[cdi-dev] do we like to define a DeploymentException type?

[Mark Struberg]

Yes, was thinking in the same direction. We would still need to do this container specific detection for CDI-1.0 containers, but at least we would get a well defined behaviour for CDI-1.1.

This would take a while until getting picked up though, as most Extensions will also aim to run on CDI-1.0 for a long time ...

LieGrue,
strub



----- Original Message -----
> From: Jason Porter <lightguard.jp at gmail.com>
> To: Mark Struberg <struberg at yahoo.de>
> Cc: cdi-dev <cdi-dev at lists.jboss.org>
> Sent: Friday, June 15, 2012 5:31 PM
> Subject: Re: [cdi-dev] do we like to define a DeploymentException type?
> 
> Having CDI define (even one) exception would be nice and would make testing 
> extensions for portability much easier. As it currently stands you have to test 
> for implementation specific exceptions. 
> 
> Sent from my iPhone
> 
> On Jun 15, 2012, at 1:48, Mark Struberg <struberg at yahoo.de> wrote:
> 
>>  Hi!
>> 
>>  While reviewing the DeltaSpike MessageBundleExtension, I (again) came to 
> the point where I didn't knew which Exception to take. 
>> 
>> 
>>  It might be nice to have a DeploymentException which indicates that a CDI 
> Extension detected a logical problem which is _not_ caused by some technical 
> problems.
>> 
>>  E.g. if a certain annotation must not be used in conjunction with another 
> one. In our case @MessageBundle must only be used on Interfaces. We collect all 
> those problems during ProcessAnnotatedType and add them as 
> AfterBeanDiscovery#addDefinitionError.
>> 
>>  A well defined DeploymentException could help indicating such 
> 'logical' problems, wdyt?
>> 
>>  LieGrue,
>>  strub
>> 
>>  _______________________________________________
>>  cdi-dev mailing list
>>  cdi-dev at lists.jboss.org
>>  https://lists.jboss.org/mailman/listinfo/cdi-dev
>