从Oracle表
我在Oracle中有两个用于表示各种树的表，这些表是：“PARTS_TREE_ENTRIES”，其中存储了所有节点（包括父节点和子节点），“PARTS_ITEMS”用于描述节点之间的关系
在表格TREE_ITEMS中，列COMPONENT_ID表示父亲，COMPONENT_ITEMID表示其子女
有多个树，并且所有树的节点都在同一个表中“TREE_ENTRIES”为了更容易理解，这是几棵树的表示：
这些是他们在表格中的条目：
正如您在表TREE_ITEMS中看到的那样，作为分支根的节点具有COMPONENT_ID的值“A”
我需要帮助来构建一个查询，以获取最后一级的所有节点及其父节点及其ID的列表，输出应类似于以下内容：
我已经阅读了“连接方式”这一条款，但我从未使用它，我不知道从哪里开始。
非常感谢你！
1 个答案:
答案 0 :(得分：0)
我意识到你通过查询明确询问了连接，但我认为我们可以练习使用Recursive With SQL，因为它在一些附加功能上做同样的工作。
根据您的样本结果，我最好的猜测是这样的。
我不清楚你是想要连接数据还是分隔列，所以对这两种情况都有建议。
with tree_view(tree_id, component_id, component_itemid, id, part_tree_id, componentid, name, lvl, tree_path, root_id) as (
select t.tree_id, t.component_id, t.component_itemid, t.id, e.part_tree_id, e.componentid, e.name, 1, e.name, component_itemid
from tree_items t,
tree_entries e
where t.component_itemid = e.componentid
and t.tree_id = e.part_tree_id
and t.component_id = 'A'
union all
select t.tree_id, t.component_id, t.component_itemid, t.id, e.part_tree_id, e.componentid, e.name, tv.lvl + 1, tv.tree_path || '=>' || e.name, tv.root_id
from tree_items t,
tree_entries e,
tree_view tv
where to_char(tv.component_itemid) = to_char(t.component_id)
and to_char(e.componentid) = to_char(t.component_itemid)
and tv.tree_id = t.tree_id
) -- end of hierarchy view
search depth first by lvl set order1
select tree_path,
name,
componentid,
regexp_substr(tree_path, '[[:alpha:]]+', 1, 1) lvl1_part,
regexp_substr(tree_path, '[[:alpha:]]+', 1, 2) lvl2_part,
regexp_substr(tree_path, '[[:alpha:]]+', 1, 3) lvl3_part -- add more if there are further levels down the tree
from (
select tree_id, component_id, component_itemid, id, part_tree_id, componentid, name, lvl, tree_path, root_id, order1,
case when lvl - lead(lvl) over (order by order1) < 0 then 0 else 1 end is_leaf
from tree_view
)
where is_leaf = 1;
以下是使用您提供的数据在Oracle上执行的示例：
Connected to Oracle Database 11g Enterprise Edition Release 11.2.0.4.0
Connected as sitja@orasitja
SQL>
SQL> col tree_path format a40
SQL> col lvl1_part format a20
SQL> col lvl2_part format a20
SQL> col lvl3_part format a20
SQL> drop table tree_entries;
Table dropped
SQL> create table tree_entries as
2 with tree_entries(part_tree_id, componentid, name) as (
3 select 1, 101, 'CLOCK' from dual union all
4 select 1, 102, 'WATCH' from dual union all
5 select 1, 105, 'BAND' from dual union all
6 select 1, 113, 'MATERIAL' from dual union all
7 select 1, 114, 'COLOR' from dual union all
8 select 1, 106, 'CASE' from dual union all
9 select 1, 115, 'MATERIAL' from dual union all
10 select 1, 116, 'SHAPE' from dual union all
11 select 1, 107, 'BEZEL' from dual union all
12 select 1, 117, 'MATERIAL' from dual union all
13 select 1, 118, 'TEXTURE' from dual union all
14 select 1, 108, 'FACE' from dual union all
15 select 1, 119, 'SHAPE' from dual union all
16 select 1, 120, 'DESIGN' from dual union all
17 select 2, 103, 'RELOJ' from dual union all
18 select 2, 104, 'RELOJPULSERA' from dual union all
19 select 2, 109, 'CORREA' from dual union all
20 select 2, 121, 'MATERIAL' from dual union all
21 select 2, 122, 'COLOR' from dual union all
22 select 2, 110, 'CAJA' from dual union all
23 select 2, 123, 'MATERIAL' from dual union all
24 select 2, 124, 'FORMA' from dual union all
25 select 2, 111, 'BISEL' from dual union all
26 select 2, 125, 'MATERIAL' from dual union all
27 select 2, 126, 'TEXTURA' from dual union all
28 select 2, 112, 'CARATULA' from dual union all
29 select 2, 127, 'FORMA' from dual union all
30 select 2, 128, 'DISEÑO' from dual
31 )
32 select * from tree_entries;
Table created
SQL> drop table tree_items;
Table dropped
SQL> create table tree_items as
2 with tree_items(tree_id, component_id, component_itemid, id) as (
3 select 1, 'A', 101, 1 from dual union all
4 select 1, 'A', 102, 2 from dual union all
5 select 1, '101', 107, 3 from dual union all
6 select 1, '101', 108, 4 from dual union all
7 select 1, '102', 105, 5 from dual union all
