This seems wrong to me. I can't find a spec anywhere on how to
properly interpret validations, but the way i read this, the
following is the requirement to use this component:
(items bound && nameKey bound && valueKey bound) || dataset bound
so to figure out when we show that validation error, we negate that,
which gives us
(items unbound || nameKey unbound || valueKey unbound) && dataset
unbound
which gives us:
<and>
<or>
<unbound name = "items"/>
<unbound name = "nameKey"/>
<unbound name = "valueKey"/>
</or>
<unbound name = "dataset"/>
</and>
Because right now, if only dataset is bound, then all three of items,
nameKey, and valueKey would be unbound, which would trip the first
part of the original <or>.
But my brain hurts every time I try to read these things. Something
about the validation requiring the negation forces me to focus all my
energy to figure out what the rules are supposed to be.
ms
On Sep 15, 2005, at 8:33 PM, Greg wrote:
> <validation message="Either items, nameKey and valueKey or
> dataset must be bound">
> <or>
> <and>
> <unbound name="items"/>
> <unbound name="nameKey"/>
> <unbound name="valueKey"/>
> </and>
> <unbound name="dataset"/>
> </or>
> </validation>
>
This archive was generated by hypermail 2.0.0 : Thu Sep 15 2005 - 22:28:16 EDT