make Point2D and Point3D constexpr (#8882)

* constexpr Point, Point2D, and Point3D

* constexpr Vector

* cleanup

* rollback overzealous constexpring

* dial back the constexpr to get windows builds working

the math stuff isn't constexpr with MSVC++

* add [[nodiscard]]
This commit is contained in:
Greg Landrum
2025-10-23 10:57:33 +02:00
committed by GitHub
parent 20907d3ba9
commit e625bdb95c
2 changed files with 84 additions and 84 deletions

View File

@@ -1,5 +1,5 @@
//
// Copyright (C) 2003-2021 Greg Landrum and other RDKit contributors
// Copyright (C) 2003-2025 Greg Landrum and other RDKit contributors
//
// @@ All Rights Reserved @@
// This file is part of the RDKit.
@@ -9,8 +9,8 @@
//
#include <RDGeneral/export.h>
#ifndef __RD_POINT_H__
#define __RD_POINT_H__
#ifndef RD_POINT_H
#define RD_POINT_H
#include <cmath>
#include <vector>
#include <map>
@@ -28,7 +28,7 @@ namespace RDGeom {
class RDKIT_RDGEOMETRYLIB_EXPORT Point {
// this is the virtual base class, mandating certain functions
public:
virtual ~Point() {}
constexpr virtual ~Point() {}
virtual double operator[](unsigned int i) const = 0;
virtual double &operator[](unsigned int i) = 0;
@@ -38,7 +38,7 @@ class RDKIT_RDGEOMETRYLIB_EXPORT Point {
virtual double lengthSq() const = 0;
virtual unsigned int dimension() const = 0;
virtual Point *copy() const = 0;
[[nodiscard]] virtual Point *copy() const = 0;
};
#ifndef _MSC_VER
// g++ (at least as of v9.3.0) generates some spurious warnings from here.
@@ -56,19 +56,21 @@ class RDKIT_RDGEOMETRYLIB_EXPORT Point3D : public Point {
double y{0.0};
double z{0.0};
Point3D() {}
Point3D(double xv, double yv, double zv) : x(xv), y(yv), z(zv) {}
constexpr Point3D() {}
constexpr Point3D(double xv, double yv, double zv) : x(xv), y(yv), z(zv) {}
~Point3D() override = default;
constexpr ~Point3D() override {};
Point3D(const Point3D &other)
constexpr Point3D(const Point3D &other)
: Point(other), x(other.x), y(other.y), z(other.z) {}
Point *copy() const override { return new Point3D(*this); }
[[nodiscard]] constexpr Point *copy() const override {
return new Point3D(*this);
}
inline unsigned int dimension() const override { return 3; }
constexpr unsigned int dimension() const override { return 3; }
inline double operator[](unsigned int i) const override {
constexpr double operator[](unsigned int i) const override {
switch (i) {
case 0:
return x;
@@ -82,7 +84,7 @@ class RDKIT_RDGEOMETRYLIB_EXPORT Point3D : public Point {
}
}
inline double &operator[](unsigned int i) override {
constexpr double &operator[](unsigned int i) override {
switch (i) {
case 0:
return x;
@@ -96,7 +98,7 @@ class RDKIT_RDGEOMETRYLIB_EXPORT Point3D : public Point {
}
}
Point3D &operator=(const Point3D &other) {
constexpr Point3D &operator=(const Point3D &other) {
if (&other == this) {
return *this;
}
@@ -106,35 +108,35 @@ class RDKIT_RDGEOMETRYLIB_EXPORT Point3D : public Point {
return *this;
}
Point3D &operator+=(const Point3D &other) {
constexpr Point3D &operator+=(const Point3D &other) {
x += other.x;
y += other.y;
z += other.z;
return *this;
}
Point3D &operator-=(const Point3D &other) {
constexpr Point3D &operator-=(const Point3D &other) {
x -= other.x;
y -= other.y;
z -= other.z;
return *this;
}
Point3D &operator*=(double scale) {
constexpr Point3D &operator*=(double scale) {
x *= scale;
y *= scale;
z *= scale;
return *this;
}
Point3D &operator/=(double scale) {
constexpr Point3D &operator/=(double scale) {
x /= scale;
y /= scale;
z /= scale;
return *this;
}
Point3D operator-() const {
constexpr Point3D operator-() const {
Point3D res(x, y, z);
res.x *= -1.0;
res.y *= -1.0;
@@ -142,7 +144,7 @@ class RDKIT_RDGEOMETRYLIB_EXPORT Point3D : public Point {
return res;
}
void normalize() override {
constexpr void normalize() override {
double l = this->length();
if (l < zero_tolerance) {
