When I try this piece of code the outcome seems (to me at least) not as
it should be. Is this is a bug or am I just missing something in the way
parameters are passed?
(Print function based on code from Sean L. Palmer as posted in the
'va_list / rolling your own printf / Outbuffer' thread. I couldn't find
any other information on it so there might just be something else which
would explain for the behaviour).


import c.stdio;
import string;

void Print( char[] msg, ... )
{
	va_list ap;
	ap = cast(va_list)&msg;
	ap += msg.size;

	char[] buf;
	buf.length = 1024;
	vsprintf(buf, msg, ap);
	buf.length = strlen(buf);

	printf( "%.*s", buf );
}

int main( char [] [] args )
{
	float test = 1.2345;
	Print( "%f\n", 1.2345 );
	Print( "%f\n", test );

	return 1;
}

This outputs:

1.234500
0.000000

- Marcel

Try changing test from float to double, or put an 'f' on the end of your 1.2345.  Should make clear what's going on.

I don't think D promotes va_list arguments automatically the same way C does.

Also I think it's lame how you're having to typecast to va_list and skip the msg buffer manually.  There has to be a better way.  We should be able to name the '...' parameter and refer to it directly as a va_list.

Sean

"Marcel Strik" <mars_888@hotmail.com> wrote in message news:Xns941FE6B8F27D5marsvulcanushotmail@63.105.9.61...
> When I try this piece of code the outcome seems (to me at least) not as
> it should be. Is this is a bug or am I just missing something in the way
> parameters are passed?
> (Print function based on code from Sean L. Palmer as posted in the
> 'va_list / rolling your own printf / Outbuffer' thread. I couldn't find
> any other information on it so there might just be something else which
> would explain for the behaviour).
>
>
> import c.stdio;
> import string;
>
> void Print( char[] msg, ... )
> {
> va_list ap;
> ap = cast(va_list)&msg;
> ap += msg.size;
>
> char[] buf;
> buf.length = 1024;
> vsprintf(buf, msg, ap);
> buf.length = strlen(buf);
>
> printf( "%.*s", buf );
> }
>
> int main( char [] [] args )
> {
> float test = 1.2345;
> Print( "%f\n", 1.2345 );
> Print( "%f\n", test );
>
> return 1;
> }
>
> This outputs:
>
> 1.234500
> 0.000000
>
> - Marcel

Thanks for the quick reply and solution. Casting the values to doubles was about the only thing I hadn't tried yet. I guess this type of behaviour is just an artifact of interfacing D with a C function.

It would indeed be nice if there was better support for multiple arguments in D or at least a better way to do the same (The C/C++ way seems a bit outdated to me).

- Marcel

"Sean L. Palmer" <palmer.sean@verizon.net> wrote in news:bnen7u$1g2m$1@digitaldaemon.com:

> Try changing test from float to double, or put an 'f' on the end of your 1.2345.  Should make clear what's going on.
> 
> I don't think D promotes va_list arguments automatically the same way C does.
> 
> Also I think it's lame how you're having to typecast to va_list and skip the msg buffer manually.  There has to be a better way.  We should be able to name the '...' parameter and refer to it directly as a va_list.
> 
> Sean
> 
> "Marcel Strik" <mars_888@hotmail.com> wrote in message news:Xns941FE6B8F27D5marsvulcanushotmail@63.105.9.61...
>> When I try this piece of code the outcome seems (to me at least) not
>> as it should be. Is this is a bug or am I just missing something in
>> the way parameters are passed?
>> (Print function based on code from Sean L. Palmer as posted in the
>> 'va_list / rolling your own printf / Outbuffer' thread. I couldn't
>> find any other information on it so there might just be something
>> else which would explain for the behaviour).
>>
>>
>> import c.stdio;
>> import string;
>>
>> void Print( char[] msg, ... )
>> {
>> va_list ap;
>> ap = cast(va_list)&msg;
>> ap += msg.size;
>>
>> char[] buf;
>> buf.length = 1024;
>> vsprintf(buf, msg, ap);
>> buf.length = strlen(buf);
>>
>> printf( "%.*s", buf );
>> }
>>
>> int main( char [] [] args )
>> {
>> float test = 1.2345;
>> Print( "%f\n", 1.2345 );
>> Print( "%f\n", test );
>>
>> return 1;
>> }
>>
>> This outputs:
>>
>> 1.234500
>> 0.000000
>>
>> - Marcel
> 
> 
>