Re: Scheme bug in internal define

From: Keith Wright <>
Date: Sun, 9 Nov 97 21:35 EST

> "or display error message." Reject the program if you must, but
> substuting one variable for a different one without a message is
> no good.

Let me reword that. I recommend rejecting the program with an error
message. Internal define is described in R4RS by translation into a
letrec. I can think of at least three different ways to translate
this. (One letrec or two? Which one is the middle expression in?)
There is no reason not to ask the programmer to re-write to remove
the ambiguity.

This is not an idle exercise. I unintentionally created a much longer
and more complicated program of this form. It took me most of a day
to figure out why it kept crashing and whittle it down to the simple
form I posted. The variable changes between evaluating the call and
entering the procedure body. At first I thought it must be a garbage
collector bug, or due to the twisted way I was using Tk widgets.

This mail message sent by GNU emacs and Linux.
Power to the people. Linux is here.
Food, Shelter, Source code.
Received on Mon Nov 10 1997 - 03:35:19 CET

This archive was generated by hypermail 2.3.0 : Mon Jul 21 2014 - 19:38:59 CEST