasetMetadata
    parameters:
    - name: Authorization
      in: header
      value: "Bearer $inputs.apiToken"
    - name: dataset_id
      in: path
      value: $inputs.datasetId
    successCriteria:
    - condition: $statusCode == 200
    outputs:
      metadata: $response.body
  - stepId: readSnapshot
    description: >-
      Retrieve a snapshot of the dataset rows in the requested format.
    operationId: getDatasetSnapshot
    parameters:
    - name: Authorization
      in: header
      value: "Bearer $inputs.apiToken"
    - name: snapshot_id
      in: path
      value: $inputs.snapshotId
    - name: format
      in: query
      value: $inputs.format
    successCriteria:
    - condition: $statusCode == 200
    outputs:
      rows: $response.body
  - stepId: deliverSnapshot
    description: >-
      Schedule delivery of the snapshot to the configured cloud destination in
      the requested format.
    operationId: deliverDatasetSnapshot
    parameters:
    - name: Authorization
      in: header
      value: "Bearer $inputs.apiToken"
    - name: snapshot_id
      in: path
      value: $inputs.snapshotId
    requestBody:
      contentType: application/json
      payload:
        destination:
          type: $inputs.destinationType
          bucket: $inputs.bucket
          credentials: $inputs.credentials
        format: $inputs.format
    successCriteria:
    - condition: $statusCode == 200
    outputs:
      deliveryResult: $response.body
  outputs:
    metadata: $steps.readMetadata.outputs.metadata
    deliveryResult: $steps.deliverSnapshot.outputs.deliveryResult