Miscelaneous

Basic types

class chemfiles::Vector3D : private std::array<double, 3>

3D vector for basic data storage in chemfiles.

This type defines the following operators, with the usual meaning:

  • Comparison operators: == and != performs strict float equality comparison;

  • Mathematical operators: + and - for addition and substraction of vectors, * and / for multiplication and division by double values. Multiplication with * is also defined for multiplication by a Matrix3D.

// equality and inequality
assert(Vector3D() == Vector3D(0, 0, 0));
assert(Vector3D(1, 2, 3) != Vector3D(0, 0, 0));

// Indexing
auto u = Vector3D(1.5, 2.0, -3.0);
assert(u[0] == 1.5);
assert(u[1] == 2.0);
assert(u[2] == -3.0);

// sum and difference
auto sum = Vector3D(1, 0, 1) + Vector3D(0, 2, 1);
assert(sum == Vector3D(1, 2, 2));

auto diff = Vector3D(1, 0, 1) - Vector3D(0, 2, 1);
assert(diff == Vector3D(1, -2, 0));

// product and division by a scalar
auto rhs = Vector3D(1, 0, 1) * 5;
assert(rhs == Vector3D(5, 0, 5));

auto lhs = 5 * Vector3D(1, 0, 1);
assert(lhs == Vector3D(5, 0, 5));

auto div = Vector3D(2, 0, 3) / 3;
assert(div == Vector3D(2.0/3, 0, 1));

Public Functions

Vector3D()

Create a Vector3D with all components equal to 0.

auto v = Vector3D();
assert(v[0] == 0.0);
assert(v[1] == 0.0);
assert(v[2] == 0.0);

Vector3D(double x, double y, double z)

Create a Vector3D from the three components x, y, and z.

auto u = Vector3D(1.5, 2.0, -3.0);
assert(u[0] == 1.5);
assert(u[1] == 2.0);
assert(u[2] == -3.0);

double norm() const

Compute the euclidean norm of this Vector3D.

auto v = Vector3D(1.0, 1.0, 1.0);
assert(v.norm() == sqrt(3));

Vector3D &operator+=(const Vector3D &rhs)

Compound addition of two vectors.

Vector3D &operator-=(const Vector3D &rhs)

Compound subtraction of two vectors.

Vector3D &operator*=(double rhs)

Compound multiplication of a vector and a scalar.

Vector3D &operator/=(double rhs)

Compound division of a vector by a scalar.

double chemfiles::dot(const Vector3D &lhs, const Vector3D &rhs)

Compute the dot product of the vectors lhs and rhs.

auto u = Vector3D(0, 1, 2);
auto v = Vector3D(1, 0, 2);

assert(dot(u, v) == 4);

Vector3D chemfiles::cross(const Vector3D &lhs, const Vector3D &rhs)

Compute the cross product of the vectors lhs and rhs.

auto u = Vector3D(0, 1, 2);
auto v = Vector3D(1, 0, 2);

assert(cross(u, v) == Vector3D(2, 2, -1));

class chemfiles::Matrix3D : private std::array<std::array<double, 3>, 3>

A 3x3 matrix class.

This type defines the following operators, with the usual meaning:

  • Comparison operators: == and != performs strict float equality comparison;

  • Mathematical operators: only multiplication between matrixes and between matrix and vector are defined, using the * operator.

// equality and inequality
assert(Matrix3D::zero() == Matrix3D(0, 0, 0, 0, 0, 0, 0, 0, 0));
assert(Matrix3D::unit() != Matrix3D::zero());

// Indexing
auto M = Matrix3D(
    11, 12, 13,
    21, 22, 23,
    31, 32, 33
);
assert(M[0][0] == 11);
assert(M[0][1] == 12);
assert(M[0][2] == 13);

assert(M[1][0] == 21);
assert(M[1][1] == 22);
assert(M[1][2] == 23);

assert(M[2][0] == 31);
assert(M[2][1] == 32);
assert(M[2][2] == 33);

