Re: Data validation, mapping, and error handling

["Aaron Bartell" <aaronbartell@xxxxxxxxx>]

> The trick is to know the tool before you do the design.

Very true.  I guess that is why I am here critiquing EGL so much - so best
practices can be developed to save others time.

Aaron Bartell
http://mowyourlawn.com


On Mon, Jun 9, 2008 at 1:31 PM, Joe Pluta <joepluta@xxxxxxxxxxxxxxxxx>
wrote:

> Aaron Bartell wrote:
> > > I haven't found the JSF keyword (if one exists) that will tell a field
> to
> > > use an alternate style if the field is in error.
> >
> > Here is an example of applying a custom class to a JSF field in error:
> >
> > JSF controller logic code:
> Again, I know how to do it in the controller... it's incredibly easy.  I
> can do it in EGL using the UIViewRoot and all is good.  The trick is
> trying to get the tool to do as much as possible, and that's why using
> validation is a good technique.
>
> > The issue I had is that JSF will do the built in validations first and if
> > any of them fail it will exit the JSF lifecycle and send a response back
> to
> > the screenbefore it gets to your business logic edits.
> Correct.  This is the exact same thing as using things like RANGE or
> VALUES on a display file.  For simple edits, it works great, for more
> complex edits it doesn't.  The other thing that often happens, at least
> in less mature frameworks, is that the inline editing presents a sort of
> generic error message, and often it's not nice enough for production work.
>
> > This was enough to
> > make me not use "the validation framework" unless the validations were
> > incredibly simple.  These are the types of "bummers" that bite you in the
> > butt 3 months after you have made the decision to go with a framework and
> > are the same types of "bummers" I am watching out for in EGL.  Not that I
> > won't choose EGL for some project eventually, it would just be great to
> not
> > learn the hard way.
> The trick is to know the tool before you do the design.  That's why I
> was willing to chase this down a little.  I don't consider this a
> "bummer" any more than I consider the VALUES keyword in DDS a bummer.
> It's a matter of understanding how it works before designing the system.
>
> Joe
> --
> This is the Web Enabling the AS400 / iSeries (WEB400) mailing list
> To post a message email: WEB400@xxxxxxxxxxxx
> To subscribe, unsubscribe, or change list options,
> visit: http://lists.midrange.com/mailman/listinfo/web400
> or email: WEB400-request@xxxxxxxxxxxx
> Before posting, please take a moment to review the archives
> at http://archive.midrange.com/web400.