example-data-1d.cc

/*
Copyright (c) 2013 Daniel Stahlke

Permission is hereby granted, free of charge, to any person obtaining a copy
of this software and associated documentation files (the "Software"), to deal
in the Software without restriction, including without limitation the rights
to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
copies of the Software, and to permit persons to whom the Software is
furnished to do so, subject to the following conditions:

The above copyright notice and this permission notice shall be included in
all copies or substantial portions of the Software.

THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
THE SOFTWARE.
*/

// This demonstrates all sorts of data types that can be plotted using send1d().  It is not
// meant as a first tutorial; for that see example-misc.cc or the project wiki.

#if (__cplusplus >= 201103)
#define USE_CXX
#endif

#include <vector>
#include <complex>
#include <cmath>

#include <boost/tuple/tuple.hpp>
#include <boost/array.hpp>
#include <boost/range/adaptor/transformed.hpp>
#include <boost/range/irange.hpp>
#include <boost/bind.hpp>

#ifdef USE_ARMA
#include <armadillo>
#endif

#ifdef USE_BLITZ
#include <blitz/array.h>
#endif

#include "gnuplot-iostream.h"

#ifndef M_PI
#	define M_PI 3.14159265358979323846
#endif

const int num_steps = 100;

double get_x(int step, double shift) {
	double theta = 2.0*M_PI*step/(num_steps-1);
	return std::cos(theta) * (1 + 0.3*std::cos(3.0*theta+2.0*M_PI*shift));
}

double get_y(int step, double shift) {
	double theta = 2.0*M_PI*step/(num_steps-1);
	return std::sin(theta) * (1 + 0.3*std::cos(3.0*theta+2.0*M_PI*shift));
}

double get_z(int step, double shift) {
	double theta = 2.0*M_PI*step/(num_steps-1);
	return 0.3*std::sin(3.0*theta+2.0*M_PI*shift);
}

// This doesn't have to be a template.  It's just a template to show that such things are
// possible.
template <typename T>
struct MyTriple {
	MyTriple() : x(0), y(0), z(0) { }
	MyTriple(T _x, T _y, T _z) : x(_x), y(_y), z(_z) { }

	T x, y, z;
};

// Tells gnuplot-iostream how to print objects of class MyTriple.
namespace gnuplotio {
	template<typename T>
	struct BinfmtSender<MyTriple<T> > {
		static void send(std::ostream &stream) {
			BinfmtSender<T>::send(stream);
			BinfmtSender<T>::send(stream);
			BinfmtSender<T>::send(stream);
		}
	};

	template <typename T>
	struct BinarySender<MyTriple<T> > {
		static void send(std::ostream &stream, const MyTriple<T> &v) {
			BinarySender<T>::send(stream, v.x);
			BinarySender<T>::send(stream, v.y);
			BinarySender<T>::send(stream, v.z);
		}
	};

	// We don't use text mode in this demo.  This is just here to show how it would go.
	template<typename T>
	struct TextSender<MyTriple<T> > {
		static void send(std::ostream &stream, const MyTriple<T> &v) {
			TextSender<T>::send(stream, v.x);
			stream << " ";
			TextSender<T>::send(stream, v.y);
			stream << " ";
			TextSender<T>::send(stream, v.z);
		}
	};
}

int main() {
	Gnuplot gp;
	// for debugging, prints to console
	//Gnuplot gp(stdout);

	int num_examples = 11;
#ifdef USE_ARMA
	num_examples += 4;
#endif
#ifdef USE_BLITZ
	num_examples += 3;
#endif
#ifdef USE_CXX
	num_examples += 3;
#endif

	double shift = 0;

	gp << "set zrange [-1:1]\n";

	// I use temporary files rather than stdin because the syntax ends up being easier when
	// plotting several datasets.  With the stdin method you have to give the full plot
	// command, then all the data.  But I would rather give the portion of the plot command for
	// the first dataset, then give the data, then the command for the second dataset, then the
	// data, etc.

	gp << "splot ";

	{
		std::vector<std::pair<std::pair<double, double>, double> > pts;
		for(int i=0; i<num_steps; i++) {
			pts.push_back(std::make_pair(std::make_pair(get_x(i, shift), get_y(i, shift)), get_z(i, shift)));
		}
		gp << gp.binFile1d(pts, "record") << "with lines title 'vector of nested std::pair'";
	}

	gp << ", ";
	shift += 1.0/num_examples;

	{
		// complex is treated as if it were a pair
		std::vector<std::pair<std::complex<double>, double> > pts;
		for(int i=0; i<num_steps; i++) {
			pts.push_back(std::make_pair(std::complex<double>(get_x(i, shift), get_y(i, shift)), get_z(i, shift)));
		}
		gp << gp.binFile1d(pts, "record") << "with lines title 'vector of pair of cplx and double'";
	}

	gp << ", ";
	shift += 1.0/num_examples;

	{
		std::vector<boost::tuple<double, double, double> > pts;
		for(int i=0; i<num_steps; i++) {
			pts.push_back(boost::make_tuple(get_x(i, shift), get_y(i, shift), get_z(i, shift)));
		}
		gp << gp.binFile1d(pts, "record") << "with lines title 'vector of boost::tuple'";
	}

	gp << ", ";
	shift += 1.0/num_examples;