throw std::runtime_error("Cannot normalize a zero length vector");
@@ -158,13 +160,13 @@ class RDKIT_RDGEOMETRYLIB_EXPORT Point3D : public Point {
return sqrt(res);
}
double lengthSq() const override {
constexpr double lengthSq() const override {
// double res = pow(x,2) + pow(y,2) + pow(z,2);
double res = x * x + y * y + z * z;
return res;
}
double dotProduct(const Point3D &other) const {
constexpr double dotProduct(const Point3D &other) const {
double res = x * (other.x) + y * (other.y) + z * (other.z);
return res;
}
@@ -223,7 +225,7 @@ class RDKIT_RDGEOMETRYLIB_EXPORT Point3D : public Point {
* The order is important here
* The result is "this" cross with "other" not (other x this)
*/
Point3D crossProduct(const Point3D &other) const {
constexpr Point3D crossProduct(const Point3D &other) const {
Point3D res;
res.x = y * (other.z) - z * (other.y);
res.y = -x * (other.z) + z * (other.x);
@@ -257,7 +259,6 @@ class RDKIT_RDGEOMETRYLIB_EXPORT Point3D : public Point {
res.x = 1;
}
double l = res.length();
POSTCONDITION(l > 0.0, "zero perpendicular");
res /= l;
return res;
}
@@ -285,19 +286,22 @@ class RDKIT_RDGEOMETRYLIB_EXPORT Point2D : public Point {
double x{0.0};
double y{0.0};
Point2D() {}
Point2D(double xv, double yv) : x(xv), y(yv) {}
~Point2D() override = default;
constexpr Point2D() {}
constexpr Point2D(double xv, double yv) : x(xv), y(yv) {}
constexpr ~Point2D() override {};
Point2D(const Point2D &other) : Point(other), x(other.x), y(other.y) {}
constexpr Point2D(const Point2D &other)
: Point(other), x(other.x), y(other.y) {}
//! construct from a Point3D (ignoring the z coordinate)
Point2D(const Point3D &p3d) : Point(p3d), x(p3d.x), y(p3d.y) {}
constexpr Point2D(const Point3D &p3d) : Point(p3d), x(p3d.x), y(p3d.y) {}
Point *copy() const override { return new Point2D(*this); }
[[nodiscard]] constexpr Point *copy() const override {
return new Point2D(*this);
}
inline unsigned int dimension() const override { return 2; }
constexpr unsigned int dimension() const override { return 2; }
inline double operator[](unsigned int i) const override {
constexpr double operator[](unsigned int i) const override {
switch (i) {
case 0:
return x;
@@ -309,7 +313,7 @@ class RDKIT_RDGEOMETRYLIB_EXPORT Point2D : public Point {
}
}
inline double &operator[](unsigned int i) override {
constexpr double &operator[](unsigned int i) override {
switch (i) {
case 0:
return x;
@@ -321,37 +325,37 @@ class RDKIT_RDGEOMETRYLIB_EXPORT Point2D : public Point {
}
}
Point2D &operator=(const Point2D &other) {
constexpr Point2D &operator=(const Point2D &other) {
x = other.x;
y = other.y;
return *this;
}
Point2D &operator+=(const Point2D &other) {
constexpr Point2D &operator+=(const Point2D &other) {
x += other.x;
y += other.y;
return *this;
}
Point2D &operator-=(const Point2D &other) {
constexpr Point2D &operator-=(const Point2D &other) {
x -= other.x;
y -= other.y;
return *this;
}
Point2D &operator*=(double scale) {
constexpr Point2D &operator*=(double scale) {
x *= scale;
y *= scale;
return *this;
}
Point2D &operator/=(double scale) {
constexpr Point2D &operator/=(double scale) {
x /= scale;
y /= scale;
return *this;
}
Point2D operator-() const {
constexpr Point2D operator-() const {
Point2D res(x, y);
res.x *= -1.0;
res.y *= -1.0;
@@ -368,7 +372,7 @@ class RDKIT_RDGEOMETRYLIB_EXPORT Point2D : public Point {
y /= ln;
}
void rotate90() {
constexpr void rotate90() {
double temp = x;
x = -y;
y = temp;
@@ -380,12 +384,12 @@ class RDKIT_RDGEOMETRYLIB_EXPORT Point2D : public Point {
return sqrt(res);
}
double lengthSq() const override {
constexpr double lengthSq() const override {
double res = x * x + y * y;
return res;
}
double dotProduct(const Point2D &other) const {
constexpr double dotProduct(const Point2D &other) const {
double res = x * (other.x) + y * (other.y);
return res;
}

