Compiling C++ with a batch file

I do not like compiling in visual studio’s IDE. I like compiling C++ with a batch file instead. This used to be a lot easier. Or at least it felt like it was 15 years ago. But Microsoft seems to make this harder to do with each updated version of visual studio. If you’re on Linux this may not be a problem for you. But my hobbies and work keep me on windows most of the time. So this is how I compile C++ with a batch file.

So, over the last few years I been slowing building and changing my script for compiling and here is what I am using right now.

compiling c++ with a batch file

Update: Here is my batch code for 2018

At some point Windows 10 started getting weird about the pushd command so I had to reorder the code. I also added support for Visual Studio 2017.

compiling c++ with a batch file

Line by line explanation

The first thing I do is store all the compiler flags I want to use in a batch file variable.

set CommonCompilerFlags=-Zi /Wv:18

Then I do the same thing with linker flags.

set CommonLinkerFlags=user32.lib gdi32.lib opengl32.lib

Then because I don’t want to have to worry about setting paths on every computer I work on I just run the file that comes with visual studio that sets the path for this session. This location will most likely change as new versions of visual studio come out.

For Visual Studio 2017:

call "C:\Program Files (x86)\Microsoft Visual Studio\2017\Community\VC\Auxiliary\Build\vcvarsall.bat" x64

For Visual Studio14:

call "C:\Program Files (x86)\Microsoft Visual Studio 14.0\VC\vcvarsall.bat" x64

Now I like my compiled files to be saved neatly into a build folder so here I make the folder. The %~dp is just a shortcut for the current directory the script is running in so you do not have to hardcode it.

mkdir %~dp0build

This will change the current directory the script is working in to the new build folder.


Then I move any assets that the program needs to run. In this case I have shader files. So I use robocopy to put them in a Shader folder inside my new build folder.

robocopy ..\Shaders Shaders /MIR

I also move some other files into a data folder inside the build folder.

 ..\data data /MIR

Now we run the compiler. We add our compiler and linker flags and the new compiled file should show up in the build directory.

cl %CommonCompilerFlags% "..\OpenGL.cpp" %CommonLinkerFlags%'

Reset and go back to the original current directory.


Sometimes I add the pause and sometimes I don’t. It depends on how much debugging I am doing.


So, the full build.bat script looks something like this and I have it saved in with my source files.

@echo off
REM Notes: I had to add "/Wv:18" to compile in new Visual Studio
set CommonCompilerFlags=-Zi /Wv:18
set CommonLinkerFlags=user32.lib gdi32.lib opengl32.lib
REM if the cl command is not working we will need to set up the path by running vcvarsall.bat
call "C:\Program Files (x86)\Microsoft Visual Studio\2017\Community\VC\Auxiliary\Build\vcvarsall.bat" x64
REM If 2017 is not found try the old location
call "C:\Program Files (x86)\Microsoft Visual Studio 14.0\VC\vcvarsall.bat" x64
mkdir "%~dp0build"
pushd "%~dp0build"
robocopy ..\Shaders Shaders /MIR
robocopy ..\data data /MIR
cl %CommonCompilerFlags% "..\OpenGL.cpp" %CommonLinkerFlags%

So how do you debug with Visual Studio if you are compiling with a script? That is easy. I have another script call debug.bat and it looks like this.

REM Open program in Visual Studio for debugging.
"C:\Program Files (x86)\Microsoft Visual Studio 14.0\Common7\IDE\" build\OpenGL.exe openGL.cpp

Just point it to where the exe file was saved. In this case, it was saved in the build directory. Then you can add cpp files on for it to open automatically when the debugger opens.

And that is compiling C++ with a batch file. This is how I compile and debug all my code. If you wanted to get fancy you could also add some macros or shortcuts to the scripts so that it is all seamless.

Share this post?

Leave a Reply

Your email address will not be published. Required fields are marked *