I have a table with the following columns.
ID (auto-inc)
When (datetime)
id1 (number)
id2 (number)
The combination of id1 and id2 can be unique or duplicated many times.
I need a query that returns the earliest record (by When) for each unique combination of id1+id2.
Example data:
ID
When
id1
id2
1
1-Jan-2020
4
5
2
1-Jan-2019
4
5
3
1-Jan-2021
4
5
4
1-Jan-2020
4
4
5
1-Jan-2019
4
4
6
1-Jan-2021
4
6
I need this to return rows 2, 5 and 6
I cannot figure out how to do this with an SQL query.
I have tried Group By on the concatenation of id1 & id2, and I have tried "Distinct id1, id2", but neither return the entire row of the record with the earliest When value.
If the result set can just return the ID that is fine also, I just need to know the rows that match these two requirements.
Okay, I had a few minutes to kill:
SELECT Data.* FROM Data WHERE ID IN (
SELECT TOP 1 ID FROM Data AS D
WHERE D.id1=Data.id1 AND D.id2=Data.id2 ORDER BY When);
or
SELECT Data.* FROM Data INNER JOIN (
SELECT id1, id2, Min(When) AS MW FROM Data
GROUP BY id1, id2) AS D
ON Data.When = D.MW AND Data.id1=D.id1 AND Data.id2=D.id2;
ID
When
id1
id2
2
1/1/2019
4
5
5
1/1/2019
4
4
6
1/1/2021
4
6
Related
I have a table with the following columns.
ID (auto-inc)
When (datetime)
id1 (number)
id2 (number)
The combination of id1 and id2 can be unique or duplicated many times.
I need a query that returns the earliest record (by When) for each unique combination of id1+id2.
Example data:
ID
When
id1
id2
1
1-Jan-2020
4
5
2
1-Jan-2019
4
5
3
1-Jan-2021
4
5
4
1-Jan-2020
4
4
5
1-Jan-2019
4
4
6
1-Jan-2021
4
6
I need this to return rows 2, 5 and 6
I cannot figure out how to do this with an SQL query.
I have tried Group By on the concatenation of id1 & id2, and I have tried "Distinct id1, id2", but neither return the entire row of the record with the earliest When value.
If the result set can just return the ID that is fine also, I just need to know the rows that match these two requirements.
Okay, I had a few minutes to kill:
SELECT Data.* FROM Data WHERE ID IN (
SELECT TOP 1 ID FROM Data AS D
WHERE D.id1=Data.id1 AND D.id2=Data.id2 ORDER BY When);
or
SELECT Data.* FROM Data INNER JOIN (
SELECT id1, id2, Min(When) AS MW FROM Data
GROUP BY id1, id2) AS D
ON Data.When = D.MW AND Data.id1=D.id1 AND Data.id2=D.id2;
ID
When
id1
id2
2
1/1/2019
4
5
5
1/1/2019
4
4
6
1/1/2021
4
6
Closed. This question needs details or clarity. It is not currently accepting answers.
Want to improve this question? Add details and clarify the problem by editing this post.
Closed 4 months ago.
Improve this question
I have 2 Tables.
Table 1 holds ID1 and ID2.
Table 2 holds ID2 and ID3.
Table 1 has unique cases for ID1 and multiple cases for ID2.
TABLE 1:
ID1 | ID2
1 1
2 2
3 3
4 3
5 4
6 5
7 5
8 6
9 7
10 6
Table 2 has unique cases for ID2 and multiple cases for ID3
TABLE 2:
ID2 | ID3
1 1
2 1
3 2
4 3
5 2
6 4
7 5
I want 1 unique case of ID3.
I need remove duplicate ID2s from Table 1 picking to remove the duplicate ID2s based on the smaller ID1
So Table 1 now looks like:
TABLE 1:
ID1 | ID2
1 1
2 2
4 3
5 4
7 5
9 7
10 6
Now I want to go to Table 2 and remove any duplicate ID3s based on the smaller ID2
TABLE 2:
ID2 | ID3
2 1
4 3
5 2
6 4
7 5
So my end result should be (I am joining the tables because both of them have other relevant information I need to combine but these are the IDs I am sorting and filtering to get the correct row):
Final Table:
ID1 | ID2 | ID3
2 2 1
7 5 2
5 4 3
10 6 4
9 7 5
Where now I have a single case for each ID3 based on the largest ID1 and ID2 associated with that ID3.
