Re: Tough One: How do u denormailize a table via a view, if column is text not numb

Wow a function what is that?

Of course that that what I need, I was asking for an algorithum to implement it. Grouping requires some sort of aggreate function, a sum, max, etc. it doesn't work if u dont use that. I need a technique to get around that

In article <jhCD7.117316$5h5.47365719_at_news3.rdc2.on.home.com>, "Paul Quenneville" <paulq_at_home.com> wrote:
>write a function to return the full concatenatedstring from table A
>
>"johnthan" <jthn342_at_hotmail.com> wrote in message
>news:3bdee6fe$0$14833$45beb828_at_newscene.com...
>>
>> We have a table A as such:
>>
>> REQ LINE_NUM DESCP
>>
>> 9 10000 PED IS OFF
>> 9 20000 SO PROCEED
>> 9 30000 WITH X
>>
>>
>> that needs to be joined to a table B as such
>>
>>
>> REQ location date
>>
>> 9 LA 01/01/01
>>
>> to produce
>>
>>
>> REQ location date DESCP
>>
>> 9 LA 01/01/01 PED IS OFF SO PROCEED
>WITH X
>>
>>
>> (the app that creates the table A only allows x number of characters per
>> column if DESCP is longer than that it creates additional rows until whole
>> DESCP is stored)
>>
>> I know how to denormalize a table if the column that is denormalized is a
>> number e.g.
>> SELECT a,b,
>> sum(decode(y, z, 99, 0)),
>> sum(decode(y, z, 88, 0))
>> FROM TB1
>> GROUP BY a,b
>>
>> but what if its a varchar?
>>
>> We need this done dynamically via a view
>>
>> Any ideas?
>>
>
>
Received on Tue Oct 30 2001 - 14:52:09 CST