I need to make a query over an SQL Server table but I don't know exactly how.
Consider this table (the real table is much more complex, Ord1 and Ord2 are dates that could be null, but i simplified it to this case):
Data of MyTable
ID MaqID Ord1 Ord2
------------------------
1 144 4 3
2 144 2 1
3 12 2 3
4 144 3 5
5 12 3 1
6 144 4 2
7 12 2 4
8 144 2 3
9 12 1 5
10 12 3 2
I need records for specific MaqID in Specific Order. I get that with this Query:
SELECT * FROM myTable WHERE MaqID=144 ORDER BY MaqID, Order1 DESC, Order2
Wich give me:
ID MaqID Ord1 Ord2
------------------------
6 144 4 2
1 144 4 3
4 144 3 5
2 144 2 1
8 144 2 3
Now, I need a single query that, for each MaqID, return the first ID for each subquery following above order. The result should be:
Expected result
MaqID ID
-----------
144 6
12 5
I have already try distinct conbination of TOP a MAX, but TOP result only one result and i need one for each MaqID, and for Max I have not field to maximize.
To sumarize: I need the first ID for each MaqID from a subquery in a specific order
Any ideas? Thanks!
You can do this using row_number():
select t.*
from (select t.*,
row_number() over (partition by macid Order1 DESC, Order2) as seqnum
from mytable t
) t
where seqnum = 1;
Related
My current T-SQL query provides the following results:
Query:
WITH CTE AS
(
SELECT SubscriberID, sum(valueMB) as ValuesMB
FROM dbo.InternetNetwork
GROUP BY SubscriberID
),
CTE2 AS (
SELECT ab.planID, a.SubscriberID, MAX(ValuesMB) as MaximumValue
FROM CTE AS a
left join
Subscriber as ab on a.SubscriberID= ab.SubscriberID
GROUP BY ab.planID, a.SubscriberID
)
select *
FROM CTE2 as b
ORDER BY b.MaximumValue desc
Output:
planID | SubscriberID | MaxValue
19 1555 97536.00
18 3528 97478.00
2 4029 93413.00
Query #2:
WITH CTE AS
(
SELECT SubscriberID, sum(valueMB) as ValuesMB
FROM dbo.InternetNetwork
GROUP BY SubscriberID
),
CTE2 AS(
SELECT ab.planID, MAX(ValuesMB) as MaximumValue
FROM CTE AS a
left join
Subscriber as ab on a.SubscriberID= ab.SubscriberID
GROUP BY ab.planID
)
SELECT pl.OperatorID, MAX(b.MaximumValue) as Super
FROM CTE2 as b
left join
Plan as pl on b.planID= pl.planID
GROUP BY pl.operatorID
ORDER BY pl.operatorID
Output #2:
OperatorID | Value
1 93413.00
2 86017.00
3 97536.00
I would like to also include a subscriberID, but I'm unable to figure out a way to do so, as the only way to do it, is including in the last SELECT and adding to GROUP BY, which when done, makes a mess of a result which is not accurate.
My desired output:
OperatorID | Value | SubscriberID
1 93413.00 4029
2 86017.00 164
3 97536.00 1544
internet network data:
SubscriberID ValuesMB
1 28
1 27
2 27
2 27
2 27
3 29
3 28
3 27
3 27
4 27
4 27
4 29
Subscriber Data:
SubscriberID PersonID PlanID
1 1 3
2 2 10
3 2 6
4 3 14
5 3 1
6 4 18
7 5 5
8 5 1
9 5 9
10 5 16
11 6 13
12 6 13
13 6 20
14 6 16
15 7 4
Plan data
PlanID OperatorID
1 1
2 1
3 2
4 2
5 2
6 2
7 2
8 2
9 2
10 2
11 2
12 3
13 3
14 3
15 3
16 3
17 3
18 3
19 3
20 3
The tables are somewhat like this related InternetNetwork-> Subscriber -> Plan. InternetNetwork contains how much each Subscribed has used. Each Subscriber has Plan associated with him. Each Plan contains a different Operator, there are only three. I wish to list all three operators, the data transferred by the subscriber of the plan that has the operator and Subscriber ID.
Window functions allow you to have fields in your select along with aggregate functions. You can do something like this
;WITH CTE AS
(
SELECT I.SubscriberID,
S.PlanID,
SUM(ValuesMB) OVER(PARTITION BY i.SubscriberID)as ValuesMB
FROM dbo.InternetNetwork I
JOIN Subscriber S
ON I.SubscriberID = S.SubscriberID
),
CTE2 AS
(
SELECT p.operatorID,
a.SubscriberID,
a.ValuesMB,
ROW_NUMBER() OVER(PARTITION BY p.operatorID ORDER BY a.ValuesMB DESC) as rn
FROM CTE a
join [Plan] P
on a.planID = P.planID
)
SELECT operatorID,
ValuesMB,
SubscriberID
FROM CTE2
where rn = 1
Hi I have a table that looks like this
StudentID
ParentID
Rank
1
11
1
1
15
5
1
16
6
2
21
1
2
22
2
3
31
1
3
37
7
3
38
8
4
41
1
4
42
2
So I want to pull only the top 2 ranks per student the out come will look like:
StudentID
ParentID
Rank
1
11
1
1
15
5
2
21
1
2
22
2
3
31
1
3
37
7
4
41
1
4
42
2
What would be the best way to do this? What makes it complicated is that every student has a parent ranked 1 but not every student has a parent ranked 2. What sql statement would I use to pull just the next ranked parent after 1?
