Re: [Axiom-developer] Unit package proposals and questions

[C Y]

--- William Sit <address@hidden> wrote:

> Just a quick answer for now:
> 
> You can use
> 
>  s:=setDimUnit("Length", "cm",('s)::S)
> 
> if s is an unassigned symbol. The quote in the argument tells Axiom
> to use s as a symbol temporarily, but then assign it to the domain
> SIUnitSystem(S) provided S has RetractableTo(Symbol) such as POLY
> INT.

Hmm.  OK.  As a design consideration I would prefer to hide this logic
within the definition of the set functions, but that's just a detail. 
The important thing is it's possible.

> So for "variables" (symbols), that is not an issue. For expressions,
> you need not assign it to a variable first:
> 
>   s:=setDimUnit("Length","cm", P*V)
> 
> is quite ok (assuming you want the aggregate variable PV).  

Right :-).  I think that use is much less common in practice than using
the same name for the nondim + dimension object and the nondim
variable, but it's good it can be done at need.  (Heck, I never
realized you COULD do something like that - I thought of x [m] as being
inseparable.)

> In our discussion, we did not use any specific expression and 
> use "a" to stand for one, so I have to use a new identifier "ua", 
> assuming "a" is already bound.

Ah :-).  Opps.  Should have used "t" from the x=x0+v0t+1/2at^2 and time
- that would have been a better example.  My bad.
 
> I'll respond to other questions later tonight. Got to go to school.

Sounds good!  I'll start going over the discussions and see if I can
distill it down over the next few days, which probably means a lot more
not-seeing-things-clearly questions :-/.  I've got to think about how
to structure this sucker first, since it is clear there will be
extensions to add for quite some time.

Cheers,
CY

__________________________________________________
Do You Yahoo!?
Tired of spam?  Yahoo! Mail has the best spam protection around 
http://mail.yahoo.com