RFR: 8145461 changes to @Deprecated annotation

Stephen Colebourne scolebourne at joda.org
Fri Apr 1 13:06:41 UTC 2016


My only (very minor) comment now is that it makes more logical sense
for the since() method to precede the forRemoval() method in the
source code (as since will be applied before forRemoval).

thanks
Stephen


On 1 April 2016 at 06:56, Stuart Marks <stuart.marks at oracle.com> wrote:
>
>
> On 3/30/16 6:03 AM, David M. Lloyd wrote:
>>
>> As I said in another forum, I think "condemned" is a bit too obtuse to
>> read,
>> linguistically (not to mention humorously ambiguous, as you yourself
>> pointed
>> out).  I prefer "plain langauge"; something more like "forRemoval" would
>> be
>> better IMO.  Also I agree with Stephen about the usefulness of specifying
>> an
>> "until" version, but I understand the counterpoints and this could also
>> just as
>> easily be done by a third-party library, especially now that JavaDoc is
>> presumably going to be much nicer (and thus hopefully more pluggable).
>
>
> Yes, we've had our bit of fun with "condemned." I've gotten a fair amount of
> in-person feedback that has been in the form of wrinkled noses and knitted
> eyebrows and comments like "isn't there a better word?" I admit that was my
> first impression as well, but I've gotten used to it. Nonetheless, it's time
> to change it.
>
> There doesn't seem to single word that expresses "we have decided to remove
> this in the future" unambiguously and without excess baggage.
>
> So, a compound word seems necessary. After kicking around "toBeRemoved" or
> "subjectToRemoval" I've come back around to David's suggestion here of
> "forRemoval". It seems a bit odd when considered by itself, but in actual
> usage it reads pretty well:
>
>     @Deprecated(since="9", forRemoval=true)
>     public void foo() { ... }
>
>
>
> For example, one say "The foo() method has been deprecated for removal."
>
> Here's an updated webrev with the rename, as well as other edits in response
> to feedback:
>
> http://cr.openjdk.java.net/~smarks/reviews/8145461/webrev.1/
>
> s'marks
>


More information about the jdk9-dev mailing list