[jdom-interest] Validation strategy

[Søren Faltz]

I like the idea of a superschema, however all the schemas are not in
disjoint namespaces...

What do you meen by tell the rules for choosing a schema?

Yes the processing pipeline would be:
1 Put xml messages in a FIFO buffer
2. process one message at a time
3. validate the message
4. process the data

best regards
lars

2006/5/10, Michael Kay <mike at saxonica.com>:
>
> It rather depends how the messages are distinguished. If they are all in
> disjoint namespaces, you could simply write a superschema that imports all
> the others, and then do strict validation against that. If the schemas have
> overlapping definitions, it's a bit more complicated. But if you can't tell
> us the rules for choosing a schema, then no-one else can!
>
> The fact that you are asking on a JDOM list also suggests that you have
> made some decisions about the processing pipeline, and these decisions may
> constrain the answer.
>
> Michael Kay
> http://www.saxonica.com/
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://www.jdom.org/pipermail/jdom-interest/attachments/20060510/b8f5b6b0/attachment.htm