[seek-dev] Re: gwsdl for put

[Matt Jones]

Hi Bing,

Thanks for the progress update.  A couple of issues that you might 
consider in the next draft of the gwsdl put document...

1) We had intended to have multiple implementation levels. Some of our 
services support put today (e.g., Metacat/SRB) while others aren't 
oriented that way currently (eg DiGIR, Xanthoria).  We need to 
accomodate both.  That is why we called the query service ' Query Level 
I', intending to have separate, more advance 'Query Level II' and 'Query 
Level III' services as well as the 'Write Service'.  Our previous 
meeting notes from Seattle outline these different levels. I would 
create a new service for the put interface that is distinct from the the 
level I query interface so that providers can choose whether or not to 
implement put.

2) Your parameters in0, in1, in2, in3 in the 'putRequest' and 
'putResponse' message type are not named well enough to understand their 
function.  You also should document the use of the parameters in the xsd 
if possible.

3) Why don't you have an input parameter of xsd:base64binary that is 
parallel to the getResponseElement from the get operation?  Seems like 
they are inverse operations, and the signatures should match.  If 
EcoGridObjDataType is intended to hold the data, how will it be encoded 
as a series of ints?  Why can't you use the native xsd types that 
support binary bytes?  Why not use base64binary?

4) The putResponse value on the put operation is an integer -- is this 
intended to be a status code?  Why not use the Fault, as we do in the 
other operations?  I thought we agreed in Edinburgh that using the fault 
mechanism was preferred?

Thanks.  Nice job.  I'm looking forward to the revision.

Matt

Bing Zhu wrote:
> Matt, Jing, Rod and others,
> 
> I put gwsdl description of our 'put' function into the our level I gwsdl
> file.
> Attached please find the file for the new put function within Ecogrid.
> 
> Notice that I used 'anyType' for the first parameter in out
> since this object can be either a byte array or a metadata object
> (e.g. an array of metadata objects).
> 
> Please review it and send me your feedback. Thanks.
> 
> Bing
> 

-- 
-------------------------------------------------------------------
Matt Jones                                     jones at nceas.ucsb.edu
http://www.nceas.ucsb.edu/    Fax: 425-920-2439    Ph: 907-789-0496
National Center for Ecological Analysis and Synthesis (NCEAS)
University of California Santa Barbara
Interested in ecological informatics? http://www.ecoinformatics.org
-------------------------------------------------------------------
-------------- next part --------------
A non-text attachment was scrubbed...
Name: smime.p7s
Type: application/x-pkcs7-signature
Size: 3407 bytes
Desc: S/MIME Cryptographic Signature
Url : http://mercury.nceas.ucsb.edu/ecoinformatics/pipermail/seek-dev/attachments/20040726/5200b4ba/smime.bin