RI 1.0 test Registry
KevinBenson
kmb at mssl.ucl.ac.uk
Wed Jul 12 04:03:09 PDT 2006
Just to follow up on my last e-mail. I was wrong the OAI from RofR is
correct since it does a xmlns="".
cheers,
Kevin
KevinBenson wrote:
> Some in line comments. Ray see my first paragraph about RofR and oai
> namespace.
>
> Ray Plante wrote:
>> Hey Kevin,
>>
>> On Mon, 10 Jul 2006, KevinBenson wrote:
>>
>>> Located at the below url you will find a test registry that conforms
>>> to 1.0. Feel free to try out various jsp pages, you can use the
>>> external** pages (bottom of menu) to see the full SOAP message
>>> coming back.
>>
>> Excellent! I'm working on the client search library now.
>>
>>
>>> OAI url as you can see from the Registry Type is:
>>> http://msslxt.mssl.ucl.ac.uk:8080/astrogrid-regtest-1_0/OAIHandlerv1_0
>>>
>>
>> This is looking good. (Mind if I use it to test my validater? ;-)
>>
> Sure go for it, let me know if I have missed anything :) Possible
> Note for those implementors with OAI using 3rd party tools, take a
> close look at the OAI namespace several times they put a default
> namespace at the top. Now that we don't use namespaces for our
> Resource if you look at the OAI it's as if the Resource element takes
> on the oai namespace :( I had to modify the 3rd party tool I was
> using. I believe the RofR you have Ray is also putting a default
> namespace, but when I harvest I tend to not look at it. Just Noting
> this because it could be a potential problem.
>> I was able to discover your Search WebService interface; however,
>> it's not returning a WSDL via ?wsdl.
> Ahha many thanks yes I had some paths all messed up since I pull wsdls
> and schemas from jars. It should be corrected now and I ran a
> WSDL2java with axis and it seemed to go.
>>
>>> Final Note the adqls->adqlx is a much older translation I am now
>>> trying to work in the 1.0 version. You should be able to still send
>>> 0.7.4, 0.9, and 1.0 type adqlx to the registry.
>>>
>>
>> So I'm having trouble supporting ADQL v1.0; there doesn't seem to be
>> a valid schema file ready yet. What are you using for an ADQL schema
>> (which is why I was groping for your WSDL)? Is it the same as
>> http://www.ivoa.net/xml/ADQL/v1.0?
>>
> Yes there was a time I saw a ADQL v1.0 (granted I think it was just
> 0.9 and the namespace changed). But having a little trouble finding
> it on the ivoa website. I believe Aurelien put up a xsl for
> adqlx->sql if anybodys interested (same location as the RI spec) . You
> can see mine here:
> http://msslxt.mssl.ucl.ac.uk:8080/astrogrid-regtest-1_0/schema/adql/ADQL/v1.0/ADQL.xsd
>
>
> cheers,
> Kevin
>> I'm going to see if I can make the best of what's there...
>>
>>
>>> Also Ray your samples were mostly good, had one type of validation
>>> error with xerces (maybe you might know on how to fix that):
>>> cvc-id.2: There are multiple occurrences of ID value 'UTC-FK5-TOPO'.
>>> cvc-attribute.3: The value 'UTC-FK5-TOPO' of attribute 'id' on
>>> element 'AstroCoordSystem' is not valid with respect to its type, 'ID'.
>>> at
>>> org.astrogrid.test.AstrogridAssert$AstrogridAssertDefaultHandler.endDocument(AstrogridAssert.java:344)
>>>
>>> at
>>> org.apache.xerces.parsers.AbstractSAXParser.endDocument(Unknown Source)
>>>
>>
>> Hmm. I'm not seein' this using Xerces. Which file was it? I just
>> rechecked the ones posted at RegUpgradeSummer2006 (including the
>> individual ones and the one's in the tar/zip balls) and they came up
>> fine. Do you have the latest?
>> cheers,
>> Ray
>>
>>
More information about the registry
mailing list