	{
		std::vector<double> x_pts, y_pts, z_pts;
		for(int i=0; i<num_steps; i++) {
			x_pts.push_back(get_x(i, shift));
			y_pts.push_back(get_y(i, shift));
			z_pts.push_back(get_z(i, shift));
		}
		gp << gp.binFile1d(boost::make_tuple(x_pts, y_pts, z_pts), "record") << "with lines title 'boost::tuple of vector'";
	}

	gp << ", ";
	shift += 1.0/num_examples;

	{
		std::vector<boost::array<double, 3> > pts(num_steps);
		for(int i=0; i<num_steps; i++) {
			pts[i][0] = get_x(i, shift);
			pts[i][1] = get_y(i, shift);
			pts[i][2] = get_z(i, shift);
		}
		gp << gp.binFile1d(pts, "record") << "with lines title 'vector of boost::array'";
	}

	gp << ", ";
	shift += 1.0/num_examples;

	{
		std::vector<std::vector<double> > pts(num_steps);
		for(int i=0; i<num_steps; i++) {
			pts[i].push_back(get_x(i, shift));
			pts[i].push_back(get_y(i, shift));
			pts[i].push_back(get_z(i, shift));
		}
		gp << gp.binFile1d(pts, "record") << "with lines title 'vector of vector'";
	}

	gp << ", ";
	shift += 1.0/num_examples;

	{
		std::vector<std::vector<double> > pts(3);
		for(int i=0; i<num_steps; i++) {
			pts[0].push_back(get_x(i, shift));
			pts[1].push_back(get_y(i, shift));
			pts[2].push_back(get_z(i, shift));
		}
		gp << gp.binFile1d_colmajor(pts, "record") << "with lines title 'vector of vector (colmajor)'";
	}

	gp << ", ";
	shift += 1.0/num_examples;

	{
		std::vector<MyTriple<double> > pts;
		for(int i=0; i<num_steps; i++) {
			pts.push_back(MyTriple<double>(get_x(i, shift), get_y(i, shift), get_z(i, shift)));
		}
		gp << gp.binFile1d(pts, "record") << "with lines title 'vector of MyTriple'";
	}

	gp << ", ";
	shift += 1.0/num_examples;

	{
		// Note: C style arrays seem to work, but are a bit fragile since they easily decay to
		// pointers, causing them to forget their lengths.  It is highly recommended that you
		// use boost::array or std::array instead.  These have the same size and efficiency of
		// C style arrays, but act like STL containers.
		double pts[num_steps][3];
		for(int i=0; i<num_steps; i++) {
			pts[i][0] = get_x(i, shift);
			pts[i][1] = get_y(i, shift);
			pts[i][2] = get_z(i, shift);
		}
		gp << gp.binFile1d(pts, "record") << "with lines title 'double[N][3]'";
	}

	gp << ", ";
	shift += 1.0/num_examples;

	{
		// Note: C style arrays seem to work, but are a bit fragile since they easily decay to
		// pointers, causing them to forget their lengths.  It is highly recommended that you
		// use boost::array or std::array instead.  These have the same size and efficiency of
		// C style arrays, but act like STL containers.
		double pts[3][num_steps];
		for(int i=0; i<num_steps; i++) {
			pts[0][i] = get_x(i, shift);
			pts[1][i] = get_y(i, shift);
			pts[2][i] = get_z(i, shift);
		}
		gp << gp.binFile1d_colmajor(pts, "record") << "with lines title 'double[N][3] (colmajor)'";
	}

	gp << ", ";
	shift += 1.0/num_examples;

	{
		// Note: C style arrays seem to work, but are a bit fragile since they easily decay to
		// pointers, causing them to forget their lengths.  It is highly recommended that you
		// use boost::array or std::array instead.  These have the same size and efficiency of
		// C style arrays, but act like STL containers.
		double x_pts[num_steps];
		double y_pts[num_steps];
		double z_pts[num_steps];
		for(int i=0; i<num_steps; i++) {
			x_pts[i] = get_x(i, shift);
			y_pts[i] = get_y(i, shift);
			z_pts[i] = get_z(i, shift);
		}
		gp << gp.binFile1d(boost::make_tuple(x_pts, y_pts, z_pts), "record") <<
			"with lines title 'boost::tuple of double[N]'";
	}

#ifdef USE_ARMA
	gp << ", ";
	shift += 1.0/num_examples;

	{
		arma::mat pts(num_steps, 3);
		for(int i=0; i<num_steps; i++) {
			pts(i, 0) = get_x(i, shift);
			pts(i, 1) = get_y(i, shift);
			pts(i, 2) = get_z(i, shift);
		}
		gp << gp.binFile1d(pts, "record") << "with lines title 'armadillo N*3'";
	}

	gp << ", ";
	shift += 1.0/num_examples;

