Jaime,
thank you very much for your hints. At last I got what I wanted. Here are the
working lines:
(%i1) a:1;
(%i2) b:1;
(%i3) c:1;
(%i4) e:1;
(%i5) f:0.5;
(%i6) g:1;
(%i7) r:a*x+b*y+c;
(%i8) s:e*x+f*y+g;
(%i9) P:linsolve([r,s],[x,y]);
(%i10) ex:r+K*s;
(%i20) define (Z(x,K), rhs ( (solve(ex,y)[1])));
(%i21) F:makelist(Z(x,K),K,0,10);
(%i22) plot2d(F,[x,-5,5]);
It is not that I insist on using functions... I tried both:
Z: rhs ( first(solve(ex,y)));
and
define (Z(x,K), rhs ( (solve(ex,y)[1])));
but only the latter worked. For sure I must get hold of a good book on Maxima.
Pier
On Sat, 04 Feb 2012 13:34:17 +0000, Jaime Villate wrote
> On 02/03/2012 08:48 PM, Pier wrote:
> > (%i6) a:1;
> > (%i7) b:1;
> > (%i8) c:1;
> > (%i9) e:1;
> > (%i10) f:0.5;
> > (%i11) g:1;
> > (%i12) r:a*x+b*y+c;
> > (%i13) s:e*x+f*y+g;
> > (%i14) P:linsolve([r,s],[x,y]); centre point of the star
> > (%i15) ex:r+K*s;
> > (%i16) solve(ex,y); copy and paste the rhs of the output as rhs of %i17
> > (%i17) Z(x,K):=-((2*x+2)*K+2*x+2)/(K+2);
> > (%i18) F:makelist(Z(x,K),K,0,10);
> > (%i19) plot2d(F,[x,-5,5]);
> >
> > One thing I haven't been able to do is to avoiding copying-pasting the rhs
> > output of %i16
> >
> > y=-((2*x+2)*K+2*x+2)/(K+2)
> >
> > as the rhs in %i17. I tried with:
> >
> > Z(ex,K):=solve(ex,y);
> >
> > but that didn't work.
> I guess you meant "Z(x,K) := solve(ex,y)", but that will fail by two
> reasons. The first reason,
> already pointed out by Ray, is that solve will give you a list of
> equations, even if there is
> only one solution. You want just the right hand side of the first
> equation in the list returned
> by solve: rhs(first(solve(ex, y)))
> (or use [1] instead of first).
>
> The second reason is that := will not evaluate its arguments. That
> means that in the command "Z(x,K) := rhs (first (solve (ex, y)))" ex
> will not be solved to find y, but that step will be delayed until
> you try to evaluate Z(x,k) at some later time; that might work but
> it might also give you unintended results; for instance:
>
> (%i15) Z(x,K) := rhs ( first(solve(ex,y)));
> (%o15) Z(x, K) := rhs(first(solve(ex, y)))
> (%i16) y:3;
> (%o16) 3
> (%i17) Z(2,4);
>
> solve: all variables must not be numbers.
> #0: Z(x=2,k=4)
> -- an error. To debug this try: debugmode(true)
>
> by the time Z tried to use solve, y was no longer an undefined
> variable so it failed.
>
> As I pointed out in my previous message you should better use expressions:
>
> Z: rhs ( first(solve(ex,y)));
>
> but if you insist on using functions, then use define() instead of :=
>
> define (Z(x,K), rhs ( first(solve(ex,y))));
>
> the difference between f(x):=exp and define(f(x),exp) is that define
> will first evaluate its arguments
> before defining the function. In the case above, solve, first and
> rhs will be executed before Z(x,K) is defined.
>
> I also use Maxima extensively in my Physics lectures; I'm interested
> to know what courses you are teaching.
>
> Cheers,
> Jaime
>
> _______________________________________________
> Maxima mailing list
> Maxima at math.utexas.edu
> http://www.math.utexas.edu/mailman/listinfo/maxima
---------------------------------------------------------------------------------------------------------------------------------------------
....../\ /\
......0 0
.....=O=
< Pierp >
---------------------------------------------------------------------------------------------------------------------------------------------
Linux Gentoo Ker. 2.6.37/Ker. 2.6.19.2 RTAI Vulcano
Web: http://pier.unirc.eu
skype: pier_carla
irc: irc.unirc.eu #linux
icq: 325474062
--------------------