Re: Non-null pointers in C/C++?

angel_tsankov wrote:
> On 15 =D0=9C=D0=B0=D0=B9, 19:00, Hans-Bernhard Br=C3=B6ker <HBBroe...@[EMAIL PROTECTED]
> wrote:
>> Note: I will ignore you mentioned C++ at all.
>>
>> angel_tsankov wrote:
>>> It's been some time since I started wondering why the C/C++ languages
>>> do not have a non-null pointer (something like a copyable C++
>>> reference)?
>> Because it wouldn't work.  Just because a pointer can't be NULL doesn'=
t
>> mean it's safer to use.  It could still be invalid in any number of
>> other ways.
>>
>>> -provide compilers with further semantic information, thus enabling
>>> them to perform more optimizations;
>> There's no optimization you could do with a non-null pointer that you
>> would be forbidden to do with a NULL pointer --- optimizations are
>> already completely free to pick among various different kinds of
>> undefined behaviour.  C code never has to check a pointer for NULL
>> before using it.

C code had better check that the return value from malloc() et al is
not=20
null before using it.  C code had better check that the return value=20
from fopen() et al is not null before using it.  C code had better
check=20
that the return value from fgets() is not null before using it.  (C
code=20
had better not be using gets(), so there's no need to check its return=20
value!)

> Sure, there would be optimizations. For instance, you can safely skip
> the null checks. The standard library could do this also, e.g. in
> functinos where this matters (free for example).

Since free(0) is defined behaviour - nothing happens - it isn't clear=20
that there is any optimization available.

--=20
Jonathan Leffler                   #include <disclaimer.h>
Email: jleffler@[EMAIL PROTECTED]
 jleffler@[EMAIL PROTECTED]
 of DBD::Informix v2008.0229 -- http://dbi.perl.org/

publictimestamp.org/ptb/PTB-3317 ripemd160 2008-05-24 03:00:05
851B441D3F34BE2EBCF36BD250B03C07F300D8D7
-- 
comp.lang.c.moderated - moderation address: clcm@[EMAIL PROTECTED]
 -- you must
have an appropriate newsgroups line in your header for your mail to be
seen,
or the newsgroup name in square brackets in the subject line.  Sorry.