Can someone point me to the initial discussion around the original
ConfigurationSource changes? I have some concerns but I'd like to be
able to study up to be sure I understand the decisions that were made
(despite this thread's implying that the changes aren't set in stone).
On Tue, Jun 12, 2012 at 3:50 PM, Gunnar Morling <gunnar(a)hibernate.org> wrote:
Any preferences or other suggestions?
2012/6/2 Gunnar Morling <gunnar(a)hibernate.org>:
> 2012/6/1 Emmanuel Bernard <emmanuel(a)hibernate.org>:
>> We can also imagine a specific provider offering an alternative source of
configuration and using this interface to expose it.
>> That was the rational behind the use of "typically".
> Maybe some bits of explantation would make it clearer?
> "Return information retrieved from the user-provided default
> configuration. While loaded from the <i>META-INF/validation.xml</i>
> file by default, implementations are free to provide other means of
> default configuration such as a properties file."
>>> * What do I return if there is no validation.xml? null? Or the class names of
the implementations default implementations?
>> Good question, I'm in favor of null. anyone against that?
> +1 for returning null.
>>> * Last but not least, why ConfigurationSource?
>> Technically this is the configuration but we already have a programmatic
>> - XmlConfiguration
>> - StaticConfiguration
>> - ConfigurationSource
>> - ConfigurationParameters
>> - ?
> Some more ideas:
> - DefaultConfiguration
> - UserDefaultConfiguration
> - ExternalConfiguration
> - BootstrapConfiguration
> I find the term "source" a bit mis-leading, as I would regard
> something like "XML" or "properties file" as source, but the
> object from that source as "configuration". I think I'm preferring
> (User)DefaultConfiguration, as it IMO clearly describes the object's
> I've created BVAL-293  for any clarifications/refinements around
>  https://hibernate.onjira.com/browse/BVAL-293
beanvalidation-dev mailing list