如何设计产品分类的表结构才是合理

父子树形结构大部分设置为  id ,name ,parentid  这个好结构比较清晰
而且现在树形递归查询算法比较多容易查询树形显示
--测试数据
DECLARE @t TABLE(ID char(3),PID char(3),Name nvarchar(10))
INSERT @t SELECT '001',NULL ,'山东省'
UNION ALL SELECT '002','001','烟台市'
UNION ALL SELECT '004','002','招远市'
UNION ALL SELECT '003','001','青岛市'
UNION ALL SELECT '005',NULL ,'四会市'
UNION ALL SELECT '006','005','清远市'
UNION ALL SELECT '007','006','小分市'--深度排序显示处理
--生成每个节点的编码累计(相同当单编号法的编码)
DECLARE @t_Level TABLE(ID char(3),Level int,Sort varchar(8000))
DECLARE @Level int
SET @Level=0
INSERT @t_Level SELECT ID,@Level,ID
FROM @t
WHERE PID IS NULL
WHILE @@ROWCOUNT>0
BEGIN
SET @Level=@Level+1
INSERT @t_Level SELECT a.ID,@Level,b.Sort+a.ID
FROM @t a,@t_Level b
WHERE a.PID=b.ID
AND b.Level=@Level-1
END--显示结果
SELECT SPACE(b.Level*2)+'|--'+a.Name
FROM @t a,@t_Level b
WHERE a.ID=b.ID
ORDER BY b.Sort
/*--结果
|--山东省
  |--烟台市
    |--招远市
  |--青岛市
|--四会市
  |--清远市
    |--小分市
--*/
--2005的方法
declare @T table (ID int,pid int,NAME varchar(6))
insert into @T
select 1,0,'上衣' union all
select 2,0,'鞋子' union all
select 3,0,'裤子' union all
select 4,1,'毛衣' union all
select 5,1,'衬衫' union all
select 6,2,'球鞋' union all
select 7,2,'皮鞋' union all
select 8,3,'西裤' union all
select 9,3,'筒裤' union all
select 10,4,'羊毛衣' union all
select 11,4,'牛毛衣' union all
select 12,5,'白衬衫' union all
select 13,5,'黑衬衫'
;with
depts as(    select * from @T
    where ID = 1
    union all
    select a.*
    from @T a, depts b
    where a.pid = b.ID
)
select * from depts
/*
ID          pid         NAME
----------- ----------- ------
1           0           上衣
4           1           毛衣
5           1           衬衫
12          5           白衬衫
13          5           黑衬衫
10          4           羊毛衣
11          4           牛毛衣
*/
------------------------------------------------------
CREATE TABLE [dbo].[levelTable](
[ID] [int] IDENTITY(1,1) NOT NULL,
[parentID] [int] NOT NULL CONSTRAINT [DF_levelTable_parentID]  DEFAULT ((0)),
[name] [nvarchar](50) NULL
)
insert into levelTable(parentID,[NAME]) values(0,'开发部')
insert into levelTable(parentID,[NAME]) values(0,'客户服务部')
insert into levelTable(parentID,[NAME]) values(0,'行政部')
insert into levelTable(parentID,[NAME]) values(1,'开发一部')
insert into levelTable(parentID,[NAME]) values(1,'开发二部')
insert into levelTable(parentID,[NAME]) values(2,'后勤服务部')
insert into levelTable(parentID,[NAME]) values(2,'大厅服务部')
insert into levelTable(parentID,[NAME]) values(3,'总裁部')
insert into levelTable(parentID,[NAME]) values(3,'人力资源部')
insert into levelTable(parentID,[NAME]) values(9,'员工管理部')
insert into levelTable(parentID,[NAME]) values(9,'人员招聘部')
insert into levelTable(parentID,[NAME]) values(5,'开发一部项目A部') go
DECLARE @t_Level TABLE(ID char(3),Level int,Sort varchar(100))
DECLARE @Level int
SET @Level=0
INSERT @t_Level SELECT id,@Level,right('1000'+ltrim(ID),3)
FROM levelTable
WHERE parentID =0 -------modify
WHILE @@ROWCOUNT>0
BEGIN
SET @Level=@Level+1
INSERT @t_Level SELECT a.ID,@Level,ltrim(b.Sort)+right('1000'+ltrim(a.ID),3)
FROM levelTable a,@t_Level b
WHERE a.parentID=b.ID
AND b.Level=@Level-1
END
--显示结果
SELECT a.*
FROM levelTable a,@t_Level b
WHERE a.ID=b.ID
ORDER BY b.Sort
/*
ID          parentID    name
----------- ----------- --------------------------------------------------
1           0           开发部
4           1           开发一部
5           1           开发二部
12          5           开发一部项目A部
2           0           客户服务部
6           2           后勤服务部
7           2           大厅服务部
3           0           行政部
8           3           总裁部
9           3           人力资源部
10          9           员工管理部
11          9           人员招聘部*/查父节点子节点方法--测试数据
if OBJECT_ID('tb') is not null
drop table tb
go
CREATE TABLE tb(ID char(3),PID char(3),Name nvarchar(10))
