Shell script for C/C++ projects

1. Download the script or the repo in a new project folder.
2. do chmod 755 run.sh to let you execute it if you don't have the permissions for.
3. do ./run.sh -g to generate the project's structure. By default it's for cpp. See ./run.sh -l command.
4. do ./run.sh to compile the code and execute it.

Run ./run.sh --patch to download latest online version of the script.

Script usage

$ ./run.sh -h # shows every possible usages (help command)

The script applies the right executable's extension depending on which OS you are :

• none for Linux
• .exe for Windows
• .app for MacOS

Structure

Project's mode 0

| bin/ # executable folder
    | release/ # folder where there's the release executable version
    | debug/ # folder where there's the debug executable version
| build/ # follows the sub-folders structure you made in your src/ folder.
    | # all object files (.o) are stored here
| include/
    | # every header files are stored here (.h/.hpp/.inl)
| src/
    | # every source files are stored here (.c, .cpp)
    | main.cpp
.gitignore
License
Makefile
README.md
run.sh

Project's mode 1

| bin/ # executable folder
    | release/ # folder where there's the release executable version
    | debug/ # folder where there's the debug executable version
| build/ # follows the sub-folders structure you made in your src/ folder.
    | # all object files (.o) are stored here
| src/
    | # every source and header files are stored here (.c, .cpp, .h, .hpp, .inl)
    | Engine/
        | Engine.cpp
        | Engine.h
    | main.cpp
.gitignore
License
Makefile
README.md
run.sh

The makefile will compile every source it will find on the src/ folder. It is also adding every sub-folders of include/ (or src/ if prject's mode = 1) folder so when you're doing an include, you just have to write the filename (with .h at the end) without its path.

Example :

#include "../../utils/something.hpp" // useless !

Use this instead :

#include "something.hpp"

So you don't have to be worried about the Makefile, just code and ./run.sh !

Tip : you can add ./run.sh as an alias or symlink so you just have to write run instead.

Script for libraries

You can use the script to build static and/or shared libraries :

./run.sh --static or ./run.sh --shared.

This will create the object files in the build folder, then create the .a, .so or .dll (on Windows) in the bin/lib/ folder.
Plus, it will copy/paste all the header files you have (.h, .hpp and .inl) in bin/lib/include/{Project_Name}/, reorganizing these as follow :

• All files header files that are alone in their folder will be up to their parent, recusivly. This avoids a ton of subfolders just for a file.
• All #include "my_project_file" pathes will be transformed so it will search with a relative path. You can then use it everywhere without worrying about the configuration.