CoCoA » CoCoALib

I know this is not pretty, but for the time being there is this workaround (if you can put your points in an affine space)

/**/ Use R ::= QQ[x,y,z];
/**/ AffPolyRing := NewPolyRing(QQ, first(IndetSymbols(R),2));
/**/ phi := PolyAlgebraHom(AffPolyRing, R, first(indets(R),2));

/**/ Pts := [[0,0,1],[1/2,1,1],[0,1,2]];
/**/ AffPts := [ [P[1]/P[3], P[2]/P[3]] | P In Pts];

/**/ AffI := IdealOfPoints(AffPolyRing, Mat(AffPts));
/**/ I := ideal([homog(phi(F), last(indets(R))) | F in gens(AffI)]);
/**/ I;
ideal(y^2 -x*z +(-1/2)*y*z, x*y -x*z, x^2 +(-1/2)*x*z)

Anna has done the work, but it gives obviously wrong result (not homog).

Here is a failing example:

use P ::= QQ[x,y,z];
L := [[1,1,2],[1,2,4]];
I := IdealOfProjectivePoints(P, mat(L));
ideal(y +(-1/2)*z,  x^2 +(-3/4)*x +(1/8)*z)
--> result is not homog

The result is "almost right": the coeffs are correct, but the PPs are wrong. I compared with C-4.7.6.
Correct answer is:

ideal(y +(-1/2)*z,  x^2 +(-3/4)*x*z +(1/8)*z^2)

There was just one test. I have just tried it, and it worked. So closing.