arazzo: 1.0.1
info:
  title: Wolfram|Alpha Query Recognizer Routing
  summary: Classify a query with the Fast Query Recognizer, then route accepted queries to Full Results.
  description: >-
    The Fast Query Recognizer classifies a natural language query in under ten
    milliseconds, reporting whether Wolfram|Alpha is likely to return a useful
    response, the detected domain, and a result significance score. This
    workflow uses that classification as a routing gate: it first recognizes the
    query, and only when the query is accepted does it spend the heavier call to
    the Full Results API for the complete computed pods. When the query is not
    accepted, the flow ends without invoking the expensive backend. Every step
    spells out its request inline so the flow can be read and executed without
    opening the underlying OpenAPI descriptions.
  version: 1.0.0
sourceDescriptions:
- name: fastQueryRecognizerApi
  url: ../openapi/wolframalpha-fast-query-recognizer-api-openapi.yml
  type: openapi
- name: fullResultsApi
  url: ../openapi/wolframalpha-full-results-api-openapi.yml
  type: openapi
workflows:
- workflowId: query-recognizer-routing
  summary: Recognize a query and route accepted queries to the Full Results API.
  description: >-
    Submits the query to the Fast Query Recognizer. When the classification is
    accepted, the flow branches to the Full Results API for the complete pod
    data; when it is not accepted, the flow ends after recognition.
  inputs:
    type: object
    required:
    - appid
    - input
    properties:
      appid:
        type: string
        description: Wolfram|Alpha AppID used to authenticate every request.
      input:
        type: string
        description: The natural language query to classify and, if accepted, compute.
      mode:
        type: string
        description: Recognition mode, either Default for typed input or Voice for spoken input.
  steps:
  - stepId: recognize
    description: >-
      Classify the query with the Fast Query Recognizer to decide whether
      Wolfram|Alpha is likely to return a useful response before spending a
      heavier backend call.
    operationId: recognizeQuery
    parameters:
    - name: appid
      in: query
      value: $inputs.appid
    - name: mode
      in: query
      value: $inputs.mode
    - name: i
      in: query
      value: $inputs.input
    - name: output
      in: query
      value: json
    successCriteria:
    - condition: $statusCode == 200
    outputs:
      accepted: $response.body#/query/accepted
      domain: $response.body#/query/domain
      resultSignificanceScore: $response.body#/query/resultsignificancescore
    onSuccess:
    - name: queryAccepted
      type: goto
      stepId: computeFullResults
      criteria:
      - context: $response.body
        condition: $.query.accepted == true
        type: jsonpath
    - name: queryRejected
      type: end
      criteria:
      - context: $response.body
        condition: $.query.accepted == false
        type: jsonpath
  - stepId: computeFullResults
    description: >-
      Run the accepted query against the Full Results API and return the
      complete computed pods in JSON.
    operationId: queryFullResults
    parameters:
    - name: appid
      in: query
      value: $inputs.appid
    - name: input
      in: query
      value: $inputs.input
    - name: output
      in: query
      value: json
    - name: format
      in: query
      value: plaintext,image
    successCriteria:
    - condition: $statusCode == 200
    - condition: $response.body#/queryresult/success == true
    outputs:
      success: $response.body#/queryresult/success
      numpods: $response.body#/queryresult/numpods
      pods: $response.body#/queryresult/pods
  outputs:
    accepted: $steps.recognize.outputs.accepted
    domain: $steps.recognize.outputs.domain
    pods: $steps.computeFullResults.outputs.pods