multiple resultsets: everybody

Showing posts with label everybody. Show all posts

Friday, March 23, 2012

Multiply record fields with same mainID.

Hi everybody,
What's the most efficient way to get the following result:
I have a number of records with a MainID and a Value fields.
eg:
SubID MainID Value
1 1 1
2 1 1
3 1 2
4 1 3
5 2 1
6 2 1
7 3 2
8 3 2
I need the product of the Value field for each MainID.
The result has to be like this:
MainID PrValue
1 6 (1*1*2*3)
2 1 (1*1)
3 4 (2*2)
TIA,
Martin.select MainID, sum(value) as PrValue
from table
group by MainID
"martin" <kashaan007@.hotmail.com> wrote in message
news:Of%23nDPnRFHA.3144@.tk2msftngp13.phx.gbl...
> Hi everybody,
> What's the most efficient way to get the following result:
> I have a number of records with a MainID and a Value fields.
> eg:
> SubID MainID Value
> 1 1 1
> 2 1 1
> 3 1 2
> 4 1 3
> 5 2 1
> 6 2 1
> 7 3 2
> 8 3 2
> I need the product of the Value field for each MainID.
> The result has to be like this:
> MainID PrValue
> 1 6 (1*1*2*3)
> 2 1 (1*1)
> 3 4 (2*2)
> TIA,
> Martin.
>
>|||Try,
use northwind
go
create table t (
SubID int,
MainID int,
Value int
)
go
insert into t values(1, 1, 1)
insert into t values(2, 1, 1)
insert into t values(3, 1, 2)
insert into t values(4, 1, 3)
insert into t values(5, 2, 1)
insert into t values(6, 2, 1)
insert into t values(7, 3, 2)
insert into t values(8, 3, 2)
go
select
MainID,
POWER(10, SUM(LOG10(Value)))
from
t
group by
MainID
drop table t
go
I took the idea from:
The T-SQL Banker
http://msdn.microsoft.com/library/default.asp?url=/library/en-us/dnsqlmag02/html/TheT-SQLBanker.asp
AMB
"martin" wrote:
> Hi everybody,
> What's the most efficient way to get the following result:
> I have a number of records with a MainID and a Value fields.
> eg:
> SubID MainID Value
> 1 1 1
> 2 1 1
> 3 1 2
> 4 1 3
> 5 2 1
> 6 2 1
> 7 3 2
> 8 3 2
> I need the product of the Value field for each MainID.
> The result has to be like this:
> MainID PrValue
> 1 6 (1*1*2*3)
> 2 1 (1*1)
> 3 4 (2*2)
> TIA,
> Martin.
>
>|||Hi,AMB
It gives me a wrong output for the col1=2
CREATE TABLE #Test
(
col1 INT NOT NULL,
col2 INT NOT NULL
)
INSERT INTO #Test VALUES (1,10)
INSERT INTO #Test VALUES (1,2)
INSERT INTO #Test VALUES (1,3)
INSERT INTO #Test VALUES (2,4)
INSERT INTO #Test VALUES (2,2)
SELECT col1,POWER(10, SUM(LOG10(col2)))
FROM #Test GROUP BY col1
"Alejandro Mesa" <AlejandroMesa@.discussions.microsoft.com> wrote in message
news:2DA4CABB-2DBD-43CA-B0DE-995B8B644F14@.microsoft.com...
> Try,
> use northwind
> go
> create table t (
> SubID int,
> MainID int,
> Value int
> )
> go
> insert into t values(1, 1, 1)
> insert into t values(2, 1, 1)
> insert into t values(3, 1, 2)
> insert into t values(4, 1, 3)
> insert into t values(5, 2, 1)
> insert into t values(6, 2, 1)
> insert into t values(7, 3, 2)
> insert into t values(8, 3, 2)
> go
> select
> MainID,
> POWER(10, SUM(LOG10(Value)))
> from
> t
> group by
> MainID
> drop table t
> go
> I took the idea from:
> The T-SQL Banker
>
