plgpsql - 是否可以声明 setof 记录或表变量？

Question

我有以下 - 不工作 - 功能：

CREATE FUNCTION permission_cache_update(affected_user_list int[])
  RETURNS TABLE(user_id INT4, permission_id INT4)
AS
  $BODY$
  DECLARE
    current_relations SETOF RECORD;
  BEGIN

  WITH
  affected_user AS (
     SELECT unnest(affected_user_list) AS u_id
  ),
  affected_relations AS (
    SELECT user_role.user_id, role_permission.permission_id
      FROM   user_role
      JOIN   role_permission ON user_role.role_id = role_permission.role_id
      JOIN   affected_user ON affected_user.u_id = user_role.user_id
    UNION
    SELECT user_permission.user_id, user_permission.permission_id
      FROM   user_permission
      JOIN   affected_user ON affected_user.u_id = user_permission.user_id
  )
  SELECT affected_relations.user_id, affected_relations.permission_id FROM affected_relations
  INTO current_relations;

  DELETE FROM permission_cache WHERE ARRAY[user_id] <@ affected_user_list;
  INSERT INTO permission_cache (user_id, permission_id) SELECT user_id, permission_id FROM current_relations;

  END
  $BODY$
LANGUAGE plpgsql;

我想将当前的用户权限关系存储为current_relations（INT4，INT4）。是否可以使用没有循环和临时表的变量来执行此操作？

稍后我将使用类似的东西，所以我真的需要它作为变量，而不是子查询

DELETE FROM permission_cache WHERE ARRAY[user_id] <@ affected_user_list AND NOT IN (SELECT user_id, permission_id FROM current_relations);

INSERT INTO permission_cache (user_id, permission_id) SELECT user_id, permission_id FROM current_relations WHERE NOT EXIST (SELECT user_id, permission_id FROM permission_cache);

我认为表 -> 二维数组转换是可能的，但这很复杂，所以如果可以使用记录，那就更好了......

解决方案：

使用一维数组创建循环比使用记录或二维数组更容易：

CREATE FUNCTION permission_cache_update(
  IN affected_user_list INT4 []
)
  RETURNS VOID
AS
  $BODY$
  DECLARE
    user_index                    INT4;
    current_user_id INT4;
    current_permission_relations  INT4 [];
    deleted_permission_relations  INT4 [];
    inserted_permission_relations INT4 [];
  BEGIN
    FOR user_index IN 1 .. array_upper(affected_user_list, 1) LOOP
      current_user_id := affected_user_list[user_index];

      WITH
          user_permission_summary AS
        (
          SELECT
            role_permission.permission_id
          FROM user_role, role_permission
          WHERE role_permission.role_id = user_role.role_id AND user_role.user_id = current_user_id
          UNION
          SELECT
            user_permission.permission_id
          FROM user_permission
          WHERE user_permission.user_id = current_user_id
        )
      SELECT
        array_agg(permission_id)
      FROM user_permission_summary
      INTO current_permission_relations;

    SELECT
      array_agg(permission_cache.permission_id)
    FROM permission_cache
    WHERE permission_cache.user_id = current_user_id AND ( current_permission_relations IS NULL OR
          NOT (ARRAY [permission_cache.permission_id] <@ current_permission_relations) )
    INTO deleted_permission_relations;

    SELECT
      array_agg(inserted_permission_id)
    FROM unnest(current_permission_relations) AS inserted_permission_id
    WHERE NOT EXISTS(SELECT
                       1
                     FROM permission_cache
                     WHERE permission_cache.user_id = current_user_id AND
                           permission_cache.permission_id = inserted_permission_id)
    INTO inserted_permission_relations;

    DELETE FROM permission_cache
    WHERE permission_cache.user_id = current_user_id AND
          permission_cache.permission_id = ANY (deleted_permission_relations);

    INSERT INTO permission_cache (user_id, permission_id)
      SELECT
        current_user_id,
        inserted_permission_id
      FROM unnest(inserted_permission_relations) AS inserted_permission_id;

    END LOOP;
  END;
  $BODY$
LANGUAGE plpgsql VOLATILE;

off： Pplpgsql 是一种非常糟糕的语言，文档甚至很糟糕，我不喜欢它...抱歉代码格式化，我的 IDE 中的自动格式化不太好...：S

Answer 1

我认为这个简化的SQL 函数可能会满足您的需求：

CREATE FUNCTION permission_cache_update(affected_user_list int[])
  RETURNS void AS
$func$

DELETE FROM permission_cache p
USING (SELECT unnest($1) AS u_id) a
WHERE  p.user_id = a.u_id;

INSERT INTO permission_cache (user_id, permission_id)
SELECT u.user_id, r.permission_id
FROM   user_role u
JOIN   role_permission r USING (role_id)
JOIN  (SELECT unnest($1) AS u_id) a ON a.u_id = u.user_id
UNION
SELECT p.user_id, p.permission_id
FROM   user_permission p
JOIN  (SELECT unnest($1) AS u_id) a ON a.u_id = p.user_id;

$func$ LANGUAGE sql;

可写 CTE（又名数据修改 CTE）将极大地简化这一过程，但仅在 Postgres 9.1 中引入。过时的版本再次成为障碍。

按名称引用参数仅在 Postgres 9.2 中为 SQL 函数引入（早期为 plpgsql）。所以我使用位置参数$1。