Windows embeddable package python это

Posted: 6th November 2016
Category: Python
Tags: python, windows

Those who have been following Python development on Windows recently will be aware that I’ve been actively redeveloping the installer. And if you’ve been watching closely you’ll know that there are now many more ways to install the official python.org release than in the past, not even including distributions such as WinPython or Anaconda.

In this post, I’m going to discuss each of the ways you can install the official releases of Python (since version 3.5), provide some context on when and why you would choose one over another, and discuss the positives and negatives of each approach.

History

Historically, there was one single MSI installer available that was intended to cover the needs of all Python users.

Python 3.4.4 installer

This installer would allow you to select a target directory and some features from its user interface or the command line (if you know the magic words), and would generally install the full distribution with all entry points (shortcuts, etc.).

Unfortunately, due to the nature of how MSIs work, there are some limitations that affect the user experience. The most major of these is the fact that MSIs cannot decide whether elevate as part of the install — it has to be hardcoded. As a result, the old installer always requires administrative privileges just in case you choose to install for all users. This prevents installation of Python on machines where you do not have full control over the system.

Secondly, while Python is often seen as one monolithic package, it is actually made up of a number of unrelated components. For example, the test suite is often not required for correct operation, nor is the documentation and often the development headers and libraries. While MSIs do support optional features, they tend to encounter issues when performing upgrades between versions (such as forgetting which options you had selected), and in general you always need to carry around the optional components even if you’re never going to install them.

Finally, some operations that are not simple file installations can be complicated. For example, when pip is installed or the standard library is precompiled, the MSI executes a background task rather than normally installing files. Without careful configuration of the MSI, these files will not be properly uninstalled or repaired, and issues in the extraction process can cause the entire task to fail. At worst, the uninstall step could fail, which can make it impossible to uninstall Python.

Modern Era

The issues described above have been addressed by the installers available since Python 3.5. However, there are also other uses for Python that do not lend themselves to a regular installer. For example, applications that want to include Python as a runtime dependency may not want to install a global copy of Python, build machines may require semi-public but non-conflicting installs of different versions, and platform-as-a-service web hosts may not allow normal installers.

Since Python 3.5.2, the official Python releases have been made available as executable installers, embeddable ZIP packages, nuget packages and Azure site extensions. There are also a range of third-party distributions that include the official Python binaries, along with other useful tools or libraries.

How do I know if a third-party distribution has the official binaries? Find the install directory, right-click each of the exe, dll or pyd files and select Properties, then Digital Signatures. If the signature is from the Python Software Foundation, it’s the official binary and has not been modified. If there is a different signature or no signature, it may not be the same as what is released on python.org.

The Python Software Foundation certificate

Executable installers

The executable installers are the main way that users download Python, and are the featured downloads at python.org. I think of these as the Python Developer Kit.

Python 3.5.0 installer

These installers provide the most flexible user interface, include all dependencies such as system updates and the Python launcher, generate shortcuts for the interpreter, the manuals and the IDLE editor, and correctly support upgrades without forgetting about feature selection.

Two versions of the executable installer are available for any given release — one labelled “executable installer” and the other “web-based installer”.

The web-based installer is typically a small initial download (around 1MB), which gets you the installer UI shown above. After you have selected or deselected optional components, the minimum set of packages necessary to install Python will be downloaded and installed. This makes it easy to minimize overall download size since unused or unnecessary components are never downloaded, though it does require that you be connected to the internet at install time. (There’s also a command-line option to download all the packages you may ever need, which will then be used later instead of downloading them over and over again.)

The other installer includes the default set of features in the EXE itself. As a result, the initial download is around 30MB, but in most cases you can install without requiring any further internet access. For a single installation, your download will likely be 3-5MB larger compared to using the web-based installer, but if you use it to install on multiple machines then you’ll likely come out ahead.

Both executable installers result in identical installations and can be automated with identical command-line options. As I mentioned above, I think of this as the Python Developer Kit, which is why there are optional features to download debugging symbols or a complete debug build, which are not available in any other options. The Python Developer Kit provides everything necessary for someone to develop a complete Python application.

