Contents
- Compilers Installation and configuration
Even though Python is an interpreted language, you may need to install Windows C++ compilers in some cases. Unlike Linux, compilers for Windows are not included by default in the OS.
For example, you will need to use them if you wish to:
Install a non-pure Python package from sources with Pip (if there is no Wheel package provided).
Compile a Cython or Pyrex file.
Microsoft provides official C++ compilers called Visual C++, you can find them bundled with Visual Studio or, for some versions, in standalone distributions. Some alternative compilers exist like MinGW, but incompatibilities may occur with a CPython official distribution that is built with Microsoft Visual C++.
Install Python OpenCV library for Visual StudioInstall NumPy, SciPy, Matplotlib for PythonPython download: Numpy,Scipy. This video shows you the basics of creating a python project in VS2017. I show you how to add existing code and files to a new project and also how to set th. Visual Studio 2017 contains Visual C 14.1 compiler. The setuptools Python package version must be at least 34.4.0. Visual Studio 2017 was upgraded by Microsoft to Visual Studio 2019. See the previous paragraph to install it.
The compiler's architecture must be the same as Python's (for example: if you use Python 64bit, you have to use an x64 compiler).
Each Python version uses a specific compiler version (e.g. CPython 2.7 uses Visual C++ 9.0, CPython 3.3 uses Visual C++ 10.0, etc). So, you need to install the compiler version that corresponds to your Python version :
Visual C++ | CPython |
14.X | 3.5, 3.6, 3.7, 3.8 |
10.0 | 3.3, 3.4 |
9.0 | 2.6, 2.7, 3.0, 3.1, 3.2 |
If the package's setup.py (still) uses distutils rather than the recommendedsetuptools, you may need extra steps:
distutils only supports the very minimum of compiler setups. The sections in this guide corresponding to them explicitly mention distutils.
For other setups, you need to run the compilation from the 'SDK prompt' of the corresponding toolchain and set the DISTUTILS_USE_SDK environment variable to a non-empty value.
Compatible architectures are specified for each compiler in brackets.
Before do anything, install or upgrade the Setuptools Python package. It contain compatibility improvements and add automatic use of compilers:
Microsoft Visual C++ 14.2 standalone: Build Tools for Visual Studio 2019 (x86, x64, ARM, ARM64)
This is a standalone version of Visual C++ 14.2 compiler, you don't need to install Visual Studio 2019.
Install Microsoft Build Tools for Visual Studio 2019.
In Build tools, install C++ build tools and ensure the latest versions of MSVCv142 - VS 2019 C++ x64/x86 build tools and Windows 10 SDK are checked.
The setuptools Python package version must be at least 34.4.0.
Build Tools also allows to install any previous Visual C++ 14 version (Including 2015, 2017 ones).
Microsoft Visual C++ 14.2 with Visual Studio 2019 (x86, x64, ARM, ARM64)
Visual Studio 2019 contains Visual C++ 14.2 compiler. The setuptools Python package version must be at least 34.4.0.
Microsoft Visual C++ 14.1 standalone: Build Tools for Visual Studio 2017 (x86, x64, ARM, ARM64)
Python Visual Studio 2017
This is a standalone version of Visual C++ 14.1 compiler, you don't need to install Visual Studio 2017.
Install Microsoft Build Tools for Visual Studio 2017.
The setuptools Python package version must be at least 34.4.0.
Build Tools for Visual Studio 2017 was upgraded by Microsoft to Build Tools for Visual Studio 2019. See the previous paragraph to install it.
Microsoft Visual C++ 14.1 with Visual Studio 2017 (x86, x64, ARM, ARM64)
Visual Studio 2017 contains Visual C++ 14.1 compiler. The setuptools Python package version must be at least 34.4.0.
Visual Studio 2017 was upgraded by Microsoft to Visual Studio 2019. See the previous paragraph to install it.
Microsoft Visual C++ 14.0 standalone: Visual C++ Build Tools 2015 (x86, x64, ARM)
This is a standalone version of Visual C++ 14.0 compiler, you don't need to install Visual Studio 2015.