// Matrix-vector multiplication
auto mat_vec = Matrix3D::unit() * Vector3D(1, 3, 2);
assert(mat_vec == Vector3D(1, 3, 2));

// Matrix-Matrix multiplication
auto A = Matrix3D(
    0.5, 0, 0,
      0, 2, 0,
      0, 0, 1
);
auto mat_mat = Matrix3D::unit() * A;
assert(mat_mat == Matrix3D(0.5, 0, 0, 0, 2, 0, 0, 0, 1));

// Scalar-Matrix multiplication
assert(2 * A == Matrix3D(1.0, 0, 0, 0, 4, 0, 0, 0, 2));
assert(A * 2 == Matrix3D(1.0, 0, 0, 0, 4, 0, 0, 0, 2));

Public Functions

Matrix3D(double m11, double m12, double m13, double m21, double m22, double m23, double m31, double m32, double m33)

Create a Matrix3D by explicitly specifying all the m_ij components of the matrix.

auto A = Matrix3D(
    11, 12, 13,
    21, 22, 23,
    31, 32, 33
);
assert(A[0][0] == 11);
assert(A[0][1] == 12);
assert(A[0][2] == 13);

assert(A[1][0] == 21);
assert(A[1][1] == 22);
assert(A[1][2] == 23);

assert(A[2][0] == 31);
assert(A[2][1] == 32);
assert(A[2][2] == 33);

double determinant() const

Compute the determinant of this matrix.

auto A = Matrix3D(
    3, 0, 0,
    0, 2, 0,
    0, 0, 1
);
assert(A.determinant() == 6);

Matrix3D invert() const

Compute the inverse of this matrix.

auto A = Matrix3D(
    3, 0, 0,
    0, 2, 0,
    0, 0, 1
);
auto inverse = Matrix3D(
    1.0/3.0,    0,    0,
      0,     1.0/2.0, 0,
      0,       0,     1.0);
assert(A.invert() == inverse);

Exceptions
  • Error: if the matrix is not invertible (i.e. if the determinant is zero)

Matrix3D transpose() const

Transpose the matrix.

auto A = Matrix3D(
    3, 0, 5,
    1, 2, 6,
    2, 0, 1
);
auto transposed = Matrix3D(
    3, 1, 2,
    0, 2, 0,
    5, 6, 1
);
assert(A.transpose() == transposed);

Matrix3D &operator+=(const Matrix3D &rhs)

Compound addition of two matrices.

Matrix3D &operator-=(const Matrix3D &rhs)

Compound subtraction of two matrices.

Matrix3D &operator*=(double rhs)

Compound multiplication of a matrix and a scalar.

Matrix3D &operator/=(double rhs)

Compound division of a matrix by a scalar.

Public Static Functions

Matrix3D zero()

Create a Matrix3D with all components set to zero.

auto C = Matrix3D::zero();
assert(C[0][0] == 0);
assert(C[0][1] == 0);
assert(C[0][2] == 0);

assert(C[1][0] == 0);
assert(C[1][1] == 0);
assert(C[1][2] == 0);

assert(C[2][0] == 0);
assert(C[2][1] == 0);
assert(C[2][2] == 0);

Matrix3D unit()

Create an unit Matrix3D (a diagonal matrix with all diagonal values set to 1).

auto B = Matrix3D::unit();
assert(B[0][0] == 1);
assert(B[0][1] == 0);
assert(B[0][2] == 0);

assert(B[1][0] == 0);
assert(B[1][1] == 1);
assert(B[1][2] == 0);

assert(B[2][0] == 0);
assert(B[2][1] == 0);
assert(B[2][2] == 1);

Format list and Metadata

std::vector<std::reference_wrapper<const FormatMetadata>> chemfiles::formats_list()

Get the list of formats chemfiles knows about, and all associated metadata

auto list = chemfiles::formats_list();
for (auto format: list) {
    auto& metadata = format.get();
    std::cout << metadata.name << " (" << metadata.extension.value_or("<no extension>") << ")" << std::endl;
}

class chemfiles::FormatMetadata

Metadata associated with a format.

