- From: Rui Lopes <rlopes@di.fc.ul.pt>
- Date: Tue, 13 Nov 2007 17:12:05 +0000
- To: Norman Walsh <ndw@nwalsh.com>
- Cc: public-xml-processing-model-wg@w3.org
- Message-ID: <4739DAE5.7070205@di.fc.ul.pt>
Norman Walsh wrote: > * Attempt to support both XPath 1.0 and XPath 2.0; there's more to > be done, but see Section 2.8, “XPaths in XProc”. On a quick read of this part of the spec, I've come across the following: "An XProc implementation can use either [XPath 1.0] or [XPath 2.0] to evaluate these expressions. [...]" Which means that, within a pipeline, each expression may be evaluated against different XPath engines (read: versions). Correct? If so, when I read on the spec the following system property: "p:xpath-version Returns the version of XPath implemented by the processor for evaluating XPath expressions on XProc elements." Which value(s) should an XProc implementation return, if it provides two XPath engines? "1.0", "2.0", "1.0 2.0" ? I propose that we should split the p:xpath-version property feature into: * a p:xpath-version-available(version) function, to check the availability of a specific version; * a p:xpath-versions property akin to p:xpath-version, but returning an xs:token value; * a p:xpath-version-default property, returning the default XPath version to be used when the p:xpath-version attribute isn't explicitly specified. Nevertheless, I would prefer to have the p:xpath-version available only at top-level elements (p:pipeline and p:pipeline-library). IMHO, this whould reduce the confusion of users when looking at a pipeline and read several XPath expressions written in different versions - on the same document. Isn't just top-level enough? Thoughts? Cheers, Rui
Received on Tuesday, 13 November 2007 17:15:20 UTC