I do agree with you on the fact that Intake is not much good for
anything except validating HTML forms.  However, that is a common
problem that must be solved for all web based (HTML) applications.  For
this, Intake is a great tool.  That is why I am contributing my time to
maintain and improve upon it.

Setting a field to a default value is incredibly easy if you populate
the group from a business object.  I use the default value in the XML
file only for default selection in select boxes.m  However, your point
it taken.  This is not obvious to someone new to Intake.

I do like the idea of having a more generic validation layer that is not
coupled to ValueParser or Velocity.  Part of my internal application
imports data from fixed format files (uploaded through a web interface).
I would love to have a service that will allow me to define validation
rules for that type of input.  The same would go for XML submission of
data as well.

> -----Original Message-----
> From: Henning P. Schmiedehausen [mailto:[EMAIL PROTECTED]] 
> Sent: Friday, January 17, 2003 11:11 AM
> To: [EMAIL PROTECTED]
> Subject: Re: Proposed change to intake validation.
> 
> 
> "Quinton McCombs" <[EMAIL PROTECTED]> writes:
> 
> >Notice that doAssertValidity is called after testing the 
> required rule 
> >bute before any other others.  Is there a reason for this?  It seems 
> >that it might be better for the basic forms of validation to occur 
> >before the subclass begins its validation.
> 
> >Thoughts?
> 
> Well, while I really appreciate your work on the intake, to 
> me it is some sort of "beating a dead horse". Intake really 
> is a mess. You can use it exactly as it was designed but 
> everything else is pretty much obfuscated. Intake works 
> exactly on one class of things (HTML forms), is heavily tied 
> to the VelocityService (tool) and Turbine RunData
> (fields) and if you want to preset a field with anything else 
> but a fixed value from the XML form: Good luck. =:-)
> 
> I did read a (german) article about field validation quite a 
> while ago and there was a generic XML framework presented to 
> validate input fields from a bean, so we could get a really 
> thin layer which represents an arbitrary HTML form as a bean, 
> then call this framework to validate the fields and put them 
> into the business objects.
> 
> And then there is commons-validator which, while having some 
> really gross problems (look at 
> o.a.commons.validator.GenericTypeValidator.
> Each of these methods should be in a separate class derived 
> from a common base type) it is a common framework for a job 
> we need again and again. And I would really love to have a 
> framework which can validate arbitrary beans (hey, what did I 
> get from the XML-RPC call just a millisecond ago) and have 
> intake just a thin layer for form parsing and bean 
> representation (Something which should be quite easy with 
> commons-beanutils).
> 
> But these are ideas that I kept under covers because I won't 
> find any time in the near future to implement such a service 
> (unless someone needs this and is willing to pay. =;-)
> 
>       Regards
>               Henning
> -- 
> Dipl.-Inf. (Univ.) Henning P. Schmiedehausen       -- 
> Geschaeftsfuehrer
> INTERMETA - Gesellschaft fuer Mehrwertdienste mbH     [EMAIL PROTECTED]
> 
> Am Schwabachgrund 22  Fon.: 09131 / 50654-0   [EMAIL PROTECTED]
> D-91054 Buckenhof     Fax.: 09131 / 50654-20   
> 
> --
> To unsubscribe, e-mail:   
> <mailto:turbine-dev-> [EMAIL PROTECTED]>
> For 
> additional commands, 
> e-mail: <mailto:[EMAIL PROTECTED]>
> 
> 

--
To unsubscribe, e-mail:   <mailto:[EMAIL PROTECTED]>
For additional commands, e-mail: <mailto:[EMAIL PROTECTED]>

Reply via email to