FieldJacobiElliptic.java
/*
* Licensed to the Hipparchus project under one or more
* contributor license agreements. See the NOTICE file distributed with
* this work for additional information regarding copyright ownership.
* The Hipparchus project licenses this file to You under the Apache License, Version 2.0
* (the "License"); you may not use this file except in compliance with
* the License. You may obtain a copy of the License at
*
* https://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
package org.hipparchus.special.elliptic.jacobi;
import org.hipparchus.CalculusFieldElement;
import org.hipparchus.special.elliptic.carlson.CarlsonEllipticIntegral;
import org.hipparchus.special.elliptic.legendre.LegendreEllipticIntegral;
import org.hipparchus.util.FastMath;
/** Computation of Jacobi elliptic functions.
* The Jacobi elliptic functions are related to elliptic integrals.
* @param <T> the type of the field elements
* @since 2.0
*/
public abstract class FieldJacobiElliptic<T extends CalculusFieldElement<T>> {
/** Parameter of the function. */
private final T m;
/** Simple constructor.
* @param m parameter of the function
*/
protected FieldJacobiElliptic(final T m) {
this.m = m;
}
/** Get the parameter of the function.
* @return parameter of the function
*/
public T getM() {
return m;
}
/** Evaluate the three principal Jacobi elliptic functions with pole at point n in Glaisher’s Notation.
* @param u argument of the functions
* @return copolar trio containing the three principal Jacobi
* elliptic functions {@code sn(u|m)}, {@code cn(u|m)}, and {@code dn(u|m)}.
*/
public abstract FieldCopolarN<T> valuesN(T u);
/** Evaluate the three principal Jacobi elliptic functions with pole at point n in Glaisher’s Notation.
* @param u argument of the functions
* @return copolar trio containing the three principal Jacobi
* elliptic functions {@code sn(u|m)}, {@code cn(u|m)}, and {@code dn(u|m)}.
*/
public FieldCopolarN<T> valuesN(final double u) {
return valuesN(m.newInstance(u));
}
/** Evaluate the three subsidiary Jacobi elliptic functions with pole at point s in Glaisher’s Notation.
* @param u argument of the functions
* @return copolar trio containing the three subsidiary Jacobi
* elliptic functions {@code cs(u|m)}, {@code ds(u|m)} and {@code ns(u|m)}.
*/
public FieldCopolarS<T> valuesS(final T u) {
return new FieldCopolarS<>(valuesN(u));
}
/** Evaluate the three subsidiary Jacobi elliptic functions with pole at point s in Glaisher’s Notation.
* @param u argument of the functions
* @return copolar trio containing the three subsidiary Jacobi
* elliptic functions {@code cs(u|m)}, {@code ds(u|m)} and {@code ns(u|m)}.
*/
public FieldCopolarS<T> valuesS(final double u) {
return new FieldCopolarS<>(valuesN(u));
}
/** Evaluate the three subsidiary Jacobi elliptic functions with pole at point c in Glaisher’s Notation.
* @param u argument of the functions
* @return copolar trio containing the three subsidiary Jacobi
* elliptic functions {@code dc(u|m)}, {@code nc(u|m)}, and {@code sc(u|m)}.
*/
public FieldCopolarC<T> valuesC(final T u) {
return new FieldCopolarC<>(valuesN(u));
}
/** Evaluate the three subsidiary Jacobi elliptic functions with pole at point c in Glaisher’s Notation.
* @param u argument of the functions
* @return copolar trio containing the three subsidiary Jacobi
* elliptic functions {@code dc(u|m)}, {@code nc(u|m)}, and {@code sc(u|m)}.
*/
public FieldCopolarC<T> valuesC(final double u) {
return new FieldCopolarC<>(valuesN(u));
}
/** Evaluate the three subsidiary Jacobi elliptic functions with pole at point d in Glaisher’s Notation.
* @param u argument of the functions
* @return copolar trio containing the three subsidiary Jacobi
* elliptic functions {@code nd(u|m)}, {@code sd(u|m)}, and {@code cd(u|m)}.
*/
public FieldCopolarD<T> valuesD(final T u) {
return new FieldCopolarD<>(valuesN(u));
}
/** Evaluate the three subsidiary Jacobi elliptic functions with pole at point d in Glaisher’s Notation.
* @param u argument of the functions
* @return copolar trio containing the three subsidiary Jacobi
* elliptic functions {@code nd(u|m)}, {@code sd(u|m)}, and {@code cd(u|m)}.
*/
public FieldCopolarD<T> valuesD(final double u) {
return new FieldCopolarD<>(valuesN(u));
}
/** Evaluate inverse of Jacobi elliptic function sn.
