declare variable - bug or poor design?

[Ken Cole]

Joe,

Look at it this way, since variables must be pre-defined in filePro, so
they have a size and edit in the token table (tok file),  that is just the
way filePro has always worked, how else can rclerk use this syntax except
to re-tokenise each and every time it loads which means it would actually
be dclerk? :-)

If the token table concept didn't exist we could have array's with dynamic
length as well but we don't! :-(

Cheers

Ken

On Thu, Feb 14, 2013 at 10:24 AM, Joe Chasan <joe at magnatechonline.com>wrote:

> i understand why its happening, i just think its poor implentation if
> fp allows for this feature but it is only available in dclerk.
>
> -joe
>
> On Thu, Feb 14, 2013 at 10:19:01AM +1100, Ken Cole wrote:
> > Joe,
> >
> > I see this as the "normal" state for compiled/tokenised code as is
> required
> > for rclerk.
> >
> > Since dclerk technically in memory re-tokenises every time you run the
> > processing table, basically the only difference between rclerk and
> dclerk,
> > I fully understand why dclerk works and rclerk doesn't until
> re-tokenised.
> >
> > rclerk has created the token table (tok file) entry for test_field the
> last
> > time it was tokenised and that definition will stay in the token table
> > until re-tokenised.
> >
> > The code you used is a short cut definition, not dynamic definition,
> unless
> > you are using dclerk.
> >
> > I don't see this as a bug or poor design.
> >
> > Regards
> >
> > Ken
> >
> > On Thu, Feb 14, 2013 at 9:26 AM, Joe Chasan <joe at magnatechonline.com>
> wrote:
> >
> > > consider the following line of code:
> > > declare local test_field(len(1),edit(1))
> > >
> > > then you code some routines around test_field and figured you have the
> > > system beat, you won't have to change dummy field length/type if field
> > > #1 changes at all.  great for reusable code, libraries, etc, or so I
> > > thought.
> > >
> > > today i changed field length.
> > >
> > > in define processing, if you press <F6><L> it correctly reports the
> > > current field length/type of test_field
> > >
> > > BUT - if you do not save the processing table and retokenize, rclerk
> > > still has the old length/type (dclerk works).
> > >
> > > If this construct is not adjusted at runtime (even with quickstart),
> > > what is the point of this "feature"?
> > >
> > > Bug?  Poor design?
> > >
> > > (5.6.11/sco - reproducable on sco5 and sco6)
> > >
> > > --
> > > -Joe Chasan-                           Magnatech Business Systems, Inc.
> > > joe - at - magnatechonline -dot- com   Hicksville, NY - USA
> > > http://www.MagnatechOnline.com         Tel.(516) 931-4444/Fax.(516)
> > > 931-1264
> > > _______________________________________________
> > > Filepro-list mailing list
> > > Filepro-list at lists.celestial.com
> > > Subscribe/Unsubscribe/Subscription Changes
> > > http://mailman.celestial.com/mailman/listinfo/filepro-list
> > >
> > -------------- next part --------------
> > An HTML attachment was scrubbed...
> > URL:
> http://mailman.celestial.com/pipermail/filepro-list/attachments/20130214/63d4a4ae/attachment.html
> > _______________________________________________
> > Filepro-list mailing list
> > Filepro-list at lists.celestial.com
> > Subscribe/Unsubscribe/Subscription Changes
> > http://mailman.celestial.com/mailman/listinfo/filepro-list
> >
> --
> -Joe Chasan-                           Magnatech Business Systems, Inc.
> joe - at - magnatechonline -dot- com   Hicksville, NY - USA
> http://www.MagnatechOnline.com         Tel.(516) 931-4444/Fax.(516)
> 931-1264
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.celestial.com/pipermail/filepro-list/attachments/20130214/e8a944d2/attachment.html