Re: brief review on protocol

Axel Polleres wrote:
> Hi,
> 
> this is a short review only for protocol (didn't find time to have a closer look, but I hope it's better than nothing).

It is, thanks Axel.

> From my side, there is no objections to publish the wsdl and xsd files  "as is".
> 
> 
> However, the protocol spec file has some open questions that should be easily fixed:
> 
> 1) there are still links to the old protocol-types.xsd, these should be fixed

I changed all the links to be relative; we'll also ensure these are 
right during publication.

> 2) the 
>     MalformedQuery and QueryRequestRefused fault messages seem to have disappeared in section 2.1.3, they should 
>   remain as in the previous version.

I restored this as 2.1.1.3 and the update faults are 2.1.2.3.

> 3) the 
>    GraphDoesNotExist and GraphAlreadyExists
>    errors were questioned in the last TC, i.e. 
>     HTTP error codes? 
>     under which circumstances these should arise? cf. also ISSUE-20
>  cf. http://www.w3.org/2009/sparql/meeting/2010-01-05#line0108
>  It was suggested to add a more generic semantic error (instead?) ... this should at least be marked with an Editor's note.

I'd like the protocol document to delegate these sorts of errors to 
whatever the Update spec ends up saying. I put this in:

<ednote><edtext>The final set of Update Fault Messages will largely 
depend on error conditions defined by the SPARQL 1.1 Update 
specification.</edtext></ednote>


> Change summary:
> 
> Compared with FPWD and the original spec, a new WSDL file (http://www.w3.org/2009/sparql/docs/protocol-1.1/protocol-query.wsdl) and XSD types declarations file (http://www.w3.org/2009/sparql/docs/protocol-1.1/protocol-types.xsd) have been added with the new update operation and faults.
> 

Added a "since previous WD" bit in the SotD.

other small changes I made:

   + replaced the XXUPDATEXX and XXQUERYXX tokens (for operation names) 
with <code>update</code> and <code>query</code>.
   + Added a placeholder section for an Update out message, noting that 
the Working Group has not yet discussed this.

Lee

Received on Sunday, 10 January 2010 16:03:48 UTC