Hello, thanks for your answer. I want to ask the other way round. Is this behaviour bpel2-compliant? The problem is that we don´t want to develop the processes only for the ODE BPEL Engine... We have to be sure that other bpel engines expect the same or this does not lead to an error...
Is there an other way for being compliant and make semantically the same? We don´t want to give up the flexibility to create complex types... Jens Alex Boisvert wrote: > > On 9/11/07, Jens Goldhammer <[EMAIL PROTECTED]> wrote: >> >> Isn´t it possible that ODE makes this initialisation? I saw in the old >> processes which we have made with activeBpel contain this function ... >> and >> it worked very well. >> Can you change the behaviour of ODE in this case? > > > > It's certainly possible to initialize variables with a minimal XML > template > that would be type-compatible in many cases... although the rules and > bounds > for creating the template must be easily understandable otherwise you > would > be trading one problem for another. > Feel free to open a Jira enhancement for it. I think we should organizing > enhancements according to level of difficulty, scope and also have people > vote for new features. I'm also hoping it would incite people to start > hacking on Ode... > > alex > > -- View this message in context: http://www.nabble.com/Problems-with-ODE-1.1-Release-tf4355659.html#a12630554 Sent from the Apache Ode User mailing list archive at Nabble.com.