http://msdn.microsoft.com/library/default.asp?url=/library/en-us/dnsqlmag02/html/TheT-SQLBanker.asp
>
> AMB
> "martin" wrote:
> > Hi everybody,
> >
> > What's the most efficient way to get the following result:
> >
> > I have a number of records with a MainID and a Value fields.
> >
> > eg:
> >
> > SubID MainID Value
> > 1 1 1
> > 2 1 1
> > 3 1 2
> > 4 1 3
> > 5 2 1
> > 6 2 1
> > 7 3 2
> > 8 3 2
> >
> > I need the product of the Value field for each MainID.
> >
> > The result has to be like this:
> >
> > MainID PrValue
> > 1 6 (1*1*2*3)
> > 2 1 (1*1)
> > 3 4 (2*2)
> >
> > TIA,
> >
> > Martin.
> >
> >
> >
> >|||Alejandro's solution is good for values greater than zero. Here's a
more general solution for all integers (this one adapted from Celko and
others):
SELECT mainid,
CAST(ROUND(
COALESCE(EXP(SUM(LOG(ABS(NULLIF(value,0))))),0)
* SIGN(MIN(ABS(value)))
* (COUNT(NULLIF(SIGN(value),1))%2*-2+1)
,0) AS INTEGER) AS product
FROM T
GROUP BY mainid
--
David Portas
SQL Server MVP
--|||Thanks a lot.
Works great.
Martin.
"Alejandro Mesa" <AlejandroMesa@.discussions.microsoft.com> wrote in message
news:2DA4CABB-2DBD-43CA-B0DE-995B8B644F14@.microsoft.com...
> Try,
> use northwind
> go
> create table t (
> SubID int,
> MainID int,
> Value int
> )
> go
> insert into t values(1, 1, 1)
> insert into t values(2, 1, 1)
> insert into t values(3, 1, 2)
> insert into t values(4, 1, 3)
> insert into t values(5, 2, 1)
> insert into t values(6, 2, 1)
> insert into t values(7, 3, 2)
> insert into t values(8, 3, 2)
> go
> select
> MainID,
> POWER(10, SUM(LOG10(Value)))
> from
> t
> group by
> MainID
> drop table t
> go
> I took the idea from:
> The T-SQL Banker
>
http://msdn.microsoft.com/library/default.asp?url=/library/en-us/dnsqlmag02/
html/TheT-SQLBanker.asp
>
> AMB
> "martin" wrote:
> > Hi everybody,
> >
> > What's the most efficient way to get the following result:
> >
> > I have a number of records with a MainID and a Value fields.
> >
> > eg:
> >
> > SubID MainID Value
> > 1 1 1
> > 2 1 1
> > 3 1 2
> > 4 1 3
> > 5 2 1
> > 6 2 1
> > 7 3 2
> > 8 3 2
> >
> > I need the product of the Value field for each MainID.
> >
> > The result has to be like this:
> >
> > MainID PrValue
> > 1 6 (1*1*2*3)
> > 2 1 (1*1)
> > 3 4 (2*2)
> >
> > TIA,
> >
> > Martin.
> >
> >
> >
> >|||If we add +0.001 it works fine
SELECT col1,POWER(10, SUM(LOG10(col2))+0.001)
FROM #Test GROUP BY col1
"Uri Dimant" <urid@.iscar.co.il> wrote in message
news:uYVS5cnRFHA.1172@.TK2MSFTNGP12.phx.gbl...
> Hi,AMB
> It gives me a wrong output for the col1=2
> CREATE TABLE #Test
> (
> col1 INT NOT NULL,
> col2 INT NOT NULL
> )
> INSERT INTO #Test VALUES (1,10)
> INSERT INTO #Test VALUES (1,2)
> INSERT INTO #Test VALUES (1,3)
> INSERT INTO #Test VALUES (2,4)
> INSERT INTO #Test VALUES (2,2)
>
> SELECT col1,POWER(10, SUM(LOG10(col2)))
> FROM #Test GROUP BY col1
>
>
>
> "Alejandro Mesa" <AlejandroMesa@.discussions.microsoft.com> wrote in
message
> news:2DA4CABB-2DBD-43CA-B0DE-995B8B644F14@.microsoft.com...
> > Try,
> >
> > use northwind
> > go
> >
> > create table t (
> > SubID int,
