[sc34wg3] TMCL: 4.4.8 Subject Identifier Constraint

[Graham Moore]

One key point:

>> If not, then the TMQL expression could look like this

>>every $t in // $$topictype
>> satisfies
>>     at least $$cardmin $si in $t ~
>>        satisfies $si =~ $$regexp
>>   & at least $$cardmax $si in $t ~
>>        satisfies $si =~ $$regexp

>> I am using $$ to prefix the Meta-variables. They become constants in
the >> TMQL query after inserting.


This $$ stuff isn't necessary and kind of breaks the intent. These
variables are in the constraint structure, so they should be located via
the tmcl expression. 

If $this is the constraint being evaluated then the cardmin is found via


$this >> occurrences [cardmin]

We want to avoid the extra prose description of how the tm structure for
the constraint maps to variables. 

Graham


--------------------------------------------
Graham Moore, Director, Networked Planet Limited
Editor XTM 1.0, ISO13250 (TopicMaps) -2,-3, TMCL
e: graham.moore at networkedplanet.com
w: www.networkedplanet.com
t: +44 1865 811131 
m: +44 7769658611 (UK)
m: +47 45271713   (Norway)

Networked Planet Limited is registered in England and Wales, no. 5273377
 

-----Original Message-----
From: sc34wg3-bounces at isotopicmaps.org
[mailto:sc34wg3-bounces at isotopicmaps.org] On Behalf Of Robert Barta
Sent: 15 February 2008 05:08
To: Discussion of ISO/IEC 13250 Topic Maps
Subject: [sc34wg3] TMCL: 4.4.8 Subject Identifier Constraint

Ad 4.4.8 in

   http://kill.devc.at/system/files/tmcl.pdf

First, the example uses

  subjectidentifier-constraint

but that seems to be the tmcl: type. The template's type is
SubjectIdentifierConstraint.

--

Then, you are using ?sic1 as identifier for this constraint invocation,
but according to CTM this is _global_, i.e. there could only be one!

  http://kill.devc.at/system/files/ctm.html#sec-topic-ref

  "If the named-wildcard is used again, the same topic is reused."

--

Then, '*' is actually not a regular expression, is it?  That is
globbing.  '.*' would be regexp.

--

Then, for this Subject Identifier Constraint (4.4.8), is it actually
possible to omit parts of it? Say,

  SubjectIdentifierConstraint (Person, 3, , '.*')

to omit the upper bound. Or to omit the regexp?

--

If not, then the TMQL expression could look like this

every $t in // $$topictype
satisfies
     at least $$cardmin $si in $t ~
        satisfies $si =~ $$regexp
   & at least $$cardmax $si in $t ~
        satisfies $si =~ $$regexp

I am using $$ to prefix the Meta-variables. They become constants in
the TMQL query after inserting.

\rho
_______________________________________________
sc34wg3 mailing list
sc34wg3 at isotopicmaps.org
http://www.isotopicmaps.org/mailman/listinfo/sc34wg3