Oracle高级查询sql

Wed, 14 Oct 2026 00:00:00 +0000

`pivot`列转行

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11


WITH a AS
 (SELECT 'darcy' na, '语文' ke, 23 score FROM dual),
b AS
 (SELECT 'tom', '语文', 23 FROM dual),
c AS
 (SELECT 'darcy', '数学', 24 FROM dual),
d AS
 (SELECT * FROM a UNION ALL SELECT * FROM b UNION ALL SELECT * FROM c)
SELECT na, 语文, 数学
 FROM (SELECT na, ke, score FROM d)
PIVOT (SUM(score) FOR ke IN('语文' AS 语文, '数学' AS 数学));

`unpivot`行转列

1
2
3


WITH a AS
 (SELECT 'darcy' na, '90' 语文, '80' 数学, '87' 英语 FROM dual)
SELECT na, ke, score FROM a UNPIVOT(score for ke in(语文, 数学, 英语))

keep 聚合的基础上再取最值

1
2
3
4
5


SELECT 
tc_bmj03,tc_bmj04,MAX(tc_bmj01) KEEP(dense_rank LAST order BY tc_bmj09) max01
FROM tc_bmj_file
WHERE tc_bmj11 ='2'
GROUP BY tc_bmj03,tc_bmj04

over 分析

语法：rank()/dense_rank over(partition by A order by B)

over是一种分析函数，根据字段A对结果进行分区，在各分区内按照字段进行排序；

over不能单独使用，需要与row_number()，rank()和dense_rank，lag()和lead(),sum()等配合使用

说明：

over()在什么条件之上
partition by 按哪个字段划分组
order by 按哪个字段排序注意：
使用rank()/dense_rank() 时，必须要带order by否则非法
rank(): 跳跃排序，如果有两个第一级时，接下来就是第三级
dense_rank(): 连续排序，如果有两个第一级时，接下来仍然是第二级

1
2
3
4
5
6
7
8
9


SELECT fan01,
 fan03,
 fan04,
 fan15,
 dense_rank() OVER(PARTITION BY fan01 ORDER BY fan03 DESC, fan04 DESC) r
 FROM fan_file
 WHERE fan041 IN ('0', '1')
 AND fan05 IN ('1', '2')
 AND fan01 = 'FLYG200001'

lag 偏移函数

可以查询上几条数据,下几条数据,适合查上下站资料

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11


SELECT fan01,
 fan03,
 fan04,
 fan15,
			 LAG(fan03||fan04) OVER (ORDER BY fan03 DESC ,fan04 DESC) a1,
			 LAG(fan03||fan04,2,0) OVER (ORDER BY fan03,fan04) a2
 FROM fan_file
 WHERE fan041 IN ('0', '1')
 AND fan05 IN ('1', '2')
 AND fan01 = 'FLYG200001'
	 ORDER BY fan03,fan04

`instr` 字符串匹配函数

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11


SELECT 
 INSTR('123456789','2'), -- 返回：2
 INSTR('123456654321','66'), -- 返回：6
 INSTR('123456654321','66',6), -- 返回：6
 INSTR('123456654321','66',8), -- 返回：0
 INSTR('11223344','2',1,2), -- 返回：4
 INSTR('11223344','2',2,3), -- 返回：0
 INSTR('11223344','2',-1,2), -- 返回：3
 INSTR('11223344','2',-6,1), -- 返回：3
 INSTR('11223344','2',-6,2) -- 返回：0
 FROM DUAL ;

istagg 聚合为一条数据

1
2


LISTAGG(字段名,',') WITHIN GROUP( ORDER BY 字段名)
listagg(字段名,',') within GROUP (order by 字段名) over (partition by 字段名) rank

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13


with temp as(
select 500 population, '中国' nation ,'江苏' city from dual union all
select 1500 population, '中国' nation ,'上海' city from dual union all
select 500 population, '中国' nation ,'北京' city from dual union all
select 1000 population, '美国' nation ,'纽约' city from dual union all
select 500 population, '美国' nation ,'波士顿' city from dual union all
select 500 population, '日本' nation ,'东京' city from dual
)
select population,
nation,
city,
listagg(city,',') within GROUP (order by city) over (partition by nation) rank
from temp

递归

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11


SELECT LEVEL,
 CONNECT_BY_ISLEAF 是否无下阶料,
 bmb01,
 bmb03,
 SYS_CONNECT_BY_PATH(bmb01, '/') 路径,
 CONNECT_BY_ROOT(bmb01) root
 FROM bmb_file
 WHERE bmb04 <= SYSDATE
 AND (bmb05 IS NULL OR bmb05 > SYSDATE)
 START WITH bmb01 LIKE 'AA0011A3CR' --从这个资料开始
CONNECT BY PRIOR bmb03 = bmb01 --下笔资料的bmb03 和当前bmb01一样

Pivot on darcyjoven