[Product-Developers] Re: formlib vs z3cform which one to use?

Martin Aspeli optilude at gmx.net
Wed Jul 30 22:19:35 UTC 2008


Daniel Nouri wrote:
> Martin Aspeli writes:
> 
>> David Bain wrote:
>>> Now that plone.app.z3cform is documented and reasonable usable should we
>>> consider replacing formlib with z3cform?
> 
> I'd recommend plone.z3cform.  A couple of people are using it in
> production and it works very well.  It's seen a lot of bug fixes and
> real life use.

I'm about six months off recommending it everyone. :) I think it's 
really good, and I use it personally for some things, but it's fair to 
point out it's a bit cutting edge.

>> Depends. z3c.form is still cutting edge. Last week, Daniel refactored
>> and broke most existing forms. :-) I suspect we're a month or two off
>> a beta or rc quality Plone/z3c.form integration package that does
>> everything you want.
> 
> What's there is very stable.  No reason to believe it's not "beta or rc
> quality".  What's missing is inline KSS edit; inline validation on the
> other hand is there.  And probably a couple of other things that we
> never needed.

I'd also miss a few contributed widgets (like captcha and 
UberSelectionWidget, lack-of-finished-UI-notwithstanding). I think it's 
mainly documentation and amount of available examples that'd put me off 
recommending it to someone who wanted to stay on the boring side of the 
technology curve.

> Whatever "properly supported" means.  I've run across a lot of bugs in
> the past working with Five's formlib integration.  Like bugs with
> internationalization (IE6 and non-ASCII form values...) and subforms
> support.

Right. Formlib certainly has its quirks. But Plone depends on it, which 
means that Plone needs it to at least work somewhat well. :-)

> I guess you're referring to an exotic feature or two that the formlib
> integration supports that plone.app.z3cform does not yet.  I doubt that
> you can say it's better documented (in or outside Plone), given the
> excellent documentation that comes with z3c.form.  The fact that almost
> no Plone specific documentation is necessary for z3c.form is in fact an
> advantage.

Please don't get me wrong - like I said, I love plone.z3cform and I 
think z3c.form is a better library than formlib ever was. It's just fair 
to point out that this is quite new in the Plone world.

Martin

-- 
Author of `Professional Plone Development`, a book for developers who
want to work with Plone. See http://martinaspeli.net/plone-book





More information about the Product-Developers mailing list