working code

[Henry Spencer]

In article <10235 at xanth.cs.odu.edu> kremer at cs.odu.edu (Lloyd Kremer) writes:
> [problems with ANSIitizing `#define foo(c) ('c'&...)']
>P.S. I support the adoption of the Standard, but it is disheartening to keep
>seeing its introduction cause working code to break...

In this context, "working code" is more properly written "code that worked
when compiled under one particular compiler, but wouldn't under others".
The trick that macro was relying on was undocumented and highly unportable.
No C compiler implemented based solely on K&R -- as some were -- would
compile it properly.

X3J11 tried very hard, and fairly successfully, not to break code that
worked because it conformed to existing definitions of C.  Code that worked
by accident (e.g. quirks of particular implementations) quite properly got
lower priority.
-- 
A bit of tolerance is worth a  |     Henry Spencer at U of Toronto Zoology
megabyte of flaming.           | uunet!attcan!utzoo!henry henry at zoo.toronto.edu