Sebastian wrote:
Since this error is on the aggregate-widget and that one is not actually on the template, there is no position where the (!) should be next to.
Therefor you need to explitely position it with a separate <ft:validation-error />
see also here: http://marc.theaimsgroup.com/?l=xml-cocoon-dev&m=108176649831199&w=2
and here:
http://marc.theaimsgroup.com/?l=xml-cocoon-dev&m=107486651220426&w=2
and let's thank Vadim :-)
Hmm. Very hacky :) Anyway. I think there is another bug there :) I don't know
if it's well known or if I should populate it on bugzilla.
from what I get now this sounds like a good idea
Let me show my beautiful example :)
it truly is a work of art!
<fd:aggregatefield id="nip" required="false">
<fd:label>NIP:</fd:label>
<fd:datatype base="string">
<fd:validation>
<fd:nip>
<fd:failmessage>Invalid NIP number.</fd:failmessage>
</fd:nip>
</fd:validation>
</fd:datatype>
<fd:split pattern="([0-9]{3})-([0-9]{2,3})-([0-9]{2})-([0-9]{2,3})">
<fd:map group="1" field="part1"/>
<fd:map group="2" field="part2"/>
<fd:map group="3" field="part3"/>
<fd:map group="4" field="part4"/>
<fd:failmessage>Not a valid 10-digit NIP number.</fd:failmessage>
</fd:split>
<fd:combine expression='Concat(part1, "-", part2, "-", part3, "-", part4)'/>
<fd:widgets>
<fd:field id="part1">
<fd:label>NIP:</fd:label>
<fd:datatype base="string"/>
</fd:field>
<fd:field id="part2">
<fd:label>-</fd:label>
<fd:datatype base="string"/>
</fd:field>
<fd:field id="part3">
<fd:label>-</fd:label>
<fd:datatype base="string"/>
</fd:field>
<fd:field id="part4">
<fd:label>-</fd:label>
<fd:datatype base="string"/>
</fd:field>
</fd:widgets>
</fd:aggregatefield>
"fd:split pattern" doesn't work correctly. For example. If I'll fill just single
field then I get a message "Not a valid 10-digit NIP number.". That's correct.
If I fill all fields with integers I get "Invalid NIP number.". That's correct
and not correct :) In fact it is an invalid NIP, but I can provide value
1-1-1-1 which not match the pattern and I should get "Not a valid 10-digit NIP
hm, supposing I understand what you are saying here my guess is that the validation on subwidgets isn't triggering the concattenation and subsequent validation on the aggregate level?
hm, this poses another interesting use case to the fact of setvalue/validation sequence (haven't looked in detail yet, maybe someone else has a better idea)
if my guess is close then you could probably work around this by adding additional validation rules on the split fields: i.e. adding validation rules enforcing the split-patterns on those substrings
(or even make their datatypes int and validate for minimum values or so, will not work for leading zeros though)
HTH -marc=
number.". Moreover if I will fill all fields but some of it will be a text instead of number I will get an exception from my NIP validator :) because of course it'll be validated against NIP rule not pattern. I've tried to change data types for parts to integer but in this case it doesn't validate at all and always returns "Not a valid 10-digit NIP number.". Bug or not bug? And is it a well known bug? :)
Kindly regards,
-- Thought by thought we see our own mistakes.
mailto: [EMAIL PROTECTED]
-- Marc Portier http://outerthought.org/ Outerthought - Open Source, Java & XML Competence Support Center Read my weblog at http://blogs.cocoondev.org/mpo/ [EMAIL PROTECTED] [EMAIL PROTECTED]
--------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