Install Microsoft Visual C++ Build Tools 2015. Check Windows 8.1 SDK and Windows 10 SDK options.
The setuptools Python package version must be at least 24.0.
Visual C++ Build Tools 2015 was upgraded by Microsoft to Build Tools for Visual Studio 2017. See the previous paragraph to install it.
Microsoft Visual C++ 14.0 with Visual Studio 2015 (x86, x64, ARM)
Visual Studio 2015 contains Visual C++ 14.0 compiler. Distutils will automatically detect the compiler and use it.
Visual Studio 2015 was upgraded by Microsoft to Visual Studio 2017. See the previous paragraph to install it.
Microsoft Visual C++ 10.0 standalone: Windows SDK 7.1 (x86, x64, ia64)
This is a standalone version of Visual C++ 10.0 compiler, you don't need to install Visual Studio 2010.
Uninstall Microsoft Visual C++ 2010 Redistributable if present (all versions and architectures). If present, it can cause an error on Windows SDK 7.1 installation.
Install Microsoft .NET Framework 4 if not present.
Install Microsoft Windows SDK for Windows 7 and .NET Framework 4. Check Windows headers and libraries, Visual C++ Compilers and Windows Native Code DevelopmentTools options only.
Install Microsoft Visual C++ 2010 Service Pack 1 Compiler Update for the Windows SDK 7.1. This updates the compiler to Visual C++ 10.0 SP1.
reinstall Microsoft Visual C++ 2010 Redistributable (for all previously installed architectures).
The setuptools Python package version must be at least 24.0.
Microsoft Visual C++ 10.0 with Visual Studio 2010 (x86, x64, ia64)
Visual Studio 2010 contains Visual C++ 10.0 compiler. Distutils will automatically detect the compiler and use it. The Express edition of Visual Studio 2010 only bundles a compiler for x86.
Microsoft Visual C++ 9.0 standalone: Visual C++ Compiler for Python 2.7 (x86, x64)
This is a standalone version of Visual C++ 9.0 compiler, you don't need to install Visual Studio 2008.
Install Microsoft Visual C++ Compiler for Python 2.7.
The setuptools Python package version must be at least 6.0.
Even though this package's name refers to Python 2.7 specifically, you can use it with all Python versions that use Visual C++ 9.0.
This package always installs its start menu shortcuts for the installing user (i.e. an administrator) only. To get them for all users, run the installation like this: msiexec /i <full path to .msi> ALLUSERS=1.
Microsoft Visual C++ 9.0 standalone: Windows SDK 7.0 (x86, x64, ia64)
This is a standalone version of Visual C++ 9.0 compiler, you don't need to install Visual Studio 2008.
The use of Microsoft Visual C++ Compiler for Python 2.7 is recommended (If you don't need to compile for ia64). See the previous paragraph to install it.
Opencv Python Visual Studio 2017
Install Microsoft .NET Framework 3.5 SP1 if not present.
Install Microsoft Windows SDK for Windows 7 and .NET Framework 3.5 SP1. Check Windows headers and libraries, Visual C++ Compilers and Win32 Development Tools options only.
The setuptools Python package version must be at least 24.0.
Python Visual Studio 2017 Tutorial
Microsoft Visual C++ 9.0 standalone: Windows SDK 6.1 (x86, x64, ia64)
This is a standalone version of Visual C++ 9.0 compiler, you don't need to install Visual Studio 2008.
Windows SDK 6.1 was upgraded by Microsoft to Windows SDK 7.0. See the previous paragraph to install it.
Install Microsoft .NET Framework 3.5 SP1 if not present.
Install Microsoft Windows SDK for Windows Server 2008 and .NET Framework 3.5. Check Windows headers and libraries, Visual C++ Compilers and Win32 Development Tools options only.
The setuptools Python package version must be at least 24.0.
Microsoft Visual C++ 9.0 with Visual Studio 2008 (x86, x64, ia64)
Visual Studio 2008 contains Visual C++ 9.0 compiler. Distutils will automatically detect the compiler and use it. The Express edition of Visual Studio 2008 only bundles a compiler for x86.
