Mailing-List: contact dev-help@cxf.apache.org; run by ezmlm
Precedence: bulk
Reply-To: dev@cxf.apache.org
Received-SPF: pass (athena.apache.org: local policy)
Message-ID: <001e01c9a331$991ba2e0$0c02050a@emea.progress.com>
From: "Sergey Beryozkin" <sberyozk@progress.com>
To: <dev@cxf.apache.org>
References: 
 <D33A5FB583A9F843BB5A128BE3FD457E90F1874F64@LNWEXMBX0144.msad.ms.com>
 <01eb01c9a326$a1226380$0c02050a@emea.progress.com>
 <e77cb25c0903120912j15f3343mf7a53f5572b47278@mail.gmail.com>
Subject: Re: Multiple HTTP methods per method in JAXRS (Re: Possibly retarded
 question)
Date: Thu, 12 Mar 2009 16:42:55 -0000
MIME-Version: 1.0
Content-Type: text/plain;
	format=flowed;
	charset="iso-8859-1";
	reply-type=original
Content-Transfer-Encoding: 7bit

> > Don't make Roy Fielding have to come around to your house!

This is funny :-)

> I think if you use GET and POST on the same URI  then you are
> innappropriately 'crossing the streams' and making it ambiguous
> as to whether the invocation is idempotent or not.

That is true.
That said, one way of looking at having say both GET and POST at the same method is that a user is saying : I know
that some new clients call GET and while others (legacy ones) call POST (with POST clients in this example providing an unexpected 
HTTP method), which I reckon what Gary is after in this case. Typically users write filters to do it or provide multiple methods 
delegating to a common one. Supporting multiple HTTP methods on a single method would be just another way to achieve the same fairly 
specific goal.

Cheers, Sergey

----- Original Message ----- 
From: "Oisin Hurley" <oisin.hurley@gmail.com>
To: <dev@cxf.apache.org>
Sent: Thursday, March 12, 2009 4:12 PM
Subject: Re: Multiple HTTP methods per method in JAXRS (Re: Possibly retarded question)


>> I don't think it's a bad idea per se. JAXRS does not prohibit multiple
>> annotations per se (AFAIK), it's just undefined what happens.
>
> I think if you use GET and POST on the same URI  then you are
> innappropriately 'crossing the streams' and making it ambiguous
> as to whether the invocation is idempotent or not.
>
> Don't make Roy Fielding have to come around to your house!
>
> --oh