std.rational -- update and progress towards review

[Joseph Rushton Wakeling]

On 05/10/13 00:40, Brad Roberts wrote:
> That's not an argument against splitting the changes up and getting the
> dependencies handled first.  It's an argument to do a good job with both sets of
> changes, which I agree with.

Understood, but given that currently the supposedly independent changes are 
_only_ used by std.rational, there is a case for presenting everything in 
context together, because people may have alternative suggestions for how to 
handle these cases which don't rely on new templates.

Yes, I am trying to avoid a situation where I get caught up in weeks of pull 
request feedback before I can even think about submitting std.rational for 
review, but it's not _all_ about saving me time. :-)