Re: qt-2004Feb0418-01: [XQuery] 'local' namespace

Hello Jonathan,

I think than in general, this explanation is acceptable.
But it may be better to explain this more explicitly in the spec.
Also, it may be good to point out that this leaves open the
possibility of name conflicts, e.g. when combining two queries
into one. Namespaces are used to avoid conflicts, but in this
case, may not work.

Regards,    Martin.

At 19:27 04/08/24 -0400, Jonathan Robie wrote:
>Hi Martin,
>
>This is a response from the XML Query Working Group, which has
>considered your request (which was possibly meant only as a
>question?):
>
>http://lists.w3.org/Archives/Public/public-qt-comments/2004Feb/0418.html
>
>The local namespace was introduced to clarify what namespace locally
>declared functions are found in.
>
>If a function call or function declaration is not prefixed, then it is
>in the default function namespace. If the query does not set the
>default function namespace explicitly, then it is the namespace of the
>Functions and Operators library, which makes it convenient to call
>these functions.
>
>But this led to an inconsistency when users declare functions without
>a prefix, which was the most common way to do so. Clearly, the
>functions written by a user in a given query should not belong to the
>Functions and Operators namespace. Clearly, the conventions used to
>specify the name of a function should be the same for a function
>declaration or a function call found in the same module.
>
>So unprefixed functions were a problem in the default configuration of
>XQuery. We felt that many beginning users are confused enough about
>namespaces that they would not know how to decide what URI to declare
>if we made them declare their own, so we predeclared a namespace for
>local functions found in the main module.
>
>Please let me know if you feel this resolution is unacceptable.
>
>Thanks!
>
>Jonathan
>
>On behalf of the XML Query Working Group

Received on Tuesday, 14 September 2004 06:40:36 UTC