openapi: 3.0.3
info:
  title: ParseHub API
  description: >-
    ParseHub is a visual web scraping tool. The v2 API allows you to start
    runs, check run status, retrieve scraped data, and manage projects and runs.
  version: '2.0'
  contact:
    name: ParseHub Support
    url: https://www.parsehub.com/docs/ref/api/v2/
servers:
  - url: https://www.parsehub.com/api/v2
    description: ParseHub API v2
security:
  - apiKey: []
paths:
  /projects:
    get:
      summary: List projects
      description: Returns a paginated list of projects in the user account.
      operationId: listProjects
      parameters:
        - name: api_key
          in: query
          required: true
          schema:
            type: string
        - name: offset
          in: query
          schema:
            type: integer
            default: 0
        - name: limit
          in: query
          schema:
            type: integer
            default: 20
            maximum: 20
        - name: include_options
          in: query
          schema:
            type: integer
            enum: [0, 1]
      responses:
        '200':
          description: Project list
          content:
            application/json:
              schema:
                type: object
                properties:
                  projects:
                    type: array
                    items:
                      $ref: '#/components/schemas/Project'
                  total_projects:
                    type: integer
  /projects/{project_token}:
    get:
      summary: Get a project
      operationId: getProject
      parameters:
        - name: project_token
          in: path
          required: true
          schema:
            type: string
        - name: api_key
          in: query
          required: true
          schema:
            type: string
        - name: offset
          in: query
          schema:
            type: integer
      responses:
        '200':
          description: Project details
          content:
            application/json:
              schema:
                $ref: '#/components/schemas/Project'
  /projects/{project_token}/run:
    post:
      summary: Start a run
      description: Starts a new run for a project, optionally with custom start URL, value overrides, or webhook.
      operationId: startRun
      parameters:
        - name: project_token
          in: path
          required: true
          schema:
            type: string
      requestBody:
        required: true
        content:
          application/x-www-form-urlencoded:
            schema:
              type: object
              required: [api_key]
              properties:
                api_key:
                  type: string
                start_url:
                  type: string
                start_template:
                  type: string
                start_value_override:
                  type: string
                send_email:
                  type: integer
                  enum: [0, 1]
      responses:
        '200':
          description: Run object created
          content:
            application/json:
              schema:
                $ref: '#/components/schemas/Run'
  /projects/{project_token}/last_ready_run/data:
    get:
      summary: Get data for the last ready run
      description: Returns the data extracted by the most recent ready run of a project.
      operationId: getLastReadyRunData
      parameters:
        - name: project_token
          in: path
          required: true
          schema:
            type: string
        - name: api_key
          in: query
          required: true
          schema:
            type: string
        - name: format
          in: query
          schema:
            type: string
            enum: [json, csv]
            default: json
      responses:
        '200':
          description: Run data
          content:
            application/json:
              schema:
                type: object
            text/csv:
              schema:
                type: string
  /runs/{run_token}:
    get:
      summary: Get a run
      operationId: getRun
      parameters:
        - name: run_token
          in: path
          required: true
          schema:
            type: string
        - name: api_key
          in: query
          required: true
          schema:
            type: string
      responses:
        '200':
          description: Run details
          content:
            application/json:
              schema:
                $ref: '#/components/schemas/Run'
    delete:
      summary: Delete a run
      description: Deletes a run and its associated data.
      operationId: deleteRun
      parameters:
        - name: run_token
          in: path
          required: true
          schema:
            type: string
        - name: api_key
          in: query
          required: true
          schema:
            type: string
      responses:
        '200':
          description: Run deleted
  /runs/{run_token}/data:
    get:
      summary: Get run data
      description: Returns data extracted by a specific run, in JSON or CSV.
      operationId: getRunData
      parameters:
        - name: run_token
          in: path
          required: true
          schema:
            type: string
        - name: api_key
          in: query
          required: true
          schema:
            type: string
        - name: format
          in: query
          schema:
            type: string
            enum: [json, csv]
            default: json
      responses:
        '200':
          description: Run data
          content:
            application/json:
              schema:
                type: object
            text/csv:
              schema:
                type: string
  /runs/{run_token}/cancel:
    post:
      summary: Cancel a run
      operationId: cancelRun
      parameters:
        - name: run_token
          in: path
          required: true
          schema:
            type: string
      requestBody:
        required: true
        content:
          application/x-www-form-urlencoded:
            schema:
              type: object
              required: [api_key]
              properties:
                api_key:
                  type: string
      responses:
        '200':
          description: Run cancelled
components:
  securitySchemes:
    apiKey:
      type: apiKey
      in: query
      name: api_key
  schemas:
    Project:
      type: object
      properties:
        token:
          type: string
        title:
          type: string
        templates_json:
          type: string
        main_template:
          type: string
        main_site:
          type: string
        options_json:
          type: string
        last_run:
          $ref: '#/components/schemas/Run'
        last_ready_run:
          $ref: '#/components/schemas/Run'
    Run:
      type: object
      properties:
        project_token:
          type: string
        run_token:
          type: string
        status:
          type: string
          enum: [initialized, queued, running, complete, cancelled, error]
        data_ready:
          type: string
        start_time:
          type: string
        end_time:
          type: string
        pages:
          type: integer
        md5sum:
          type: string
        start_url:
          type: string
        start_template:
          type: string
        start_value_override:
          type: object