Date: Mon, 7 Oct 1996 11:16:25 -0700 (MST) From: Terry Lambert <terry@lambert.org> To: j@uriah.heep.sax.de (J Wunsch) Cc: freebsd-hackers@freebsd.org Subject: Re: I plan to change random() for -current (was Re: rand() and random()) Message-ID: <199610071816.LAA14477@phaeton.artisoft.com> In-Reply-To: <199610071752.TAA15133@uriah.heep.sax.de> from "J Wunsch" at Oct 7, 96 07:52:33 pm
next in thread | previous in thread | raw e-mail | index | archive | help
> The requirements for rand() are being set straight in the ANSI and ISO > documents. There are no implementation details, and i don't see why > we are obligued to keep a buggy implementation just since some people > on the earth might be used to this one. > > If some particular package relies on the pseudo-unrandomness of a > particular implementation, they should ship this particular > implementation along with their sources (and give it a name that > doesn't clash with the standard). There's nothing more they could > expect from a standard-conforming implementation than to conform to > the standard. 1) Please *strongly* document this change. 2) Please provide a library "orand" function in libcompat and mention in the documentation; feel free to remove the orand (or replace it) in the next release. If you do this, then I will be happy (or at least grudgingly nonopposed). I would prefer that old code linked against a difference version of libc kick out a printf when they call the changed random, since this is a possibly high impact version mismatch. Terry Lambert terry@lambert.org --- Any opinions in this posting are my own and not those of my present or previous employers.
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199610071816.LAA14477>