Specifications for the abstract ring of multivariate polynomials, with several bases
Ticket 6629
First micro draft
Setup the framework for MultivariatePolynomials with several bases:
Let us work over `F=\QQ(q,t)` (will be needed for Macdonald polynomials)::
sage: F = FractionField(QQ['q,t']); (q,t) = F.gens(); F.rename('QQ(q,t)')
We construct an (abstract) ring of multivariate polynomials over F::
sage: P = AbstractPolynomialRing(F, 'x0,x1,x2'); P
The abstract ring of multivariate polynomials in x0, x1, x2 over QQ(q,t)
See
[[http://mupad-combinat.svn.sourceforge.net/viewvc/mupad-combinat/trunk/MuPAD-Combinat/lib/EXAMPLES/MultivariatePolynomials.mu?view=markup|examples::MultivariatePolynomials]] ([[http://mupad-combinat.svn.sourceforge.net/viewvc/mupad-combinat/trunk/MuPAD-Combinat/lib/EXAMPLES/TEST/MultivariatePolynomials.tst?view=markup|Tests]]
for a preliminary implementation in MuPAD-Combinat
This ring has several bases, starting with the usual monomial basis::
sage: m = P.m
Multivariate Polynomial Ring in x0, x1, x2 over QQ(q,t)
sage: m.basis().keys()
Integer vectors of length 3
sage: x0,x1,x2 = m.gens()
sage: m.term([3,1,2]) + x2^3 + 3
x0^3*x1*x2^2 + x2^3 + 3
sage:: m is MultivariatePolynomialRing(F, 'x0,x1,x2')
True
sage: m.print_style(style = "vectors")
sage: m.term([3,1,2]) + x2^3 + 3
x[3,1,2] + x[0,0,3] + 3*x[0,0,0]
The Schubert basis::
sage: Y = P.schubert()
sage: Y
Multivariate polynomials in the Schubert basis
sage: Y.basis().keys()
Integer vectors of length 3
sage: Y[1,0,0] * Y[0,1,0]
Y[1,1,0] + Y[2,0,0]
sage: m(Y[1,0,0])
x0
# TODO: add larger examples!
# Design to be discussed in the long run
# One can optionaly specify an alphabet, whose elements should leave in the ground ring::
# sage: Y = P.schubert([x2,x1,x0])
# If the ground ring allows for it, the default alphabet should be [y0,y1,y2]
The key polynomials (Demazure characters)::
sage: K = P.key_polynomials(type = "A")
sage: K
Multivariate polynomials in the key polynomial basis
sage: K.basis().keys()
Integer vectors of length 3
sage: K(x0 * (x0+x1))
K[0,2,0] - K[2,0,0]
sage: K[1,0,0] * K[0,1,0]
K[1,1,0] + K[2,0,0]
sage: m(K[0,2,0] - K[2,0,0])
x0^2 + x0*x1
# TODO: add larger examples computed with ACE!
The Demazure atoms::
sage: HK = P.demazure_atoms(type = "A")
Or possibly::
sage: HK = K.dual()
sage: HK
Multivariate polynomials in the demazure atoms basis
sage: HK.dual()
Multivariate polynomials in the key polynomial basis
sage: scalar(HK[5,2,4], K[4,2,5]) # Watch for the reversal of the vector (weight)
1
sage: HK.basis().keys()
Integer vectors of length 3
sage: H
sage: HK[1,0,0] * HK[0,1,0]
K[1,1,0] + K[2,0,0]
sage: m(Y[1,0,0])
x0
# TODO: add larger examples!
The SchurSchubert basis (see MuPAD)::
sage:: P.SchurSchubert()
This is the free module over Schur polynomials with basis Schubert
polynomials; the later are indexed by (the code of) permutations
of `S_n`.
sage:: P.coeffRing()
Symmetric polynomials in the Schur basis over QQ(q,t)
sage:: P.basis().keys()
Permutations of S_n ? or Codes ?
sage:: P.basis().cardinality()
6
Other bases in MuPAD-Combinat:
* NonSymmetricHL, NonSymmetricHLdual
* UniversalDecompositionAlgebra (free module over symmetric
functions in the e basis, with monomial below the stair as basis
* FreeModule over symmetric functions in the e basis over t, with
descent monomials as basis.
Non symmetric Macdonald polynomials (should recycle the current sage.combinat.sf.ns_macdonald)
sage: Macdo = P.MacdonaldPolynomials(q, t)
sage: E = Macdo.E(pi = [3,1,2]); E
Multivariate Polynomial Ring in x0, x1, x2 over QQ(q,t), in the Macdonald E basis, with basement [3,1,2]
sage: E[1,0,0]
E[1,0,0]
sage: m(E[1,0,0])
x0}}}