IIRC, about the template, it is good about expanding and not expanding
portions of the URL and query string. This might be interesting for
<link>s where you publish the URI parameters of the href, but not the
format of the href. Then the user just uses something like UriBuilder +
the template extension to create the URL to invoke on. Always the
server to have a lot of rich options for URL creation, yet keeps the URL
template opaque.
On 10/29/10 1:34 PM, Sergey Beryozkin wrote:
> Hi
>
> I'd like to start a new discussion with the purpose of trying to figure out
> what concrete advantages providing an alternative URI Template
> implementation [1] can bring to CXF JAXRS users.
>
> Andy and Bill have agreed that it can be good for generating concrete URIs.
>
> However using it for generating URIs only would basically make it impossible
> for users to use expressions possible with [1] in JAXRS Path annotations
> which is really a main point of providing an alternative mechanism which
> users could try while still relying on the JAXRS runtime to deal with the
> invocations.
>
> So I'd really like to see what concrete advantages users can get while using
> [1] for matching, as opposed to using the default JAXRS spec-compliant
> URITemplate.
>
> Comments are welcome and I'm hoping to contribute as well :-)
> thanks, Sergey
>
>
>
> [1] http://tools.ietf.org/html/draft-gregorio-uritemplate-04
>
--
Bill Burke
JBoss, a division of Red Hat
http://bill.burkecentral.com
|