RE: FW: DN-FO-09: 15.1.11 fn:distinct-values

The WGs considered your request in the meeting on 9/16/2003 and decided
not to change the semantics of fn:distinct-values to return values in
order of first appearance.  The feeling was that such a constraint would
inhibit optimization of this function.  

All the best, Ashok

> -----Original Message-----
> From: public-qt-comments-request@w3.org [mailto:public-qt-comments-
> request@w3.org] On Behalf Of Dimitre Novatchev
> Sent: Monday, August 04, 2003 9:00 AM
> To: Kay, Michael; public-qt-comments@w3.org
> Cc: dnovatchev@europe.com
> Subject: RE: FW: DN-FO-09: 15.1.11 fn:distinct-values
> 
> 
> As usual, Mike expressed the problem in the most precise manner.
> 
> Please, just note that exactly the same argumentation applies to any
other
> function
> returning a sequence, the order of results of which is "implementation
> defined".
> 
> Please, therefore modify these to return a sequence in "order of first
> appearance".
> 
> Thank you!
> 
> Dimitre Novatchev.
> 
> --- "Kay, Michael" <Michael.Kay@softwareag.com> wrote:
> > For the record, I am in favour of distinct-values being defined to
> retain
> > the first occurrence of each set of duplicates, and for the result
to
> be
> > in "order of first appearance".
> >
> > I think that making this implementation-defined will lead to
> > interoperability problems. People will discover that their favourite
> > implementation returns results in a predictable order, and they will
> come to
> > rely on it, which will give them problems moving to a different
product.
> >
> > Our data model provides sequences of values, not sets of values, and
we
> > should therefore assume that by default, the order of the values is
> > significant. They can always wrap the result in a call to
fn:unordered()
> if
> > they want a higher-performance implementation that doesn't retain
order.
> >
> > Michael Kay
> >
> >
> > > -----Original Message-----
> > > From: Dimitre Novatchev [mailto:dnovatchev@yahoo.com]
> > > Sent: 25 July 2003 22:34
> > > To: public-qt-comments@w3.org
> > > Cc: dnovatchev@europe.com
> > > Subject: Re: FW: DN-FO-09: 15.1.11 fn:distinct-values
> > >
> > >
> > >
> > > > We have clarified the text to indicate that order of the output
> > > > sequence if implementation dependent.  We have also changed the
> > > > wording of the example as you suggested.
> > >
> > > Thank you.
> > >
> > > >
> > > > If the user wants the output in a particular order, he can sort
the
> > > > values by document order by using $x/. or in any other
> > > order by using
> > > > the order-by clause.
> > > >
> > > > Please let us know if this is satisfactory.
> > >
> > > Document order is defined for nodes. We may want to preserve
> > > the *sequence order* between the distinct elements that would
> > > be returned.
> > >
> > > The example I gave before:
> > >
> > > I have the following sequence of random numbers:
> > >
> > > 3, 5, 2, 8, 6, 4, 9, 7, 1, 5, 2, 10, 9, 3
> > >
> > > I want to get only the (first occuring) distinct numbers with
> > > their ordering
> > > preserved:
> > >
> > > 3, 5, 2, 8, 6, 4, 9, 7, 1, 10
> > >
> > > This could be possible if the distinct-values() function was
> > > defined with an additional argument "preserve-order"
> > > (possibly with default value of false).
> > >
> > >
> > > fn:distinct-values($srcval as xs:anyAtomicType*,
> > >                    $preserve-order as xs:boolean? ) as
> > > xs:anyAtomicType*
> > >
> > >
> 8> > Cheers,
> > > Dimitre Novatchev.
> > >
> > >
> > >
> > > __________________________________
> > > Do you Yahoo!?
> > > Yahoo! SiteBuilder - Free, easy-to-use web site design
> > > software http://sitebuilder.yahoo.com
> > >
> >
> 
> 
> =====
> Cheers,
> 
> Dimitre Novatchev.
> http://fxsl.sourceforge.net/ -- the home of FXSL
> 
> __________________________________
> Do you Yahoo!?
> Yahoo! SiteBuilder - Free, easy-to-use web site design software
> http://sitebuilder.yahoo.com
>

Received on Monday, 22 September 2003 17:17:04 UTC