select multiple count set, sql server 2008 r2 - sql

Using SQL Server 2008 R2, i have following result set returned by a query -
QID QcID QtID QsID
21 1 SC 3
4 1 SC 1
8 1 MC 1
2 1 SC 1
23 1 SC 3
24 1 SC 3
5 1 SC 1
22 1 SC 3
1 1 SC 1
29 1 MC 3
10 1 MC 1
30 1 MC 3
26 1 MC 3
25 1 SC 3
6 1 MC 1
27 1 MC 3
7 1 MC 1
3 1 SC 1
28 1 MC 3
9 1 MC 1
Now i want to find a random set of 15 QID, which must includes say -
9 QsID having QsID = 1
6 QsID having QsID = 3
9 QtID having QtID = SC
6 QtID having QtID = MC
15 QsID having QtID = 1
how it can be done keeping performance in mind as it may have tens of thousand records.
# Damien_The_Unbeliever the expected output could be -
21 1 SC 3
4 1 SC 1
8 1 MC 1
2 1 SC 1
23 1 SC 3
24 1 SC 3
5 1 SC 1
1 1 SC 1
10 1 MC 1
25 1 SC 3
6 1 MC 1
27 1 MC 3
7 1 MC 1
3 1 SC 1
28 1 MC 3
Also forget the random, how it be possible to select the set satisfying all the conditions.


EDIT2:
Well, how about using a stored procedure for that?
Assumption is that you have 4 sets of conditions you sample data with. Deduced from the output you provided. It might not be right but then again you can adjust it as you like.
Parameters:
#SIZE - is the size of sampling result set in rows
#P1-#P3 - percent of the sampling result set that should be filled with random rows for particular set of conditions.
#P4=#SIZE-(#N1+#N2+#N3)
CREATE PROCEDURE sqlsampling
#SIZE INT, #P1 DECIMAL(6,4), #P2 DECIMAL(6,4), #P3 DECIMAL(6,4)
AS
DECLARE #N1 INT, #N2 INT, #N3 INT, #N4 INT;
SET #N1=CEILING(#SIZE*#P1*0.01);
SET #N2=CEILING(#SIZE*#P2*0.01);
SET #N3=CEILING(#SIZE*#P3*0.01);
SET #N4=#SIZE-(#N1+#N2+#N3);
CREATE TABLE #sample(QID INT, QcID INT, QtID CHAR(2), QsID INT);
INSERT INTO #sample
SELECT TOP(#N1) * FROM mytable
WHERE QtID = 'MC' AND QsID = 1
ORDER BY CHECKSUM(NEWID());
INSERT INTO #sample
SELECT TOP(#N2) * FROM mytable
WHERE QtID = 'MC' AND QsID = 3 AND QID NOT IN(SELECT QID FROM #sample)
ORDER BY CHECKSUM(NEWID());
INSERT INTO #sample
SELECT TOP(#N3) * FROM mytable
WHERE QtID = 'SC' AND QsID = 1 AND QID NOT IN(SELECT QID FROM #sample)
ORDER BY CHECKSUM(NEWID());
INSERT INTO #sample
SELECT TOP(#N4) * FROM mytable
WHERE QtID = 'SC' AND QsID = 3 AND QID NOT IN(SELECT QID FROM #sample)
ORDER BY CHECKSUM(NEWID());
SELECT * FROM #sample;
DROP TABLE #sample;
GO
If we execute it on your sample data like this
EXEC sqlsampling #SIZE=15, #P1=26.666, #P2=13.333, #P3=33.333;
It will give us the output:
QID QCID QTID QSID
10 1 MC 1
9 1 MC 1
7 1 MC 1
6 1 MC 1
27 1 MC 3
30 1 MC 3
1 1 SC 1
4 1 SC 1
3 1 SC 1
5 1 SC 1
2 1 SC 1
21 1 SC 3
23 1 SC 3
25 1 SC 3
24 1 SC 3
Other considerations:
Correct indexes for your sets of conditions should help
CHECKSUM(NEWID()) helps us a bit with the penalty of using NEWID()
Original answer:
You can do something like this:
SELECT TOP 15 * FROM
(
SELECT * FROM (SELECT TOP 9 QID FROM mytable
WHERE QsID = 1
ORDER BY CHECKSUM(NEWID())) a
UNION
SELECT * FROM (SELECT TOP 6 QID FROM mytable
WHERE QsID = 3
ORDER BY CHECKSUM(NEWID())) b
...
) z ORDER BY CHECKSUM(NEWID())