What about having a single MSI installer? There’s a section coming up about this. Just keep reading.

Embeddable package

If the executable installer is the Python Developer Kit, then the embeddable package is the Python runtime redistributable. Rather than trying to be an easy-to-use installer, this package is a simple ZIP file containing the bare minimum of Python required to run applications. This includes the python[w].exe executables, the python35.dll (or later) and python3.dll modules, the standard library extension modules (*.pyd), and a precompiled copy of the standard library stored in another ZIP file.

The resulting package is about 7MB to download and around 12MB when extracted. Documentation, tools, and shortcuts are not included, and the embeddable package does not reliably build and install packages. However, once your application is ready, rather than instructing users to install Python themselves, you can include the contents of this package in your own installer. (For example, Microsoft’s command-line tools for Azure will likely do this, and installers created using pynsist can include this package automatically.)

Using the embeddable package allows you to distribute applications on Windows that use Python as a runtime without exposing it to your users. By default, a configuration file is also included to force the use of isolated mode and prevents environment variables and registry settings from affecting it (python36._pth on Python 3.6; pyvenv.cfg for Python 3.5). On Python 3.6 this file can also specify additional search paths. If your application is hosting Python, you can also choose not to distribute python.exe or any extension modules that are not used in your application.

There is no support for pip, setuptools or distutils in the embeddable package, since the idea is that you will develop against the Python Developer Kit and then lock your dependencies when you release your application. Depending on the installer technology you are using for your application, you will probably vendor any third-party packages by copying them directly into the directory with your Python code.

See this blog post for more information about how to take advantage of the embeddable distribution.

Nuget package

Nuget is a packaging technology typically used on Windows to manage development dependencies. There are many packages available as source code or pre-built binaries, mostly for .NET assemblies, as well as build tools and extensions.

There are four Python packages available on nuget, released under my name (steve.dower) but built as part of the official python.org releases. The packages are:

  • python — the latest 3.x 64-bit
  • pythonx86 — the latest 3.x 32-bit
  • python2 — the latest 2.x 64-bit
  • python2x86 — the latest 2.x 32-bit

These may be referenced by projects in Visual Studio or directly using nuget.exe to easily install a copy of Python into a build directory. It will typically install into a directory like packages\python.3.5.2\tools\python.exe, though this can often be customised.

rem Install Python 2.7
nuget.exe install -OutputDirectory packages python2

rem Add -Prerelease to get Python 3.6
nuget.exe install -OutputDirectory packages -Prerelease python

rem More options are available
nuget.exe install -Help

The contents of the nuget package is somewhere between the full installation and the embeddable package. The headers, libs and pip are included so that you can install dependencies or build your own modules. The standard library is not zipped, but also does not include the CPython test suite or libraries intended for user interaction. Operating system updates are not included, so you will need to ensure your build machine is up to date before using these packages.

There is no configuration in these packages to restrict search paths or environment variables, as these are very important to control in build definitions. As a result, there is a high likelihood that a regular installation of Python may conflict with these packages. In general, it’s best to avoid installing Python on build machines where you are using these packages. If you need a full installation, avoid using the nuget packages or test for conflicts thoroughly. (Note that conflicts typically only occur within the same x.y version, so you can safely install 2.7 and use the 3.5 nuget packages.)

Azure Site Extensions

Update 2019: These packages have been deprecated and removed. This section is of historical interest only.

Note: This particular package is released by Microsoft and is managed by my team there. The Python Software Foundation is not responsible for this package.

Azure App Service is a platform-as-a-service offering for web services (including web apps, mobile backends, and triggered jobs). It uses site extensions to customise and enhance your web services, including a range of Python versions to simplify configuration of Python-based servers.

