[RBASE-L] - Re: SELECT DISTINCT

[jan johansen]

Ok,

This is just to weird. Trust me when I say that there is a column there
but I am typing it in now instead of pasting.

SELECT COUNT(DISTINCT WOBILLID) INTO vCountWOBILLID FROM Invoiceing +
   WHERE Status = 'D' AND BillingType = 'G'

If I change it to 
SELECT COUNT(DISTINCT WOBILLID,CustPO) INTO vCountWOBILLID FROM Invoiceing +
   WHERE Status = 'D' AND BillingType = 'G'
it throws an error.

So it looks like a 2 column DISTINCT works in a CURSOR but not for COUNT
which is probably normal since COUNT is an aggregate function.

So I am still looking for a way to count the unique values in rows based on 
2 columns.

Jan

 


-----Original Message-----
From: "jan johansen" <j...@jjcalibrations.com>
To: rbase-l@rbase.com (RBASE-L Mailing List)
Date: Mon, 07 Dec 2009 09:51:36 -0800
Subject: [RBASE-L] - Re: SELECT DISTINCT


Bill, 
 
You are absolutely right!
That was weird. I just copied the command and
the column(s) disappeared.
 
Here they are again
 
SELECT
(DISTINCT ) INTO vCountWOBILLID FROM Invoiceing +
WHERE Status = 'D' AND BillingType = 'G'
DECLARE Invoice CURSOR FOR SELECT +
DISTINCT WOBILLID , CustPO FROM Invoiceing + 
WHERE Status = 'D' AND BillingType = 'G' COUNT WOBILLID
 
 
 
So if I change count to   
SELECT COUNT(DISTINCT ) INTO vCountWOBILLID FROM Invoiceing +
WHERE Status = 'D' AND BillingType 'G'it throws an error

 
 
WOBILLID,CustPO
Jan 
 

 
-----Original Message-----
From: Bill Downall <bdown...@downallconsulting.com>
To: rbase-l@rbase.com (RBASE-L Mailing List)
Date: Mon, 7 Dec 2009 12:43:17 -0500
Subject: [RBASE-L] - Re: SELECT DISTINCT

Jan, 

 
DISTINCT needs a column name after it.

 
Bill


On Mon, Dec 7, 2009 at 12:39 PM, jan johansen < j...@jjcalibrations.com> 
wrote:

Group,
 
This morning as I was troubleshooting a process for generating invoices, I 
ran into challenge
 
I have a cursor that works fine and generated 39 invoices. The problem was 
that my method
to determine how many invoices would print calculated 33 invoices. While a 
minor issue, I
need a better to count.
 
My cursor has the following;
DECLARE
+
DISTINCT WOBILLID , CustPO FROM Invoiceing + 
WHERE Status = 'D' AND BillingType = 'G' Invoice CURSOR FOR SELECT
 
My calculation has;
SELECT
(DISTINCT ) INTO vCountWOBILLID FROM Invoiceing +
WHERE Status = 'D' AND BillingType = 'G'   COUNT WOBILLID
I know, I know. They are different. However when I tried to change my SELECT 
to
 
 
 
SELECT COUNT(DISTINCT ) INTO vCountWOBILLID FROM Invoiceing +
WHERE Status = 'D' AND BillingType = 'G'
WOBILLID,CustPO
I get an error. 

 
My suspicion is that the aggregate COUNT doesn't like it. I just found it 
interesting that I could declare a
distinct cursor on 2 columns but not count distinct on 2 columns.
 
The reason for the distinct is that a customer could send several different 
po's during a billing cycle
Also the possibility exists that 2 different customers could use the same 
PO.
 
Any suggestions are appreciated.
 
Jan
 
 
 
-----Original Message-----
From: Lawrence Lustig <larrylus...@yahoo.com>
To: rbase-l@rbase.com (RBASE-L Mailing List)
Date: Mon, 7 Dec 2009 09:22:30 -0800 (PST)
Subject: [RBASE-L] - Re: 7.6 Entry/Edit form


<<
I do not want the user to be able to add additional rows when in EDIT mode.
>>

Try (in the AFTER START EEP):

IF RBTI_FORM_MODE = 'EDIT' THEN
  PROPERTY TABLE YourTableName 'DISABLE_ADD_NEW_ROWS'
ENDIF

--
Larry