Hm, should we leave out the annotations from 3.6 to have more time for
building consensus? Otherwise this will delay the 3.6 release...
Benedikt
Matt Sicker schrieb am Mi. 19. Apr. 2017 um 01:33:
> You can work around annotation permissiveness by implementing an annotation
> processor to fail c
You can work around annotation permissiveness by implementing an annotation
processor to fail compilations with improper combinations of annotations,
but I feel that the annotation+enum approach is a lot simpler to enforce.
On 18 April 2017 at 04:40, sebb wrote:
> I think ThreadingBehavior is ve
I think ThreadingBehavior is very confusing as an enum.
I expect enum values to be mutually exclusive, but STATELESS is also IMMUTABLE.
And of course IMMUTABLE is SAFE.
How do I know which enum to use?
I don't think this is a good solution to the potential problem of
someone using both @ThreadSa
Hi,
> Am 18.04.2017 um 08:46 schrieb Gary Gregory :
>
> Because if you use multiple annotations, you (or tools) can write silly
> things like:
>
> @ThreadSafe
> @NotThreadSafe
> public class Foo {
> …
> }
Thank you for the clarification. It makes complete sense to me. Given this
reason I’m f
Because if you use multiple annotations, you (or tools) can write silly
things like:
@ThreadSafe
@NotThreadSafe
public class Foo {
...
}
Gary
On Mon, Apr 17, 2017 at 11:27 PM, Benedikt Ritter
wrote:
> Hello Gary,
>
> Can you please explain why you think it is better to use a single
> parame
Hello Gary,
Can you please explain why you think it is better to use a single parameterized
annotation over several individual annotations?
Thank you!
Benedikt
> Am 17.04.2017 um 20:54 schrieb ggreg...@apache.org:
>
> Repository: commons-lang
> Updated Branches:
> refs/heads/master 5242157df
