Path: utzoo!utgpu!news-server.csri.toronto.edu!rpi!usc!jarthur!ucivax!milne
From: milne@ics.uci.edu (Alastair Milne)
Newsgroups: comp.lang.pascal
Subject: Re: wish i could figure this out for myself
Message-ID: <28276C38.19606@ics.uci.edu>
Date: 8 May 91 03:10:47 GMT
References: <1991May3.140719.5931@javelin.sim.es.com> <1991May3.150243.3793@e2big.mko.dec.com> <1991May3.170232.7710@javelin.sim.es.com> <ricki.673528313@rwthi3>
Distribution: na
Organization: UC Irvine Department of ICS
Lines: 22

In <ricki.673528313@rwthi3> ricki@john.informatik.rwth-aachen.de (Richard Breuer) writes:
>FUNCTION ValidChar(Ch:Char):Boolean;
>  CONST
>    ValidChars = 'ABCDEFHIJKLMNO'; (* and so on... *)
>  BEGIN
>    ValidChar := Pos(Ch,ValidChars)>0;
>  END;

  If it is faster to
  - push a string parameter (actually 2 of them)
  - call a function
  - scan the string doing substring checks
  - return the result

  than to do set inclusion, which should be done with a bitmask,
  then I will be quite surprised.

  Sets are one of Pascal's greatest features (greater, of course,
  in dialects that let you have nice big ones).  Use them in good health.


  Alastair Milne