Hi Franck,

Thanks for your comments and for your input about the suggested changes.

I only wrote a few XSLTs for TAPIR responses and it was easy to handle namespaces. You basically just need to declare them in the root element and use the corresponding prefixes when referencing nodes. It worked with the two main browsers, so I don't know if there can be issues with other XSLT processors.

Please note that, as it is now, "omit-ns" can only be used with the search operation when the TAPIR envelope is turned off. In all other situations, clients will need to handle namespaces anyway.

We need to be careful about forcing providers to implement protocol features that may never be really used. Personally I was only expecting to keep these two features if: 1) there can be serious usability or interoperability issues if we don't keep them; 2) there are clients that are already making use of them; or 3) someone spent considerable time implementing them in a provider software besides TapirLink.

Best Regards, -- Renato

Hi Renato,

Congratulation to you and Markus for all your work!

Removing the "partial" parameter seems also a good solution to me as it would certainly make maintenance of the intermediate provider easier by simplifying the code it needs.

But I would maybe keep the "omit-ns" parameter if we plan to use XSLT stylesheets to convert responses from the provider into webpages for Internet browsers. Namespace can be sometimes difficult to parse in xslt processor and keeping this attribute would allow a better compatibility between search responses and the "exclude-result-prefixes" xsl attribute.

Best regards,

Franck

-----Original Message----- From: tdwg-tapir-bounces@lists.tdwg.org [mailto:tdwg-tapir-bounces@lists.tdwg.org] On Behalf Of renato@cria.org.br Sent: mercredi 21 janvier 2009 21:45 To: tdwg-tapir@lists.tdwg.org Subject: Re: [tdwg-tapir] "partial" and "omit-ns" parameters

Thanks, Tim.

Actually the same provider can explicitly declare that it supports more than one output model, so you don't need to have separate access points.

Anyway, the most important thing is that clients/networks can create as many response structures as necessary if they want to work with smaller pieces of the same big schema. So I don't think we will lose so much if we remove the "partial" parameter.

Best Regards,

Renato

...

Hi Renato,

Congratulations with all your work.

...

I can see two solutions for this: 1) Specify that the "partial" parameter only needs to be supported by TAPIR Full, which will make the parameter specification more intricate. 2) Remove the "partial" parameter from the protocol.

Since there appears no strong demand, I would be in favor of it removed. For TAPIR Intermediate, a possible work around solution could

...

be 2 access points each with differing output models, one being a subset of the other?

...

Markus is also suggesting to remove the "omit-ns" parameter from the protocol. "omit-ns" is used to indicate that search responses should not include any namespaces at all. If this is not being used by any network or client, I also don't mind removing it.

I personally have never had a need to use it. Most XML clients I use have a "namespaceAware" parameter anyway.

...

Please let me know if you have any feelings about this.

Well done again Renato

Tim