Re: Embedded SQL For Screen Load

[Booth Martin <booth@xxxxxxxxxxxx>]

Fwiw, this is an important design issue in my opinion.  The programmer 
can consider how the data will be used and design the flow so that a 
user is never presented with thousands of rows. That is an unusable data 
pile and should be avoided.  Design a flow that gets relevant data to 
the user in 200 rows or less, even if it adds a step or two.  Load-all 
subfiles, with a limit set at 250 rows or so, is what users are used to 
seeing.


On 12/15/2016 4:30 PM, Charles Wilt wrote:
> ​Ah, no..
>
> Scroll-able cursors only scroll within the result set.  If I've got L,N,O
> in the data and I open a cursor, WHERE KEY >= 'M'
>
> Then the result set starts with N, you can't do a scroll -1 and get to L.
>
>
> ​Charles​