Re: Aggregating function which returns more than one row

2002-12-05 Thread Felix LUNGU
No. It is not what I want.
This is a normal aggregation.
The sum function will be called once for each
group (SalespersonNm) and it will return one row.

What I have asked is : if I have 10 rows, can I return 12?
It's rude, but that's the idea.

Thanks,
Felix


DL Neil wrote:

 Felix,

  Is is it possible to write an aggregation function which returns
  more than one row?

 Yes - if I have understood your requirement correctly.

 It would have been helpful if you had given some idea of the
 application/what you want to achieve. Here's an example from out of my
 imagination: if there is a table holding details of all of the sales
 (invoiced lines) for a particular month's trading, and if each is
 recorded/credited against a particular salesperson. Let's say there are one
 million rows in the table, but only ten sales staff. You can ask MySQL to
 return the sum of (aggregate) all sales attributed to each salesperson, and
 to count the number of sales/lines for each person:

 SELECT SalespersonNm-or-ID,
  COUNT( * ) AS NumberOfSales,
  SUM( ExtendedValue) AS TValueOfSales
 FROM =tbl=
 WHERE =this-month=
 GROUP BY SalespersonNm

 This will analyse the million rows but only output ten lines of
 results/salespersons' names (and the NumberOfSales column's entries will add
 up to one million/the number of rows in the table).

 What you wanted?
 =dn


-
Before posting, please check:
   http://www.mysql.com/manual.php   (the manual)
   http://lists.mysql.com/   (the list archive)

To request this thread, e-mail [EMAIL PROTECTED]
To unsubscribe, e-mail [EMAIL PROTECTED]
Trouble unsubscribing? Try: http://lists.mysql.com/php/unsubscribe.php




Re: Aggregating function which returns more than one row

2002-12-05 Thread Felix LUNGU
The idea is simple:
I want to implement some temporal computation
on timeseries like normalization and interpolation
to smooth the data and fill the missing values.

So, for example if have a table like that:
starttime,endtime,counter
12:00,12:30,34
13:00,14:00,25

you see that for 12:00 the measurement was for
30 minutes and a segement is missing (12:30-13:00).

What I want is to write some aggregation functions
the will do this kind of computations like the following:

select nisum(starttime,endtime,counter,interval,trashold) from my_table
where 

That's it,
Felix


---
sql, query

Victor Pendleton wrote:

 Felix,
 Can you please give an example of what you are attempting to do?

 -Original Message-
 From: Felix LUNGU [mailto:[EMAIL PROTECTED]]
 Sent: Wednesday, December 04, 2002 9:22 AM
 To: [EMAIL PROTECTED]
 Subject: Aggregating function which returns more than one row

 Is is it possible to write an aggregation function which returns
 more than one row?

 Thanks,
 felix

 Mysql sql query

 -
 Before posting, please check:
http://www.mysql.com/manual.php   (the manual)
http://lists.mysql.com/   (the list archive)

 To request this thread, e-mail [EMAIL PROTECTED]
 To unsubscribe, e-mail
 [EMAIL PROTECTED]
 Trouble unsubscribing? Try: http://lists.mysql.com/php/unsubscribe.php


-
Before posting, please check:
   http://www.mysql.com/manual.php   (the manual)
   http://lists.mysql.com/   (the list archive)

To request this thread, e-mail [EMAIL PROTECTED]
To unsubscribe, e-mail [EMAIL PROTECTED]
Trouble unsubscribing? Try: http://lists.mysql.com/php/unsubscribe.php




RE: Aggregating function which returns more than one row

2002-12-05 Thread Grant Cooper


-Original Message-
From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]]On
Behalf Of Felix LUNGU
Sent: Thursday, December 05, 2002 12:27 AM
To: Victor Pendleton; [EMAIL PROTECTED]
Subject: Re: Aggregating function which returns more than one row


The idea is simple:
I want to implement some temporal computation
on timeseries like normalization and interpolation
to smooth the data and fill the missing values.

So, for example if have a table like that:
starttime,endtime,counter
12:00,12:30,34
13:00,14:00,25

you see that for 12:00 the measurement was for
30 minutes and a segement is missing (12:30-13:00).

What I want is to write some aggregation functions
the will do this kind of computations like the following:

select nisum(starttime,endtime,counter,interval,trashold) from my_table
where 

That's it,
Felix


---
sql, query

Victor Pendleton wrote:

 Felix,
 Can you please give an example of what you are attempting to do?

 -Original Message-
 From: Felix LUNGU [mailto:[EMAIL PROTECTED]]
 Sent: Wednesday, December 04, 2002 9:22 AM
 To: [EMAIL PROTECTED]
 Subject: Aggregating function which returns more than one row

 Is is it possible to write an aggregation function which returns
 more than one row?

 Thanks,
 felix

 Mysql sql query

 -
 Before posting, please check:
http://www.mysql.com/manual.php   (the manual)
http://lists.mysql.com/   (the list archive)

 To request this thread, e-mail [EMAIL PROTECTED]
 To unsubscribe, e-mail
 [EMAIL PROTECTED]
 Trouble unsubscribing? Try: http://lists.mysql.com/php/unsubscribe.php


-
Before posting, please check:
   http://www.mysql.com/manual.php   (the manual)
   http://lists.mysql.com/   (the list archive)

To request this thread, e-mail [EMAIL PROTECTED]
To unsubscribe, e-mail [EMAIL PROTECTED]
Trouble unsubscribing? Try: http://lists.mysql.com/php/unsubscribe.php


-
Before posting, please check:
   http://www.mysql.com/manual.php   (the manual)
   http://lists.mysql.com/   (the list archive)

To request this thread, e-mail [EMAIL PROTECTED]
To unsubscribe, e-mail [EMAIL PROTECTED]
Trouble unsubscribing? Try: http://lists.mysql.com/php/unsubscribe.php




Re: Aggregating function which returns more than one row

2002-12-04 Thread DL Neil
Felix,

 Is is it possible to write an aggregation function which returns
 more than one row?


Yes - if I have understood your requirement correctly.

It would have been helpful if you had given some idea of the
application/what you want to achieve. Here's an example from out of my
imagination: if there is a table holding details of all of the sales
(invoiced lines) for a particular month's trading, and if each is
recorded/credited against a particular salesperson. Let's say there are one
million rows in the table, but only ten sales staff. You can ask MySQL to
return the sum of (aggregate) all sales attributed to each salesperson, and
to count the number of sales/lines for each person:

SELECT SalespersonNm-or-ID,
 COUNT( * ) AS NumberOfSales,
 SUM( ExtendedValue) AS TValueOfSales
FROM =tbl=
WHERE =this-month=
GROUP BY SalespersonNm

This will analyse the million rows but only output ten lines of
results/salespersons' names (and the NumberOfSales column's entries will add
up to one million/the number of rows in the table).

What you wanted?
=dn


-
Before posting, please check:
   http://www.mysql.com/manual.php   (the manual)
   http://lists.mysql.com/   (the list archive)

To request this thread, e-mail [EMAIL PROTECTED]
To unsubscribe, e-mail [EMAIL PROTECTED]
Trouble unsubscribing? Try: http://lists.mysql.com/php/unsubscribe.php