Hi Stian,
Yes, I think you're correct in pointing out this. I believe the intention was for the returned URI to have only a single "../". I'm not sure what happened to your issue (I don't think anyone removed it on purpose), so please file it again.
We'll try to address it as soon as possible. Thanks.
-- Santiago
On Apr 21, 2013, at 9:48 PM, Stian Soiland-Reyes <stian_at_soiland-reyes.com> wrote:
> The issue seems to have been deleted from Jira. That does not make it go away.
>
> The examples in the javadoc of "Proposed Final Draft" downloaded from http://jcp.org/en/jsr/detail?id=339 is different from the one online, but also wrong:
>
> Request URI: http://host:port/app/root/a/b/c
> Supplied URI: a/d/e
> Returned URI: ../../d/e
>
> This would resolve to http://host:port/app/root/d/e
>
> Request URI: http://host:port/app/root/a/b/c
> Supplied URI: http://host:port/app/root/a/d/e
> Returned URI: ../../d/e
>
> This would resolve to http://host:port/app/root/d/e
>
>
>
> On Mon, Apr 22, 2013 at 2:38 AM, Stian Soiland-Reyes <stian_at_soiland-reyes.com> wrote:
>
> Hi, I am new to the JCP process and just recently started using JAX-RS through CXF. Sorry for just jumping in..
>
>
> I just reported this issue about the javadoc for UriInfo.relativize() having a misleading (and wrong) example:
>
> https://java.net/jira/browse/JAX_RS_SPEC-398
>
> and then realised you had scheduled JAX-RS 2.0 for tomorrow! :-)
>
> I don't know if this would be blocking the release - but as it led to the wrong implementation in CXF I thought you might want to check it, and so set it for Critical.
>
> The quickest fix is just to add the terminating / at the example, so a very small change.
>
>
> --
> Stian Soiland-Reyes
> Manchester, UK
> http://soiland-reyes.com/
>
>
>
> --
> Stian Soiland-Reyes
> Manchester, UK
> http://soiland-reyes.com/