如何为多表关联查询创建带触发器函数自动刷新的MATERIALIZED VIEW

现有4张业务数据表，DDL如下：

CREATE TABLE public.type
(
  id serial,
  name text,
  groups_id integer,
  comment text
);
CREATE TABLE public.berand
(
  id serial,
  name text,
  type_id integer,
  comment text
);
CREATE TABLE public.model
(
  id serial,
  name text,
  berand_id integer,
  comment text
);
CREATE TABLE public.versions
(
  id serial,
  name text,
  model_id integer,
  comment text
);

多层左关联查询性能较差，对应查询语句如下：

SELECT id,name tname, bname, mname, vname 
FROM type 
left join(
  select  berand.type_id, berand.name bname,mname,vname 
  from berand 
    left join (
     select   model.name mname, versions.name vname,model.berand_id 
     from model 
      left join  versions on versions.model_id = model.id
    ) m on berand.id = m.berand_id
) b on  b.type_id = type.id;

通过物化视图预计算关联结果，直接查询物化视图可大幅降低耗时，搭配触发器实现源表数据变更时自动刷新物化视图，完整实现代码如下：

-- 创建预关联结果物化视图
CREATE MATERIALIZED VIEW v_tbmv
AS 
SELECT id,name tname, bname, mname, vname FROM type left join(select  berand.type_id, berand.name bname,mname,vname from berand left join (select   model.name mname, versions.name vname,model.berand_id from model left join  versions on versions.model_id = model.id) m on berand.id = m.berand_id) b on  b.type_id = type.id;

-- 定义物化视图刷新函数
CREATE OR REPLACE FUNCTION fn_refresh_v_tbmv()
RETURNS trigger LANGUAGE plpgsql AS $$
BEGIN
    REFRESH MATERIALIZED VIEW v_tbmv;
    RETURN NULL;
END;
$$;

-- 为四张源表绑定数据变更触发规则
CREATE TRIGGER tg_refresh_v_tbmv_type AFTER INSERT OR UPDATE OR DELETE
ON type
FOR EACH STATEMENT EXECUTE PROCEDURE fn_refresh_v_tbmv();

CREATE TRIGGER tg_refresh_v_tbmv_berand AFTER INSERT OR UPDATE OR DELETE
ON berand
FOR EACH STATEMENT EXECUTE PROCEDURE fn_refresh_v_tbmv();

CREATE TRIGGER tg_refresh_v_tbmv_model AFTER INSERT OR UPDATE OR DELETE
ON model
FOR EACH STATEMENT EXECUTE PROCEDURE fn_refresh_v_tbmv();

CREATE TRIGGER tg_refresh_v_tbmv_versions AFTER INSERT OR UPDATE OR DELETE
ON versions -- *原代码误将该触发器绑定到type表，此处已修正*
FOR EACH STATEMENT EXECUTE PROCEDURE fn_refresh_v_tbmv();

补充提示：如果数据量较大、对刷新时效要求不高的场景，也可改为定时任务调度刷新，避免高频数据变更触发多次刷新影响写入性能。

内容的提问来源于stack exchange，提问作者yasdd dsfsdf