Releasing of the multiple schedules using async postgres datastore produce an error

[Reskov]

Hello! 👋 I am trying to use the latest app scheduler 4.x master branch with an async Postgres datastore.

I have a problem with releasing multiple schedules at the same time:
when the scheduler tries to release more than one value it fails with

sqlalchemy.exc.ResourceClosedError: This result object does not return rows. It has been closed automatically.

on the line

apscheduler/src/apscheduler/datastores/async_sqlalchemy.py

Lines 298 to 299 in 82992cd

	if self._supports_update_returning:
	update = update.returning(self.t_schedules.c.id)

The problem occurs when the number of the items at update_args is greater than 1, because sqlalchemy will use executemany under the hood

Suddenly, it is not possible to use asyncpg executemany with returning statement

MagicStack/asyncpg#137

code to reproduce

import asyncio

from sqlalchemy import Integer, Column, bindparam, MetaData, String, Table
from sqlalchemy.ext.asyncio import create_async_engine
from sqlalchemy.sql.elements import BindParameter

meta = MetaData()

t1 = Table("t1", meta, Column("id", Integer, primary_key=True), Column("name", String))


async def async_main():
    engine = create_async_engine(
        "postgresql+asyncpg://postgres:postgres@localhost/test", echo=True
    )

    async with engine.begin() as conn:
        await conn.run_sync(meta.drop_all)
        await conn.run_sync(meta.create_all)

        await conn.execute(t1.insert(), [{"name": "name1"}, {"name": "name2"}])

        p_id: BindParameter = bindparam("p_id")
        p_name: BindParameter = bindparam("p_name")

        stmt_args = [
            {"p_id": 1, "p_name": f"name1_up"}, # works with only one item
            {"p_id": 2, "p_name": f"name2_up"},
        ]
        stmt = t1.update().where(t1.c.id == p_id).values(name=p_name).returning(t1.c.id)
        result = await conn.execute(stmt, stmt_args)
        print(list(result.scalars()))


asyncio.run(async_main())