[xsl] Re: XSLT 3.0: Are keys really components that a package should care about?

Subject: [xsl] Re: XSLT 3.0: Are keys really components that a package should care about?
From: "Michael Kay mike@xxxxxxxxxxxx" <xsl-list-service@xxxxxxxxxxxxxxxxxxxxxx>
Date: Sun, 16 Nov 2014 23:19:50 -0000
I have recorded this as a comment on the spec at


Michael Kay
+44 (0) 118 946 5893

On 16 Nov 2014, at 23:11, Dimitre Novatchev <dnovatchev@xxxxxxxxx> wrote:

> In Section 3.6.3 "Named Components in Packages"
> (http://www.w3.org/TR/2014/WD-xslt-30-20141002/#named-components ) of
> the XSLT 3.0 2nd Last Call, we read:
> "This section discusses the use of named components in packages:
> specifically functions, named templates, attribute sets, modes,
> accumulators, keys, modes, and global variables and parameters. Some
> of the provisions in this section also apply to named modes, but there
> are differences noted in 3.6.4 Overriding Template Rules from a Used
> Package. The section is largely concerned with details of the rules
> that affect references from one component to another by name, whether
> the components are in the same package or in different packages. The
> rules are designed to meet a number of requirements:"
> Besides the mentioning of modes twice in this list (and this also
> happens at another place of the document), which I hope resulted from
> duplication and not from omission, I am wondering why keys are in this
> list of named components at all.
> As we discover later in the document, an xsl:key always has private
> visibility -- therefore it cannot be exposed to using packages,
> because a using-package cannot reference a private component from a
> used-package.
> Is this an error... or if not, could someone explain, please?
> -- 
> Cheers,
> Dimitre Novatchev

Current Thread