arb_fmpz_poly.h – extra methods for integer polynomials

This module provides methods for FLINT polynomials with integer and rational coefficients (fmpz_poly_t) and (fmpq_poly_t) requiring use of Arb real or complex numbers.

Some methods output real or complex numbers while others use real and complex numbers internally to produce an exact result. This module also contains some useful helper functions not specifically related to real and complex numbers.

Note that methods that combine Arb polynomials and FLINT polynomials are found in the respective Arb polynomial modules, such as arb_poly_set_fmpz_poly() and arb_poly_get_unique_fmpz_poly().

Evaluation

void _arb_fmpz_poly_evaluate_arb_horner(arb_t res, const fmpz * poly, slong len, const arb_t x, slong prec)
void arb_fmpz_poly_evaluate_arb_horner(arb_t res, const fmpz_poly_t poly, const arb_t x, slong prec)
void _arb_fmpz_poly_evaluate_arb_rectangular(arb_t res, const fmpz * poly, slong len, const arb_t x, slong prec)
void arb_fmpz_poly_evaluate_arb_rectangular(arb_t res, const fmpz_poly_t poly, const arb_t x, slong prec)
void _arb_fmpz_poly_evaluate_arb(arb_t res, const fmpz * poly, slong len, const arb_t x, slong prec)
void arb_fmpz_poly_evaluate_arb(arb_t res, const fmpz_poly_t poly, const arb_t x, slong prec)
void _arb_fmpz_poly_evaluate_acb_horner(acb_t res, const fmpz * poly, slong len, const acb_t x, slong prec)
void arb_fmpz_poly_evaluate_acb_horner(acb_t res, const fmpz_poly_t poly, const acb_t x, slong prec)
void _arb_fmpz_poly_evaluate_acb_rectangular(acb_t res, const fmpz * poly, slong len, const acb_t x, slong prec)
void arb_fmpz_poly_evaluate_acb_rectangular(acb_t res, const fmpz_poly_t poly, const acb_t x, slong prec)
void _arb_fmpz_poly_evaluate_acb(acb_t res, const fmpz * poly, slong len, const acb_t x, slong prec)
void arb_fmpz_poly_evaluate_acb(acb_t res, const fmpz_poly_t poly, const acb_t x, slong prec)

Evaluates poly (given by a polynomial object or an array with len coefficients) at the given real or complex number, respectively using Horner’s rule, rectangular splitting, or a default algorithm choice.

Utility methods

ulong arb_fmpz_poly_deflation(const fmpz_poly_t poly)

Finds the maximal exponent by which poly can be deflated.

void arb_fmpz_poly_deflate(fmpz_poly_t res, const fmpz_poly_t poly, ulong deflation)

Sets res to a copy of poly deflated by the exponent deflation.

Polynomial roots

void arb_fmpz_poly_complex_roots(acb_ptr roots, const fmpz_poly_t poly, int flags, slong prec)

Writes to roots all the real and complex roots of the polynomial poly, computed to prec accurate bits. The real roots are written first in ascending order (with the imaginary parts set exactly to zero). The following nonreal roots are written in arbitrary order, but with conjugate pairs grouped together (the root in the upper plane leading the root in the lower plane).

The input polynomial must be squarefree. For a general polynomial, compute the squarefree part \(f / \gcd(f,f')\) or do a full squarefree factorization to obtain the multiplicities of the roots:

fmpz_poly_factor_t fac;
fmpz_poly_factor_init(fac);
fmpz_poly_factor_squarefree(fac, poly);

for (i = 0; i < fac->num; i++)
{
    deg = fmpz_poly_degree(fac->p + i);
    flint_printf("%wd roots of multiplicity %wd\n", deg, fac->exp[i]);
    roots = _acb_vec_init(deg);
    arb_fmpz_poly_complex_roots(roots, fac->p + i, 0, prec);
    _acb_vec_clear(roots, deg);
}

fmpz_poly_factor_clear(fac);

All roots are refined to a relative accuracy of at least prec bits. The output values will generally have higher actual precision, depending on the precision used internally by the algorithm.

This implementation should be adequate for general use, but it is not currently competitive with state-of-the-art isolation methods for finding real roots alone.

The following flags are supported:

  • ARB_FMPZ_POLY_ROOTS_VERBOSE

Special polynomials

Note: see also the methods available in FLINT (e.g. for cyclotomic polynomials).

void arb_fmpz_poly_cos_minpoly(fmpz_poly_t res, ulong n)

Sets res to the monic minimal polynomial of \(2 \cos(2 \pi / n)\). This is a wrapper of FLINT’s fmpz_poly_cos_minpoly, provided here for backward compatibility.

void arb_fmpz_poly_gauss_period_minpoly(fmpz_poly_t res, ulong q, ulong n)

Sets res to the minimal polynomial of the Gaussian periods \(\sum_{a \in H} \zeta^a\) where \(\zeta = \exp(2 \pi i / q)\) and H are the cosets of the subgroups of order \(d = (q - 1) / n\) of \((\mathbb{Z}/q\mathbb{Z})^{\times}\). The resulting polynomial has degree n. When \(d = 1\), the result is the cyclotomic polynomial \(\Phi_q\).

The implementation assumes that q is prime, and that n is a divisor of \(q - 1\) such that n is coprime with d. If any condition is not met, res is set to the zero polynomial.

This method provides a fast (in practice) way to construct finite field extensions of prescribed degree. If q satisfies the conditions stated above and \((q-1)/f\) additionally is coprime with n, where f is the multiplicative order of p mod q, then the Gaussian period minimal polynomial is irreducible over \(\operatorname{GF}(p)\) [CP2005].