INSERT tb SELECT '001',NULL ,'山东省'
UNION ALL SELECT '002','001','烟台市'
UNION ALL SELECT '004','002','招远市'
UNION ALL SELECT '003','001','青岛市'
UNION ALL SELECT '005',NULL ,'四会市'
UNION ALL SELECT '006','005','清远市'
UNION ALL SELECT '007','006','小分市'
GO
--2000的方法--查询指定节点及其所有子节点的函数
CREATE FUNCTION f_Cid(@ID char(3))
RETURNS @t_Level TABLE(ID char(3),Level int)
AS
BEGIN
declare @Level int
set @level=1
insert @t_level select @id,@level
while @@rowcount>0
begin
    set @level=@level+1
    insert @t_Level select tb.id,@level
     from tb join @t_level t on tb.pid=t.id
     where t.level+1=@level
   end
   return
endselect tb.*
from tb join dbo.f_cid('002') b
on tb.ID=b.id
/*
ID   PID  Name
---- ---- ----------
002  001  烟台市
004  002  招远市*/
go
--2005的方法（CTE）declare @n varchar(10)
set @n='002'
;with
jidian as
(
  select * from tb where ID=@n
  union all
  select t.* from jidian j join tb t on j.ID=t.PID
)
select * from jidian
go
/*
ID   PID  Name
---- ---- ----------
002  001  烟台市
004  002  招远市
*/
go
--查找指定节点的所有父节点(标准树形,即一个子节点只有一个父节点)
CREATE FUNCTION f_Pid(@ID char(3))
RETURNS @t_Level TABLE(ID char(3))
AS
BEGIN
INSERT @t_Level SELECT @ID
SELECT @ID=PID FROM tb
WHERE ID=@ID
AND PID IS NOT NULL
WHILE @@ROWCOUNT>0
BEGIN
INSERT @t_Level SELECT @ID
SELECT @ID=PID FROM tb
WHERE ID=@ID
AND PID IS NOT NULL
END
RETURN
END
select tb.*
from tb join dbo.f_Pid('004') b
on tb.ID=b.id
/*
ID   PID  Name
---- ---- ----------
001  NULL 山东省
002  001  烟台市
004  002  招远市
*/
go
--2005的方法
declare @n varchar(10)
set @n='004'
;with
fujidian as
(
select * from tb where ID=@n and PID is not null
union all
select a.* from tb a join fujidian f on a.ID=f.PID
)
select * from fujidian order by ID
/*
ID   PID  Name
---- ---- ----------
001  NULL 山东省
002  001  烟台市
004  002  招远市
*/
父子树形结构还是比较好的是现在用得比较多的一种方法
BOM结构用得多
可以参考下面的一些代码:
CREATE TABLE BOM(PID INT,ID INT)
INSERT INTO BOM SELECT 801,101
INSERT INTO BOM SELECT 801,102
INSERT INTO BOM SELECT 801,103
INSERT INTO BOM SELECT 801,601
INSERT INTO BOM SELECT 601,101
INSERT INTO BOM SELECT 601,105
INSERT INTO BOM SELECT 601,501
INSERT INTO BOM SELECT 501,106
INSERT INTO BOM SELECT 501,121
GOCREATE FUNCTION F_GETROOT(@PID INT)
RETURNS INT
AS
BEGIN
    DECLARE @ID INT
    WHILE EXISTS(SELECT 1 FROM BOM WHERE ID=@PID)
    BEGIN
        SET @ID=@PID
        SELECT @PID=PID FROM BOM WHERE ID=@ID
    END
    RETURN @PID
END
GOSELECT PID=DBO.F_GETROOT(PID),ID FROM BOM
GO/*
PID         ID
----------- -----------
801         101
801         102
801         103
801         601
801         101
801         105
801         501
801         106
801         121
*/
DROP FUNCTION F_GETROOT
DROP TABLE BOM
GO
--生成测试数据
create table BOM_1(Item int,bom_head varchar(20),bom_child varchar(20),number int,products_attribute  varchar(20))
insert into BOM_1 select 1 ,'A' ,'A1',1,'采购'
insert into BOM_1 select 2 ,'A' ,'A2',2,'生产'
insert into BOM_1 select 3 ,'A2','A3',3,'生产'
insert into BOM_1 select 4 ,'A2','A4',2,'采购'
insert into BOM_1 select 5 ,'A3','A5',2,'采购'
insert into BOM_1 select 6 ,'A3','A6',1,'采购'
insert into BOM_1 select 7 ,'B' ,'B1',1,'采购'
insert into BOM_1 select 8 ,'B' ,'B2',2,'生产'
insert into BOM_1 select 9 ,'B2','B3',3,'生产'
insert into BOM_1 select 10,'B2','B4',2,'采购'
insert into BOM_1 select 11,'B3','B5',2,'采购'
insert into BOM_1 select 12,'B3','B6',2,'采购'
go
   --创建用户定义函数，用于取每个父节点下子节点的采购配置信息
create function f_stock(@bom_head varchar(20))
returns @t table(bom varchar(20),number int)
as
begin
    declare @level int
    declare @a table(bom varchar(20),number int,products_attribute varchar(20),[level] int)
    set @level=1    if exists(select 1 from BOM_1 where bom_head=@bom_head)
    insert into @a
    select bom_child,number,products_attribute,@level
    from BOM_1
    where bom_head=@bom_head

