I forgot to mention in previous reply that one important constraint is
that Every single addition needs to fill out the Version value. The
default maven processing makes no use of it and proceeds as before.
Only in the IDE's subclass we will use it to throw exception or not.
If a request or parameter bean can ensure that every new addition in
the future contains the version value, then I'm fine with it.  Having
a new mandatory parameter seems like the safest way to go ahead..

That's why I also renamed some of the private member methods in the
validator implementation to make it more obvious what version is
applicable within the method..

Milos

On Fri, Jun 29, 2012 at 12:31 PM, Olivier Lamy <[email protected]> wrote:
> Agree it's hard to inject that. So that reduce possible backward comp issues.
>
> BTW what about using this bean/data structure rather than adding new
> parameters ?
>
> 2012/6/29 Milos Kleint <[email protected]>:
>> Is ModelProblemCollector intended for use outside of maven codebase?
>> MavenModelBuilder is hardcoding reference on DefaultMPC and there's a
>> few other implementations in tests or compat module only..
>>
>> Milos
>>
>> On Fri, Jun 29, 2012 at 11:49 AM, Olivier Lamy <[email protected]> wrote:
>>> Hi,
>>> The main issue I see is non backward comp for tools implementing their
>>> own ModelProblemCollector.
>>> I don't have issue to change signature but for future enhancement if
>>> needed here, I would prefer to see something more easy to change and
>>> don't break again backward comp in the future.
>>> So instead more parameters
>>>
>>> -    void add( ModelProblem.Severity severity, String message,
>>> InputLocation location, Exception cause );
>>>
>>> +   void add( ModelProblem.Severity severity, ModelProblem.Version
>>> version, String message, InputLocation location, Exception cause );
>>>
>>> I would prefer we use a bean which contains the needed informations.
>>> something like: void add( ModelProblemCollector
>>> modelProblemCollectorRequest ); (or an other name :-) ).
>>>
>>> Makes sense ?
>>>
>>> 2012/6/29 Milos Kleint <[email protected]>:
>>>> hello,
>>>>
>>>> I've prepared a patch for MavenModelBuilder and related code that
>>>> hopefully will improve the performance of NetBeans
>>>> integration/embedding. The basic idea is to have all validation
>>>> problems collected but only fail to build the Mavenproject instance
>>>> when serious base problems are found (validation level minimal).
>>>>
>>>> See http://jira.codehaus.org/browse/MNG-5306 for details and links to
>>>> patch. I haven't submitted to maven codebase for a while so I'd like
>>>> to have a review before integrating, Thanks.
>>>>
>>>> Milos Kleint
>>>>
>>>> ---------------------------------------------------------------------
>>>> To unsubscribe, e-mail: [email protected]
>>>> For additional commands, e-mail: [email protected]
>>>>
>>>
>>>
>>>
>>> --
>>> Olivier Lamy
>>> Talend: http://coders.talend.com
>>> http://twitter.com/olamy | http://linkedin.com/in/olamy
>>>
>>> ---------------------------------------------------------------------
>>> To unsubscribe, e-mail: [email protected]
>>> For additional commands, e-mail: [email protected]
>>>
>>
>> ---------------------------------------------------------------------
>> To unsubscribe, e-mail: [email protected]
>> For additional commands, e-mail: [email protected]
>>
>
>
>
> --
> Olivier Lamy
> Talend: http://coders.talend.com
> http://twitter.com/olamy | http://linkedin.com/in/olamy
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: [email protected]
> For additional commands, e-mail: [email protected]
>

---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]

Reply via email to