Perhaps you may try percent with top.. this is not the complete answer, but to throw some light into the direction..
select * from demo where qid in
(select top 40 percent qid
from demo order by newid())
;
Here is also a Reference to tablesample: http://msdn.microsoft.com/en-us/library/ms189108.aspx

Related

SQL: subset data: select id when time_id for id satisfy a condition from another column

I have a data (dt) in SQL like the following:
ID time_id act rd
11 1 1 1
11 2 4 1
11 3 7 0
12 1 8 1
12 2 2 0
12 3 4 1
12 4 3 1
12 5 4 1
13 1 4 1
13 2 1 0
15 1 3 1
16 1 8 0
16 2 8 0
16 3 8 0
16 4 8 0
16 5 8 0
and I want to take the subset of this data such that only ids (and their corresponding time_id, act, rd) that has time_id == 5 is retained. The desired output is the following
ID time_id act rd
12 1 8 1
12 2 2 0
12 3 4 1
12 4 3 1
12 5 4 1
16 1 8 0
16 2 8 0
16 3 8 0
16 4 8 0
16 5 8 0
I know I should use having clause somehow but have not been successful so far (returns me empty outputs). below is my attempt:
SELECT * FROM dt
GROUP BY ID
Having min(time_id) == 5;

This query:
select id from tablename where time_id = 5
returns all the ids that you want in the results.
Use it with the operator IN:
select *
from tablename
where id in (select id from tablename where time_id = 5)

You can use a correlated subquery with exists:
select t.*
from t
where exists (select 1 from t t2 where t2.id = t.id and t2.time_id = 5);

WITH temp AS
(
SELECT id FROM tab WHERE time_id = 5
)
SELECT * FROM tab t join temp tp on(t.id=tp.id);

check this query
select * from table t1 join (select distinct ID from table t where time_id = 5) t2 on t1.id =t2.id;

how to reset the column value

i have table with 9 records that 2 records with _id = 1 was deleted:
id | name | index | _id
1 a 1 1
8 b 3 1
9 c 7 1
10 d 4 1
15 e 2 1
16 d 1 2
17 e 2 2
and I want to reset the index of _id = 1 like this:
id | name | index | _id
1 a 1 1
8 b 2 1
9 c 3 1
10 d 4 1
15 e 5 1
16 d 1 2
17 e 2 2
i 'd use this query
declare #_idCount int = (select count(*) from tbl where _id = 1),
#index int = 1
while(#_idCount > 0)
begin
update tbl
set code = #index
where _id = 1
set #index = #index + 1
set #picCount = #picCount - 1
end
but this code is set all of record with same code.
what can i do to solve it?

You can simply use ROW_NUMBER:
WITH CTE AS
(
SELECT id,
name,
[index],
[_id],
new_index = ROW_NUMBER() OVER(PARTITION BY [_id]
ORDER BY [id])
FROM dbo.tbl
)
UPDATE CTE
SET [index] = new_index;

