Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!sun-barr!newstop!sun!quintus!dave
From: dave@quintus.UUCP (David Bowen)
Newsgroups: comp.lang.prolog
Subject: Re: Deterministic predicate question
Message-ID: <1420@quintus.UUCP>
Date: 28 Aug 90 18:53:41 GMT
References: <90239.224751F0O@psuvm.psu.edu> <3631@goanna.cs.rmit.oz.au>
Reply-To: dave@quintus.UUCP (David Bowen)
Organization: Quintus Computer Systems, Inc.
Lines: 18

In article <3631@goanna.cs.rmit.oz.au> Richard A. O'Keefe writes:
>Many Prolog compilers operate a clause at a time (in some, such as ALS
>Prolog and QP3.0, assert _is_ "compile"), so won't spot this.  

It is not correct that `assert is "compile"' in Quintus Prolog Release 3.0.
Dynamic code in Quintus Prolog continues to be represented in a form that is
optimized for speed of assert/retract rather than for speed of execution.

The change that may have misled Richard is that we no longer distinguish
between "compile" and "consult"; that is, there is no special way that you
have to load your code in order to be able to debug it - all code is
debuggable.

Since release 2.5, it is possible to access the clause-compiler via a built-in
predicate called multifile_assertz which compiles a clause and adds it to the
end of a predicate.  There is no corresponding retract operation (only
abolish).  Also, multifile_assertz is not available in runtime systems, since
the built-in compiler is not included in those systems.