On 6/13/06, Bill de hÓra <[EMAIL PROTECTED]> wrote:
Tim Bray wrote:
> On Jun 13, 2006, at 12:30 PM, Bill de hÓra wrote:
>
>> In short I'm looking to constrain what can result in a 2xx.
>
> It's my impression that the HTTP spec has an, if not perfect, pretty
> workable description of the appropriate usage of HTTP status codes.
> Would you be OK with a pointer to that, underlined for emphasis? -Tim
Let's make the discussion concrete since I think I may have
gotten lost along the way. Let's presume there is a bit of foreign
markup that a client wishes to store in an entry. If I PUT such an entry
<entry xmlns="http://www.w3.org/2005/Atom";
xmlns:myns="http//bitworking.org/mynamespace/">
...
<link rel="edit" href="http://example.org/entry/1/"/>
...
<myns:stuff>my special stuff</myns:stuff>
</entry>
to
http://example.org/entry/1/
which is a server that knows nothing about the 'myns' namespace
I can still receive a 200 Ok from the server. Doing a GET later on
http://example.org/entry/1/
returns
<entry xmlns="http://www.w3.org/2005/Atom";
xmlns:myns="http//bitworking.org/mynamespace/">
...
<link rel="edit" href="http://example.org/entry/1/"/>
...
</entry>
Note the absence of the myns:stuff element. The client can do this
all day long and receive 200 Ok every time and never get that namespaced
element to 'stick' in the entry. Is this the situation you are worried about?
-joe
--
Joe Gregorio http://bitworking.org