SQL Recursive CTE unexpectedly returns alternating sets

I am trying to get the use recursive CTE to repeat the same pattern over and over, resetting when "Scenario" increases in value. RowNumber repeats 1-21 (as desired), but whenever "Scenario" is an even number, there are too few items in the "Vals" column to feed into "Value". I can't figure out which part of the code is causing me to be 1 short for only even Scenarios.
Below are the results of the code I'm using at the bottom.
Scenario RowNumber Value Vals
1 1 A A,A,A,A,A,A,A,A,A,A,A,A,A,A,A,A,A,A,B,C
1 2 A A,A,A,A,A,A,A,A,A,A,A,A,A,A,A,A,A,B,C
1 3 A A,A,A,A,A,A,A,A,A,A,A,A,A,A,A,A,B,C
1 4 A A,A,A,A,A,A,A,A,A,A,A,A,A,A,A,B,C
1 5 A A,A,A,A,A,A,A,A,A,A,A,A,A,A,B,C
1 6 A A,A,A,A,A,A,A,A,A,A,A,A,A,B,C
1 7 A A,A,A,A,A,A,A,A,A,A,A,A,B,C
1 8 A A,A,A,A,A,A,A,A,A,A,A,B,C
1 9 A A,A,A,A,A,A,A,A,A,A,B,C
1 10 A A,A,A,A,A,A,A,A,A,B,C
1 11 A A,A,A,A,A,A,A,A,B,C
1 12 A A,A,A,A,A,A,A,B,C
1 13 A A,A,A,A,A,A,B,C
1 14 A A,A,A,A,A,B,C
1 15 A A,A,A,A,B,C
1 16 A A,A,A,B,C
1 17 A A,A,B,C
1 18 A A,B,C
1 19 A B,C
1 20 B C
1 21 C
2 1 A A,A,A,A,A,A,A,A,A,A,A,A,A,A,A,A,B,B,C
2 2 A A,A,A,A,A,A,A,A,A,A,A,A,A,A,A,B,B,C
2 3 A A,A,A,A,A,A,A,A,A,A,A,A,A,A,B,B,C
2 4 A A,A,A,A,A,A,A,A,A,A,A,A,A,B,B,C
2 5 A A,A,A,A,A,A,A,A,A,A,A,A,B,B,C
2 6 A A,A,A,A,A,A,A,A,A,A,A,B,B,C
2 7 A A,A,A,A,A,A,A,A,A,A,B,B,C
2 8 A A,A,A,A,A,A,A,A,A,B,B,C
2 9 A A,A,A,A,A,A,A,A,B,B,C
2 10 A A,A,A,A,A,A,A,B,B,C
2 11 A A,A,A,A,A,A,B,B,C
2 12 A A,A,A,A,A,B,B,C
2 13 A A,A,A,A,B,B,C
2 14 A A,A,A,B,B,C
2 15 A A,A,B,B,C
2 16 A A,B,B,C
2 17 A B,B,C
2 18 B B,C
2 19 B C
2 20 C
2 21 A A,A,A,A,A,A,A,A,A,A,A,A,A,A,A,A,A,B,B,C
3 1 A A,A,A,A,A,A,A,A,A,A,A,A,A,A,A,A,A,B,C,C
3 2 A A,A,A,A,A,A,A,A,A,A,A,A,A,A,A,A,B,C,C
3 3 A A,A,A,A,A,A,A,A,A,A,A,A,A,A,A,B,C,C
3 4 A A,A,A,A,A,A,A,A,A,A,A,A,A,A,B,C,C
3 5 A A,A,A,A,A,A,A,A,A,A,A,A,A,B,C,C
3 6 A A,A,A,A,A,A,A,A,A,A,A,A,B,C,C
3 7 A A,A,A,A,A,A,A,A,A,A,A,B,C,C
3 8 A A,A,A,A,A,A,A,A,A,A,B,C,C
3 9 A A,A,A,A,A,A,A,A,A,B,C,C
3 10 A A,A,A,A,A,A,A,A,B,C,C
3 11 A A,A,A,A,A,A,A,B,C,C
3 12 A A,A,A,A,A,A,B,C,C
3 13 A A,A,A,A,A,B,C,C
3 14 A A,A,A,A,B,C,C
3 15 A A,A,A,B,C,C
3 16 A A,A,B,C,C
3 17 A A,B,C,C
3 18 A B,C,C
3 19 B C,C
3 20 C C
3 21 C
4 1 A A,A,A,A,A,A,A,A,A,A,A,A,A,A,A,B,B,B,C
4 2 A A,A,A,A,A,A,A,A,A,A,A,A,A,A,B,B,B,C
4 3 A A,A,A,A,A,A,A,A,A,A,A,A,A,B,B,B,C
4 4 A A,A,A,A,A,A,A,A,A,A,A,A,B,B,B,C
4 5 A A,A,A,A,A,A,A,A,A,A,A,B,B,B,C
4 6 A A,A,A,A,A,A,A,A,A,A,B,B,B,C
4 7 A A,A,A,A,A,A,A,A,A,B,B,B,C
4 8 A A,A,A,A,A,A,A,A,B,B,B,C
4 9 A A,A,A,A,A,A,A,B,B,B,C
4 10 A A,A,A,A,A,A,B,B,B,C
4 11 A A,A,A,A,A,B,B,B,C
4 12 A A,A,A,A,B,B,B,C
4 13 A A,A,A,B,B,B,C
4 14 A A,A,B,B,B,C
4 15 A A,B,B,B,C
4 16 A B,B,B,C
4 17 B B,B,C
4 18 B B,C
4 19 B C
4 20 C
This is the code I used to generate the above sample. Where am I going wrong?
CREATE TABLE #temp3
(
Scenario INT
,Vals VARCHAR(64)
,LEN INT
)
;
WITH vals AS
(
SELECT
v.*
FROM
(VALUES ('A'), ('B'), ('C')) v(x)
),
CTE AS
(
SELECT CAST('A' AS VARCHAR(MAX)) AS STR, 0 AS LEN
UNION ALL
SELECT (CTE.STR + ',' + vals.x), CTE.LEN + 1
FROM
CTE
JOIN vals
ON vals.x >= RIGHT(CTE.STR, 1)
WHERE CTE.LEN < 19
)
INSERT INTO #temp3
SELECT
ROW_NUMBER() OVER(ORDER BY STR + ',C') AS Scenario
,STR + ',C' AS Vals
,LEN
FROM
CTE
WHERE
STR + 'C' LIKE '%B%'
AND LEN = 19
;
-- Split strings created above into individual characters
WITH cte(Scenario, Value, Vals) AS
(
SELECT
Scenario
,CAST(LEFT(Vals, CHARINDEX(',',Vals+',')-1) AS VARCHAR(10)) AS Value
,STUFF(Vals, 1, CHARINDEX(',',Vals+','), '') AS Vals
FROM #temp3
UNION ALL
SELECT
Scenario
,CAST(LEFT(Vals, CHARINDEX(',',Vals+',')-1) AS VARCHAR(10))
,STUFF(Vals, 1, CHARINDEX(',',Vals+','), '')
FROM cte
WHERE Vals > ''
)
SELECT
Scenario
,ROW_NUMBER() OVER (PARTITION BY Scenario ORDER BY Scenario) RowNumber
,Value
,Vals
FROM cte t