> > MainID int,
> > Value int
> > )
> > go
> >
> > insert into t values(1, 1, 1)
> > insert into t values(2, 1, 1)
> > insert into t values(3, 1, 2)
> > insert into t values(4, 1, 3)
> > insert into t values(5, 2, 1)
> > insert into t values(6, 2, 1)
> > insert into t values(7, 3, 2)
> > insert into t values(8, 3, 2)
> > go
> >
> > select
> > MainID,
> > POWER(10, SUM(LOG10(Value)))
> > from
> > t
> > group by
> > MainID
> >
> > drop table t
> > go
> >
> > I took the idea from:
> >
> > The T-SQL Banker
> >
>
http://msdn.microsoft.com/library/default.asp?url=/library/en-us/dnsqlmag02/html/TheT-SQLBanker.asp
> >
> >
> > AMB
> >
> > "martin" wrote:
> >
> > > Hi everybody,
> > >
> > > What's the most efficient way to get the following result:
> > >
> > > I have a number of records with a MainID and a Value fields.
> > >
> > > eg:
> > >
> > > SubID MainID Value
> > > 1 1 1
> > > 2 1 1
> > > 3 1 2
> > > 4 1 3
> > > 5 2 1
> > > 6 2 1
> > > 7 3 2
> > > 8 3 2
> > >
> > > I need the product of the Value field for each MainID.
> > >
> > > The result has to be like this:
> > >
> > > MainID PrValue
> > > 1 6 (1*1*2*3)
> > > 2 1 (1*1)
> > > 3 4 (2*2)
> > >
> > > TIA,
> > >
> > > Martin.
> > >
> > >
> > >
> > >
>|||Cast 10 to float.
SELECT
col1,
cast(POWER(cast(10 as float), SUM(LOG10(col2))) as decimal(5))
FROM #Test GROUP BY col1
go
AMB
"Uri Dimant" wrote:
> Hi,AMB
> It gives me a wrong output for the col1=2
> CREATE TABLE #Test
> (
> col1 INT NOT NULL,
> col2 INT NOT NULL
> )
> INSERT INTO #Test VALUES (1,10)
> INSERT INTO #Test VALUES (1,2)
> INSERT INTO #Test VALUES (1,3)
> INSERT INTO #Test VALUES (2,4)
> INSERT INTO #Test VALUES (2,2)
>
> SELECT col1,POWER(10, SUM(LOG10(col2)))
> FROM #Test GROUP BY col1
>
>
>
> "Alejandro Mesa" <AlejandroMesa@.discussions.microsoft.com> wrote in message
> news:2DA4CABB-2DBD-43CA-B0DE-995B8B644F14@.microsoft.com...
> > Try,
> >
> > use northwind
> > go
> >
> > create table t (
> > SubID int,
> > MainID int,
> > Value int
> > )
> > go
> >
> > insert into t values(1, 1, 1)
> > insert into t values(2, 1, 1)
> > insert into t values(3, 1, 2)
> > insert into t values(4, 1, 3)
> > insert into t values(5, 2, 1)
> > insert into t values(6, 2, 1)
> > insert into t values(7, 3, 2)
> > insert into t values(8, 3, 2)
> > go
> >
> > select
> > MainID,
> > POWER(10, SUM(LOG10(Value)))
> > from
> > t
> > group by
> > MainID
> >
> > drop table t
> > go
> >
> > I took the idea from:
> >
> > The T-SQL Banker
> >
> http://msdn.microsoft.com/library/default.asp?url=/library/en-us/dnsqlmag02/html/TheT-SQLBanker.asp
> >
> >
> > AMB
> >
> > "martin" wrote:
> >
> > > Hi everybody,
> > >
> > > What's the most efficient way to get the following result:
> > >
> > > I have a number of records with a MainID and a Value fields.
> > >
> > > eg:
> > >
> > > SubID MainID Value
> > > 1 1 1
> > > 2 1 1
> > > 3 1 2
> > > 4 1 3
> > > 5 2 1
> > > 6 2 1
> > > 7 3 2
> > > 8 3 2
> > >
> > > I need the product of the Value field for each MainID.
> > >
> > > The result has to be like this:
> > >
> > > MainID PrValue
> > > 1 6 (1*1*2*3)
> > > 2 1 (1*1)
> > > 3 4 (2*2)
> > >
> > > TIA,
> > >
> > > Martin.
> > >
> > >
> > >
> > >
>
>|||Good catch!!!
AMB
"Uri Dimant" wrote:
> Hi,AMB
> It gives me a wrong output for the col1=2
> CREATE TABLE #Test
> (
> col1 INT NOT NULL,
> col2 INT NOT NULL
> )
> INSERT INTO #Test VALUES (1,10)
> INSERT INTO #Test VALUES (1,2)
> INSERT INTO #Test VALUES (1,3)
> INSERT INTO #Test VALUES (2,4)
> INSERT INTO #Test VALUES (2,2)
>
> SELECT col1,POWER(10, SUM(LOG10(col2)))
> FROM #Test GROUP BY col1
>
>
>
> "Alejandro Mesa" <AlejandroMesa@.discussions.microsoft.com> wrote in message
> news:2DA4CABB-2DBD-43CA-B0DE-995B8B644F14@.microsoft.com...
> > Try,
> >
> > use northwind
> > go
> >
> > create table t (
> > SubID int,
> > MainID int,
> > Value int
> > )
> > go
> >
> > insert into t values(1, 1, 1)
> > insert into t values(2, 1, 1)
> > insert into t values(3, 1, 2)
> > insert into t values(4, 1, 3)
> > insert into t values(5, 2, 1)
> > insert into t values(6, 2, 1)
> > insert into t values(7, 3, 2)
> > insert into t values(8, 3, 2)
> > go
> >
> > select
> > MainID,
> > POWER(10, SUM(LOG10(Value)))
> > from
> > t
> > group by
> > MainID
> >
> > drop table t
> > go
> >
> > I took the idea from:
> >
> > The T-SQL Banker
> >
> http://msdn.microsoft.com/library/default.asp?url=/library/en-us/dnsqlmag02/html/TheT-SQLBanker.asp
> >
> >
> > AMB
> >
> > "martin" wrote:
> >
> > > Hi everybody,
> > >
> > > What's the most efficient way to get the following result:
> > >
> > > I have a number of records with a MainID and a Value fields.
> > >
> > > eg:
> > >
> > > SubID MainID Value
> > > 1 1 1
> > > 2 1 1
> > > 3 1 2
> > > 4 1 3
> > > 5 2 1
> > > 6 2 1
> > > 7 3 2
> > > 8 3 2
> > >
> > > I need the product of the Value field for each MainID.
> > >
> > > The result has to be like this:
> > >
> > > MainID PrValue
> > > 1 6 (1*1*2*3)
> > > 2 1 (1*1)
> > > 3 4 (2*2)
> > >
> > > TIA,
> > >
> > > Martin.
> > >
> > >
> > >
> > >
>
>|||David,
This is really a good one.
AMB
"David Portas" wrote:
> Alejandro's solution is good for values greater than zero. Here's a
> more general solution for all integers (this one adapted from Celko and
> others):
> SELECT mainid,
> CAST(ROUND(
> COALESCE(EXP(SUM(LOG(ABS(NULLIF(value,0))))),0)
> * SIGN(MIN(ABS(value)))
> * (COUNT(NULLIF(SIGN(value),1))%2*-2+1)
> ,0) AS INTEGER) AS product
> FROM T
> GROUP BY mainid
> --
> David Portas
> SQL Server MVP
> --
>|||Use better the one posted by David Portas, and if you decide to use this one,
read the correction in my answer to Uri.
AMB
"martin" wrote:
> Thanks a lot.
> Works great.
> Martin.
>
> "Alejandro Mesa" <AlejandroMesa@.discussions.microsoft.com> wrote in message
> news:2DA4CABB-2DBD-43CA-B0DE-995B8B644F14@.microsoft.com...
> > Try,
> >
> > use northwind
> > go
> >
> > create table t (
> > SubID int,
> > MainID int,
> > Value int
> > )
> > go
> >
> > insert into t values(1, 1, 1)
> > insert into t values(2, 1, 1)
> > insert into t values(3, 1, 2)
> > insert into t values(4, 1, 3)
