Re: Idempotent partial updates

2012/2/29 Henrik Nordström <henrik@henriknordstrom.net>:
> ons 2012-02-29 klockan 19:59 +0000 skrev Mike Kelly:
>
>> Right, it is closely coupled - as is any other interaction driven by
>> shared understanding between the client and the server. Not everything
>> needs, or can be, visible on the network and is better off dealt with
>> via shared understanding between client and server. What is it about
>> the non-partial'ness of PUT that requires it to be made visible by
>> HTTP across the web as a whole?
>
> Simply the fact that HTTP/1.x do not allow upgrades of defined methods
> in this manner.
>
> If this "partial PUT" is sent to a server that do not understand it
> would result in breakage, therefore it can not be accepted as an
> extension of the existing PUT method. Instead it needs to use a new
> method, making sure that breakage do not occur if the request is
> attepmted somewhere where it's not supported.

Upgrading PUT so it is less specific about replacement wouldn't result
in this breakage. Clients don't make requests to servers arbitrarily,
they make them according to whatever application they are fulfilling.
i.e. if an application is operating on the basis that PUT requests to
its resources are replacements, then HTTP relaxing the semantics of
PUT to permit partials would not create breakage.

Cheers,
Mike

Received on Wednesday, 29 February 2012 23:52:42 UTC