    while exists(select 1 from @a where [level]=@level and products_attribute='生产')
    begin
        set @level=@level+1
        insert into @a(bom,number,products_attribute,[level])
        select a.bom_child,a.number,a.products_attribute,@level
        from BOM_1 a,@a b
        where a.bom_head=b.bom and b.[level]=@level-1
    end

    insert into @t(bom,number) select bom,number from @a where products_attribute='采购'
    return
end
go
--执行调用，取父节点'A'一个标准配置分解的采购信息及数量
select * from dbo.f_stock('A')
--生成测试数据
create table BOM(ID INT,PID INT,MSG VARCHAR(1000))
insert into BOM select 1,0,NULL
insert into BOM select 2,1,NULL
insert into BOM select 3,1,NULL
insert into BOM select 4,2,NULL
insert into BOM select 5,3,NULL
insert into BOM select 6,5,NULL
insert into BOM select 7,6,NULL
go--创建用户定义函数用于取每个父节点下子节点的采购配置信息
create function f_getChild(@ID VARCHAR(10))
returns @t table(ID VARCHAR(10),PID VARCHAR(10),Level INT)
as
begin
    declare @i int
    set @i = 1
    insert into @t select ID,PID,@i from BOM where PID = @ID

    while @@rowcount<>0
    begin
        set @i = @i + 1

        insert into @t
        select
            a.ID,a.PID,@i
        from
            BOM a,@t b
        where
            a.PID=b.ID and b.Level = @i-1
    end
    return
end
go--执行查询
select ID from dbo.f_getChild(3)
go--输出结果
/*
ID
----
5
6
7
*/--删除测试数据
drop function f_getChild
drop table BOM创建用户定义函数,每个子节点de父节点的信息
--生成测试数据
create table BOM(ID int,parentID int,sClassName varchar(10))
insert into BOM values(1,0,'1111'      )
insert into BOM values(2,1,'1111_1'    )
insert into BOM values(3,2,'1111-1-1'  )
insert into BOM values(4,3,'1111-1-1-1')
insert into BOM values(5,1,'1111-2'    )go--创建用户定义函数,每个子节点de父节点的信息
create function f_getParent(@ID int)
returns varchar(40)
as
begin
    declare @ret varchar(40)    while exists(select 1 from BOM where ID=@ID and parentID<>0)
    begin
        select @ID=b.ID,@ret=','+rtrim(b.ID)+isnull(@ret,'')
        from
            BOM a,BOM b
        where
            a.ID=@ID and b.ID=a.parentID
    end

    set @ret=stuff(@ret,1,1,'')
    return @ret
end
go--执行查询
select ID,isnull(dbo.f_getParent(ID),'') as parentID from BOM
go--输出结果
/*
ID          parentID
----------- ----------------------------------------
1
2           1
3           1,2
4           1,2,3
5           1
*/--删除测试数据
drop function f_getParent
drop table BOM
go
父子树形结构还有一种巧妙设置为  id ,name ,parentid , IDList  查询超级方便IDList存放的是所有父级的ID列表，如,1（爷爷级）,3（父级）,4(自己),
这样爷爷级要查自己的子孙，只要跟IDList查询含有,1,的就可以

调试易

如何设计产品分类的表结构才是合理

解决方案 »