[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]
Subject: RE: [xacml-comment] More questions about new version 2.0
Diego - Check out Section 5.42, where it says ... "The <AttributeSelector> element's RequestContextPath XML attribute SHALL contain a legal XPath expression whose context node is the <xacml-context:Request> element." And, yes. We will use a new namespace. All the best. Tim. -----Original Message----- From: Polar Humenn [mailto:firstname.lastname@example.org] Sent: Wednesday, July 07, 2004 11:36 AM To: Diego M. Gonzalez Cc: email@example.com Subject: Re: [xacml-comment] More questions about new version 2.0 Hi diego, comments inline. On Tue, 6 Jul 2004, Diego M. Gonzalez wrote: > I finished the reviewing of the WD12 and I have the following > questions. > > > AttributeSelector > When I was working on the implementation for the 1.1 spec I found that > there is no information about the current node for the Xpath > expressions. The Xpath expressions can make reference to absolute root > (/doc/node@attrib) or a relative root (../parent@attrib) nodes. The > Xpath spec calls the "context node". Since the AttributeSelector > contains an Xpath its important to stablish which is the "context > node" specially if relative location paths are supported. Are you > planing adding this to the spec? I'm not an expert on this Xpath stuff. Somebody more proficient in this may be able to answer. > Missing VariableReference > What is the expected behavior when the VariableReference references to > a variable that does not exists? Is there any MissinVariable element > in the Result? This is a malformed policy and should not type check. If you don't do typechecking, I guess the "behavior" would be to raise an Indeterminate result. > > Versioning > How are you planing to manage the different version of the schemas. > Are you going to define a new schema with a new namespace xx:xx:2.0 or > you will create a new schema that extends the previous one with the > extesions or restrictions to the elements? I think there will be a new schema identiifer, correct? Cheers, -Polar > > Thanks, > Diego Gonzalez > Lagash Systems SA > > To unsubscribe from this list, send a post to > firstname.lastname@example.org, or visit > http://www.oasis-open.org/mlmanage/. > To unsubscribe from this list, send a post to email@example.com, or visit http://www.oasis-open.org/mlmanage/.