«

#include <stdio.h>
#define p printf
#define r0 return 0;
#define P int main(void)\
{\
p("Hello world! NULL is %p\n", (void*)NULL);\
r0\
}
P

/*
I think it is because Microsoft and gcc say so!

Cheers!
--
Citizen C
*/

Citizen C wrote:
> #include <stdio.h>
> #define p printf
> #define r0 return 0;
> #define P int main(void)\
> {\
> p("Hello world! NULL is %p\n", (void*)NULL);\
> r0\
> }
> P

Once you get rid of the stupid preprocessor tricks, this expands (apart
from any expansion of <stdio.h> to

#include <stdio.h>
int main(void){ printf("Hello world! NULL is %p\n", (void*)NULL); return 0;}

or, in a sane world,

#include <stdio.h>
int main(void)
{
printf("Hello world! NULL is %p\n", (void *) NULL);
return 0;
}


Now, tell me, do _you_ think this is a standard portable program?
Stupid preprocessor tricks are for infants.

On Jan 6, 6:36 am, Citizen C <alex566...@aol.com> wrote:
> #include <stdio.h>
> #define p printf
> #define r0 return 0;
> #define P int main(void)\
> {\
> p("Hello world! NULL is %p\n", (void*)NULL);\
> r0\
> }
>
> P

Yes

On 6 Jan, 04:36, Citizen C <alex566...@aol.com> wrote:
> #include <stdio.h>
> #define p printf
> #define r0 return 0;
> #define P int main(void)\
> {\
> * *p("Hello world! NULL is %p\n", (void*)NULL);\
> * *r0\}
>
> P
>
> /*
> I think it is because Microsoft and gcc say so!

Why did you suspect it wasn't? That is why did you needd to "test"
it with MS and gcc?

Note 1: using a couple of compilers does *not* prove a program
standard and portable.

Note 2: the style is really crap (it would fail reveiew in any
reasonable development environment)


--
Nick Keighley

#include <stdio.h>
#define HAN int main (void)
#define HAS { printf(
#define A "Hello world! "
#define WART "I'm a piece of ****!\n");
#define ON return
#define HIS 0;
#define DICK }

HAN HAS A WART ON HIS DICK

writes:

> On 6 Jan, 04:36, Citizen C <alex566...@aol.com> wrote:
>> #include <stdio.h>
>> #define p printf
>> #define r0 return 0;
>> #define P int main(void)\
>> {\
>> Â* Â*p("Hello world! NULL is %p\n", (void*)NULL);\
>> Â* Â*r0\}
>>
>> P
>>
>> /*
>> I think it is because Microsoft and gcc say so!
>
> Why did you suspect it wasn't? That is why did you needd to "test"
> it with MS and gcc?

Err, they are the compilers he uses?

>
> Note 1: using a couple of compilers does *not* prove a program
> standard and portable.

Actually it does prove its portable - between the machines of interest
for those compiler versions he mentions.

>
> Note 2: the style is really crap (it would fail reveiew in any
> reasonable development environment)
>
>
> --
> Nick Keighley

--
I'm not a person who particularly had heros when growing up.
- Dennis Ritchie when asked about the hero worship coming from c.l.c

On Jan 6, 5:49*am, Han from China - Master Troll <autistic-
lamer...@comp.lang.c> wrote:
> > HAN HAS A WART ON HIS DICK
>
> I thought it was a nice way to add some girth.

thats kind of hack and technically invokes UB, phalloplasty is at
least standard-comformant.

> "Only entropy comes easy." -- Anton Chekhov

if i ever have a daughter, im going name her "entropy" just so i can
say that same thing about her.

----- ---- - - - ---- - -+
SIGNATURE

On Jan 6, 12:58*pm, Richard <rgrd...@gmail.com> wrote:
> nick_keighley_nos...@hotmail.com writes:
> > Note 1: using a couple of compilers does *not* prove a program
> > standard and portable.
>
> Actually it does prove its portable - between the machines of interest
> for those compiler versions he mentions.

You're wrong. UB is UB. It may compile once without any errors and
never compile again with a lot of errors. Compiling proves nothing.

wrote:
> On Jan 6, 12:58 pm, Richard <rgrd...@gmail.com> wrote:
>> nick_keighley_nos...@hotmail.com writes:
>>> Note 1: using a couple of compilers does *not* prove a program
>>> standard and portable.
>> Actually it does prove its portable - between the machines of interest
>> for those compiler versions he mentions.
>
> You're wrong. UB is UB. It may compile once without any errors and
> never compile again with a lot of errors. Compiling proves nothing.

True; only by studying the documentation for the implementations in
question can you discover whether they provide a definition of the
behavior that the standard leaves undefined. If both implementations
provide definitions compatible with your code, then it is portable
between those two implementations, at least. However, if either
implementation's documentation is silent on a relevant issue, all bets
are off.

Citizen C wrote:
> #include <stdio.h>
> #define p printf
> #define r0 return 0;
> #define P int main(void)\
> {\
> p("Hello world! NULL is %p\n", (void*)NULL);\
> r0\
> }
> P
>
> /*
> I think it is because Microsoft and gcc say so!
>
> Cheers!

A diagnostic is required because the posted code violates
the constraint of 5.1.1.1p3: "[...] A source file shall not end
in a partial preprocessing token or in a partial comment. [...]"
Other than that, I see nothing incorrect about it -- silly, yes,
but not actually incorrect.

--
Eric Sosman
lid