[英]MYSQL Query With Select Having Distinct with LIMIT Condition
目前,我有一個下表,我想提取ID為帶有Estimate_date的記錄,這些記錄被分配給update_index值的最小值,如果該值為null,我想去分配給下一個update_index的值,
例如,用於記錄
輸入表
+----+--------------+---------------------+
| id | update_index | estimated_date |
+----+--------------+---------------------+
| 73 | 1 | 2017-06-13 00:00:00 |
+----+--------------+---------------------+
| 73 | 2 | 2017-01-13 00:00:00 |
+----+--------------+---------------------+
| 73 | 3 | 2017-05-13 00:00:00 |
+----+--------------+---------------------+
| 73 | 4 | NULL |
+----+--------------+---------------------+
| 75 | 1 | 2017-01-01 00:00:00 |
+----+--------------+---------------------+
| 75 | 2 | NULL |
+----+--------------+---------------------+
| 75 | 3 | 2019-01-01 00:00:00 |
+----+--------------+---------------------+
| 76 | 1 | NULL |
+----+--------------+---------------------+
| 76 | 2 | 2018-06-01 00:00:00 |
+----+--------------+---------------------+
輸出表
+----+--------------+---------------------+
| id | update_index | estimated_date |
+----+--------------+---------------------+
| 73 | 1 | 2017-06-13 00:00:00 |
+----+--------------+---------------------+
| 75 | 1 | 2017-01-01 00:00:00 |
+----+--------------+---------------------+
| 76 | 2 | 2018-06-01 00:00:00 |
+----+--------------+---------------------+
我嘗試了以下值,但始終只能得到一個記錄,您能幫我嗎?
SELECT id,update_index,estimated_date
FROM tablename where estimated_date = (
SELECT DISTINCT estimated_date
FROM tablename
where estimated_date is not null
ORDER BY id, udpate_index ASC
LIMIT 1);
`
我將使用相關的子查詢:
select t.*
from tablename t
where t.update_index = (select t2.update_index
from tablename t2
where t2.id = t.id and t2.estimated_date is not null
order by t2.update_index asc
limit 1
);
請嘗試這個。
select t2.*
from
(select id, min(update_index) updated_index
from input_table
where estimated_date is not null
group by 1 )t1
left join input_table t2
on t1.id = t2.id
and t1.update_index = t2.update_index
您不能嘗試這樣:
從表名t選擇t.id,t.update_index,t.estimated_date,其中t.update_index =(從表名t1選擇MIN(update_index),其中t1.id = t.id並且t1.estimated_date不為空)按t.id分組
使用帶有LIMIT
子句的子查詢:
select t.*
from table t
where t.update_index = (select t2.update_index
from table t2
where t2.id = t.id and t2.estimated_date is not null
order by t2.estimated_date desc
limit 1
);
但是,您的樣本數據進一步提示了我:
select t.*
from table t
where t.estimated_date = (select max(t2.estimated_date)
from table t2
where t2.id = t.id and t2.estimated_date is not null
);
如果您與estimated_date
有聯系,則可以采用第一種方法。
聲明:本站的技術帖子網頁,遵循CC BY-SA 4.0協議,如果您需要轉載,請注明本站網址或者原文地址。任何問題請咨詢:yoyou2525@163.com.