>> You know full well that the design isn't going to be improved in a >> subsequent release unless there's an issue that's a real stopper, because >> people expect backwards compatibility. > > http://exist.2174344.n4.nabble.com/Signature-of-fn-filter-doesn-t-look-like-it-conforms-to-the-spec-td4662991.html >
I am not sure that this is making the point you think it is! We (eXist) chose to implement a *draft* standard in it's very early days and make that available to our users. We previously made the same decision with a draft version of XQuery Update. That is always a risky decision as the draft may well change, in fact we almost certainly expect it to. The Working Group are not expected to maintain backwards compatibility through drafts, otherwise nothing would ever be improved! eXist was an early implementer of the draft standard and it allowed us to provide feedback to the Working Group about any issues with implementation, and/or suggestions based on real-world use of where it may be improved. We requested that the function signatures were changed, based on our experience. As it was a draft standard that was acceptable to do. However, when the standard hits "Recommendation" status, it is final, and it is at this point that backward compatibility must be preserved as it is expected by XQuery developers; hence being unable to change the functionality of doc. Anyway, I am sure you realise that you could write your own function in XSLT, let's call it 'doc-or-empty' and use it in place of doc to achieve exactly what you want. -- Adam Retter skype: adam.retter tweet: adamretter http://www.adamretter.org.uk _______________________________________________ [email protected] http://x-query.com/mailman/listinfo/talk
