[boost] Re: math_constants parameterisation and return by ref

From: Deane Yang (deane_yang_at_[hidden])
Date: 2004-01-22 09:54:23

Paul A. Bristow wrote:

> There was widespread agreement that no solution requiring pi() would be widely
> acceptable.
> 

I'm sorry if my recollection is incorrect, but I only remember seeing
messages from you stating this. I do not remember seeing anyone else
on the boost mailing list expressing this view.

I agree that the syntax is less than ideal, but so is the syntax for
MPL and many other things in C++.

It appears to me that the efforts to replace "pi()" by "pi" are
both testing the limits of current compilers AND delaying the review
of the constants library. Why can't we get moving on the constants
library with a correct C++ interface and, if a simpler interface is
developed later, modify the library later (with backwards compatibility,
which should not be a problem)?

I also do not understand your characterization of the "no macros policy"
as being "Politically Correct". There are good reasons for this policy,
and they have been articulated quite clearly by those who ask for it.

I vote for the easier-to-implement interface (i.e., "pi()")
and no macros and let's get this constants library out the door.
It IS very useful.