	{
		arma::mat pts(3, num_steps);
		for(int i=0; i<num_steps; i++) {
			pts(0, i) = get_x(i, shift);
			pts(1, i) = get_y(i, shift);
			pts(2, i) = get_z(i, shift);
		}
		gp << gp.binFile1d_colmajor(pts, "record") << "with lines title 'armadillo 3*N (colmajor)'";
	}

	gp << ", ";
	shift += 1.0/num_examples;

	{
		arma::Row<double> x_pts(num_steps);
		arma::Col<double> y_pts(num_steps);
		arma::Col<double> z_pts(num_steps);
		for(int i=0; i<num_steps; i++) {
			x_pts(i) = get_x(i, shift);
			y_pts(i) = get_y(i, shift);
			z_pts(i) = get_z(i, shift);
		}
		gp << gp.binFile1d(boost::make_tuple(x_pts, y_pts, z_pts), "record")
			<< "with lines title 'boost tuple of arma Row,Col,Col'";
	}

	gp << ", ";
	shift += 1.0/num_examples;

	{
		arma::field<boost::tuple<double,double,double> > pts(num_steps);
		for(int i=0; i<num_steps; i++) {
			pts(i) = boost::make_tuple(
				get_x(i, shift),
				get_y(i, shift),
				get_z(i, shift)
			);
		}
		gp << gp.binFile1d(pts, "record") << "with lines title 'armadillo field of boost tuple'";
	}
#endif

#ifdef USE_BLITZ
	gp << ", ";
	shift += 1.0/num_examples;

	{
		blitz::Array<blitz::TinyVector<double, 3>, 1> pts(num_steps);
		for(int i=0; i<num_steps; i++) {
			pts(i)[0] = get_x(i, shift);
			pts(i)[1] = get_y(i, shift);
			pts(i)[2] = get_z(i, shift);
		}
		gp << gp.binFile1d(pts, "record") << "with lines title 'blitz::Array<blitz::TinyVector<double, 3>, 1>'";
	}

	gp << ", ";
	shift += 1.0/num_examples;

	{
		blitz::Array<double, 2> pts(num_steps, 3);
		for(int i=0; i<num_steps; i++) {
			pts(i, 0) = get_x(i, shift);
			pts(i, 1) = get_y(i, shift);
			pts(i, 2) = get_z(i, shift);
		}
		gp << gp.binFile1d(pts, "record") << "with lines title 'blitz<double>(N*3)'";
	}

	gp << ", ";
	shift += 1.0/num_examples;

	{
		blitz::Array<double, 2> pts(3, num_steps);
		for(int i=0; i<num_steps; i++) {
			pts(0, i) = get_x(i, shift);
			pts(1, i) = get_y(i, shift);
			pts(2, i) = get_z(i, shift);
		}
		gp << gp.binFile1d_colmajor(pts, "record") << "with lines title 'blitz<double>(3*N) (colmajor)'";
	}
#endif

#ifdef USE_CXX
	gp << ", ";
	shift += 1.0/num_examples;

	{
		std::function<boost::tuple<double,double,double>(int)> f = [&shift](int i) {
			return boost::make_tuple(get_x(i, shift), get_y(i, shift), get_z(i, shift)); };

		auto pts = boost::irange(0, num_steps) | boost::adaptors::transformed(f);

		gp << gp.binFile1d(pts, "record") << "with lines title 'boost transform to tuple'";
	}

	gp << ", ";
	shift += 1.0/num_examples;

	{
		auto steps = boost::irange(0, num_steps);

		gp << gp.binFile1d(boost::make_tuple(
				steps | boost::adaptors::transformed(boost::bind(get_x, _1, shift)),
				steps | boost::adaptors::transformed(boost::bind(get_y, _1, shift)),
				steps | boost::adaptors::transformed(boost::bind(get_z, _1, shift))
			), "record") << "with lines title 'tuple of boost transform'";
	}

	gp << ", ";
	shift += 1.0/num_examples;

	{
		// Note: C style arrays seem to work, but are a bit fragile since they easily decay to
		// pointers, causing them to forget their lengths.  It is highly recommended that you
		// use boost::array or std::array instead.  These have the same size and efficiency of
		// C style arrays, but act like STL containers.
		double x_pts[num_steps];
		double y_pts[num_steps];
		double z_pts[num_steps];
		for(int i=0; i<num_steps; i++) {
			x_pts[i] = get_x(i, shift);
			y_pts[i] = get_y(i, shift);
			z_pts[i] = get_z(i, shift);
		}
		// Note: std::make_tuple doesn't work here since it makes the arrays decay to pointers,
		// and as a result they forget their lengths.
		gp << gp.binFile1d(std::tie(x_pts, y_pts, z_pts), "record") <<
			"with lines title 'std::tie of double[N]'";
	}
#endif

	gp << std::endl;

	shift += 1.0/num_examples;
	//std::cout << shift << std::endl;
	assert(std::fabs(shift - 1.0) < 1e-12);

#ifdef _WIN32
	// For Windows, prompt for a keystroke before the Gnuplot object goes out of scope so that
	// the gnuplot window doesn't get closed.
	std::cout << "Press enter to exit." << std::endl;
	std::cin.get();
#endif

	return 0;
}