I have tried creating subqueries in the WHERE function to remove the duplicates but my understanding of SQL is not good enough to really figure out what is happening.
Group By and DISTINCT does not work for this case.
Decision Tree
I added a Decision Tree to help visualize the problem. Essentially, each ID3 can potentially have multiple ID2s, which can potentially have multiple ID1s.
I want to keep only the largest ID1, which gives me the correct ID2 associated with that ID3.
with t1 as (
select ID1, ID2
from
(
select *
,row_number() over(partition by ID2 order by ID1 desc) as rn
from t
) t
where rn = 1
),
t3 as (
select ID2, ID3
from
(
select *
,row_number() over(partition by ID3 order by ID2 desc) as rn
from t2
) t
where rn = 1
)
select t1.ID1
,t1.ID2
,t3.ID3
from t1 join t3 on t3.ID2 = t1.ID2
order by ID3
ID1
ID2
ID3
2
2
1
7
5
2
5
4
3
10
6
4
9
7
5
Fiddle
There is a table with values as below,
Id Value
1 1
2 1
3 2
4 2
5 3
6 4
7 4
now need to write a query to retrieve value from the table and output should look as
ID Value
1 1
3 2
5 3
6 4
any suggestion ?
The query you want is nothing to do with being distinct, it's a simple aggregation of value with the minimum ID for each:
select Min(id) Id, value
from table
group by value
I am trying to exclude rows where a value exists in another column of other row.
select * from TABLE1
ID1 ID2 VALUE
1 1 HIGH
2 2 MEDIUM
3 3 LOW
4 4 HIGH
5 4 HIGH
6 6 MEDIUM
All the data is coming from the same table what I want is to exclude ID1 = 4 because the value 4 exists in column ID2 in row 5. The final desired result is as follows:
ID1 ID2 VALUE
1 1 HIGH
2 2 MEDIUM
3 3 LOW
6 6 MEDIUM
I tried using a simple query such as:
Select * from TABLE1 Where ID1 = ID2
But this will wrongly also include row 4 as below since I need to exclude it because the value exists in another row but in ID2 column:
ID1 ID2 VALUE
1 1 HIGH
2 2 MEDIUM
3 3 LOW
4 4 HIGH
6 6 MEDIUM
You just have to add, this will exclude the records where you see more than 1 ids.
and id2 not in (Select id2 from table1 group by id2 having count(*) > 1)
Similarly add for id1 with OR
You can use the logic in the query below.
select * from t T1
Where 2 > (Select count(1) from t T2 where T2.id2 = T1.id2);
I want all the specific Ids which have a common other Id. I am sending the data through an user-defined table type.
CREATE TYPE rangeType AS TABLE (
ID2 int NOT NULL,
StartRange int NULL,
EndRange int NULL
);
The table is Like the following
ID1 ID2 Value
11 2 3
12 2 4
12 3 8.9
15 3 10
15 2 4
The value I will send will be of the form
DECLARE #temp_table rangeType
Insert INTO #temp_table values (2,4,10)
INSERT INTO #temp_table values (3,5,10)
So I want Output to be all those ID1's which have both the value of ID2 as 2 and 3 and the rows which have ID2 as 2 should have a value between 4 and 10 and all those rows which have ID2 as 3 should have a value between 5 and 10.
So my output, in this case, should be
ID1
12
15
as the ID1 12 and 15 maps both 2 and 3 and have the ranges between the specified respective ranges.
I Tried an inner join on the table followed by a BETWEEN operator. Which is giving me a correct value the operation which is performed is OR operation rather than an AND operation which I want.
You can use below query
SELECT ID1 FROM TABLE
WHERE ID2 IN (2,3)
AND
CASE WHEN ID2 = 2 AND VALUE >= 4 AND VALUE >=10 THEN 1
WHEN ID2 = 3 AND VALUE >= 5 AND VALUE >=10 THEN 1
ELSE 0
END = 1;