> ## Documentation Index
> Fetch the complete documentation index at: https://docs.aiderx.io/llms.txt
> Use this file to discover all available pages before exploring further.

# [임시] Get recommendation bucket hourly metrics

> [임시] 추천 버킷 시간별 지표를 저장된 그대로 돌려준다.

추천 엔티티에는 아직 관리 API 가 없는데 e2e 테스트가 싱커의 기록을 확인해야 해서
임시로 둔다. 집계 없이 행을 그대로 내보내는 것도 같은 이유다 — 리포팅 엔드포인트가
어떤 모양이어야 하는지는 제품 결정이라 여기서 앞질러 정하지 않는다.



## OpenAPI

````yaml /openapi/adm.openapi.en-dev.json get /api/manager/metric/recommendation-bucket
openapi: 3.1.0
info:
  description: >
    ## Introduction

    The Data Management APIs comprehensively oversees the **diverse data** used
    by advertising and recommendation systems, ensuring that the Decision API
    runs accurately and efficiently.


    From registering, updating, and deleting campaigns and placements to **log
    and statistics analysis**, it provides the insights necessary for effective
    system operation.


    The resulting data and metrics are used by the Decision API to **optimize ad
    and content decisions**, ultimately boosting the overall performance of the
    advertising and recommendation system.
  title: Data Management APIs
  version: dev
servers: []
security: []
tags:
  - description: >-
      Authentication endpoints. All roles (admin, operator, advertiser) use
      these to log in, refresh tokens, and manage their own credentials.
    name: Auth
  - description: >-
      Advertiser account management. Operators and admins onboard advertisers,
      manage their budgets, and deactivate accounts. Advertisers can only read
      their own profile and balance — they cannot create, delete, or modify
      another advertiser's account.
    name: Advertisers
  - description: >-
      Internal user (operator / admin) account management. Admin only. Covers
      creating operator accounts, changing roles, toggling activation status,
      and hard-deleting accounts. Operators have no access to this resource.
    name: Users
  - description: >-
      The AdUnit Management API provides comprehensive functionalities for
      managing ad units and related resources. AdUnits are organized in a
      hierarchical path structure derived from their code and parent path. Only
      admin and operator roles have access — advertisers cannot access raw
      inventory. References:
      [Placement](https://docs.aiderx.io/guides/about-a2/en/placement)
    name: AdUnit
  - description: >-
      Ad catalog management. CRUD for ad catalogs plus public shared-view access
      via a share link.
    name: Ad Catalogs
  - description: >-
      Ad source (demand partner) management. CRUD for external ad sources that
      supply demand to ad units.
    name: AdSource
  - description: >-
      Allocation management. An allocation binds a line item to a serving target
      such as an ad unit. Covers CRUD, status counts, and single-allocation
      lookup.
    name: Allocations
  - description: >-
      Associated items pool (admin). Bulk upsert/delete and lookup of
      associated-items pools used by the recommendation pipeline.
    name: AssociatedItemsPool
  - description: >-
      Custom targeting segment management. CRUD and bulk operations for custom
      targeting segments.
    name: Custom Targeting Segment
  - description: Audit log access. Read the system audit log of administrative actions.
    name: Audit Logs
  - description: >-
      Creative block list. List, block, and unblock creatives to exclude them
      from serving.
    name: Blocked Creative
  - description: Catalog item management. List and update catalog (product/data) items.
    name: Catalogs
  - description: >-
      Creative asset management. Operators and admins manage creatives;
      advertisers see only their own. Deletion is blocked while any line-item
      association (LICA) still exists.
    name: Creatives
  - description: >-
      Dashboard aggregates. Summary data, inventory, and line-item metric
      rollups for the console dashboard.
    name: Dashboard
  - description: Experiment management. CRUD for A/B and optimization experiments.
    name: Experiment
  - description: >-
      External creative inspection. Read-only access to creatives ingested from
      external demand sources.
    name: External Creative
  - description: >-
      Third-party integration management. Manage integration sources (external
      data/serving connectors), probe connectivity, and read public
      (sensitive-config-redacted) configs.
    name: Integration
  - description: >-
      Line Item–Creative Associations (LICA). Associates creatives with line
      items, including batch create and delete (max 100, partial failure
      allowed) and dashboard counts.
    name: LICAs
  - description: License information. Read available license features and usage metrics.
    name: License
  - description: >-
      Line item management within an order. Covers create, update, delete, batch
      operations (max 100, partial failure allowed), and status transitions. A
      line item defines budget, goal, schedule, and targeting under its parent
      order.
    name: LineItems
  - description: Media library. List, upload, and delete media files used by creatives.
    name: Media Library
  - description: >-
      Metric query endpoints. Retrieve aggregated metrics for ad units,
      creatives, line items, and related entities.
    name: Metric
  - description: >-
      Advertiser order management. Operators and admins create and manage orders
      that group line items under a single budget; advertisers can read their
      own. Includes batch operations (max 100) and draft-only deletion rules.
    name: Orders
  - description: >-