> > insert into t values(5, 2, 1)
> > insert into t values(6, 2, 1)
> > insert into t values(7, 3, 2)
> > insert into t values(8, 3, 2)
> > go
> >
> > select
> > MainID,
> > POWER(10, SUM(LOG10(Value)))
> > from
> > t
> > group by
> > MainID
> >
> > drop table t
> > go
> >
> > I took the idea from:
> >
> > The T-SQL Banker
> >
> http://msdn.microsoft.com/library/default.asp?url=/library/en-us/dnsqlmag02/
> html/TheT-SQLBanker.asp
> >
> >
> > AMB
> >
> > "martin" wrote:
> >
> > > Hi everybody,
> > >
> > > What's the most efficient way to get the following result:
> > >
> > > I have a number of records with a MainID and a Value fields.
> > >
> > > eg:
> > >
> > > SubID MainID Value
> > > 1 1 1
> > > 2 1 1
> > > 3 1 2
> > > 4 1 3
> > > 5 2 1
> > > 6 2 1
> > > 7 3 2
> > > 8 3 2
> > >
> > > I need the product of the Value field for each MainID.
> > >
> > > The result has to be like this:
> > >
> > > MainID PrValue
> > > 1 6 (1*1*2*3)
> > > 2 1 (1*1)
> > > 3 4 (2*2)
> > >
> > > TIA,
> > >
> > > Martin.
> > >
> > >
> > >
> > >
>
>|||Ok thanks.
Martin.
"Alejandro Mesa" <AlejandroMesa@.discussions.microsoft.com> wrote in message
news:210DC41A-BF9B-4211-9E16-995A554C0374@.microsoft.com...
> Use better the one posted by David Portas, and if you decide to use this
one,
> read the correction in my answer to Uri.
>
> AMB
> "martin" wrote:
> > Thanks a lot.
> >
> > Works great.
> >
> > Martin.
> >
> >
> > "Alejandro Mesa" <AlejandroMesa@.discussions.microsoft.com> wrote in
message
> > news:2DA4CABB-2DBD-43CA-B0DE-995B8B644F14@.microsoft.com...
> > > Try,
> > >
> > > use northwind
> > > go
> > >
> > > create table t (
> > > SubID int,
> > > MainID int,
> > > Value int
> > > )
> > > go
> > >
> > > insert into t values(1, 1, 1)
> > > insert into t values(2, 1, 1)
> > > insert into t values(3, 1, 2)
> > > insert into t values(4, 1, 3)
> > > insert into t values(5, 2, 1)
> > > insert into t values(6, 2, 1)
> > > insert into t values(7, 3, 2)
> > > insert into t values(8, 3, 2)
> > > go
> > >
> > > select
> > > MainID,
> > > POWER(10, SUM(LOG10(Value)))
> > > from
> > > t
> > > group by
> > > MainID
> > >
> > > drop table t
> > > go
> > >
> > > I took the idea from:
> > >
> > > The T-SQL Banker
> > >
> >
http://msdn.microsoft.com/library/default.asp?url=/library/en-us/dnsqlmag02/
> > html/TheT-SQLBanker.asp
> > >
> > >
> > > AMB
> > >
> > > "martin" wrote:
> > >
> > > > Hi everybody,
> > > >
> > > > What's the most efficient way to get the following result:
> > > >
> > > > I have a number of records with a MainID and a Value fields.
> > > >
> > > > eg:
> > > >
> > > > SubID MainID Value
> > > > 1 1 1
> > > > 2 1 1
> > > > 3 1 2
> > > > 4 1 3
> > > > 5 2 1
> > > > 6 2 1
> > > > 7 3 2
> > > > 8 3 2
> > > >
> > > > I need the product of the Value field for each MainID.
> > > >
> > > > The result has to be like this:
> > > >
> > > > MainID PrValue
> > > > 1 6 (1*1*2*3)
> > > > 2 1 (1*1)
> > > > 3 4 (2*2)
> > > >
> > > > TIA,
> > > >
> > > > Martin.
> > > >
> > > >
> > > >
> > > >
> >
> >
> >

