---
title: "Returns paginated SpanAnnotation groups."
method: POST
path: "/search/span_annotation/groups"
tags: ["search"]
---

# Returns paginated SpanAnnotation groups.

`POST /search/span_annotation/groups`

## Request body

- GroupQueryRequestSpanColumns
  - `project_id` integer, required — Project the search runs in
  - `search_query` string — Full-text query applied before grouping
  - `filter` FilterSpanColumnsInput, required
    - `id` string, required — Unique id of this node within the filter tree
    - `items` union[], required — Child expressions and/or nested filter nodes
      - union
        - FilterExpressionSpanColumns
          - `id` string, required — Unique id of this expression within the filter tree
          - `column` union, required — The column to filter on: an enum member, or an int id referring to a project-metadata column
            - 'SP_SPAN_TEXT' | 'SP_CODE_ID_LIST_RECURSIVE' | 'SP_USER_ID' | 'SP_MEMO_CONTENT' | 'SP_SOURCE_SOURCE_DOCUMENT_NAME' | 'SP_TAG_ID_LIST_RECURSIVE' | 'SP_FOLDER_ID_LIST_RECURSIVE'
            - integer
          - `operator` union, required — The comparison operator applied to the column
            - 'ID_EQUALS' | 'ID_NOT_EQUALS'
            - 'NUMBER_EQUALS' | 'NUMBER_NOT_EQUALS' | 'NUMBER_GT' | 'NUMBER_LT' | 'NUMBER_GTE' | 'NUMBER_LTE'
            - 'STRING_CONTAINS' | 'STRING_EQUALS' | 'STRING_NOT_EQUALS' | 'STRING_STARTS_WITH' | 'STRING_ENDS_WITH'
            - 'ID_LIST_CONTAINS' | 'ID_LIST_NOT_CONTAINS'
            - 'IDLR_CONTAINS' | 'IDLR_NOT_CONTAINS' | 'IDLR_CONTAINS_RECURSIVE'
            - 'LIST_CONTAINS' | 'LIST_NOT_CONTAINS'
            - 'DATE_EQUALS' | 'DATE_GT' | 'DATE_LT' | 'DATE_GTE' | 'DATE_LTE'
            - 'BOOLEAN_EQUALS' | 'BOOLEAN_NOT_EQUALS'
            - 'ATTACHED_OBJECT_TYPE_EQUALS' | 'ATTACHED_OBJECT_TYPE_NOT_EQUALS' — Filters on the attached object's type (a single AttachedObjectType value).
            - 'ATTACHED_OBJECT_EQUALS' | 'ATTACHED_OBJECT_NOT_EQUALS' — Compares an attached object as a (type, id) pair. The column is a tuple ``(type_expr, id_expr)`` and the value is a two-element list ``[type, id]`` (e.g. ``["tag", "5"]``). Filtering by the raw id alone is meaningless because ids collide across entity types (tag 5, code 5, sdoc 5 are different objects), so the type is always part of the comparison.
            - 'SPAN_ANNOTATION_CONTAINS' | 'SPAN_ANNOTATION_NOT_CONTAINS' — Filters by a single span annotation, given as a [code_id, span_text] pair. The column is an aggregated 2-D array of ``[code_id, span_text]`` string pairs (e.g. ``subquery_dict.SPAN_ANNOTATIONS``), so membership is tested by checking whether the pair is an element of that array.
          - `value` union, required — The value the column is compared against
            - boolean
            - string
            - integer
            - string[]
            - array[]
              - …
        - FilterSpanColumnsInput — recursive
    - `logic_operator` 'or' | 'and', required — This tells our filter how to combine multiple column expressions.
  - `group_by` GroupConfigSpanColumns, required
    - `field` 'SP_SPAN_TEXT' | 'SP_CODE_ID_LIST_RECURSIVE' | 'SP_USER_ID' | 'SP_MEMO_CONTENT' | 'SP_SOURCE_SOURCE_DOCUMENT_NAME' | 'SP_TAG_ID_LIST_RECURSIVE' | 'SP_FOLDER_ID_LIST_RECURSIVE', required
    - `date_granularity` 'day' | 'week' | 'month' | 'year' — Bucket size used when grouping a date column.
  - `page_number` integer — Zero-based page index
  - `page_size` integer — Number of groups per page

## Response `200`

Successful Response

- GroupPage — A paginated list of groups.
  - `items` GroupSummary[], required — The groups on the requested page
    - `key` string, required — Stable identity of the group (an id, date bucket, first letter, or boolean). Defines the partition — grouping is always by key, never by label. The frontend sends this back as `group_key` to drill into the group.
    - `label` string, required — Human-readable name shown as the group header. Functionally dependent on `key`; used for display and alphabetical ordering.
    - `total_results` integer, required — Number of rows in this group
    - `target_id` integer, nullable — ID of the object this group points to (for drill-down navigation), if the group corresponds to a navigable object.
    - `target_type` string, nullable — Type of the target object (an AttachedObjectType), if set.
  - `total_results` integer, required — Total number of groups (unpaginated), used for pagination

## Other responses

- `422` — Validation Error

## Changes

- **2026-08-20** `28b2f0e8bc01` — 1 info
  - endpoint added

[Change history](https://skmtc.dev/uhh-lt/apis/discourse-analysis-tool-suite-api/changes/search/span_annotation/groups/post.md)

---

[API](https://skmtc.dev/uhh-lt/apis/discourse-analysis-tool-suite-api.md) · [All operations](https://skmtc.dev/uhh-lt/apis/discourse-analysis-tool-suite-api/llms.txt) · [OpenAPI document](https://skmtc-service-production.skmtc.workers.dev/v1/apis/uhh-lt/discourse-analysis-tool-suite-api/revisions/a84d27ef7a0d/schema)