-------- Original Message -------- Subject: Re: About PPL Date: Fri, 18 Mar 2005 06:05:50 -0800 (PST) From: gulsah yilmaz <gulsahyilmaz20@yahoo.com> To: Roberto Bagnara <bagnara@cs.unipr.it> Mr. Bagnara, First of all, thanks for your attention. Since Polylib uses this representation I think you will be familiar with the representation. I will exlplain what I want to do: I have inequalities like: x+M+N+5 >= 0 y+z+1 >= 0 x+y >= 0 x+z+1 >= 0 z+M+2>=0 and an inequality which include unknown coefficients like: (a-b)x+cz+(d+e)M+N >= 0 and by combining these inequalities I want to know the relation between unknowns like: a+b>1 d-e+c>M .... Thanks Gulsah Yilmaz --- Roberto Bagnara <bagnara@cs.unipr.it> wrote:

Dear Gulsah,

if the file `Makefile' is missing, it is quite likely you did not configure your build directory. Please follow the instructions in the files `README.configure' and `INSTALL' (they are part of the library's documentation).

...

An example to my problem: The numbers are faik.

x y z M N Constant 1 1 0 0 1 1 5 1 0 1 1 0 0 1 1 1 1 0 0 0 0 1 1 0 1 0 0 1 1 0 0 1 1 0 2

and 1 a-b 0 c d+e 1 0

I want to know the relation between the unknown coefficient a,b,c,d,e. I ask you whether this is possible or not by using PPL.

Sorry, but the table above means nothing to me: I can imagine the columns named `x', `y' and so forth contain coefficients of the variables with the same names; I can imagine the column named `Constant' gives the inhomogeneous term; but I cannot make any sense of the first, unnamed column; then I don't know if that is meant to represent a set of constraints (equalities? non-strict inequalities? <=? >=? something else?) or a set of generators.

Can you please explain your problem in the language of linear algebra? Chances are that we will be able to answer your question then. Cheers,

__________________________________________________ Do You Yahoo!? Tired of spam? Yahoo! Mail has the best spam protection around http://mail.yahoo.com -- Prof. Roberto Bagnara Computer Science Group Department of Mathematics, University of Parma, Italy http://www.cs.unipr.it/~bagnara/ mailto:bagnara@cs.unipr.it