
Boost Users : 
From: Jeff Garland (jeff_at_[hidden])
Date: 20050122 10:51:07
On Wed, 19 Jan 2005 13:54:22 +0100, Erik Thiele wrote
> hi
Hi  sorry to be slow on the response...anther busy week...
> the docs of time_duration say that minutes() returns 0..59 and
> seconds() return 0..59. this is not true, the correct answer is,
> minutes() returns 59..+59 and seconds() also.
Yes, you are correct we need to fix the docs.
> the handling of negative durations is not fully documented. i.e. is
> it true that the sign of hours(),minutes() and seconds() is always
> equal to each others sign?
Yes  the sign is reflected in all the components.
> x=time_duration(anything...);
> y=x.invert_sign();
>
> is the following true for all values of x:
>
> abs(x.hours()) == abs(y.hours())
> abs(x.minutes()) == abs(y.minutes())
> abs(x.seconds()) == abs(y.seconde())
Yes.
> but whatever way you document it, it should somehow be documented.
Agree.
> to continue on my work i now make the assumption that in the
> following assertions are true for all values of x:
>
> if (x.is_negative()) x = x.invert_sign();
> assert(x.hours()>=0);
> assert(0<=x.minutes()<=59);
> assert(0<=x.seconds()<=59);
>
> i hope this assertions are correct for all values of x !!!!! :)
Yes, they are. We'll work on fixing the docs  thx for the report.
Jeff
Boostusers list run by williamkempf at hotmail.com, kalb at libertysoft.com, bjorn.karlsson at readsoft.com, gregod at cs.rpi.edu, wekempf at cox.net