Re: SetGT/ReadPE method Exception

On Thu, 2005-08-25 at 10:04 -0700, Tony Carolla wrote:
> Thanks to all for your attention to this matter. I have learned alot about 
> SETGT and READPE today ;-) . I guess I thought that, if the SETGT did not 
> find a key greater than the one supplied, and positioned the file to EOF, 
> the subsequent READPE would then find the prior equal-key record for the 
> account (which would be the very last record in the file), get the sequence 
> number, add 1, and write the record successfully. If SETGT is performing as 
> per the manual, then it must be the EOF condition that prevents the READPE 
> from working. 

No.  We aren't paying proper attention.


If %Found(Mmemo_I);
MO.Mmseq#=MI.Mmseq#+1;
Else;
MO.Mmseq#=1;
EndIf;

Oldest ILE problem in the book.  %found is not set by READPE.  Only %EOF
is set.  So, if a READPE occurs and it gets a hit, %eof will be '0'.  If
it doesn't get a match, %eof will be '1'.

How about:

if %eof(Mmemo_I);
    MO.Mmseq# = 1;
Else;
    MO.Mmseq# = MI.Mmseq#+1;
endif;

> 
> 
> On 8/25/05, Rich Duzenbury <rduz-midrange@xxxxxxxxxxxxxxxxxxx> wrote:
> 
> > The problem is that you are doing your setgt and readpe inside the loop.
> > It resets the sequence number back to 1, and then when you write again,
> > you still get the same error.
> 
> 
> 
> -- 
> "Enter any 11-digit prime number to continue..."
> "In Hebrew SQL, how do you use right() and left()?..." - Random Thought
> "If all you have is a hammer, all your problems begin to look like nails"
Regards,
Rich