GCC - MinGW-w64 (x86, x64)
MinGW-w64 is an alternative C/C++ compiler that works with all Python versions up to 3.4.
Install Win-builds into C:MinGW_w64.
Open Win-builds, switch to install at least binutils, gcc, gcc-g++, getext, mingw-w64, win-iconv, winpthreads, zlib, and click Process.
Add C:MinGW_w64bin to the PATH environment variable.
Create a distutils.cfg file with the following contents in the folder Libdistutils in Python install directory :
GCC - MinGW (x86)
MinGW is an alternative C/C++ compiler that works with all Python versions up to 3.4.
Install Minimalist GNU For Windows into C:MinGW.
Open MinGW Installation Manager, check mingw32-base and mingw32-gcc-g++, and Apply Changes in the Installation menu.
Add C:MinGWbin to the PATH environment variable.
Create a distutils.cfg file with the following contents in the folder Libdistutils in Python install directory :
Python is a popular programming language that is reliable, flexible, easy to learn, free to use on all operating systems, and supported by both a strong developer community and many free libraries. The language supports all manners of development, including web applications, web services, desktop apps, scripting, and scientific computing and is used by many universities, scientists, casual developers, and professional developers alike.
Visual Studio provides first-class language support for Python. This tutorial guides you through the following steps:
Prerequisites
- Visual Studio 2017 with the Python workload installed. For instructions, see Work with Python in Visual Studio - Step 0.
- Visual Studio 2019 with the Python workload installed. For instructions, see Work with Python in Visual Studio - Step 0.
You can also use an earlier version of Visual Studio with the Python Tools for Visual Studio installed. See Install Python support in Visual Studio.
Step 1: Create a new Python project
A project is how Visual Studio manages all the files that come together to produce a single application, including source code, resources, configurations, and so on. A project formalizes and maintains the relationship between all the project's files as well as external resources that are shared between multiple projects. As such, projects allow your application to effortlessly expand and grow much easier than simply managing a project's relationships in ad hoc folders, scripts, text files, and even your own mind.
In this tutorial you begin with a simple project containing a single, empty code file.
In Visual Studio, select File > New > Project (Ctrl+Shift+N), which brings up the New Project dialog. Here you browse templates across different languages, then select one for your project and specify where Visual Studio places files.
To view Python templates, select Installed > Python on the left, or search for 'Python'. Using search is a great way to find a template when you can't remember its location in the languages tree.
Notice how Python support in Visual Studio includes a number of project templates, including web applications using the Bottle, Flask, and Django frameworks. For the purposes of this walkthrough, however, let's start with an empty project.
Select the Python Application template, specify a name for the project, and select OK.
After a few moments, Visual Studio shows the project structure in the Solution Explorer window (1). The default code file is open in the editor (2). The Properties window (3) also appears to show additional information for any item selected in Solution Explorer, including its exact location on disk.
Take a few moments to familiarize yourself with Solution Explorer, which is where you browse files and folders in your project.
(1) Highlighted in bold is your project, using the name you gave in the New Project dialog. On disk, this project is represented by a .pyproj file in your project folder.
(2) At the top level is a solution, which by default has the same name as your project. A solution, represented by a .sln file on disk, is a container for one or more related projects. For example, if you write a C++ extension for your Python application, that C++ project could reside within the same solution. The solution might also contain a project for a web service, along with projects for dedicated test programs.
(3) Under your project you see source files, in this case only a single .py file. Selecting a file displays its properties in the Properties window. Double-clicking a file opens it in whatever way is appropriate for that file.
(4) Also under the project is the Python Environments node. When expanded, you see the Python interpreters that are available to you. Expand an interpreter node to see the libraries that are installed into that environment (5).
Right-click any node or item in Solution Explorer to access a menu of applicable commands. For example, the Rename command allows you to change the name of any node or item, including the project and the solution.
Next step
Go deeper
- Python projects in Visual Studio.
- Python for Beginners (python.org)