>>
>> It would become more difficult to do this on chil101/child02 and
>> child11/child12 right?
> right - the appraoch would work best on groups of data - this is a
> limitation that makes authoring for cases like yours ugly.
>
I know, the businesspeople in our company always want something that is not
standard ;-)
> while i think it's a nice idea to have such partial validation it's
> maybe still not the silver bullet for all these rendering issues - and
> that's what i think some of these issues are really related to - in many
> of these use cases it's not so much a questions of *what* gets validated
> but *when* to actually *render the alerts* if any.
Oh, yes, I 1000% agree
>
> we still have the 'visited' construct in Chiba which we introduced on
> the way - the idea behind this was to say: 'don't show alerts for
> controls the user not even had a chance to edit'.
>
Hmmm... I've never seen that, That could solve our problem completely.
Validation of fields that
are visited but in error may happen anytime. I can get that 'sold' to the
business easily
> are these semantics compatible with your usecase?
>
Yes, I think so. We are still in development fase for a couple of months so no
hurries.
> currently the implementation is quite simplistic and sets 'visited' to
> true whenever updateControlValue is called. maybe this feature can be
> extended in some way to be available via javascript... ? ideas?
>
Definately sounds good. If I find the time somewhere I'll look into this.
Ronald
--
Kijk niet terug, maar kijk naar mij
Don't look back, just look at me
-------------------------------------------------------
This SF.Net email is sponsored by:
Sybase ASE Linux Express Edition - download now for FREE
LinuxWorld Reader's Choice Award Winner for best database on Linux.
http://ads.osdn.com/?ad_id=5588&alloc_id=12065&op=click
|
