【原创】POSTGRESQL交叉表的实现

这里我来演示下在POSTGRESQL里面如何实现交叉表的展示，至于什么是交叉表，我就不多说了，度娘去哦。
原始表数据如下：

点击(此处)折叠或打开

1. t_girl=# select * from score;
   
2.  name | subject | score
   
3. -------+---------+-------
   
4.  Lucy | English | 100
   
5.  Lucy | Physics | 90
   
6.  Lucy | Math | 85
   
7.  Lily | English | 95
   
8.  Lily | Physics | 81
   
9.  Lily | Math | 84
   
10.  David | English | 100
    
11.  David | Physics | 86
    
12.  David | Math | 89
    
13.  Simon | English | 90
    
14.  Simon | Physics | 76
    
15.  Simon | Math | 79
    
16. (12 rows)
    
17. 
    
18. 
    
19. Time: 2.066 ms

想要实现以下的结果：
 

点击(此处)折叠或打开

1. name | English | Physics | Math
   
2. -------+---------+---------+------
   
3.  Simon | 90 | 76 | 79
   
4.  Lucy | 100 | 90 | 85
   
5.  Lily | 95 | 81 | 84
   
6.  David | 100 | 86 | 89

大致有以下几种方法：


1、用标准SQL展现出来

点击(此处)折叠或打开

1. t_girl=# select name,
   
2. t_girl-# sum(case when subject = 'English' then score else 0 end) as "English",
   
3. t_girl-# sum(case when subject = 'Physics' then score else 0 end) as "Physics",
   
4. t_girl-# sum(case when subject = 'Math' then score else 0 end) as "Math"
   
5. t_girl-# from score
   
6. t_girl-# group by name order by name desc;
   
7.  name | English | Physics | Math
   
8. -------+---------+---------+------
   
9.  Simon | 90 | 76 | 79
   
10.  Lucy | 100 | 90 | 85
    
11.  Lily | 95 | 81 | 84
    
12.  David | 100 | 86 | 89
    
13. (4 rows)
    
14. 
    
15. 
    
16. Time: 1.123 ms

2、用PostgreSQL 提供的第三方扩展 tablefunc 带来的函数实现
以下函数crosstab 里面的SQL必须有三个字段，name, 分类以及分类值来作为起始参数，必须以name,分类值作为输出参数。

点击(此处)折叠或打开

1. t_girl=# SELECT *
   
2. FROM crosstab('select name,subject,score from score order by name desc',$$values ('English'::text),('Physics'::text),('Math'::text)$$)
   
3. AS score(name text, English int, Physics int, Math int);
   
4.  name | english | physics | math
   
5. -------+---------+---------+------
   
6.  Simon | 90 | 76 | 79
   
7.  Lucy | 100 | 90 | 85
   
8.  Lily | 95 | 81 | 84
   
9.  David | 100 | 86 | 89
   
10. (4 rows)
    
11. 
    
12. 
    
13. Time: 2.059 ms

3、用PostgreSQL 自身的聚合函数实现

点击(此处)折叠或打开

1. t_girl=# select name,split_part(split_part(tmp,',',1),':',2) as "English",
   
2. t_girl-# split_part(split_part(tmp,',',2),':',2) as "Physics",
   
3. t_girl-# split_part(split_part(tmp,',',3),':',2) as "Math"
   
4. t_girl-# from
   
5. t_girl-# (
   
6. t_girl(# select name,string_agg(subject||':'||score,',') as tmp from score group by name order by name desc
   
7. t_girl(# ) as T;
   
8.  name | English | Physics | Math
   
9. -------+---------+---------+------
   
10.  Simon | 90 | 76 | 79
    
11.  Lucy | 100 | 90 | 85
    
12.  Lily | 95 | 81 | 84
    
13.  David | 100 | 86 | 89
    
14. (4 rows)
    
15. 
    
16. 
    
17. Time: 2.396 ms

4、 存储函数实现

点击(此处)折叠或打开

1. create or replace function func_ytt_crosstab_py ()
   
2. returns setof ytt_crosstab
   
3. as
   
4. $ytt$
   
5.   for row in plpy.cursor("select name,string_agg(subject||':'||score,',') as tmp from score group by name order by name desc"):
   
6.       a = row['tmp'].split(',')
   
7.       yield (row['name'],a[0].split(':')[1],a[1].split(':')[1],a[2].split(':')[1])
   
8. $ytt$ language plpythonu;
   
9. 
   
10. 
    
11. t_girl=# select name,english,physics,math from func_ytt_crosstab_py();
    
12.  name | english | physics | math
    
13. -------+---------+---------+------
    
14.  Simon | 90 | 76 | 79
    
15.  Lucy | 100 | 90 | 85
    
16.  Lily | 95 | 81 | 84
    
17.  David | 100 | 86 | 89
    
18. (4 rows)
    
19. 
    
20. 
    
21. Time: 2.687 ms

5、 用PLPGSQL来实现

点击(此处)折叠或打开

1. t_girl=# create type ytt_crosstab as (name text, English text, Physics text, Math text);
   
2. CREATE TYPE
   
3. Time: 22.518 ms
   
4. 
   
5. 
   
6. create or replace function func_ytt_crosstab ()
   
7. returns setof ytt_crosstab
   
8. as
   
9. $ytt$
   
10.   declare v_name text := '';
    
11.                 v_english text := '';
    
12. v_physics text := '';
    
13. v_math text := '';
    
14. v_tmp_result text := '';
    
15.   declare cs1 cursor for select name,string_agg(subject||':'||score,',') from score group by name order by name desc;
    
16. begin
    
17.   open cs1;
    
18.   loop
    
19.     fetch cs1 into v_name,v_tmp_result;
    
20.     exit when not found;
    
21.     v_english = split_part(split_part(v_tmp_result,',',1),':',2);
    
22.     v_physics = split_part(split_part(v_tmp_result,',',2),':',2);
    
23.     v_math = split_part(split_part(v_tmp_result,',',3),':',2);
    
24.     return query select v_name,v_english,v_physics,v_math;
    
25.   end loop;
    
26. end;
    
27. $ytt$ language plpgsql;
    
28. 
    
29. 
    
30. t_girl=# select name,English,Physics,Math from func_ytt_crosstab();
    
31.  name | english | physics | math
    
32. -------+---------+---------+------
    
33.  Simon | 90 | 76 | 79
    
34.  Lucy | 100 | 90 | 85
    
35.  Lily | 95 | 81 | 84
    
36.  David | 100 | 86 | 89
    
37. (4 rows)
    
38. 
    
39. 
    
40. Time: 2.127 ms