On Jun 7, 2011, at 9:08 PM, Noah Davis wrote:
>
>
> On Jun 7, 2:55 pm, "Diez B. Roggisch" <[email protected]> wrote:
>>> I've got a pair of tw.forms that are children of a custom widget. All
>>> the widgets are named, so the resulting fields have names like
>>> "person_form.firstname", etc. I can pass data to the forms at
>>> instantiation time (although using unqualified names like "firstname",
>>
>> This sounds suspicious. What do you mean by instantiation time?
>>
>
> Sorry, in the template itself, not when the widget is actually
> instantiated. The widget is passed to the template through
> templ_context, which I understand is the current accepted method for
> non-data objects. In the template the widget is realized like
> tmpl_context.my_compound(dict(person_form=dict(firstname=firstname...
> etc.
Ok, that's after the book.
>>> rather than "person_form.firstname"), but the kwargs dict passed to
>>> the controller contains keys as "person_form.firstname", etc.
>>
>>> This isn't a problem, although it seems a little counter intuitive,
>>> until I add validators... at which time the validators begin to fail
>>> with errors like "The input field 'person_form' was not expected.",
>>> which I suspect is directly related to the name mangling going on with
>>> the child/parent widget relationships.
>>
>> How exactly do you add validators to the mix?
>
> class PersonForm(TableForm):
> show_error = True
> fields = [
> TextField('firstname', validator=NotEmpty, label_text='First
> Name:', show_error=True),
> TextField('lastname', validator=NotEmpty(msg="test"),
> label_text='Last Name:'),
> TextField('phone', validator=NotEmpty, label_text='Phone:'),
> #FIXME: Validator for Phone
> HiddenField('method'),
> HiddenField('term'),
> ]
> submit_text='Add Person'
>
> Then add @validate(form=pform, error_handler=...) to the appropriate
> controller methods.
Hm. Try adding a strip_name=True to the encompassing widget. Or a
strip_name=False to the form.
It looks a bit though as if there is some incongruency here, between how forms
are renderend (and what value-structure they assume), and the validator
structure. Might in fact be a bug. If I find time, I take a look into it.
Diez
--
You received this message because you are subscribed to the Google Groups
"TurboGears" group.
To post to this group, send email to [email protected].
To unsubscribe from this group, send email to
[email protected].
For more options, visit this group at
http://groups.google.com/group/turbogears?hl=en.
