Path: utzoo!utgpu!jarvis.csri.toronto.edu!mailrus!csd4.milw.wisc.edu!uxc!uxc.cso.uiuc.edu!ux1.cso.uiuc.edu!uxe.cso.uiuc.edu!mcdonald
From: mcdonald@uxe.cso.uiuc.edu
Newsgroups: comp.std.c
Subject: Re: trigraphs
Message-ID: <229900002@uxe.cso.uiuc.edu>
Date: 28 Apr 89 15:37:00 GMT
References: <2469@ogccse.ogc.edu>
Lines: 14
Nf-ID: #R:ogccse.ogc.edu:2469:uxe.cso.uiuc.edu:229900002:000:664
Nf-From: uxe.cso.uiuc.edu!mcdonald    Apr 28 10:37:00 1989


>One fact that seems not to have come out yet is that trigraphs as added
>by X3J11 were an existing scheme that was proposed for adoption, not a
>new design that simply seemed to be a good idea.  I suspect that is part

What do you mean by "existing scheme"? What significant compiler
(i.e. one selling more than 10000 copies per year) implemented
them? They seem to be probably the worst misfeature of ANSI C:
one that actually breaks working code. I have some code,
based on K&R C that uses the sequences ??(, ??), and ??! as
delimiters in a text file format - they are used in jillions
of string constants. Wouldn't trigraphs break such schemes?

Doug McDonald