Message-Id: <199801211711.TAA06183@ankara.duzen.com.tr>
Comments: Authenticated sender is <mitch AT ankara DOT duzen DOT com DOT tr>
From: "S. M. Halloran" <mitch AT duzen DOT com DOT tr>
Organization: User RFC 822- and 1123-Compliant
To: djgpp AT delorie DOT com
Date: Wed, 21 Jan 1998 19:11:57 +0200
MIME-Version: 1.0
Content-type: text/plain; charset=US-ASCII
Content-transfer-encoding: 7BIT
Subject: Re: random numbers?
CC: mdang AT delorie DOT com
In-reply-to: <34C608AE.2569@utc.fr>
Precedence: bulk

On 21 Jan 98, Van Mo Dang was found to have commented thusly:

> I would have two questions about random() and srandom() :
> 
> 1 - Are those routines portables ? I found them both in DJGPP 
> and on Solaris systems, but can they be considered to be available
> in any C standard library ?

I don't have any copy of the ANSI C standard and what functions are 
considered to be part of the library.  (Where can that be had?)  
However, I can say that your functions originated with the Berkeley 
distribution (even the DJGPP include on the functions mentions it), 
and so they may be POSIX but not ANSI.  The functions rand() and 
srand() are the equivalents in SVR4 Unix (USL), but I cannot say if 
they are standard functions;  my man page tells me that drand48() and 
random() are more "elaborate" and produce a better distribution.  
Newsgroup comp.lang.c probably is the place to look for the final 
word on portability.

> 
> 2 - Which file should be included to know their prototypes ? On
> Solaris, it seems that neither <stdio.h> nor <stdlib.h> contain any
> prototype of random() and srandom().  Actually none of the ".h"
> files in /usr/include contains their declaration.

That's the darnedest thing for me too.  Grep on my /usr/include and 
/usr/include/sys on a system running SVR4 doesn't pull out the 
prototype for those, even though they are available.  I guess you 
can just set up your own header or put the prototypes at the start of 
your source in the appropriate location.  Don't yank out the DJGPP 
prototypes and put them in however.  Check the man pages on the 
Solaris and yank them out of there.

Mitch Halloran
Research (Bio)chemist
Duzen Laboratories Group
Ankara   TURKEY
mitch AT duzen DOT com DOT tr

other job title:  Sequoia's (dob 12-20-95) daddy