Joern Turner probeerde me het volgende wijs te maken:
> Ronald,
>
> Ronald van Kuijk wrote:
>> Better schema2xforms support
>> - support namespaces
> there's a patch pending which i'll try to get into next release (if it
> works ;)
>
This is real great, I currently have to change all xsd's we get from our
standardization
department to get them working (can't wait for the next milestone of OOo)
>>
>> Second is the thing you already asked for, some more client-side validation
> yep. first thing to start by ourselves after 1.0 - if patches occur
> maybe earlier.
>
> improved usability through (generated) script support is one of our main
> priorities after 1.0
>>
>> Third is client-side adding of another element to repeats.
> definitely.
>
> a short draft of our current ideas:
> first, to enable the client to execute the interesting features of
> XForms (like repeats) we need to transport the instance-data to the
> client in some way.
The instace-date? Why? isn't it enough to know how to generate a new node id on
the client and
communicate that to the model on the server side? Can't that be done by xmlrpc?
- click on 'add an element'
- add it visually (taking into account a possible max number of elements)
- use xmlrpc (or plain http) to also add the node in the server-side model
- get a node-id back and put that in the (possibly nested) elements
>
> the client should be able to execute as much actions as possible on the
> client without the need to contact the server. to enable this we think
> of a data-island in the generated output which may be manipulated by
> javascript to set values, insert + delete nodesets and revalidate as
> much as reasonable.
Aren't you afraid that you are redeveloping the convex applet version into a
javasript version?
I'd keep it 'simple' see below. For more complex things I'd advice the applet
version.
> changes to the local model may then be synchronized
> via xmlrpc with Chiba running on server.
>
> it's still an open question how far this script support can/will
> reasonably grow.
Besides the repeats, I'd focus on simple length, pattern integer, alphanumeric
checks first. Those
are the ones our (internal) clients miss most. They are used to the fact that
for dependency
checks there currently has to be a round trip to the server.
Ronald
--
Kijk niet terug, maar kijk naar mij
Don't look back, just look at me
-------------------------------------------------------
SF email is sponsored by - The IT Product Guide
Read honest & candid reviews on hundreds of IT Products from real users.
Discover which products truly live up to the hype. Start reading now.
http://productguide.itmanagersjournal.com/
|
