RE: [boost] Re: rationals initialization bug

From: Paul Moore (gustav_at_[hidden])
Date: 2000-11-28 17:04:41

From: austern_at_[hidden]
> It seems to me that what people are really asking for is an
> algorithm, given a rational number p and an integer N, for
> finding the closest rational number q whose denominator is
> no greater than N.  That's a generally useful algorithm; it
> shouldn't be hidden away inside a constructor, and there
> should never be an implied value for N.  (Because only the
> user knows what the appropriate value for N is.)

Quite possibly. That's exactly what Reggie Seagraves kindly supplied. If
there is a general feeling that this would be worth including, I am happy to
do so.

> On the other hand, conversion from a string doesn't strike me
> as such a bad idea.  Radix conversion is easy; it's parsing that's
> hard.  If you're looking at a decimal string, it's easy to see
> that "0.1" is 1*10^-1.

Hm. I guess take the point. As long as we exclude exponential notation, it's
not too hard to do. But I still worry about edge conditions caused by
limited precision integers...

Can someone provide some real examples where a constructor from string is
really (really, really, truly :-) important? I don't want to start adding
"nice to haves" without a genuine benefit. And when you do, remember that
it's not hard to write it yourself if you need it - so why should it be part
of the class?

> If people are interested, I can extract the floating-point parsing
> code from the SGI I/O library.

Probably not necessary - unless exponential notation is important, it's not
too hard to do by hand.

Thanks,
Paul.