Re: Comments on draft-dusseault-http-patch-08, was: [Fwd: I-D ACTION:draft-dusseault-http-patch-08.txt]

James M Snell wrote:
>> Section 2.1., para. 6:
>> OLD:
>>
>>     It is RECOMMENDED that Servers provide strong ETags for all resources
>>     for which PATCH is supported.
>>
>> Again, there's a set patch operations where the previous state of the
>> resource is irrelevant, such as for "append" operations.  I think they
>> should be allowed, thus this requirement is a problem.  Just state that
>> strong etags can help making sure that you apply the patch to the
>> version you want it   apply to, and let's leave it at that.
>>
> 
> If this were a MUST, I'd agree.  There are some extremely good reasons
> why an ETag should be provided, even if there are some good cases where
> they might not be useful.

If it's foreseeable that in some use cases they aren't useful, there 
shouldn't be SHOULD-level requirement to provide them.

>> Section 2.1., para. 8:
>> OLD:
>>
>>     The server MUST NOT ignore any Content-* (e.g.  Content-Range)
>>     headers that it does not understand or implement and MUST return a
>>     501 (Not Implemented) response in such cases.
>>
>> A MUST level requirement with a wild card Just Does Not Work.  I'm not
>> even sure what this is about.  As far as I understand what we could say,
>> if at all, is that entity headers sent with the request apply to the
>> enclosed entity, not the resource being addressed.  Thus, for instance,
>> a "Content-Language" request header in PATCH describes the natural
>> language of the patch document (entity).  So what why would it be a
>> problem to ignore it?  Or, rephrasing it, what do you expect the server
>> to do with it?
>>
>>
> 
> FWIW, this language was actually taken directly from RFC2616.  For

Interesting, we'll have to clarify this in rfc2616bis then.

> something like Content-Language, ignoring it really is not all that much
> of a problem.  For Content-Range, however, ignoring the header could
> cause some serious problems with the processing of the PATCH operation.

Yes. Let's state clearly what the requirement is then. MUST NOT ignore 
Content-Range?

Best regards, Julian

Received on Wednesday, 25 July 2007 17:30:40 UTC