Affinda Upload and Parse a Document

arazzo: 1.0.1
info:
  title: Affinda Upload and Parse a Document
  summary: Upload a document for parsing, poll until processing completes, then retrieve the parsed data.
  description: >-
    The core Affinda document AI pattern. A file is uploaded for asynchronous
    parsing with wait set to false, which returns an identifier immediately. The
    workflow then polls the document endpoint until the meta.ready flag flips to
    true, and finally fetches the full parsed document so the structured data is
    available to the caller. 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: documentsApi
  url: ../openapi/affinda-documents-api-openapi.yml
  type: openapi
workflows:
- workflowId: upload-and-parse-document
  summary: Upload a file, wait for parsing to finish, and return the parsed data.
  description: >-
    Uploads a document via multipart/form-data into a workspace, creating it with
    wait=false so the API returns an identifier right away, then repeatedly checks
    the document until it is ready and reads back the parsed result.
  inputs:
    type: object
    required:
    - workspace
    - file
    properties:
      workspace:
        type: string
        description: The workspace identifier to upload the document into.
      file:
        type: string
        description: The document file contents (binary) to upload for parsing.
      fileName:
        type: string
        description: Optional file name to store the document under.
      documentType:
        type: string
        description: Optional document type identifier if already known.
  steps:
  - stepId: uploadDocument
    description: >-
      Upload the file for asynchronous parsing. Setting wait to false returns an
      identifier immediately so the document can be polled for completion.
    operationId: createDocument
    requestBody:
      contentType: multipart/form-data
      payload:
        file: $inputs.file
        workspace: $inputs.workspace
        fileName: $inputs.fileName
        documentType: $inputs.documentType
        wait: false
    successCriteria:
    - condition: $statusCode == 201
    outputs:
      identifier: $response.body#/meta/identifier
      ready: $response.body#/meta/ready
  - stepId: pollUntilReady
    description: >-
      Poll the document until its meta.ready flag becomes true, indicating that
      parsing has finished and the parsed data is available.
    operationId: getDocument
    parameters:
    - name: identifier
      in: path
      value: $steps.uploadDocument.outputs.identifier
    successCriteria:
    - condition: $statusCode == 200
    - context: $response.body
      condition: $.meta.ready == true
      type: jsonpath
    outputs:
      ready: $response.body#/meta/ready
      failed: $response.body#/meta/failed
  - stepId: getParsedDocument
    description: >-
      Fetch the completed document to return the full set of parsed data and
      metadata to the caller.
    operationId: getDocument
    parameters:
    - name: identifier
      in: path
      value: $steps.uploadDocument.outputs.identifier
    successCriteria:
    - condition: $statusCode == 200
    outputs:
      data: $response.body#/data
      meta: $response.body#/meta
  outputs:
    identifier: $steps.uploadDocument.outputs.identifier
    data: $steps.getParsedDocument.outputs.data
    meta: $steps.getParsedDocument.outputs.meta