[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [Elist Home]
Subject: combine
> > Getting rid of combine="replace" is pointless unless you get rid of > > combine="group". They are both equally order-dependent. Here's where my > > XHTML mod rewrite uses combine="group": > > I now believe we can't make RELAX NG an order-insignificant language > without significant drawback. > > But all combine="group"s in XHTML are used to combine attributes. So > actually they makes no order dependency. Right. We could have a restriction on combine="group" that you can't use it in a way which makes order significant (eg combining two <element> patterns). On the other hand that would rule out using combine="group" to provide functionality similar to derivation of complex types by extension in XML Schema (adding things on to the end of a content model). I wonder whether one could attach a priority to definitions to allow combination without order dependence (like xsl:template). With combine="replace", the pattern with the highest priority would be the one used. With combine="group", the patterns would be grouped in order of priority. You wouldn't be able to mix different kinds of combine for a single pattern. James
[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [Elist Home]
Powered by eList eXpress LLC