* @param x value of Jacobi elliptic function {@code sn(u|m)}
* @return u such that {@code x=sn(u|m)}
* @since 2.1
*/
public T arcsn(final T x) {
// p = n, q = c, r = d, see DLMF 19.25.29 for evaluating Δ(q, p) and Δ(r, p)
return arcsp(x, x.getField().getOne().negate(), getM().negate());
}
/** Evaluate inverse of Jacobi elliptic function sn.
* @param x value of Jacobi elliptic function {@code sn(u|m)}
* @return u such that {@code x=sn(u|m)}
* @since 2.1
*/
public T arcsn(final double x) {
return arcsn(getM().getField().getZero().newInstance(x));
}
/** Evaluate inverse of Jacobi elliptic function cn.
* @param x value of Jacobi elliptic function {@code cn(u|m)}
* @return u such that {@code x=cn(u|m)}
* @since 2.1
*/
public T arccn(final T x) {
// p = c, q = n, r = d, see DLMF 19.25.29 for evaluating Δ(q, p) and Δ(r, q)
return arcpqNoDivision(x, x.getField().getOne(), getM().negate());
}
/** Evaluate inverse of Jacobi elliptic function cn.
* @param x value of Jacobi elliptic function {@code cn(u|m)}
* @return u such that {@code x=cn(u|m)}
* @since 2.1
*/
public T arccn(final double x) {
return arccn(getM().getField().getZero().newInstance(x));
}
/** Evaluate inverse of Jacobi elliptic function dn.
* @param x value of Jacobi elliptic function {@code dn(u|m)}
* @return u such that {@code x=dn(u|m)}
* @since 2.1
*/
public T arcdn(final T x) {
// p = d, q = n, r = c, see DLMF 19.25.29 for evaluating Δ(q, p) and Δ(r, q)
return arcpqNoDivision(x, getM(), x.getField().getOne().negate());
}
/** Evaluate inverse of Jacobi elliptic function dn.
* @param x value of Jacobi elliptic function {@code dn(u|m)}
* @return u such that {@code x=dn(u|m)}
* @since 2.1
*/
public T arcdn(final double x) {
return arcdn(getM().getField().getZero().newInstance(x));
}
/** Evaluate inverse of Jacobi elliptic function cs.
* @param x value of Jacobi elliptic function {@code cs(u|m)}
* @return u such that {@code x=cs(u|m)}
* @since 2.1
*/
public T arccs(final T x) {
// p = c, q = n, r = d, see DLMF 19.25.29 for evaluating Δ(q, p) and Δ(r, p)
return arcps(x, x.getField().getOne(), getM().subtract(1).negate());
}
/** Evaluate inverse of Jacobi elliptic function cs.
* @param x value of Jacobi elliptic function {@code cs(u|m)}
* @return u such that {@code x=cs(u|m)}
* @since 2.1
*/
public T arccs(final double x) {
return arccs(getM().getField().getZero().newInstance(x));
}
/** Evaluate inverse of Jacobi elliptic function ds.
* @param x value of Jacobi elliptic function {@code ds(u|m)}
* @return u such that {@code x=ds(u|m)}
* @since 2.1
*/
public T arcds(final T x) {
// p = d, q = c, r = n, see DLMF 19.25.29 for evaluating Δ(q, p) and Δ(r, p)
return arcps(x, getM().subtract(1), getM());
}
/** Evaluate inverse of Jacobi elliptic function ds.
* @param x value of Jacobi elliptic function {@code ds(u|m)}
* @return u such that {@code x=ds(u|m)}
* @since 2.1
*/
public T arcds(final double x) {
return arcds(getM().getField().getZero().newInstance(x));
}
/** Evaluate inverse of Jacobi elliptic function ns.
* @param x value of Jacobi elliptic function {@code ns(u|m)}
* @return u such that {@code x=ns(u|m)}
* @since 2.1
*/
public T arcns(final T x) {
// p = n, q = c, r = d, see DLMF 19.25.29 for evaluating Δ(q, p) and Δ(r, p)
return arcps(x, x.getField().getOne().negate(), getM().negate());
}
/** Evaluate inverse of Jacobi elliptic function ns.
* @param x value of Jacobi elliptic function {@code ns(u|m)}
* @return u such that {@code x=ns(u|m)}
* @since 2.1
*/
public T arcns(final double x) {
return arcns(getM().getField().getZero().newInstance(x));
}
/** Evaluate inverse of Jacobi elliptic function dc.