8 select 1, '102', 106, 6 from dual union all
9 select 1, '107', 117, 7 from dual union all
10 select 1, '107', 118, 8 from dual union all
11 select 1, '108', 119, 9 from dual union all
12 select 1, '108', 120, 10 from dual union all
13 select 1, '105', 113, 11 from dual union all
14 select 1, '105', 114, 12 from dual union all
15 select 1, '106', 115, 13 from dual union all
16 select 1, '106', 116, 14 from dual union all
17 select 2, 'A', 103, 15 from dual union all
18 select 2, 'A', 104, 26 from dual union all
19 select 2, '103', 111, 33 from dual union all
20 select 2, '103', 112, 42 from dual union all
21 select 2, '104', 109, 54 from dual union all
22 select 2, '104', 110, 62 from dual union all
23 select 2, '111', 125, 74 from dual union all
24 select 2, '111', 126, 82 from dual union all
25 select 2, '112', 127, 91 from dual union all
26 select 2, '112', 128, 10 from dual union all
27 select 2, '109', 127, 114 from dual union all
28 select 2, '109', 122, 122 from dual union all
29 select 2, '110', 123, 3334 from dual union all
30 select 2, '110', 124, 141 from dual
31 )
32 select * from tree_items;
Table created
SQL> with tree_view(tree_id, component_id, component_itemid, id, part_tree_id, componentid, name, lvl, tree_path, root_id) as (
2 select t.tree_id, t.component_id, t.component_itemid, t.id, e.part_tree_id, e.componentid, e.name, 1, e.name, component_itemid
3 from tree_items t,
4 tree_entries e
5 where t.component_itemid = e.componentid
6 and t.tree_id = e.part_tree_id
7 and t.component_id = 'A'
8 union all
9 select t.tree_id, t.component_id, t.component_itemid, t.id, e.part_tree_id, e.componentid, e.name, tv.lvl + 1, tv.tree_path || '=>' || e.name, tv.root_id
10 from tree_items t,
11 tree_entries e,
12 tree_view tv
13 where to_char(tv.component_itemid) = to_char(t.component_id)
14 and to_char(e.componentid) = to_char(t.component_itemid)
15 and tv.tree_id = t.tree_id
16 ) -- end of hierarchy view
17 search depth first by lvl set order1
18 select tree_path,
19 name,
20 componentid,
21 regexp_substr(tree_path, '[[:alpha:]]+', 1, 1) lvl1_part,
22 regexp_substr(tree_path, '[[:alpha:]]+', 1, 2) lvl2_part,
23 regexp_substr(tree_path, '[[:alpha:]]+', 1, 3) lvl3_part -- add more if there are further levels down the tree
24 from (
25 select tree_id, component_id, component_itemid, id, part_tree_id, componentid, name, lvl, tree_path, root_id, order1,
26 case when lvl - lead(lvl) over (order by order1) < 0 then 0 else 1 end is_leaf
27 from tree_view
28 )
29 where is_leaf = 1;
TREE_PATH NAME COMPONENTID LVL1_PART LVL2_PART LVL3_PART
---------------------------------------- ------------ ----------- -------------------- -------------------- --------------------
CLOCK=>BEZEL=>MATERIAL MATERIAL 117 CLOCK BEZEL MATERIAL
CLOCK=>BEZEL=>TEXTURE TEXTURE 118 CLOCK BEZEL TEXTURE
CLOCK=>FACE=>SHAPE SHAPE 119 CLOCK FACE SHAPE
CLOCK=>FACE=>DESIGN DESIGN 120 CLOCK FACE DESIGN
WATCH=>BAND=>MATERIAL MATERIAL 113 WATCH BAND MATERIAL
WATCH=>BAND=>COLOR COLOR 114 WATCH BAND COLOR
WATCH=>CASE=>MATERIAL MATERIAL 115 WATCH CASE MATERIAL
WATCH=>CASE=>SHAPE SHAPE 116 WATCH CASE SHAPE
RELOJ=>BISEL=>MATERIAL MATERIAL 125 RELOJ BISEL MATERIAL
RELOJ=>BISEL=>TEXTURA TEXTURA 126 RELOJ BISEL TEXTURA
RELOJ=>CARATULA=>FORMA FORMA 127 RELOJ CARATULA FORMA
RELOJ=>CARATULA=>DISEÑO DISEÑO 128 RELOJ CARATULA DISEÑO
RELOJPULSERA=>CORREA=>COLOR COLOR 122 RELOJPULSERA CORREA COLOR
RELOJPULSERA=>CORREA=>FORMA FORMA 127 RELOJPULSERA CORREA FORMA
RELOJPULSERA=>CAJA=>MATERIAL MATERIAL 123 RELOJPULSERA CAJA MATERIAL
RELOJPULSERA=>CAJA=>FORMA FORMA 124 RELOJPULSERA CAJA FORMA
16 rows selected
SQL>
相关问题
最新问题
- 我写了这段代码，但我无法理解我的错误
- 我无法从一个代码实例的列表中删除 None 值，但我可以在另一个实例中。为什么它适用于一个细分市场而不适用于另一个细分市场？
- 是否有可能使 loadstring 不可能等于打印？卢阿
- java中的random.expovariate()
- Appscript 通过会议在 Google 日历中发送电子邮件和创建活动
- 为什么我的 Onclick 箭头功能在 React 中不起作用？
- 在此代码中是否有使用“this”的替代方法？
- 在 SQL Server 和 PostgreSQL 上查询，我如何从第一个表获得第二个表的可视化
- 每千个数字得到
- 更新了城市边界 KML 文件的来源？