Re: DSPF numeric field editing

[Booth Martin <booth@xxxxxxxxxxxx>]

Its pretty much working as designed.

My dimming  memory says this is how I solved this same sort of issue:  
Immediately after the exfmt, do a write of the format, then do your 
validations.

If that works, great.  If it doesn't, then my memory has gotten even worse.

On 2/8/2010 3:08 PM, Timothy Adair wrote:
> Ok, I tried PUTOVR but it didn't work.  I added (unconditional) PUTOVR at
> the record level and (unconditional) OVRDTA on the field, and recompiled the
> DSPF and the pgm.  It was as if the keywords weren't even there (yes I
> checked the compile listing to verify).
>
> It may have something to do with the fact that we're running S/36
> Environment, or the phase of the moon, or the price of tea...  anyway, you
> get the idea.  Or maybe I'm just missing something? (nah.)
>
> Not to beat the proverbial deceased equine, but if anyone has a quick
> thought as to why this wouldn't be working, please enlighten me.
>
> Otherwise, thank you all for all responses.
>
> TA
>
>
>
>
> "Timothy Adair"<tadair@xxxxxxxxxxxxxxxx>  wrote in message
> news:mailman.7074.1265648703.2580.midrange-l@xxxxxxxxxxxxxxx
>    
> > I'm doing an extra Write if the user doesn't key in a decimal point.
> >
> > write foot2;
> > exfmt scrn01;
> > if %int(futprc) = futprc;
> >     futprc = futprc / 10000;
> >     write scrn01;
> > endif;
> >
> >
> > That was before Dan's suggestion of using PUTOVR - I may try that instead;
> > it seems a bit more elegant.
> >
> >
> >
> >
> >
> >
> > "Alan Shore"<AlanShore@xxxxxxxx>  wrote in message
> > news:mailman.7071.1265647998.2580.midrange-l@xxxxxxxxxxxxxxx
> >
> > No problem
> > I'm curious.
> > What way around this have you found?
> >
> >
> >
> > Alan Shore
> > Programmer/Analyst, Direct Response
> > E:AShore@xxxxxxxx
> > P:(631) 200-5019
> > C:(631) 880-8640
> > "If you're going through Hell, keep going" - Winston Churchill
> >
> >
> >
> >              "Timothy Adair"
> >              <tadair@prairiefa
> >              rms.com>                                                    To
> >              Sent by:                  midrange-l@xxxxxxxxxxxx
> >              midrange-l-bounce                                          cc
> >              s@xxxxxxxxxxxx
> >                                                                    Subject
> >                                        Re: DSPF numeric field editing
> >              02/08/2010 11:49
> >              AM
> >
> >
> > That's it.  That's the answer I needed.
> >
> > Now that I know what's happening, I think I've found a way around this.
> >
> > Thanks for the quick response (and my wall thanks you too).
> >
> > TA
> >
> >
> >
> >
> > "Alan Shore"<AlanShore@xxxxxxxx>  wrote in message
> > news:mailman.7062.1265646857.2580.midrange-l@xxxxxxxxxxxxxxx
> >
> > Hi Timothy
> > I think that this will be the answer to your problem
> > As soon as an indicator associated with an ERRMSG is switched on, then any
> > change to the DSPF buffer is basically ignored
> > You can use debug and you will see that values on the DSPF change AFTER
> > the
> > index is switched on, but the DSPF display will NOT reflect that
> > We came across this a couple of years ago. The program acted as if it was
> > controlled by gremlins
> > If you look up ERRMSG in the ibm reference manual, this is as designed.
> > I hated it
> > We have since changed that program to use a message file instead, which
> > was
> > NOT a small project, but now that its complete - NO GREMLINS and things
> > are
> > displayed as expected
> >
> >
> > Alan Shore
> > Programmer/Analyst, Direct Response
> > E:AShore@xxxxxxxx
> > P:(631) 200-5019
> > C:(631) 880-8640
> > "If you're going through Hell, keep going" - Winston Churchill
> >
> >
> >
> >              "Timothy Adair"
> >              <tadair@prairiefa
> >              rms.com>                                                    To
> >              Sent by:                  midrange-l@xxxxxxxxxxxx
> >              midrange-l-bounce                                          cc
> >              s@xxxxxxxxxxxx
> >                                                                    Subject
> >                                        DSPF numeric field editing
> >              02/08/2010 11:26
> >              AM
> >
> >
> > I've searched the archives but have been unable to find any previous
> > reference to this issue.
> >
> > I have a DSPF with a numeric field (11,4) where the user is allowed to key
> > in a number with or without a decimal point.  If they key in a decimal
> > point, then the program takes that number as the actual value.  If not,
> > then
> > the program assumes 4 decimal places and divides the entered number by
> > 10,000.  (Please don't ask why - I've fought that battle and lost.  Yes
> > our
> >
> > users are still in the dark ages.)
> >
> > So far, so good.
> >
> > Let's say they key in the number "2" and press Enter.  The program will
> > assume four decimal places and divide by 10,000.
> >
> > So far, still good.
> >
> > If there are no errors or messages, the program will go back to the first
> > screen and allow the user to select another record to be updated.  But if
> > there is an error or message (my message, not system message) the program
> > loops around and does another EXFMT and the number shows up exactly as
> > they
> >
> > keyed it in, ignoring any editing, even though the value is correct
> > (0.0002,
> > verified in Debug).  I've tried an edit code and an edit word - neither
> > works.
> >
> > Is there some kind of quirk with a numeric display field that ignores any
> > editing if there's an ERRMSG on?
> >
> > Here is the field DDS:
> >
> > A            FUTPRC        11Y 4B 14 28COMP(GE .0000)
> > A                                      DSPATR(PC)
> > A                                      EDTWRD('0      .    ')
> > A  41                                  ERRMSG('Price Below Cost.  Press
> > F5-
> > A                                       to override.' 41)
> > A  42                                  ERRMSG('No Cost Available' 42)
> > A  49                                  ERRMSG('Both fields are required'
> > 4-
> > A                                      9)
> > A  40                                  ERRMSG('Abnormal price - press
> > ENTE-
> > A                                      R to accept.')
> > A  40                                  COLOR(RED)
> >
> >
> > Please help...  I now have a dent in my wall, and a headache.
> >      
>