Skip to content

It`s uncomfortable to use regular C++ after Python due to a little number of built-in functions... Pythonic solves this issue.

Notifications You must be signed in to change notification settings

donRumata03/pythonic

Repository files navigation

Pythonic

It`s a C++ library which provides basic functions simplifying all aspects of work I use it in all of my C++ projects and it`s absolutely irreplaceable for me.

Usage:

This library uses CMake as build system. To use this library you probably need to build it for you platform and compiler. However, there are some binary files in "cmake-build-debug" and "cmake-build-release" (.exe and .lib)

The only dependency is nlohmann json from https://github.com/nlohmann/json You should find the single-include file and provide the path to it or use ./Cmake/FindJson.cmake

There are no additional dependencies, so, once you are done with json you can simply run CMake. After that, *.lib files for Release or Debug mode appear in the corresponding locations. To use it from other libraries or executables you should type set(Pythonic_DIR ${LOCATION_OF_THE_ROOT_FOLDER_OF_PYTHONIC}) in you CMake file. And after that: find_package(Pythonic required)

"pythonic_static_library" variable will contain the name of the library you should link against. "pythonic_include_directories" variable will contain the list of directories you should add to you include directory index.

So, you should write this lines:

include_directories(${pythonic_include_directories})
target_link_libraries(${YOUR_TARGET_NAME} ${pythonic_static_library})

Features:

  • Utilities for displaying (in "./displaying/" folder):

    • Stream output operators for STL containers
    • Type name getter function, returns std::string, pays respect to pointers, references, specifiers and qualifiers
    • ANSI colors in console (see https://en.wikipedia.org/wiki/ANSI_escape_code)
    • Some library from github, which displays a number of bytes in human readable format and supports different standards
  • Smoothing:

    • It uses Gaussian window for smoothing: 1-d (std::vector) and n-d (nd_array) arrays
  • Plotting:

    • Uses matplotlib from python and .json interface between languages the path should be provided...
    • Just like in matplotlib, here are add_to_plot and show_plot functions
    • You can also add to plot different objects: lines, exponents and custom functions
    • There is also support of throwing functions: is a function throws and the corresponding argument flag is set, the graph is divided into separate parts with the same color.
    • Legend and labels is also supported
  • Statistics utilities:

    • "normal" function: outputs the value of The Normal Distribution given σ and μ
    • "Exponentating" and "Logariphmation" are used for "Exponentating" and "Logariphmation" std::vector`s of values
    • Mean functions: ariphmetic, geometric, n_ometric, harmonic, n_monic, quadratic, n_atic means are supported
    • There are also "decorators" for all the means which logoriphmate or exponentate the values before performing mean functions and execute the inverse transformation. For example, logomean<arithmetic_mean_functor>(...) is like geometric_mean(...)
    • Some other useful functions
  • Multidimensional Array class with slicing support (see n_dim/nd_array.h):

    • // TODO: write some useful characters here!
  • A collection of random generators:

    • Random number from [0, 1) with rand() and with std::mt19937
    • Random number from [x, y) with rand() and with std::mt19937
    • Choice a random element of a collection
    • Normalizer: a class which applies a hack to speed the generation of normal distributed random numbers up: it generates a number of them at the initialization and when a random number with a given σ and μ is required, it applies some linear transformations and outputs the result. It`s possible to tune the precision and memory consumption by changing the number of the generated randoms
    • "probability_choose" function chooses one or more (depending on its argument) numbers of the given set with probabilities proportional to the fitness values (you don`t need to additionally normalize them because it`s performed automatically)
  • Tools to deal with encodings (in "./encoding/" folder):

    • My C++ wrapper around WinAPI for converting between UTF-8, UTF-16 (windows system encoding), and CP-1251 (one-byte cyrillic encoding)
    • Automatic encoding determination (UTF-8, CP-1251 and pure ASCII)
    • Automatic recoding to one of those encodings (UTF-8, CP-1251 and pure ASCII)
  • Some primitives for geometry (in "./geometry/" folder):

    • Point: has stream output operator and other stuff
    • Line (has support of different formats for representing a line)
    • Circle : also supports different formats including "three points" format
  • Utilities to deal with letters: determines if a character is russian, english, space, digit. e.t.c. assumes that the character represents a cp1251-encoded code unit

All of the features ^^^ above ^^^ are covered by tests (in folder with corresponding name)

Feel free to contact me if there are some issues with understanding the library interface or even bugs: donrumata03@gmail.com

About

It`s uncomfortable to use regular C++ after Python due to a little number of built-in functions... Pythonic solves this issue.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages