perfsonar development work

[Roman Lapacz <>]

Jochen Reinwand wrote:
> Hi Maxim,
>
> This namespace stuff is normally handled by the XML library used for XML 
> handling, since this behaviour is standardised by XML.
>   

Right. In java services the work is done by the xml library so namespace 
declarations don't have to be stuck to the head element.

Roman

> I can only speak for our Perl based services: We are using XML::LibXML, the 
> Perl interface to libxml, the standard OpenSource XML library. Not only for 
> performance reason we implemented SOAP support on our own and the complete 
> XML message received via HTTP (including SOAP Envelope) is one LibXML DOM 
> object. Therefore there should not be any problems with the namespace issues 
> you mentioned and similar ones.
>
> greetings,
> Jochen
>
> On Wednesday 27 February 2008 19:58, maxim wrote:
>   
> >  Hello,
> > I am trying to get a conclusive answer from each service's developer on
> > namespace declaration. Is it true that every ps service supports
> > namespace declarations either in the head element ( message ) or
> > in the top of the element where it used  ( so it wont choke on missing
> > declaration in the message element ).For example pinger namespace
> > can be declared here :
> >
> >  <nmwg:message
> >     xmlns:pinger="http://ggf.org/ns/nmwg/tools/pinger/2.0/"; ...
> >     xmlns:nmwg="http://ggf.org/ns/nmwg/base/2.0/"; type="SetupDataResponse"
> > id="message.10186492">
> >
> > and used below
> >
> >  <nmwg:data metadataIdRef="meta2" id="data1">
> >       <nmwg:commonTime value="1198279649" type="unix">
> >        <pinger:datum value="53.258" name="minRtt"/>
> >
> > or like that ( in every data element ):
> >
> >    <nmwg:message
> >     xmlns:nmwg="http://ggf.org/ns/nmwg/base/2.0/"; type="SetupDataResponse"
> > id="message.10186492">
> >
> >  <nmwg:data metadataIdRef="meta2" id="data1"
> > xmlns:pinger="http://ggf.org/ns/nmwg/tools/pinger/2.0/";> <nmwg:commonTime
> > value="1198279649" type="unix">
> >        <pinger:datum value="53.258" name="minRtt"/>
> >  --
> > The both should be supported, right ?
> > Thanks,
> >         Maxim.
> >     
>
>