* @param x value of Jacobi elliptic function {@code dc(u|m)}
* @return u such that {@code x=dc(u|m)}
* @since 2.1
*/
public T arcdc(final T x) {
// p = d, q = c, r = n, see DLMF 19.25.29 for evaluating Δ(q, p) and Δ(r, q)
return arcpq(x, getM().subtract(1), x.getField().getOne());
}
/** Evaluate inverse of Jacobi elliptic function dc.
* @param x value of Jacobi elliptic function {@code dc(u|m)}
* @return u such that {@code x=dc(u|m)}
* @since 2.1
*/
public T arcdc(final double x) {
return arcdc(getM().getField().getZero().newInstance(x));
}
/** Evaluate inverse of Jacobi elliptic function nc.
* @param x value of Jacobi elliptic function {@code nc(u|m)}
* @return u such that {@code x=nc(u|m)}
* @since 2.1
*/
public T arcnc(final T x) {
// p = n, q = c, r = d, see DLMF 19.25.29 for evaluating Δ(q, p) and Δ(r, q)
return arcpq(x, x.getField().getOne().negate(), getM().subtract(1).negate());
}
/** Evaluate inverse of Jacobi elliptic function nc.
* @param x value of Jacobi elliptic function {@code nc(u|m)}
* @return u such that {@code x=nc(u|m)}
* @since 2.1
*/
public T arcnc(final double x) {
return arcnc(getM().getField().getZero().newInstance(x));
}
/** Evaluate inverse of Jacobi elliptic function sc.
* @param x value of Jacobi elliptic function {@code sc(u|m)}
* @return u such that {@code x=sc(u|m)}
* @since 2.1
*/
public T arcsc(final T x) {
// p = c, q = n, r = d, see DLMF 19.25.29 for evaluating Δ(q, p) and Δ(r, p)
return arcsp(x, x.getField().getOne(), getM().subtract(1).negate());
}
/** Evaluate inverse of Jacobi elliptic function sc.
* @param x value of Jacobi elliptic function {@code sc(u|m)}
* @return u such that {@code x=sc(u|m)}
* @since 2.1
*/
public T arcsc(final double x) {
return arcsc(getM().getField().getZero().newInstance(x));
}
/** Evaluate inverse of Jacobi elliptic function nd.
* @param x value of Jacobi elliptic function {@code nd(u|m)}
* @return u such that {@code x=nd(u|m)}
* @since 2.1
*/
public T arcnd(final T x) {
// p = n, q = d, r = c, see DLMF 19.25.29 for evaluating Δ(q, p) and Δ(r, q)
return arcpq(x, getM().negate(), getM().subtract(1));
}
/** Evaluate inverse of Jacobi elliptic function nd.
* @param x value of Jacobi elliptic function {@code nd(u|m)}
* @return u such that {@code x=nd(u|m)}
* @since 2.1
*/
public T arcnd(final double x) {
return arcnd(getM().getField().getZero().newInstance(x));
}
/** Evaluate inverse of Jacobi elliptic function sd.
* @param x value of Jacobi elliptic function {@code sd(u|m)}
* @return u such that {@code x=sd(u|m)}
* @since 2.1
*/
public T arcsd(final T x) {
// p = d, q = n, r = c, see DLMF 19.25.29 for evaluating Δ(q, p) and Δ(r, p)
return arcsp(x, getM(), getM().subtract(1));
}
/** Evaluate inverse of Jacobi elliptic function sd.
* @param x value of Jacobi elliptic function {@code sd(u|m)}
* @return u such that {@code x=sd(u|m)}
* @since 2.1
*/
public T arcsd(final double x) {
return arcsd(getM().getField().getZero().newInstance(x));
}
/** Evaluate inverse of Jacobi elliptic function cd.
* @param x value of Jacobi elliptic function {@code cd(u|m)}
* @return u such that {@code x=cd(u|m)}
* @since 2.1
*/
public T arccd(final T x) {
// p = c, q = d, r = n, see DLMF 19.25.29 for evaluating Δ(q, p) and Δ(r, q)
return arcpq(x, getM().subtract(1).negate(), getM());
}
/** Evaluate inverse of Jacobi elliptic function cd.
* @param x value of Jacobi elliptic function {@code cd(u|m)}
* @return u such that {@code x=cd(u|m)}
* @since 2.1
*/
public T arccd(final double x) {
return arccd(getM().getField().getZero().newInstance(x));
}
/** Evaluate inverse of Jacobi elliptic function ps.
* <p>
* Here p, q, r are any permutation of the letters c, d, n.
