On 06/03/2008, Andreas Marienborg <[EMAIL PROTECTED]> wrote:
> To follow up even further, what is the rationale behind validators
>  only running if all constraints pass? It makes for a rather
>  unintuitive process for the users? They get 3 red boxes, fix them, and
>  _then_ get another red box on something that seemed fine.
>
>  I can understand validators only running for fields where constraints
>  pass, but only after all seems a bit strange.

I originally made each step (constraints, inflators, validators,
transformers) an "all or nothing" effort because:
- firstly for inflators - they could be considered relatively
expensive, and so if there's any constraint errors on *any* fields,
you'll probably just be re-displaying the form again, and so probably
wouldn't want any inflators run.
- validators - I originally thought would be used for higher-level
"business logic" involving multiple fields. If you're needing to
validate on a single field, and it doesn't need inflated - then you
may be as well using a Constraint instead.

If you wanted validators to run if there's constraint errors on
unrelated fields, there would need to be a mechanism for the validator
to state which other fields should also be error-free (much like
others() on some constraints).
I think this would probably be a bit too much complexity, and error
prone, as the field list would need to be kept in sync with the
validation code.

Currently the only validator is HTML::FormFu::Validator::Imager::Size.
I've never yet needed to write any custom validators.

I hope this explains how we got where we are :)

Carl

_______________________________________________
HTML-FormFu mailing list
[email protected]
http://lists.scsys.co.uk/cgi-bin/mailman/listinfo/html-formfu

Reply via email to