Just one note: if you do plan a large refactoring I would suggest to
split it into two steps. In first step remove the @deprecated flag and
push changes, then we can easily release 2.5.15. And postpone the
second step (the refactoring) till 2.6.

I do not expect large changes in 2.5.x codebase


Regards
-- 
Łukasz
+ 48 606 323 122 http://www.lenart.org.pl/

2018-01-04 11:56 GMT+01:00 Yasser Zamani <[email protected]>:
>
>
> On 1/4/2018 10:58 AM, Lukasz Lenart wrote:
>> As far I understand these annotations are representation of the same
>> properties which can be used in XXX-conversion.properties files
>> https://struts.apache.org/core-developers/type-conversion.html#collection-and-map-support
>
> Thanks Łukasz! With further investigation following above link, I
> discovered those codes are copied from `DefaultConversionFileProcessor`.
> Then I saw in it ...
>
>>                     //for properties of classes
>>                     else if 
>> (!(key.startsWith(DefaultObjectTypeDeterminer.ELEMENT_PREFIX) ||
>>                             
>> key.startsWith(DefaultObjectTypeDeterminer.KEY_PREFIX) ||
>>                             
>> key.startsWith(DefaultObjectTypeDeterminer.DEPRECATED_ELEMENT_PREFIX))
>>                             ) {
>
> i.e. all ORs are NOTed. Now I understand what's the problem with
> `DefaultConversionAnnotationProcessor` :) Copier of those codes wished
> to remove NOT but wrongly just applied on first condition and rewrote it
> as tc.rule() != ConversionRule.ELEMENT while `NOT` should be refactored
> to all inside () i.e. tc.rule() != ConversionRule.ELEMENT && tc.rule()
> != ConversionRule.KEY && tc.rule() != ConversionRule.COLLECTION.
>
> Also I found a lot of XXX-conversion.properties inside test resources
> and now, I could coveralls my PR DefaultConversionAnnotationProcessor
> without knowing underlying details via copying current tests ;)
>
> Regards.
>

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

Reply via email to