Re: [Reduce-algebra-developers] Merely loading avector hurts other code

[Rainer S. <rai...@gm...>]

I found this while cleaning my mailbox....

The avector package itself uses "cross" as operator, never ^. Ie. it is safe to 
remove the redefinition of the ^ token from the package.

The package redefines rcons from rlisp/list.red, which needed an update as well.

  Rainer


On Thu, 16 Feb 2012 at 07:59 -0000, Arthur Norman wrote:

 > 1: load_package avector;
 > 
 > *** ^ redefined
 > 
 > 2: on backtrace;
 > 
 > 3: int(1/(x^6-1), x);
 > +++ Error: getv nil
 > Inside: vectorcrossprod
 > Arg1: (x 6)
 > apply: vectorcrossprod
 > Inside: vecsm!*
 > Arg1: (cross x 6)
 > Inside: vecsm!*
 > Arg1: (difference (cross x 6) 1)
 > Inside: vecsm!*
 > Arg1: (quotient 1 (difference (cross x 6) 1))
 > Inside: vecsm!*
 > Arg1: (int (quotient 1 (difference (cross x 6) 1)) x)
 > 
 > 4:
 > 
 > I am posting this so it is in the list of things to deal with sometime - 
 > is there an "avector" expert in the house? Well actually I guess this is 
 > just that avector changes the syntax so that "x^6" is a cross product not 
 > a power of x, but this really seems to me to be a horrid trap and a 
 > substantial number of the test scripts use "^" rather than "**".
 > 
 > The simplest patch I can think of at present is to make the syntax in 
 > avector use say ^^ for a cross product rather than merely ^. That would be 
 > an incompatible change that would hurt avector users but would make it 
 > easier for people who do not use avector at present to load the package 
 > and maybe sometimes benefit from it.
 > 
 > Thoughts about that - or other solutions? And suggestions for a procedure 
 > for deciding that changes that break things for some people can be 
 > accepted - it feels dangerous to take unilateral action.
 > 
 > ==============
 > 
 > load_package invbase;
 > in "packages/alg/alg.tst";
 > 
 > fails with "fluid gg invalid as array"
 > 
 > because invbase make gg fluid. I bet other packages make things fluid in 
 > ways that gratuitously pollute the name-space for users. Again does 
 > anybody have a neat suggestion for a way to improve matters? In this case
 > I can just rename the variable in invbase to be "more obscure" (eg say 
 > invbase_gg) and not feel too nervous, but it would be good to detect and
 > mend ALL such cases!
 > 
 >   Arthur
 > 
 > 
 > ------------------------------------------------------------------------------
 > Virtualization & Cloud Management Using Capacity Planning
 > Cloud computing makes use of virtualization - but cloud computing 
 > also focuses on allowing computing to be delivered as a service.
 > http://www.accelacomm.com/jaw/sfnl/114/51521223/
 > _______________________________________________
 > Reduce-algebra-developers mailing list
 > Red...@li...
 > https://lists.sourceforge.net/lists/listinfo/reduce-algebra-developers
 > 

   Rainer Schöpf