[Catalyst] State of the art in form validation; opinion poll… Model based forms/validation?

[Bill Moseley]

On Tue, Dec 04, 2007 at 05:47:27PM -0800, Ashley Pond V wrote:
> Putting the form profiles in YAML seems to work great (even though I  
> hate the fairly ridiculous arrays of arrays grammar). For example:
> 
...
> 
> As crazy as it is, I like having it out of the code and into the  
> config stuff.

Specifically into the config?  Or just out of the controller?

I think it would not be hard to make any of the form systems use YAML
to read in their config, but when I've done that I find it quickly
gets too limiting for specifying validation.


> I'm leaning more and more to having *all* of this stuff defined  
> (validation patterns, user messages for invalid input, clues for CGI  
> fields) in the Model (DBIC; I love it and never find time to try  
> Rose). Sort of a super phat Model. If it's not defined exclusively  
> there, it ends up defined there (via schema), the controller,  
> possibly the templates (messages), and even client-side in JS. I know  
> I'm showing my tremendous grasp of the obvious when I say, this is  
> just effing awful.

Again, I think you can only put so much in storage, so you need a
place for the form logic code that can't be represented in simple
regular expressions, ranges, etc.

The Rose form code is very cool.  That's what I based my forms on.

I don't find automatic form markup creation that useful -- the forms
always seem to need customization.  So, I generate the markup with
templates.  So my forms end up looking like:

http://search.cpan.org/src/HANK/Catalyst-Plugin-Form-Processor-0.05/example/FormProcessor/root/templates/user/edit.tt

Here's an example of how my forms work in Catalyst:

http://search.cpan.org/~hank/Catalyst-Plugin-Form-Processor-0.05/lib/Catalyst/Plugin/Form/Processor.pm

http://search.cpan.org/~hank/Form-Processor-0.15/lib/Form/Processor.pm


-- 
Bill Moseley
moseley at hank.org