Multiply record fields with same mainID.

Monday, February 20, 2012

Multiple rows

Hi everybody:
I have this result from a query:
category id pos name
Fixed
1 100009 999003 BLA BLA BLA 1
1 100010 999003 BLA BLA BLA 1
2 100321 999007 EX EX EX 0
2 100322 999007 EX EX EX 0
2 100323 999007 EX EX EX 0
2 100324 999007 EX EX EX 0
2 100325 999007 EX EX EX 0
2 100326 999007 EX EX EX 0
Is there any way (not using UNION) to get the rows where Fixed NOT 0
but only one from where Fixed = 0 ?<acidscan@.gmail.com> wrote in message
news:1175639463.958180.180500@.n76g2000hsh.googlegroups.com...
> Hi everybody:
> I have this result from a query:
> category id pos name
> Fixed
> 1 100009 999003 BLA BLA BLA 1
> 1 100010 999003 BLA BLA BLA 1
> 2 100321 999007 EX EX EX 0
> 2 100322 999007 EX EX EX 0
> 2 100323 999007 EX EX EX 0
> 2 100324 999007 EX EX EX 0
> 2 100325 999007 EX EX EX 0
> 2 100326 999007 EX EX EX 0
> Is there any way (not using UNION) to get the rows where Fixed NOT 0
> but only one from where Fixed = 0 ?
>
Please post DDL so that we don't have to guess what the keys are. Please
post the query that produces this result.
Explain the logic. Which row do you want to return where Fixed = 0? You mean
just some random row?
--
David Portas, SQL Server MVP
Whenever possible please post enough code to reproduce your problem.
Including CREATE TABLE and INSERT statements usually helps.
State what version of SQL Server you are using and specify the content
of any error messages.
SQL Server Books Online:
http://msdn2.microsoft.com/library/ms130214(en-US,SQL.90).aspx
--|||id pos name Fixed
100009 999003 BLA BLA BLA 1
100010 999003 BLA BLA BLA 1
100321 999007 EX EX EX 0
100322 999007 EX EX EX 0
100323 999007 EX EX EX 0
100324 999007 EX EX EX 0
100325 999007 EX EX EX 0
100326 999007 EX EX EX 0
I made it shorted to remove the format. This is a simple select
returning some elements from a table. What i would like to do is only
show the first 3 elements based on the "fixed" column, first showing
the elements where "fixed" is not 0 and later all the elements where
"fixed" is 0 grouped into one row.
I could do this making a select where fixed != 0 and a "union all"
with another select where fixed = 0 using only the first element. This
union will return:
id pos name Fixed
100009 999003 BLA BLA BLA 1
100010 999003 BLA BLA BLA 1
100321 999007 EX EX EX 0
My question is: is there any shorter way (not using the union) to do
this ?|||<acidscan@.gmail.com> wrote in message
news:1175642227.533516.273640@.y66g2000hsf.googlegroups.com...
> id pos name Fixed
> 100009 999003 BLA BLA BLA 1
> 100010 999003 BLA BLA BLA 1
> 100321 999007 EX EX EX 0
> 100322 999007 EX EX EX 0
> 100323 999007 EX EX EX 0
> 100324 999007 EX EX EX 0
> 100325 999007 EX EX EX 0
> 100326 999007 EX EX EX 0
> I made it shorted to remove the format. This is a simple select
> returning some elements from a table. What i would like to do is only
> show the first 3 elements based on the "fixed" column, first showing
> the elements where "fixed" is not 0 and later all the elements where
> "fixed" is 0 grouped into one row.
> I could do this making a select where fixed != 0 and a "union all"
> with another select where fixed = 0 using only the first element. This
> union will return:
> id pos name Fixed
> 100009 999003 BLA BLA BLA 1
> 100010 999003 BLA BLA BLA 1
> 100321 999007 EX EX EX 0
> My question is: is there any shorter way (not using the union) to do
> this ?
>
You didn't specify a key so I'll assume that ID is the key. If I'm right,
try:
SELECT id, pos, name, fixed
FROM tbl
WHERE fixed <> 0
OR id =(SELECT MIN(id)
FROM tbl
WHERE fixed = 0);
--
David Portas, SQL Server MVP
Whenever possible please post enough code to reproduce your problem.
Including CREATE TABLE and INSERT statements usually helps.
State what version of SQL Server you are using and specify the content
of any error messages.
SQL Server Books Online:
http://msdn2.microsoft.com/library/ms130214(en-US,SQL.90).aspx
--

Friday, March 23, 2012

Multiply record fields with same mainID.

Multiply record fields with same mainID.

Multiply record fields with same mainID.

Monday, February 20, 2012

Multiple rows

multiple resultsets

Blog Archive

About Me