回答編集履歴
1
参考
test
CHANGED
|
@@ -1,3 +1,61 @@
|
|
|
1
1
|
creator_idを別テーブルに正規化すれば簡単に処理できますが
|
|
2
2
|
|
|
3
3
|
いまのままではSQLで処理するには非効率的すぎると思います
|
|
4
|
+
|
|
5
|
+
|
|
6
|
+
|
|
7
|
+
# sample
|
|
8
|
+
|
|
9
|
+
|
|
10
|
+
|
|
11
|
+
- 元テーブル
|
|
12
|
+
|
|
13
|
+
```SQL
|
|
14
|
+
|
|
15
|
+
create table tbl1(ID int unique,NAME varchar(10));
|
|
16
|
+
|
|
17
|
+
insert into tbl1 values(1,'TEST1'),(2,'TEST2'),(3,'TEST3');
|
|
18
|
+
|
|
19
|
+
create table tbl2(ID int unique,uid int,creator_id varchar(10),unique(uid,creator_id));
|
|
20
|
+
|
|
21
|
+
insert into tbl2 values(1,1,'A0001'),(2,2,'A0002'),(3,2,'A0003'),(4,3,'A0004');
|
|
22
|
+
|
|
23
|
+
create table tbl3(ID int unique,CODE varchar(10) unique,NAME varchar(10));
|
|
24
|
+
|
|
25
|
+
insert into tbl3 values(1,'A0001','テスト太郎'),(2,'A0002','テスト次郎'),(3,'A0003','テスト三郎'),(4,'A0004','テスト花子');
|
|
26
|
+
|
|
27
|
+
```
|
|
28
|
+
|
|
29
|
+
|
|
30
|
+
|
|
31
|
+
- 集計1
|
|
32
|
+
|
|
33
|
+
```SQL
|
|
34
|
+
|
|
35
|
+
select t1.ID,group_concat(distinct t.1NAME) as NAME,group_concat(creator_id) as creator_id
|
|
36
|
+
|
|
37
|
+
from tbl1 as t1
|
|
38
|
+
|
|
39
|
+
inner join tbl2 as t2 on t1.ID=t2.uid
|
|
40
|
+
|
|
41
|
+
group by ID;
|
|
42
|
+
|
|
43
|
+
```
|
|
44
|
+
|
|
45
|
+
|
|
46
|
+
|
|
47
|
+
- 集計2
|
|
48
|
+
|
|
49
|
+
```SQL
|
|
50
|
+
|
|
51
|
+
select t1.ID,group_concat(distinct t1.NAME) as NAME,group_concat(t3.NAME) as creator_name
|
|
52
|
+
|
|
53
|
+
from tbl1 as t1
|
|
54
|
+
|
|
55
|
+
inner join tbl2 as t2 on t1.ID=t2.uid
|
|
56
|
+
|
|
57
|
+
inner join tbl3 as t3 on t2.creator_id=t3.CODE
|
|
58
|
+
|
|
59
|
+
group by ID;
|
|
60
|
+
|
|
61
|
+
```
|