arazzo: 1.0.1
info:
  title: Affinda Provision Collection and Ingest a Document
  summary: Create a workspace, create a collection bound to an extractor, then upload a document into it and parse it.
  description: >-
    An end-to-end onboarding flow that builds the container hierarchy and then
    proves it works by ingesting a document. A workspace is created, a collection
    bound to an extractor is created inside it, a file is uploaded into that
    collection for asynchronous parsing, and the document is polled until parsing
    completes. Every step spells out its request inline so the flow can be read and
    executed without opening the underlying OpenAPI description.
    Note: the collections endpoints are marked deprecated in the v3 specification
    but remain the documented way to bind an extractor to a document container.
  version: 1.0.0
sourceDescriptions:
- name: affindaV3Api
  url: ../openapi/affinda-v3-openapi.yml
  type: openapi
workflows:
- workflowId: provision-and-ingest-document
  summary: Build a workspace and collection, then upload and parse a document into it.
  description: >-
    Creates a workspace, creates an extractor-bound collection inside it, uploads a
    file into that collection with wait=false, and polls until the document is
    ready.
  inputs:
    type: object
    required:
    - organization
    - workspaceName
    - collectionName
    - extractor
    - file
    properties:
      organization:
        type: string
        description: The organization identifier to create the workspace under.
      workspaceName:
        type: string
        description: The display name for the new workspace.
      collectionName:
        type: string
        description: The display name for the new collection.
      extractor:
        type: string
        description: The extractor identifier the collection should use.
      file:
        type: string
        description: The document file contents (binary) to upload.
      fileName:
        type: string
        description: Optional file name to store the document under.
  steps:
  - stepId: createWorkspace
    description: Create the workspace that will hold the collection and documents.
    operationId: createWorkspace
    requestBody:
      contentType: application/json
      payload:
        organization: $inputs.organization
        name: $inputs.workspaceName
    successCriteria:
    - condition: $statusCode == 201
    outputs:
      workspaceIdentifier: $response.body#/identifier
  - stepId: createCollection
    description: Create an extractor-bound collection inside the new workspace.
    operationId: createCollection
    requestBody:
      contentType: application/json
      payload:
        name: $inputs.collectionName
        workspace: $steps.createWorkspace.outputs.workspaceIdentifier
        extractor: $inputs.extractor
    successCriteria:
    - condition: $statusCode == 201
    outputs:
      collectionIdentifier: $response.body#/identifier
  - stepId: uploadDocument
    description: >-
      Upload the file into the new collection with wait=false so an identifier is
      returned for polling.
    operationId: createDocument
    requestBody:
      contentType: multipart/form-data
      payload:
        file: $inputs.file
        workspace: $steps.createWorkspace.outputs.workspaceIdentifier
        collection: $steps.createCollection.outputs.collectionIdentifier
        fileName: $inputs.fileName
        wait: false
    successCriteria:
    - condition: $statusCode == 201
    outputs:
      identifier: $response.body#/meta/identifier
  - stepId: pollUntilReady
    description: Poll the document until meta.ready becomes true.
    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:
      data: $response.body#/data
  outputs:
    workspaceIdentifier: $steps.createWorkspace.outputs.workspaceIdentifier
    collectionIdentifier: $steps.createCollection.outputs.collectionIdentifier
    documentIdentifier: $steps.uploadDocument.outputs.identifier
    data: $steps.pollUntilReady.outputs.data