On Wed, 9 Mar 2011 17:18:10 +0100 Joachim Faulhaber <afojgo@googlemail.com> wrote:
Because n1692, the paper presented to the C++ Standards Committee in 2004 that I based XInt's interface on, has them there.
Nope, I just looked at the paper, they are free standing functions there. And they are named without the is_ prefix:
<sigh> My statement was true when I originally wrote the code. I didn't realize that the later changes took it so far away from the original design. Thanks for the correction.
you are rehabilitated ;-) There are 3 papers N1692, N2020, N2143 on the topic and in the *latest* paper http://www.open-std.org/jtc1/sc22/wg21/docs/papers/2007/n2143.pdf the functions is_odd(), is_even(), sign() are member functions just as in your implementation! So you adapted to the upcoming standard here.
That thump that you just heard was the sound of my jaw hitting the ground. I didn't know about the later papers, so I can't claim any credit for that. (I can't download either of the new-to-me ones at the moment either, apparently due to a server glitch.) If it weren't for the date on the page that lists them, I might think that Mr. Kronenburg took inspiration from XInt! :-)
It would be better for you to add a reference to the latest paper to your docs and refer to it next time :)
I've already added it to the docs. Thanks for letting me know about them. -- Chad Nelson Oak Circle Software, Inc. * * *