Because web services are sensitive to even the smallest change in a dependency, each version is available as its own package. This allows you to be confident that when your site uses one of these it is not going to change without you explicitly updating your site. The current packages available at time of writing are:

  • python352x64
  • python352x86
  • python351x64
  • python2712x64
  • python2712x86
  • python2711x64

The contents of these packages is almost entirely unmodified from the official python.org releases. Some extra files for correct installation, configuration and behaviour of the web server are included, as well as copies of pip, setuptools, and certifi. Occasionally a package will include targeted patches to fix or work around issues with the platform, but we always aim to upstream fixes as soon as possible. Under the hood, these are simply nuget packages that can also be installed using nuget.exe on any copy of Windows.

C:\> nuget.exe list python -Source https://www.siteextensions.net/api/v2/
python2711x64 2.7.11.3
python2712x64 2.7.12.2
python2712x86 2.7.12.1
python351x64 3.5.1.6
python352x64 3.5.2.2
python352x86 3.5.2.1

Visit aka.ms/PythonOnAppService for the most up-to-date information about how to use these packages on Azure App Service.

Hypothetical Futures

While that covers the current set of available installers, there are some further use-cases that are not as well served. In this section I will briefly discuss the cases that I am currently aware of and their status. There are no promises that official installation packages for these will ever be produced (bearing in mind that Python is developed almost entirely by volunteers with limited free time), but there is also nothing preventing third-parties from producing and distributing these formats.

Are you already distributing Python in any of these formats? Let me know and I’m happy to link to you, provided I’m not concerned about the contents of your distribution.

Nuget package for source/runtime dependency

Earlier I discussed the nuget packages as build tools, but the more common use of nuget packages is for build dependencies. Normally a project (typically a Visual Studio project, but nuget can also be used independently) will specify a dependency on a source or binary package and obtain build steps or configuration from a known location within the package.

Providing a nuget package containing either the Python source code or the embeddable package may simplify projects that host the runtime. These would predominantly be C/C++ projects rather than pure Python projects, but some installer toolkits may prefer a ready-to-embed nuget package rather than a plain ZIP file.

There has not been much demand for this particular format. In general, a C/C++ project can make equally good use of the current nuget packages, and would require those for the headers and libraries anyway, while the embeddable package is not always suitable for installation completely unmodified. These reduce the value of a dependency nuget package to nearly zero, which is why we currently don’t have one.

Universal Windows Platform

The Universal Windows Platform is part of Windows 10 and specifies a common API set that is available across all Windows devices. This includes PCs, tablets, phone, IoT Core, XBox, HoloLens, and likely any new Windows hardware into the future.

Providing a UWP package of Python would allow developers to distribute Python code across all of these platforms. Indeed, the team behind IoT Core have already provided their version of this package. However, as the API set is not always compatible with the Win32 API, this task requires supporting a new platform within Python (that is, sys.platform would return a value other than 'win32'). Currently nobody has completely adapted Python for UWP, added the extensions required to access new platform APIs, or fully implemented the deployment tools needed for this to be generally useful (though the IoT Core support is a huge step towards this).

Administrative Deployment

System administrators will often deploy software to some or all machines on their network using management tools such as Group Policy or System Center. While it is possible to remotely install from the executable installers, these tools often require or have enhanced functionality when the installer is a pure MSI.

Unfortunately, the issues and limitations of MSI described at the start of this post still apply. It is not possible for an MSI to install all required dependencies, create an MSI that can run without administrator privileges, and robustly ensure that upgrade and remove operations behave correctly. However, it would be possible to produce a suitable MSI and installation instructions for the limited use case of administrative deployment. Such a package would likely have these characteristics:

  • Fails if certain operating system updates are missing
  • Always requires administrator privileges
  • Only allows installation for all users
  • Only allows configuration at the command line (via msiexec)
  • Requires a separate task to precompile the standard library and install pip
  • Requires additional cleanup task when uninstalling
  • Prevents the executable installer from installing for all users

System administrators would be responsible for following the documentation associated with such an MSI, and I have no doubt that most are entirely capable of doing this. However, as this would not be a good experience for most users it cannot be the default or recommended installer. I’m aware that there are some people who are grieved by this, but interactive installs are vastly more common and so take priority when determining what to offer from python.org.

