[mercury-users] lisp macros

[Manuel Hermenegildo]

 > But for an example of the *kind* of thing that might be done this way,
 > consider Definite Clause Grammar rules, which are precisely the kind of
 > "hide the plumbing" technique I was describing.  Come to think of it,
 > the functional sublanguage is also an example.  (I and many other people
 > have term_expansion/2-based plugins that let you use functional syntax
 > in Prolog.)

This is what I meant in my previous message: term_expansion-like
predicates allow you to hide away details and plumbing and raise
the level of abstraction of your programs. Examples are: 

- functional (and higher-order) syntax, 

- feature terms (i.e., records with named arguments), 

- writing constraints in a nice syntax when the underlying plumbing
  uses attributed variables and other nasty stuff,

- other control rules (the possible meta-interpretation is hidden
  under the plumbing), 

- expert system rules, fuzzy logic, ... 

- etc.

These things can be done by the term_expansion predicate in
traditional Prologs. However, as I pointed out in the previous
message, since the first definition of term_expansion in Dec-10 Prolog
(late 70's, right Richard?) improved designs are available that are
more flexible and easier to use, and which allow you to define syntax
extensions on a per-module basis (this is the way in which Ciao Prolog
implements the extensions pointed out above).

The paper I pointed out explains in some detail (with code examples),
among other things, how to write expansions for feature terms and
fucntional syntax.  The TR corresponding roughly to this paper:

@InProceedings{ciao-modules-cl2000,
  author =       {D.~Cabeza and M.~Hermenegildo},
  title =        {{A} {N}ew {M}odule {S}ystem for {P}rolog}, 
  booktitle =	 {International Conference on Computational Logic, CL2000},
  publisher =    {Springer-Verlag},
  series =       {LNAI},
  number =       {1861},
  year = 	 2000,
  month = 	 {July},
  pages=         {131--148}
}

can be downloaded from http://www.clip.dia.fi.upm.es if anybody is
curious enough. I think this would also be a good approach for
Mercury. It would not hurt to have some kind of standard for this
among logic languages, whatever it is...

Manuel

-- 

-----------------------------------------------------------------------------
herme at fi.upm.es                      | Manuel Hermenegildo                 
+34-91-336-7435 (Work)               | Facultad de Informatica, UPM
+34-91-352-4819 or 336-7412 (FAX)    | Universidad Politecnica de Madrid   
http://www.clip.dia.fi.upm.es/~herme | 28660-Boadilla del Monte, MADRID SPAIN
-----------------------------------------------------------------------------

--------------------------------------------------------------------------
mercury-users mailing list
post:  mercury-users at cs.mu.oz.au
administrative address: owner-mercury-users at cs.mu.oz.au
unsubscribe: Address: mercury-users-request at cs.mu.oz.au Message: unsubscribe
subscribe:   Address: mercury-users-request at cs.mu.oz.au Message: subscribe
--------------------------------------------------------------------------