Public Members

const char *name = ""

Name of the format.

optional<const char*> extension = nullopt

Potential extension associated with the format.

const char *description = ""

User facing format description.

const char *reference = ""

URL pointing to the format definition/reference.

bool read = false

Is reading files in this format implemented?

bool write = false

Is writing files in this format implemented?

bool memory = false

Does this format support in-memory IO?

bool positions = false

Does this format support storing atomic positions?

bool velocities = false

Does this format support storing atomic velocities?

bool unit_cell = false

Does this format support storing unit cell information?

bool atoms = false

Does this format support storing atom names or types?

bool bonds = false

Does this format support storing bonds between atoms?

bool residues = false

Does this format support storing residues?

Errors handling

In chemfiles, any error will throw an exception. Any program using chemfiles should thus wrap any call in a try ... catch block, and handle the errors in any pertinent way.

#include <iostream>
#include <chemfiles.hpp>

int main() {
    try {
        chemfiles::Trajectory file("filename.xyz");
        auto frame = file.read();
        auto positions = frame.positions();
        // Do something here
    } catch (const chemfiles::Error& e) {
        // Basic error handling logging error to stdout
        std::cout << "Error in chemfiles:" << e.what() << std::endl;
        return -1;
    }

    return 0;
}

Any exceptions thrown by chemfiles will derive from chemfiles::Error. Catching chemfiles::Error will then catch any exception thrown by chemfiles. You also can catch any other error if you need finer grain control. chemfiles::Error derives from std::runtime_error, so it should play nicely with any existing C++ error handling.

struct Error : public std::runtime_error

Base exception class for chemfiles library.

Subclassed by chemfiles::ConfigurationError, chemfiles::FileError, chemfiles::FormatError, chemfiles::MemoryError, chemfiles::OutOfBounds, chemfiles::PropertyError, chemfiles::SelectionError

struct FileError : public chemfiles::Error

Exception for files related failures.

struct MemoryError : public chemfiles::Error

Exception for memory related failures.

struct FormatError : public chemfiles::Error

Exception for formats related failures.

struct SelectionError : public chemfiles::Error

Exception for errors in selections.

struct ConfigurationError : public chemfiles::Error

Exception for errors in configuration files.

struct OutOfBounds : public chemfiles::Error

Exception for out of bounds error when accessing atoms or residues.

struct PropertyError : public chemfiles::Error

Exception for errors related to frame and atomic properties.

Warnings

Chemfiles send warnings when it encounters malformed files, or any other condition that the user might want to know about. By default, these warnings are printed to the standard error stream. chemfiles::set_warning_callback() allow to redirect these warning by giving it a callback function to be called on each warning event.

void chemfiles::set_warning_callback(warning_callback_t callback)

Set the global callback for warning events. The default is to print them on the standard error stream.

// a function to do some work with the warnings
void work_with_warning(const std::string&);

/// Use a function as warning callback
chemfiles::set_warning_callback(work_with_warning);

/// Use a lambda as warning callback
chemfiles::set_warning_callback([](const std::string& message){
    std::cout << "got a warning from chemfiles: " << message << std::endl;
});

Parameters
  • callback: callback function that will be called on each warning

typedef std::function<void(const std::string &message)> chemfiles::warning_callback_t

Callback type used to process a warning event.

Configuration files

You can add more configuration file to chemfiles with chemfiles::add_configuration().

void chemfiles::add_configuration(const std::string &path)

Read configuration data from the file at path.

By default, chemfiles reads configuration from any file named .chemfiles.toml or chemfiles.toml in the current directory or any parent directory. This function can be used to add data from another configuration file.

This function will throw a ConfigurationError if there is no file at path, or if the file is incorrectly formatted. Data from the new configuration file will overwrite any existing data.

chemfiles::add_configuration("local-file.toml");

// Reading a trajectory will now use data from local-file.toml

Parameters
  • path: path to the configuration file to add

Exceptions