* </p>
* @param x value of Jacobi elliptic function {@code ps(u|m)}
* @param deltaQP Δ(q, p) = qs²(u|m) - ps²(u|m) (equation 19.5.28 of DLMF)
* @param deltaRP Δ(r, p) = rs²(u|m) - ps²(u|m) (equation 19.5.28 of DLMF)
* @return u such that {@code x=ps(u|m)}
* @since 2.1
*/
private T arcps(final T x, final T deltaQP, final T deltaRP) {
// see equation 19.25.32 in Digital Library of Mathematical Functions
// https://dlmf.nist.gov/19.25.E32
final T x2 = x.square();
final T rf = CarlsonEllipticIntegral.rF(x2, x2.add(deltaQP), x2.add(deltaRP));
return FastMath.copySign(1.0, rf.getReal()) * FastMath.copySign(1.0, x.getReal()) < 0 ?
rf.negate() : rf;
}
/** Evaluate inverse of Jacobi elliptic function sp.
* <p>
* Here p, q, r are any permutation of the letters c, d, n.
* </p>
* @param x value of Jacobi elliptic function {@code sp(u|m)}
* @param deltaQP Δ(q, p) = qs²(u|m) - ps²(u|m) (equation 19.5.28 of DLMF)
* @param deltaRP Δ(r, p) = rs²(u|m) - ps²(u|m) (equation 19.5.28 of DLMF)
* @return u such that {@code x=sp(u|m)}
* @since 2.1
*/
private T arcsp(final T x, final T deltaQP, final T deltaRP) {
// see equation 19.25.33 in Digital Library of Mathematical Functions
// https://dlmf.nist.gov/19.25.E33
final T x2 = x.square();
return x.multiply(CarlsonEllipticIntegral.rF(x.getField().getOne(),
deltaQP.multiply(x2).add(1),
deltaRP.multiply(x2).add(1)));
}
/** Evaluate inverse of Jacobi elliptic function pq.
* <p>
* Here p, q, r are any permutation of the letters c, d, n.
* </p>
* @param x value of Jacobi elliptic function {@code pq(u|m)}
* @param deltaQP Δ(q, p) = qs²(u|m) - ps²(u|m) (equation 19.5.28 of DLMF)
* @param deltaRQ Δ(r, q) = rs²(u|m) - qs²(u|m) (equation 19.5.28 of DLMF)
* @return u such that {@code x=pq(u|m)}
* @since 2.1
*/
private T arcpq(final T x, final T deltaQP, final T deltaRQ) {
// see equation 19.25.34 in Digital Library of Mathematical Functions
// https://dlmf.nist.gov/19.25.E34
final T x2 = x.square();
final T w = x2.subtract(1).negate().divide(deltaQP);
final T rf = CarlsonEllipticIntegral.rF(x2, x.getField().getOne(), deltaRQ.multiply(w).add(1));
final T positive = w.sqrt().multiply(rf);
return x.getReal() < 0 ? LegendreEllipticIntegral.bigK(getM()).multiply(2).subtract(positive) : positive;
}
/** Evaluate inverse of Jacobi elliptic function pq.
* <p>
* Here p, q, r are any permutation of the letters c, d, n.
* </p>
* <p>
* This computed the same thing as {@link #arcpq(CalculusFieldElement, CalculusFieldElement, CalculusFieldElement)}
* but uses the homogeneity property Rf(x, y, z) = Rf(ax, ay, az) / √a to get rid of the division
* by deltaRQ. This division induces problems in the complex case as it may lose the sign
* of zero for values exactly along the real or imaginary axis, hence perturbing branch cuts.
* </p>
* @param x value of Jacobi elliptic function {@code pq(u|m)}
* @param deltaQP Δ(q, p) = qs²(u|m) - ps²(u|m) (equation 19.5.28 of DLMF)
* @param deltaRQ Δ(r, q) = rs²(u|m) - qs²(u|m) (equation 19.5.28 of DLMF)
* @return u such that {@code x=pq(u|m)}
* @since 2.1
*/
private T arcpqNoDivision(final T x, final T deltaQP, final T deltaRQ) {
// see equation 19.25.34 in Digital Library of Mathematical Functions
// https://dlmf.nist.gov/19.25.E34
final T x2 = x.square();
final T wDeltaQP = x2.subtract(1).negate();
final T rf = CarlsonEllipticIntegral.rF(x2.multiply(deltaQP), deltaQP, deltaRQ.multiply(wDeltaQP).add(deltaQP));
final T positive = wDeltaQP.sqrt().multiply(rf);
return FastMath.copySign(1.0, x.getReal()) < 0 ?
LegendreEllipticIntegral.bigK(getM()).multiply(2).subtract(positive) :
positive;
}
}