Re: [PATCH REPOST] Trace and canelation editor changes part 4

On Fri, 2002-01-25 at 02:52, Michael Wood wrote:
> > The third param is clearer as NULL, since that param is a ptr.
> [snip]
>
> Especially since NULL is not guaranteed to be 0. (I think 0 is
> a valid pointer on VAX if I remember correctly.)

The null pointer is not guaranteed to be bitwise zero in runtime memory,
but an integer constant 0, converted to a pointer, is guaranteed to be a
null pointer. So, this is truly a question of aesthetics after all. (I
agree with Greg Stein that NULL is clearer.)

(Subversion has chosen to assume that the null pointer is bitwise zero
on the platforms it runs on anyway, so that it can initialize structures
with zeroed memory and assume that pointers are null pointers. I don't
like this practice since I'm a language purist, but I don't expect it to
cause any practical issues.)

See http://www.eskimo.com/~scs/C-faq/s5.html for more information.

---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org
Received on Sat Oct 21 14:36:59 2006