View File

@@ -1,5 +1,5 @@
//
// Copyright (C) 2004-2008 Greg Landrum and Rational Discovery LLC
// Copyright (C) 2004-2025 Greg Landrum and other RDKit contributors
//
// @@ All Rights Reserved @@
// This file is part of the RDKit.
@@ -8,8 +8,8 @@
// of the RDKit source tree.
//
#include <RDGeneral/export.h>
#ifndef __RD_VECTOR_H__
#define __RD_VECTOR_H__
#ifndef RD_VECTOR_H
#define RD_VECTOR_H
#include <RDGeneral/Invariant.h>
#include <RDGeneral/utils.h>
@@ -18,8 +18,8 @@
#include <cstdlib>
#include <cstring>
#include <ctime>
#include <memory>
#include <boost/random.hpp>
#include <boost/smart_ptr.hpp>
static constexpr double zero_tolerance = 1.e-16;
@@ -29,10 +29,10 @@ namespace RDNumeric {
template <class TYPE>
class Vector {
public:
typedef boost::shared_array<TYPE> DATA_SPTR;
typedef std::shared_ptr<TYPE[]> DATA_SPTR;
//! Initialize with only a size.
explicit Vector(unsigned int N) {
constexpr explicit Vector(unsigned int N) {
d_size = N;
TYPE *data = new TYPE[N];
memset(static_cast<void *>(data), 0, d_size * sizeof(TYPE));
@@ -40,22 +40,19 @@ class Vector {
}
//! Initialize with a size and default value.
Vector(unsigned int N, TYPE val) { //: Vector(N) {
constexpr Vector(unsigned int N, TYPE val) {
d_size = N;
TYPE *data = new TYPE[N];
unsigned int i;
for (i = 0; i < N; i++) {
data[i] = val;
d_data.reset(new TYPE[N]);
for (auto i = 0u; i < N; i++) {
d_data[i] = val;
}
d_data.reset(data);
}
//! Initialize from a smart pointer.
/*!
<b>NOTE:</b> the data is not copied in this case
*/
Vector(unsigned int N, DATA_SPTR data) { // TYPE *data) {
constexpr Vector(unsigned int N, DATA_SPTR data) {
d_size = N;
d_data = data;
}
@@ -63,48 +60,47 @@ class Vector {
//! copy constructor
/*! We make a copy of the other vector's data.
*/
Vector(const Vector &other) {
constexpr Vector(const Vector &other) {
d_size = other.size();
const TYPE *otherData = other.getData();
TYPE *data = new TYPE[d_size];
d_data.reset(new TYPE[d_size]);
memcpy(static_cast<void *>(data), static_cast<const void *>(otherData),
d_size * sizeof(TYPE));
d_data.reset(data);
memcpy(static_cast<void *>(d_data.get()),
static_cast<const void *>(otherData), d_size * sizeof(TYPE));
}
~Vector() = default;
constexpr ~Vector() = default;
//! return the size (dimension) of the vector
unsigned int size() const { return d_size; }
constexpr unsigned int size() const { return d_size; }
//! returns the value at a particular index
inline TYPE getVal(unsigned int i) const {
TYPE getVal(unsigned int i) const {
PRECONDITION(i < d_size, "bad index");
return d_data[i];
}
//! sets the index at a particular value
inline void setVal(unsigned int i, TYPE val) {
void setVal(unsigned int i, TYPE val) {
PRECONDITION(i < d_size, "bad index");
d_data[i] = val;
}
inline TYPE operator[](unsigned int i) const {
TYPE operator[](unsigned int i) const {
PRECONDITION(i < d_size, "bad index");
return d_data[i];
}
inline TYPE &operator[](unsigned int i) {
TYPE &operator[](unsigned int i) {
PRECONDITION(i < d_size, "bad index");
return d_data[i];
}
//! returns a pointer to our data array
inline TYPE *getData() { return d_data.get(); }
constexpr TYPE *getData() { return d_data.get(); }
//! returns a const pointer to our data array
inline const TYPE *getData() const {
constexpr const TYPE *getData() const {
// return dp_data;
return d_data.get();
}
@@ -146,7 +142,7 @@ class Vector {
}
//! multiplication by a scalar
Vector<TYPE> &operator*=(TYPE scale) {
constexpr Vector<TYPE> &operator*=(TYPE scale) {
unsigned int i;
