Subject: Re: lib/19638: isalpha (3) bug
To: Mike Cheponis <mac@Wireless.Com>
From: None <mcmahill@mtl.mit.edu>
List: netbsd-bugs
Date: 01/03/2003 22:25:31
On Fri, 3 Jan 2003, Mike Cheponis wrote:

> I guess, in sum, I conclude that the isxxx() functions are broken.  I know
> that's not the "official" case, just my opinion.

then write your own version that your program uses.


> Lastly, I -still- believe libc should be held to a higher standard.  Maybe
> I'm crazy on this, but it seems "reasonable" to me.

but what "standard" is this?  Unless its a published adopted standard,
this statement means nothing.  As many others have pointed out, the
standard is very very clear on this subject.  Have you made sure your
program works on solaris, irix, hp/ux?  How about different versions of
those?  In fact there's no particular reason to assume that the undefined
by the standard behaviour you're depending on will even be the same across
different versions of an OS or even different platforms of the same
version.

If you don't like the undefined behaviour, then write your own
mac_isxxx() functions that are done in a portable way and then you can
ensure they do what you want.  

-dan