Oxylabs Search Then Scrape Result

arazzo: 1.0.1
info:
  title: Oxylabs Search Then Scrape Result
  summary: Run a parsed search scrape, then scrape a discovered result URL with a second realtime job.
  description: >-
    Chains two Web Scraper API jobs into a discover-then-fetch flow. The first
    job runs a parsed, query-based search source so the structured response
    surfaces a result URL, and the workflow branches on whether a usable URL
    was returned. When a URL is present a second realtime job scrapes that
    target URL with JavaScript rendering and returns its content; when no URL
    is present the flow ends after the search. Both jobs use the single
    submitQuery operation against the Realtime server, so results arrive
    synchronously in each response. 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: oxylabsApi
  url: ../openapi/oxylabs-openapi.yml
  type: openapi
workflows:
- workflowId: search-then-scrape-result
  summary: Parse a search result then scrape the discovered URL.
  description: >-
    Submits a parsed search scrape, branches on whether a result URL was
    returned, and scrapes that URL with a second rendered realtime job.
  inputs:
    type: object
    required:
    - searchSource
    - query
    properties:
      searchSource:
        type: string
        description: Query-based search source (e.g. google_search).
      query:
        type: string
        description: Search query to run.
      fetchSource:
        type: string
        description: Source used to scrape the discovered URL (e.g. universal).
      geo_location:
        type: string
        description: Geo-targeting location string.
  steps:
  - stepId: search
    description: >-
      Run a parsed, query-based search so the structured response can surface a
      result URL.
    operationId: submitQuery
    requestBody:
      contentType: application/json
      payload:
        source: $inputs.searchSource
        query: $inputs.query
        parse: true
        geo_location: $inputs.geo_location
    successCriteria:
    - condition: $statusCode == 200
    outputs:
      resultUrl: $response.body#/results/0/url
      searchContent: $response.body#/results/0/content
    onSuccess:
    - name: haveUrl
      type: goto
      stepId: scrapeResult
      criteria:
      - context: $response.body
        condition: $.results[0].url != null
        type: jsonpath
    - name: noUrl
      type: end
      criteria:
      - context: $response.body
        condition: $.results[0].url == null
        type: jsonpath
  - stepId: scrapeResult
    description: >-
      Scrape the discovered result URL with JavaScript rendering using a second
      realtime job.
    operationId: submitQuery
    requestBody:
      contentType: application/json
      payload:
        source: $inputs.fetchSource
        url: $steps.search.outputs.resultUrl
        render: html
        geo_location: $inputs.geo_location
    successCriteria:
    - condition: $statusCode == 200
    outputs:
      content: $response.body#/results/0/content
      resultStatusCode: $response.body#/results/0/status_code
  outputs:
    resultUrl: $steps.search.outputs.resultUrl
    content: $steps.scrapeResult.outputs.content