Re: [xsl] mixed content grouping by whitespace

["G. Ken Holman" <gkholman@xxxxxxxxxxxxxxxxxxxx>]

At 2010-04-13 17:19 +0100, David Carlisle wrote:
> On 13/04/2010 16:47, Wendell Piez wrote:
> > It's really helpful to keep this distinction in mind. One sort of
> > grouping works with a key; @group-by or @group-adjacent calculates that
> > key. The other sort simply applies a match criterion to each node in the
> > group to determine whether it's the particular sort of node
> > (group-starting or group-ending) of interest for that sort of grouping.
>
> Is it just me?
> this is one of the few distinctions in xslt for which I can never 
> find a mental model (so I can never remember) I always have to look 
> this up (or more exactly I don't look it up and get it wrong half the time).
> the use of match expression sseem completely arbitrary, Why didn't 
> they always make them xpath expressions?

As I see it and have no problems remembering it:

   group-by and group-adjacent are based on values,
   group-starting-with and group-ending-with are based on names.

Which of course is incomplete, because the latter are based on 
complete match patterns which can be very elegant and not simply 
names.  But that is the easy distinction for me.  I look at the 
content of the instruction as if it were a standalone template rule 
that matches the entire group, so I write the attribute name as if it 
were that template's match= attribute that triggers the processing of 
the group.  Using that mental imagery, I find the writing of the 
instruction itself quite smooth.

Where I usually trip is over remembering to use a predicate on the 
current-group() as in current-group()[self::h1] but I'm getting 
better at mastering it.

> It would mean you had to use
> self::h1 instead of h1 to trigger grouping on an h1 element, but it 
> would seem to be a lot more consistent and more powerful as the 
> match versions of grouping are restricted to nodes, so you can't for 
> example group a sequence of integers with group-starting with 
> specifying some condition on integers...

I'm not really sure what you are trying to say there, David.  And in 
what you are trying to say are you accommodating the feature of using 
a sequence in group-by= in order to place a single population member 
into more than one group simultaneously?  I need to include a running 
example of that in the classroom to help convey that concept to students.

I hope the other comments above help.

. . . . . . . . . Ken

--
XSLT/XQuery training:         San Carlos, California 2010-04-26/30
Principles of XSLT for XQuery Writers: San Francisco,CA 2010-05-03
XSLT/XQuery training:                 Ottawa, Canada 2010-05-10/14
XSLT/XQuery/UBL/Code List training: Trondheim,Norway 2010-06-02/11
Vote for your XML training:   http://www.CraneSoftwrights.com/s/i/
Crane Softwrights Ltd.          http://www.CraneSoftwrights.com/s/
G. Ken Holman                 mailto:gkholman@xxxxxxxxxxxxxxxxxxxx
Male Cancer Awareness Nov'07  http://www.CraneSoftwrights.com/s/bc
Legal business disclaimers:  http://www.CraneSoftwrights.com/legal