[D-runtime] druntime commit, revision 410

[Jonathan M Davis]

On Friday 05 November 2010 06:06:09 Steve Schveighoffer wrote:
> That sound exactly like what I wanted, thanks!
> 
> BTW, when I said I think std.datetime should use core.time.Duration as its
> duration type, I meant that in terms of where the Duration type was
> located, not which implementation should be used.  I guess I didn't really
> make that clear. Definitely, we should use the most complete
> implementation.

Well, I figured that that's what you meant, but I'm obviously more partial to 
using my implementation unless there's something a lot better about the other 
(especially since so much of std.datetime uses it), but I really do think that 
what I have is better, and if there's something that needs to be adjusted about 
it to work better in core.time, then that can be done. Regardless, the 
suggestion that core.time and std.datetime should use the same Duration type 
makes good sense. Perhaps if implicit conversions could be easily created, then 
having two types wouldn't be all that big a deal, but they can't be easily 
created, and even then, it would still be preferable to have them use the same 
type if at all reasonable.

In any case, I don't get the impression that core.time needs anything 
particularly fancy while std.datetime has somewhat more expansive needs, so it 
makes more sense that the Duration type cater somewhat more to what std.datetime 
needs while making sure that it does what core.time needs. But for the most 
part, I expect they need the same thing - it's just that std.datetime needs a 
bit more.

- Jonathan M Davis