关于mysql中数据存储复合树形结构，查询时结果按树形结构输出

1、主要思想：根据已有数据，规则性的造数据

select * FROM(
select lId,strName,lId as lParentId,-1 as orderIdx from tbClassify WHERE lParentId  = 0
UNION ALL
(select t1.* from tbClassify t1 join
(select lId from tbClassify where lParentId=0 order by orderIdx) t2 on
t1.lParentId = t2.lId
where 1=1 order by t1.lParentId,t1.orderIdx)
) tbLast where 1=1 GROUP BY tbLast.lParentId,tbLast.lId ORDER BY tbLast.lParentId,tbLast.orderIdx
;

上面的排序有问题，最后成型：

select lId FROM(
 select lId,orderIdx as pIdx,-1 as sIdx from tbClassify WHERE lParentId  = 0
  UNION ALL
 (select a.lId,b.orderIdx pIdx,a.orderIdx sIdx from tbClassify a left JOIN tbClassify b on (a.lParentId = b.lId ) where a.lParentId != 0)
)
tbLast ORDER BY tbLast.pIdx,tbLast.sIdx
;

该方法缺陷：前提必须知道树形结构共有几层，不带有通用型，当前方法只适用于两层树形结构，重点是提供了一种解决问题的思路：根据已有数据造数据