- From: Jeni Tennison <jeni@jenitennison.com>
- Date: Wed, 10 Oct 2007 19:58:55 +0100
- To: public-xml-processing-model-comments@w3.org
Norman Walsh wrote: > / Vasil Rangelov <boen.robot@gmail.com> was heard to say: > | Any p:parameter elements override any provided on a port. And if there's > | more than one p:parameter providing value for the same parameter, they are > | considered in document order. > > More than one p:parameter with the same name is an error, I think, though > I can't put my finger on that in the text just at this moment. > > | Makes perfect sense to me. > > I'm on the fence here. I could be convinced either way. The status quo > says that the order matters. Without revisiting the whole p:parameter discussion, I seem to remember that we had use cases where it was intended that user-supplied parameters should override the (default) author-supplied parameters (to allow the user control over the parameters passed to the XSLT step), and use cases where author-supplied parameters should override any user-supplied ones (to prevent the user from controlling those particular aspects of the transformation). So I think the decision to have order matter was based on providing support for both scenarios. Unless there's a really good reason not to (and I haven't seen one), I think we should continue to support them both. Cheers, Jeni -- Jeni Tennison http://www.jenitennison.com
Received on Wednesday, 10 October 2007 18:59:15 UTC