Path: utzoo!yunexus!ists!helios.physics.utoronto.ca!news-server.csri.toronto.edu!mailrus!wuarchive!udel!burdvax!lang
From: lang@PRC.Unisys.COM (Francois-Michel Lang)
Newsgroups: comp.lang.prolog
Subject: Re: Why is bagof so slow?
Keywords: bagof, setof
Message-ID: <14004@burdvax.PRC.Unisys.COM>
Date: 31 May 90 13:48:32 GMT
Article-I.D.: burdvax.14004
References: <2556@randvax.UUCP> <3086@goanna.cs.rmit.oz.au> <2559@randvax.UUCP> <3110@goanna.cs.rmit.oz.au>
Sender: news@PRC.Unisys.COM
Organization: Unisys Corporation, Paoli Research Center; Paoli, PA
Lines: 36


This is indirectly related to the speed of bagof/3:
Do any (current or former) implementors out there
have any thoughts about the possibility of providing
built-in versions of bagof/setof that return a difference list?

E.g., given

f(1).
f(2).
f(3).
f(4).

we could do

| ?- bagof_dl(X, f(X), List, Tail).

X = _495,
List = [1,2,3,4|Tail],
Tail = _542 

| ?- 

I isn't hard to write a version of *findall* that returns
a difference list (after writing one myself I discovered
one in the library...), but I haven't tried doing the same thing
with bagof/setof.  Another problem is that anything user-defined
would have to use the ``expensive'' versions of assert/retract
instead of the ``special cheap'' versions (presumably not accessible
to users), and therefore be much slower.

Any thoughts about this?
----------------------------------------------------------------------------
Francois-Michel Lang                                (215) 648-2536
Unisys Center for Advanced Information Technology   lang@prc.unisys.com      
Dept of Comp & Info Science, U of PA                lang@linc.cis.upenn.edu