RE: [boost] Physical quantities (was Re: runtime unit-modifiablephysical-quantitiesusingpqs1-00-01)

From: Paul A. Bristow (boost_at_[hidden])
Date: 2004-01-07 18:12:38

| -----Original Message-----
| From: boost-bounces_at_[hidden]
| [mailto:boost-bounces_at_[hidden]]On Behalf Of Deane Yang
| Sent: Wednesday, January 07, 2004 4:06 AM
| To: boost_at_[hidden]
| Subject: [boost] Physical quantities (was Re: runtime
| unit-modifiablephysical-quantitiesusingpqs1-00-01)
|
| And, wouldn't this units library be much simpler to develop than the
| whole physical dimensions library? I think that Abrahams, Duncan, and
| Schabel have already pretty much described how to do it. The only
| difficulty I am aware of is the compilation time involved in handling
| lists of types that represent the different fundamental units in
| a derived unit.

I support this view.

But the compile time IS a significant drawback, though IMO not a killer
problem - faster processors will keep reducing it.

My guess is that at least 90% of applications will use only SI units.

(And not a lot will use the 'proof-of-concept' furlong/stone/fortnight units
;-) )

As the Schabel code stands, are we 're-meta/macro-processing' the same input
again and again every time we use the package in SI-only units?

Can we foresee any way of only performing the 'meta-processing' once, analogous
to taking the output from the C/CX++ pre-processor, and then using that as
'source' for the compiler proper?  Or is this the inevitable cost of using
templates?

Paul

Paul A Bristow, Prizet Farmhouse, Kendal, Cumbria, LA8 8AB  UK
[phone]   Mobile [phone]
mailto:pbristow_at_[hidden]