I'm not exactly sure what the problem you are describing is, but the ROW_NUMBER() should use an ORDER BY clause that completely orders the rows in each partition.
When you use "PARTITION BY Scenario ORDER BY Scenario" the order in which the ROW_NUMBER() values are assigned is undefined. Try something like
WITH cte(Scenario, depth, Value, Vals) AS
(
SELECT
Scenario, 0 depth
,CAST(LEFT(Vals, CHARINDEX(',',Vals+',')-1) AS VARCHAR(10)) AS Value
,STUFF(Vals, 1, CHARINDEX(',',Vals+','), '') AS Vals
FROM #temp3
UNION ALL
SELECT
Scenario, depth+1
,CAST(LEFT(Vals, CHARINDEX(',',Vals+',')-1) AS VARCHAR(10))
,STUFF(Vals, 1, CHARINDEX(',',Vals+','), '')
FROM cte
WHERE Vals > ''
)
SELECT
Scenario
,depth
,ROW_NUMBER() OVER (PARTITION BY Scenario ORDER BY depth ) RowNumber
,Value
,Vals
FROM cte t

SQL recursive hierarchy

I am struggling to get one recursive CTE to work as desired but still with no chance..
So, I have the following similar table structures:
tblMapping:
map_id | type_id | name | parent_id
1 1 A1 0
2 1 A2 0
3 1 A3 1
4 1 A4 3
5 2 B1 0
6 2 B2 5
7 2 B3 6
8 1 A5 4
9 2 B4 0
tblRoleGroup:
role_group_id | type_id | map_id | desc_id
1 1 0 null
1 2 0 null
2 1 3 1
2 2 6 0
3 1 8 1
3 2 9 1
In tblRoleGroup, the desc_id field means:
null - allow all (used only in combination with map_id=0)
0 - allow all from parent including parent
1 - allow only current node
Still in tblRoleGroup if map_id=0 then the query should get all elements from same type_id
The query result should look like this:
role_group_id | type_id | map_id | path
1 1 1 A1
1 1 2 A2
1 1 3 A1.A3
1 1 4 A1.A3.A4
1 1 8 A1.A3.A4.A5
1 2 5 B1
1 2 6 B1.B2
1 2 7 B1.B2.B3
1 2 9 B4
2 1 3 A1.A3
2 2 6 B1.B2
2 2 7 B1.B2.B3
3 1 8 A1.A3.A4.A5
3 2 9 B4
The query below solves only a part of the expected result, but I wasn't able to make it work as the expected result..
WITH Hierarchy(map_id, type_id, name, Path) AS
(
SELECT t.map_id, t.type_id, t.name, CAST(t.name AS varchar(MAX)) AS Expr1
FROM dbo.tblMapping AS t
LEFT JOIN dbo.tblMapping AS t1 ON t1.map_id = t.parent_id
WHERE (t1.parent_id=0)
UNION ALL
SELECT t.map_id, t.type_id, t.name, CAST(h.Path + '.' + t.name AS varchar(MAX)) AS Expr1
FROM Hierarchy AS h
JOIN dbo.tblMapping AS t ON t.parent_id = h.map_id
)
SELECT h.map_id, h.type_id, t.role_group_id, h.Path AS Path
FROM Hierarchy AS h
LEFT JOIN dbo.tblRoleGroup t ON t.map_id = h.map_id
Could someone help me on this?
Thank you