for (i = 0; i < d_size; i++) {
d_data[i] *= scale;
@@ -155,7 +151,7 @@ class Vector {
}
//! division by a scalar
Vector<TYPE> &operator/=(TYPE scale) {
constexpr Vector<TYPE> &operator/=(TYPE scale) {
unsigned int i;
for (i = 0; i < d_size; i++) {
d_data[i] /= scale;
@@ -164,7 +160,7 @@ class Vector {
}
//! L2 norm squared
inline TYPE normL2Sq() const {
constexpr inline TYPE normL2Sq() const {
TYPE res = (TYPE)0.0;
unsigned int i;
TYPE *data = d_data.get();
@@ -175,10 +171,10 @@ class Vector {
}
//! L2 norm
inline TYPE normL2() const { return sqrt(this->normL2Sq()); }
constexpr TYPE normL2() const { return sqrt(this->normL2Sq()); }
//! L1 norm
inline TYPE normL1() const {
constexpr TYPE normL1() const {
TYPE res = (TYPE)0.0;
unsigned int i;
TYPE *data = d_data.get();
@@ -189,7 +185,7 @@ class Vector {
}
//! L-infinity norm
inline TYPE normLinfinity() const {
constexpr TYPE normLinfinity() const {
TYPE res = (TYPE)(-1.0);
unsigned int i;
TYPE *data = d_data.get();
@@ -203,7 +199,7 @@ class Vector {
//! \brief Gets the ID of the entry that has the largest absolute value
//! i.e. the entry being used for the L-infinity norm
inline unsigned int largestAbsValId() const {
constexpr unsigned int largestAbsValId() const {
TYPE res = (TYPE)(-1.0);
unsigned int i, id = d_size;
TYPE *data = d_data.get();
@@ -217,7 +213,7 @@ class Vector {
}
//! \brief Gets the ID of the entry that has the largest value
inline unsigned int largestValId() const {
constexpr unsigned int largestValId() const {
TYPE res = (TYPE)(-1.e8);
unsigned int i, id = d_size;
TYPE *data = d_data.get();
@@ -231,7 +227,7 @@ class Vector {
}
//! \brief Gets the ID of the entry that has the smallest value
inline unsigned int smallestValId() const {
constexpr unsigned int smallestValId() const {
TYPE res = (TYPE)(1.e8);
unsigned int i, id = d_size;
TYPE *data = d_data.get();
@@ -245,7 +241,7 @@ class Vector {
}
//! returns the dot product between two Vectors
inline TYPE dotProduct(const Vector<TYPE> other) const {
TYPE dotProduct(const Vector<TYPE> other) const {
PRECONDITION(d_size == other.size(),
"Size mismatch in vector doct product");
const TYPE *oData = other.getData();
@@ -259,7 +255,7 @@ class Vector {
}
//! Normalize the vector using the L2 norm
inline void normalize() {
constexpr void normalize() {
TYPE val = this->normL2();
if (val < zero_tolerance) {
throw std::runtime_error("Cannot normalize a zero length vector");
@@ -268,7 +264,7 @@ class Vector {
}
//! Set to a random unit vector
inline void setToRandom(unsigned int seed = 0) {
void setToRandom(unsigned int seed = 0) {
// we want to get our own RNG here instead of using the global
// one. This is related to Issue285.
RDKit::rng_type generator(42u);
@@ -301,7 +297,7 @@ typedef Vector<double> DoubleVector;
//! returns the algebraic tanimoto similarity [defn' from JCIM 46:587-96 (2006)]
template <typename T>
double TanimotoSimilarity(const Vector<T> &v1, const Vector<T> &v2) {
constexpr double TanimotoSimilarity(const Vector<T> &v1, const Vector<T> &v2) {
double numer = v1.dotProduct(v2);
if (numer == 0.0) {
return 0.0;
@@ -316,8 +312,8 @@ double TanimotoSimilarity(const Vector<T> &v1, const Vector<T> &v2) {
//! ostream operator for Vectors
template <typename TYPE>
std::ostream &operator<<(std::ostream &target,
const RDNumeric::Vector<TYPE> &vec) {
constexpr std::ostream &operator<<(std::ostream &target,
const RDNumeric::Vector<TYPE> &vec) {
unsigned int siz = vec.size();
target << "Size: " << siz << " [";
unsigned int i;