如何將 JSON 數組合並為一個數組

Question

我的 PostgreSQL 數據庫。 我需要通過 sqlalchemy 擺脫 JSON 嵌套。

組列組中的數據

團體
[{“id”：5，“名稱”：“imba”}]
[{"id": 1, "name": "group1"}, {"id": 3, "name": "group3"}]

如何在 SQLAlchemy 中創建查詢？

預期結果

[{"id": 5, "name": "imba"}, {"id": 1, "name": "group1"}, {"id": 3, "name": "group3"}]

Answer 1

像這樣在 Postgresql 中給定一個表

test# table t73051318;
 id_ │                           groups                           
═════╪════════════════════════════════════════════════════════════
   1 │ [{"id": 5, "name": "imba"}]
   2 │ [{"id": 1, "name": "group1"}, {"id": 3, "name": "group3"}]
(2 rows)

我們可以通過對groups列上調用jsonb_array_elements的結果調用json_agg函數來合並 JSONB 值。

test# select json_agg(row) as aggregated
        from (select jsonb_array_elements(groups) as row
                from t73051318
        ) q;
                                      aggregated                                       
═══════════════════════════════════════════════════════════════════════════════════════
 [{"id": 5, "name": "imba"}, {"id": 1, "name": "group1"}, {"id": 3, "name": "group3"}]
(1 row)

在 SQLAlchemy 核心中，代碼如下所示：

import sqlalchemy as sa

# Create a subquery for the jsonb_array_elements call.
subq = sa.select(
    sa.func.jsonb_array_elements(tbl.c.groups).label('row')
).subquery()

# Aggregate the subquery result.
stmt = sa.select(sa.func.json_agg(subq.c.row).label('aggregated'))

with engine.connect() as conn:
    res = conn.execute(stmt)
    print(next(res).aggregated)

在 Flask-SQLAlchemy 中，代碼是

# Create a subquery for the jsonb_array_elements call.
subq = db.session.query(
    db.func.jsonb_array_elements(MyModel.groups).label('row')
).subquery()

# Aggregate the subquery result.
stmt = db.session.query(db.func.json_agg(subq.c.row).label('aggregated'))

result = stmt.first()