Gnuplot++

A simple C++ Gnuplot interface.

This is a very simple library that allows you to call the Gnuplot program from C++ code using a pipe.

Make sure to install Gnuplot before using this library (on Windows you have to add Gnuplot to PATH by selecting the option during installation or by adding the bin folder to PATH by hand). To test your gnuplot installation you can open a terminal and type gnuplot --version.

Also take a look at the Gnuplot manual and the docs and examples folders.

If you don't know how to use C++, take a look at IDontKnowCpp guide

Features

• extremely simple and easy to use
• no need of knowing gnuplot syntax
• simple std::cout like interaction for sending commands using the << operator
• modern C++ style interation using high level of abstraction functions
• no messing with Gnuplot indices
• no temporary data files needed
• easy plot customization
• MIT license

Usage

Using the Gnuplotpp class is extremely esasy.

Take a look at the Including guide to discover how you can use the Gnuplot++ library in your project, then read the Quick Start guide to rapidly start using the library

You can send commands like you would do in the gnuplot shell:

#include <gnuplotpp/gnuplotpp.hpp>
// ...
Gnuplotpp gp;
gp << "plot [-pi/2:pi] cos(x),-(sin(x) > sin(x+1) ? sin(x) : sin(x+1))" << std::endl;

or use the interface functions to easily create complex and customizable plots:

Gnuplotpp gp;
auto myPlot = gp.plot({ 1, 2, 3, 2 });
gp.render({ myPlot });

Why Gnuplot

TODO

Why not Gnuplot

TODO

Contributing

• read CODE_OF_CONDUCT.md and CONTRIBUTING.md

You can fork the project and create your own version.

If you want to tell a problem, suggest a feature or request assistance for a problem that can be shared, create an issue.

If you want to directly contribute to the main project, create a fork and then create a pull request that I will review. Please, read the rules in the CONTRIBUTING.md file before creating a pull request.

If you appreciate my work, please consider making a donation using the Donate button: STUDENTS NEED COFFEE ☕

---

---

---

---

---

---

---

---

---

---

---

---

---

---

OLD

Using as a single header library

This is the simplest option: you have to place the gnuplotpp.hpp and the gnuplotpp.cpp file in your favourite location, for example near your source file, then write:

#define _LC_GNUPLOTPP_HEADER_ONLY
#include "gnuplotpp.hpp"

Optionally, if you are using CMake as build system:

target_compile_definitions(
  your_target_name
  PUBLIC
    _LC_GNUPLOTPP_HEADER_ONLY
)

and then, in your C++ code simply:

#include "gnuplotpp.hpp"

If you are using a different build system, you can add the _LC_GNUPLOTPP_HEADER_ONLY definition in different ways.

Building and installing

This is the recommended option if you are using CMake as build system.

On linux or any other system where you can call cmake from console you can navigate to the project folder and type:

cmake .

this will configure the project, it sould take a few seconds.

Then you have to build the library, type:

cmake --build .

or

make

And after another few seconds the library will be built. To install, type:

make install

At this point you can use the library adding the header path and linking the lib. With CMake this is much simpler:

find_package(gnuplotpp REQUIRED)
target_link_libraries(your_target_name PUBLIC gnuplotpp) # or PRIVATE

Please, read the wiki Building and Installing page for more detailed info and common problems solutions.

Adding .hpp and .cpp fiels to your project

This is another simple option: you add both gnuplotpp.hpp and gnuplotpp.cpp to your project so that gnuplotpp.cpp gets compiled.