I guess what you need is not RecordC. I cannot say if you will have to revise your whole design eventually, but tuples can easily be encode as one only integer. Take a look at the class Cartesian at
http://www.ps.uni-sb.de/~niehren/Web/Vorlesungen/Oz-NL-SS01/vorlesung/node274.html On this website you'll find examples how to use it. Cheers, Jorge. Selon Raphael Collet <[EMAIL PROTECTED]>: > rmk216 wrote: > > Sorry if this question has come up before, but how can the Oz constraint > > system be used to model constraints over variables whose domains are > > tuples? For example, I've written some 3-D puzzle solving algorithms > > using JCL, and all of my work uses tuples to represent the game pieces. > > This works fine for me, but I'd like to investigate doing something > > similar under Oz. Do I have to somehow redo my tuple-domain algorithms > > to work with integer domains instead, or is there some sort of built-in > > support for making fancy constraints? > > Mozart has constraints over records, which extend tuples. Here are a > few examples of such constraints: > > %% R is a record with label 'foo' > R = foo(...) > > %% R is a record with label 'name', and at least a field 'name' > R = person(name:N ...) > > %% R is a record with a field 'name' whose value is N > R^name = N > > %% R is a record of width 3 > {RecordC.width R 3} > > Consult the documentation of module RecordC. The document > http://www.mozart-oz.org/papers/abstracts/oznlp.html also mentions > "feature" constraints, and their use for language processing. > > Cheers, > raph > > _________________________________________________________________________________ > mozart-users mailing list > [email protected] > http://www.mozart-oz.org/mailman/listinfo/mozart-users > _________________________________________________________________________________ mozart-users mailing list [email protected] http://www.mozart-oz.org/mailman/listinfo/mozart-users
