Microsoft Purview Profile a Data Asset and Poll for Results

arazzo: 1.0.1
info:
  title: Microsoft Purview Profile a Data Asset and Poll for Results
  summary: Kick off data profiling for an asset, then poll until the profiling completes.
  description: >-
    Runs and waits on a data profiling job in the Purview Data Quality service.
    The workflow initiates profiling for a data asset, then reads the profiling
    result and branches on its status: while the profiling is still InProgress
    it loops back to read the result again, and when it reaches Completed or
    Failed it ends. Every step spells out its request inline — including the
    inline OAuth2 bearer token and the required api-version query parameter — so
    the flow can be read and executed without opening the underlying OpenAPI
    description.
  version: 1.0.0
  x-realizes-capability-ids:
  - BC-610.30
  x-capability-derivation:
    method: 'deterministic join: sourceDescriptions -> per-tag OpenAPI -> tag/capability edge. No classification at this step.'
    min_confidence: 0.7
    sources:
    - capability_id: BC-610.30
      capability_name: Data Quality Management
      spec: microsoft-purview-data-profiling-api-openapi.yml
      confidence: 0.9
    model: Turbo EA Capabilities by Vincent Verdet — Turbo EA, https://github.com/vincentmakes/turbo-ea-capabilities, CC BY 4.0
sourceDescriptions:
- name: dataProfilingApi
  url: ../openapi/microsoft-purview-data-profiling-api-openapi.yml
  type: openapi
workflows:
- workflowId: profile-asset-and-poll
  summary: Start data profiling for an asset and poll until it completes.
  description: >-
    Initiates profiling for a data asset and repeatedly reads the profiling
    result, looping while it is InProgress and ending when it reaches a terminal
    status.
  inputs:
    type: object
    required:
    - authorization
    - assetId
    properties:
      authorization:
        type: string
        description: The OAuth2 bearer token value, e.g. "Bearer eyJ0...".
      apiVersion:
        type: string
        description: The Data Quality API version.
        default: '2024-03-01-preview'
      assetId:
        type: string
        description: The data asset identifier to profile.
      columns:
        type: array
        description: Optional list of column names to restrict profiling to.
        items:
          type: string
      sampleSize:
        type: integer
        description: Optional number of rows to sample during profiling.
  steps:
  - stepId: startProfiling
    description: Initiate data profiling for the asset.
    operationId: runDataProfiling
    parameters:
    - name: Authorization
      in: header
      value: $inputs.authorization
    - name: assetId
      in: path
      value: $inputs.assetId
    - name: api-version
      in: query
      value: $inputs.apiVersion
    requestBody:
      contentType: application/json
      payload:
        columns: $inputs.columns
        sampleSize: $inputs.sampleSize
    successCriteria:
    - condition: $statusCode == 202
    outputs:
      status: $response.body#/status
  - stepId: pollResult
    description: >-
      Read the profiling result and inspect its status. Loop back while it is
      still InProgress; end when it reaches Completed or Failed.
    operationId: getDataProfilingResult
    parameters:
    - name: Authorization
      in: header
      value: $inputs.authorization
    - name: assetId
      in: path
      value: $inputs.assetId
    - name: api-version
      in: query
      value: $inputs.apiVersion
    successCriteria:
    - condition: $statusCode == 200
    outputs:
      status: $response.body#/status
      rowCount: $response.body#/rowCount
      columnProfiles: $response.body#/columnProfiles
    onSuccess:
    - name: stillProfiling
      type: goto
      stepId: pollResult
      criteria:
      - context: $response.body
        condition: $.status == 'InProgress'
        type: jsonpath
    - name: finished
      type: end
      criteria:
      - context: $response.body
        condition: $.status == 'Completed' || $.status == 'Failed'
        type: jsonpath
  outputs:
    status: $steps.pollResult.outputs.status
    rowCount: $steps.pollResult.outputs.rowCount
    columnProfiles: $steps.pollResult.outputs.columnProfiles