[ckan-dev] validation library and deform.

[Seb Bacon]

I had a chat with David on the phone about this, which amounted to:

 - I think his suggested code is good
 - But: I think that the challenge of reusing things rather than
writing your own is too often ignored by developers.  If we aspire to
reuse and share and improve, I don't think we should roll our own so
often.
 - When I selected colander I thought it'd be me doing it, and that's
what I would've used.    But most importantly, it's looking like it's
David who'll be doing most of the work doing dictization and
validation, so I think it's important he works with something he
likes.  That's what counts the most.

So, apart from some very minor discomfort around point (2), I don't
feel strongly about it.  And ultimately the person doing the work
should enjoy doing it, that's the most important thing of all :)  I
would be interested to get feedback from Pyramids people who are using
colander, though.

Seb

On 2 April 2011 15:23, james at 3aims.com <james at 3aims.com> wrote:
> I strongly prefer your version so will have to take a look at how it might
> work with Deform. From what I remember though, we agreed on hand generated
> forms, so I'm not sure this will be an issue anyway.
>
> Cheers,
>
> James
>
> ----- Reply message -----
> From: "David Raznick" <kindly at gmail.com>
> Date: Fri, Apr 1, 2011 15:50
> Subject: [ckan-dev] validation library and deform.
> To: "CKAN Development Discussions" <ckan-dev at lists.okfn.org>
>
> Hello
>
> A while ago Colander was decided by Seb to use as a validation/conversion
> library.
>
> I have been doing the dictization work and in doing so came up with my own
> library to do this, called navl.  https://bitbucket.org/kindly/navl/src
>
> Obviously I prefer my own library, it is more flexible and easier to write
> validators and schemas for it.
>
> HOWEVER, colander is probably workable with. I am not 100% sure it will do
> *all* we will need and some workarounds may be needed.  I also know that I
> will find it frustrating to use.
>
> The main advantage of colander is that it Deform (a form library) needs
> Colander schemas for it to work.   Deform may save a lot of time in creating
> forms.
>
> So the question is.
>
> Will will be using deform??
>
> Seb said this was a minor point when looking for a
> validation/conversion library,  however it may be very useful for us to be
> able to use it.
>
> I am really not sure about this one.
>
> David
>
>
>
> _______________________________________________
> ckan-dev mailing list
> ckan-dev at lists.okfn.org
> http://lists.okfn.org/mailman/listinfo/ckan-dev
>
>