[bv-dev] Clarification regarding executable parameter/return value validation with group sequence redefinition

[Guillaume Smet]

Hi Matt,

On Tue, Apr 3, 2018 at 11:22 PM, Matt Benson <mbenson at apache.org> wrote:

> Hello, I am encountering failures on three tests from
> MethodValidationTest in the TCK:
>
> * #constructorParameterValidationWithRedefinedDefaultGroupSequence()
> * #methodParameterValidationWithRedefinedDefaultGroupSequence()
> * #methodReturnValueValidationWithRedefinedDefaultGroupSequence()
>
> In each case I am failing to pick up a @Size constraint from the
> Default group of an Item parameter or Order return value. However, due
> to the group sequence redefinition of Basic, Default on the bean class
> in question I am confused as to why I am expected to pick up the
> failures from the Default group after having encountered a @NotNull
> constraint violation in the Basic group. Thanks for any help here.
>

I would tend to agree with you. (The fact that Item also has a different
Basic group does not help to understand the issue though).

AFAIU the spec (
http://beanvalidation.org/2.0/spec/#constraintdeclarationvalidationprocess-validationroutine-methodconstructorvalidation),
we should do:
- For each group
  - For each parameter, apply constraints
  - For each cascading parameter, cascade

This in no particular order if there is no group sequence. If a group
sequence is defined, we should exit as soon as we have a violation for a
given group

Currently, the RI does:
- For each group
  - For each parameter, apply constraints
    - exit as soon as I have a constraint if group sequence
- For each group
  - For each cascading parameter, cascade
    - exit as soon as I have a constraint if group sequence

In the aforementioned case:
1/ we apply the Basic constraints for the parameter constraints
2/ we exit as soon as we have a violation, we don't validate the Default
group
3/ we apply cascading
   - no violation for the Basic group
   - we validate the default group
   - we exit as soon as we have a violation

As per spec, I think we should exit just after 2/

Gunnar, any insights?

-- 
Guillaume
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.jboss.org/pipermail/beanvalidation-dev/attachments/20180404/2ba5e62a/attachment-0001.html