StarRocks/starrocks

Support using JSON as GROUP BY

Open

#62,502 建立於 2025年8月29日

在 GitHub 查看
 (4 留言) (0 反應) (1 負責人)Java (5,717 star) (1,246 fork)batch import
good first issuetype/enhancement

描述

Enhancement

MySQL root@127.1:d1> select event_date from tjson group by event_date;
(1064, 'Getting analyzing error. Detail message: Type (nested) percentile/hll/bitmap/json not support group-by.')

JSON is currently not supported in the GROUP BY clause, but enabling this functionality could be beneficial for certain scenarios.

To implement GROUP BY support for JSON, a stable hash function and comparison function are required. This can be achieved using asymmetric hash mixing techniques, such as XOR, which is already implemented in JsonValue::hash.

貢獻者指南

技術棧
javasql
領域
databaseperformance
議題類型
feature
難度面向新貢獻者的預計實作難度，1 表示很小改動，5 表示專家級工作。
4
預計時間有經驗貢獻者完成調查、實作、測試並準備 pull request 的粗略時間範圍。
over 1 week
活動狀態議題目前的可參與程度：新鮮、活躍、陳舊、阻塞或等待維護者輸入。
blocked
清晰度議題是否清楚說明預期改動、驗收標準和下一步。
clear
前置要求
StarRocks internalsJSON data typesHashing techniques
新手友善度1-100 的估計分數，表示該議題對首次貢獻者的友善程度。
30
研究方向
Investigate the current implementation of JsonValue::hash in the StarRocks codebase (likely in the C++ backend). Understand how asymmetric hash mixing is implemented and extend the GROUP BY operator to support JSON type by using that hash function. Also need a comparison function. Check if any existing similar implementations for other types can serve as a template. Refer to the issue comments for any additional maintainer guidance.