The canonical solution is to use row_number():
select t.*
from (select t.*,
row_number() over (partition by studentid order by rank) as seqnum
from t
) t
where seqnum <= 2;
Under some circumstances, it might be faster to use:
select t.*
from t
where t.rank in (select top (2) t2.rank
from t t2
where t2.studentid = t.studentid
order by rank asc
) ;
I think this might have better performance if there were few students that had lots of ranks per student and you had an index on studentid, rank.
I am dealing with a huge volume of traffic data. I want to identify the vehicles which have changed their lanes, I'm Microsoft Access with VB.Net.
Traffic Data:
Vehicle_ID Lane_ID Frame_ID Distance
1 2 12 100
1 2 13 103
1 2 14 105
2 1 16 130
2 1 17 135
2 2 18 136
3 1 19 140
3 2 20 141
I have tried to distinct the Vehicle_ID and then count(distinct Lane_ID).
I could list the distinct Vehicle_ID but the it counts the total Lane_ID instead of Distinct Lane_ID.
SELECT
Distinct Vehicle_ID, count(Lane_ID)
FROM Table1
GROUP BY Vehicle_ID
Shown Result:
Vehicle_ID Lane Count
1 3
2 3
3 2
Correct Result:
Vehicle_ID Lane Count
1 1
2 2
3 2
Further to that i would like to get all Vehicle_ID who have changed their lane (all data including previous lane and new lane). Output result would be somehow like: Vehicle_ID Lane_ID Frame_ID Distance
2 1 17 135
2 2 18 136
3 1 19 140
3 2 20 141
Access does not support COUNT(DISTINCT columnname) so do this:
SELECT t.Vehicle_ID, COUNT(t.Lane_ID) AS [Lane Count]
FROM (
SELECT DISTINCT Vehicle_ID, Lane_ID FROM Table1
) AS t
GROUP BY t.Vehicle_ID
So
to identify the vehicles which have changed their lanes
you need to add to the above query:
HAVING COUNT(t.Lane_ID) > 1
SELECT
Table1.Vehicle_ID,
LANE_COUNT
FROM Table1
JOIN (
SELECT Vehicle_ID, COUNT(*) as LANE_COUNT FROM (
SELECT distinct Vehicle_ID, Lane_ID FROM Table1
) dTable1 # distinct vehicle and land id
GROUP BY Vehicle_ID # counting the distinct
) cTable1 ON cTable1.Vehicle_ID = Table1.Vehicle_ID # join the table with the counting
I think you should do one by one,
Distinct the vehicle id and land id
counting the distinct combination
and merge the result with the actual table.
If you want vehicles that have changed their lanes, then you can do:
SELECT Vehicle_ID,
IIF(MIN(Lane_ID) = MAX(Lane_ID), 0, 1) as change_lane_flag
FROM Table1
GROUP BY Vehicle_ID;
I think this is as good as counting the number of distinct lanes, because you are not counting actual "lane changes". So this would return "2" even though the vehicle changes lanes multiple times:
2 1 16 130
2 1 17 135
2 2 18 136
2 1 16 140
2 1 17 145
2 2 18 146
How can I select latest records in a a table with many-to-many relationship. The store_id,product_id is not a composite group key so they repeat many times.
id store_id product_id
1 1 1
2 2 1
3 1 1
4 3 1
5 2 1
6 3 1
The result should be like this:
id store_id product_id
3 1 1
5 2 1
6 3 1
For the sake of simplicity I only added 1 product but in the real case there are many products. So the result should be something like :
For each store_id show the latest added products_id.
e.g.
id store_id product_id
11 1 1
20 1 2
40 1 3
41 1 4
53 2 1
61 2 2
62 2 3
63 2 4
70 3 1
71 3 2
72 3 3
73 3 4
I don't see what this has to do with a pivot table. This seems like a basic aggregation:
select max(id) as id, store_id, product_id
from t
group by store_id, product_id
order by store_id, max(id);
I have a table which rank the items which i have.
I need a queries which will pick up only the top 2 ranks for a given item, the rank may not be in sequential order.
I need to fetch the item with least two ranks, there will same rank for two items as well.
Here is the snap shot of my table.
Item Id Supp Id Rank
1 2 2
1 1 7
1 7 5
1 9 11
2 67 4
2 9 14
2 10 14
2 34 4
2 25 3
2 60 3
2 79 5
my requirement is if I enter 2 i should get the result as below
Item Id Supp_id Rank
2 25 3
2 60 3
2 67 4
2 34 4
I am using oracle 10g version.
As one of the approaches it can be done as follows. Here we are using dense_rank() over() analytic function to assign a rank for a row in a ordered by rank group of rows .
select t.item_id
, t.supp_id
, t.rank
from (select item_id
, supp_id
, rank
, dense_rank() over(partition by item_id
order by rank) as rn
from t1
where item_id = 2
) t
where t.rn <= 2
Result:
ITEM_ID SUPP_ID RANK
---------- ---------- ----------
2 25 3
2 60 3
2 67 4
2 34 4
SQLFiddle Demo