Original title: https://leetcode-cn.com/problems/consecutive-numbers/
Problem-solving directory https://blog.csdn.net/weixin_42845682/article/details/105196004
Article Directory
1. Title Description
Write a SQL query to find all numbers that appear at least three times in a row.
+----+-----+
| Id | Num |
+----+-----+
| 1 | 1 |
| 2 | 1 |
| 3 | 1 |
| 4 | 2 |
| 5 | 1 |
| 6 | 2 |
| 7 | 2 |
+----+-----+
来源：力扣（LeetCode）
链接：https://leetcode-cn.com/problems/consecutive-numbers
著作权归领扣网络所有。商业转载请联系官方授权，非商业转载请注明出处。
+-----------------+
| ConsecutiveNums |
+-----------------+
| 1 |
+-----------------+
来源：力扣（LeetCode）
链接：https://leetcode-cn.com/problems/consecutive-numbers
著作权归领扣网络所有。商业转载请联系官方授权，非商业转载请注明出处。
A few points to note:
1. It appears three times in a row, not three times.
2. If more than 3 times, no matter how many times, the number can only appear once. For example: 3 appears 6 times in a row, and there can only be 1 3 in the final result.
Second, the first answer-mysql
my answer:
select
distinct(n1) ConsecutiveNums /* 因为条件2，所以需要distinct */
from(
select
l1.id i1,
l1.num n1,
l2.id i2,
l2.num n2,
l3.id i3,
l3.num n3
from logs l1
join logs l2 on l1.id+1=l2.id
join logs l3 on l1.id+2=l3.id
) tmp
where n1 = n2
and n2 = n3
It should be noted here that where must be written after: n1 = n2
and n2 = n3. Instead of n1 = n2 = n3, why? It is recommended to refer to this article I wrote: In Mysql, what is the difference between 1 = 1 and 1 = 1 = 1 and -1 = -1 and -1 = -1 = -1 and 5 = 5 and 5 = 5 = 5 . The address is: https://blog.csdn.net/weixin_42845682/article/details/105262892
3. The second answer – hive
Its practical over () can also be done. But mysql does not have over (). For the sake of convenience, I did it with hive.
First of all, for the convenience of the diagram, enable the local mode of hive:
hive (test)> set hive.exec.mode.local.auto=true;
hive (test)> set hive.exec.mode.local.auto;
hive.exec.mode.local.auto=true
Insert the data and take a look at the query.
hive (test)> select * from logs;
OK
logs.id logs.num
1 1
2 1
3 1
4 2
5 1
6 2
7 2
Time taken: 0.047 seconds, Fetched: 7 row(s)
Use over () to query:
select
distinct(n1)
from(
select
id,
num n1,
lag(num,1) over( ) n2,
lag(num,2) over( ) n3
from logs
) t1
where n1 =n2
and n2=n3
Fourth, the third answer-hive expansion
Now it is required to appear 3 times in a row. If it is required to appear 5 times in a row, 10 times in a row? Do we have to write so many joins or so many lags?
No, you have to change the wording.
Next, the process is a bit complicated, try to read it. . .
First, this is the original data:
hive (test)> select * from logs;
OK
logs.id logs.num
1 1
2 1
3 1
4 2
5 1
6 2
7 2
Time taken: 0.027 seconds, Fetched: 7 row(s)
Tired, write again after a while.