      Placement management. A placement maps ad units to a serving surface.
      Endpoints cover CRUD and replacing the ad units bound to a placement.
    name: Placements
  - description: >-
      System settings. Read and update system configuration, including site
      logo, public URL, TLS, and access-token regeneration.
    name: Settings
  - description: System resources. Public static resources such as the site logo.
    name: System
  - description: >-
      Operational/developer control endpoints. Low-level ext patches and health
      checks intended for internal operation, not general API consumers.
    name: developer
  - description: >-
      Operational allocation control. Update allocation ext or force-update
      status (admin only).
    name: developer/control/allocations
  - description: >-
      Ad-serving monitoring (operational). Read ad-serving status, ad-unit
      monitoring history, and alerts.
    name: developer/monitoring
  - description: Internal health and probe endpoints.
    name: internal
  - description: >-
      Manager-level utility endpoints (e.g. service health) not tied to a
      specific resource.
    name: manager
  - description: Manage mappings between ad units, native formats, and native styles.
    name: Ad Unit Native Style
  - description: Manage custom targeting keys, including lifecycle and bulk operations.
    name: Custom Targeting Key
  - description: >-
      Manage values belonging to custom targeting keys, including lifecycle and
      bulk operations.
    name: Custom Targeting Value
  - description: Manage native formats and the fields that define each format.
    name: Native Format
  - description: List templates available for native styles.
    name: Native Style Template
  - description: Manage native rendering styles.
    name: Native Style
paths:
  /api/manager/metric/recommendation-bucket:
    get:
      tags:
        - Metric
      summary: '[임시] Get recommendation bucket hourly metrics'
      description: |-
        [임시] 추천 버킷 시간별 지표를 저장된 그대로 돌려준다.

        추천 엔티티에는 아직 관리 API 가 없는데 e2e 테스트가 싱커의 기록을 확인해야 해서
        임시로 둔다. 집계 없이 행을 그대로 내보내는 것도 같은 이유다 — 리포팅 엔드포인트가
        어떤 모양이어야 하는지는 제품 결정이라 여기서 앞질러 정하지 않는다.
      operationId: >-
        handle_recommendation_bucket_api_manager_metric_recommendation_bucket_get
      parameters:
        - in: query
          name: auid
          required: true
          schema:
            format: uuid
            title: Auid
            type: string
        - in: query
          name: experiment_id
          required: true
          schema:
            format: uuid
            title: Experiment Id
            type: string
      responses:
        '200':
          content:
            application/json:
              schema:
                items:
                  $ref: '#/components/schemas/RecommendationBucketHourlyMetric'
                type: array
          description: Successful Response
        '422':
          content:
            application/json:
              schema:
                $ref: '#/components/schemas/HTTPValidationError'
          description: Validation Error
      security:
        - HTTPBearer: []
components:
  schemas:
    RecommendationBucketHourlyMetric:
      description: >-
        Represents a recommendation hourly metric, per experiment bucket, source
        and context.
      properties:
        auid:
          description: Id of the ad unit
          format: uuid
          title: Auid
          type: string
        bucket_id:
          title: Bucket Id
          type: string
        click:
          description: number of clicks
          title: Click
          type: integer
        conversion:
          description: number of conversions
          title: Conversion
          type: integer
        experiment_id:
          description: Id of the experiment
          format: uuid
          title: Experiment Id
          type: string
        impression:
          description: number of impressions
          title: Impression
          type: integer
        policy_id:
          description: Id of the policy the bucket applied
          format: uuid
          title: Policy Id
          type: string
        serving_context:
          description: Context queried within that source
          title: Serving Context
          type: string
        source:
          description: Candidate source the recommendation came from
          title: Source
          type: string
        timestamp:
          description: Timestamp of the metric
          format: date-time
          title: Timestamp
          type: string
      required:
        - auid
        - bucket_id
        - click
        - conversion
        - experiment_id
        - impression
        - policy_id
        - serving_context
        - source
        - timestamp
      title: RecommendationBucketHourlyMetric
      type: object
    HTTPValidationError:
      properties:
        detail:
          items:
            $ref: '#/components/schemas/ValidationError'
          title: Detail
          type: array
      title: HTTPValidationError
      type: object
    ValidationError:
      properties:
        loc:
          items:
            anyOf:
              - type: string
              - type: integer
          title: Location
          type: array
        msg:
          title: Message
          type: string
        type:
          title: Error Type
          type: string
      required:
        - loc
        - msg
        - type
      title: ValidationError
      type: object
  securitySchemes:
    HTTPBearer:
      description: >-
        Bearer authentication. Pass the access token in the `Authorization:
        Bearer <token>` header.
      scheme: bearer
      type: http

````