At first I create a function that brings all descendants of passed map_id:
CREATE FUNCTION mapping (#map_id int)
RETURNS TABLE
AS
RETURN
(
WITH rec AS (
SELECT map_id,
[type_id],
CAST(name as nvarchar(max)) as name,
parent_id
FROM tblMapping
WHERE map_id = #map_id
UNION ALL
SELECT m.map_id,
m.[type_id],
r.name+'.'+m.name,
m.parent_id
FROM rec r
INNER JOIN tblMapping m
ON m.parent_id = r.map_id
)
SELECT *
FROM rec
);
GO
Then run this:
;WITH rec AS (
SELECT map_id,
[type_id],
CAST(name as nvarchar(max)) as name,
parent_id
FROM tblMapping
WHERE parent_id=0
UNION ALL
SELECT m.map_id,
m.[type_id],
r.name+'.'+m.name,
m.parent_id
FROM rec r
INNER JOIN tblMapping m
ON m.parent_id = r.map_id
)
SELECT t.role_group_id,
r.[type_id],
r.map_id,
r.name as [path]
FROM tblRoleGroup t
CROSS JOIN rec r
WHERE r.[type_id] = CASE WHEN t.desc_id IS NULL AND t.map_id = 0 THEN t.[type_id] ELSE NULL END
OR r.map_id = CASE WHEN t.desc_id = 1 THEN t.map_id ELSE NULL END
OR r.map_id IN (
SELECT map_id
FROM dbo.mapping (CASE WHEN t.desc_id = 0 THEN t.map_id ELSE NULL END)
)
ORDER BY role_group_id, r.[type_id], r.map_id
Will give you:
role_group_id type_id map_id path
1 1 1 A1
1 1 2 A2
1 1 3 A1.A3
1 1 4 A1.A3.A4
1 1 8 A1.A3.A4.A5
1 2 5 B1
1 2 6 B1.B2
1 2 7 B1.B2.B3
1 2 9 B4
2 1 3 A1.A3
2 2 6 B1.B2
2 2 7 B1.B2.B3
3 1 8 A1.A3.A4.A5
3 2 9 B4

T-SQL Reverse Pivot on every character of a string

We have a table like below in an sql server 2005 db:
event_id staff_id weeks
1 1 NNNYYYYNNYYY
1 2 YYYNNNYYYNNN
2 1 YYYYYYYYNYYY
This is from a piece of timetabling software and is basically saying which staff members are assigned to an event (register) and the set of weeks they are teaching that register. So staff_id 1 isn't teaching the first 3 weeks of event 1 but is teaching the following 4....
Is there an easy way to convert that to an easier form such as:
event_id staff_id week
1 1 4
1 1 5
1 1 6
1 1 7
1 1 10
1 1 11
1 1 12
1 2 1
1 2 2
1 2 3
1 2 7
1 2 8
1 2 9
2 1 1
2 1 2
2 1 3
2 1 4
2 1 5
2 1 6
2 1 7
2 1 8
2 1 10
2 1 11
2 1 12

WITH cte AS
(
SELECT 1 AS [week]
UNION ALL
SELECT [week] + 1
FROM cte
WHERE [week] < 53
)
SELECT t.event_id, t.staff_id, cte.[week]
FROM your_table AS t
INNER JOIN cte
ON LEN(ISNULL(t.weeks, '')) >= cte.[week]
AND SUBSTRING(t.weeks, cte.[week], 1) = 'Y'
ORDER BY t.event_id, t.staff_id, cte.[week]