Summary

Installing Python on Windows has always been a fairly reliable process. The ability to select precisely which version you would like without fear of damaging system components allows a lot of confidence that is not always available on other platforms. Improvements in recent releases make it easier to install, upgrade and manage Python, even for non-administrator users.

We have a number of different formats in which Python may be obtained depending on your intended use. The executable installers provide the full Python Developer Kit; the embeddable package contains the runtime dependencies; nuget packages allow easy use of Python as a build tool; and site extensions for Azure App Service make it easier to manage Python as a web server dependency.

There is also potential to add new formats in the future, either through third-party distributions or as new maintainers volunteer their time towards core development. For an open-source project that is run almost entirely on volunteer time, Python is an amazing example of a robust, trustworthy product with as much flexibility as any professionally developed product.

Discussion of this post is welcome here in the comments. If you are having issues installing Python, please file an issue on bugs.python.org.

Windows Embeddable Python with virtualenv support

The repository contains code to repackage Windows Embeddable version of Python with virtualenv.

What is Windows Embeddable Python?

Windows Embeddable Python is a minimalistic Python distribution suitable for bundling Python inside applications for Windows.

The package is stored in downloads: https://www.python.org/downloads

The name of the package is: Windows x86-64 embeddable zip file

Why is it necessary to repackage ZIP with Embeddable Python?

The official version has the following limitations:

  • the package does not contain pip
  • the package does not support virtualenv
  • resolution of module path behaves slightly different
    • more details: https://dev.to/fpim/setting-up-python-s-windows-embeddable-distribution-properly-1081

How it works

  • download and unpack Windows Embeddable Python
  • extract content of pythonYY.zip to Lib, remove pythonYY.zip
  • remove pythonYY._pth file
    • this file controls behavior of imports when starting Python
  • download get-pip.py and install pip
  • pip install virtualenv
  • download or use full version of python and copy Lib\venv\scripts\nt\*.exe to python bundle
    • during the creation of virtualenv these python.exe and pythonw.exe files are being copied to virtual environment location
    • these two files are different from python.exe and pythonw.exe in the top Python directory

How to build

PowerShell:

.\Bundle-Python.ps1 -PythonVersion 3.8.6

Result Python distribution will be available in python directory.

How to build with GitHub Actions

Check the version of Python in .github/workflow/bundle-python.yaml.
Commit the change and check Actions at GitHub.com.
Zip file with the result will be available as artifact in build details.

The Python Embeddable Package (PEP) is a highly useful feature of Python that allows you to embed a Python interpreter into C and C++ applications providing a powerful scripting environment by extending its functionality.

In this article, we’ll explore what the Python Embeddable Package is and how it can be used in your own projects.

Overview of Python Embeddable Package

PEP is a lightweight package that is designed to be easily embedded within other applications. It is essentially a minimal Python environment.

This package uses Python as runtime and consists of Python Standard Library modules, a Python interpreter, a set of header files, and C APIs that can be used to interact with Python from within a host application as it provides functions to execute Python code, access Python objects, etc.

It is created to be used by developers who specifically want to integrate Python scripting capabilities into their own applications, without the need for users to install a separate version of Python. By embedding Python within your application, you can provide a flexible interface for customization and powerful extensibility.

Implementing an embeddable Python environment within existing projects opens doors to a world of possibilities, making applications more dynamic, customizable, and efficient.

Advantages of Python Embeddable Package

Advantages of Python Embeddable Package

There are several benefits of using PEP in your application, including:

  1. Powerful: The Python interpreter provides programmers with a powerful environment for scripting that can be used to extend the performance of their programs.
  2. Readability: Python code is very easy to read and understand which makes it a great option to work with complex code.
  3. Easy to use: PEP can be integrated into any of your C or C++ applications quickly and is very easy to use.
  4. Community: Python has one of the largest active communities to provide you with support and documentation.

How to use Python Embeddable Package?

The Python Embeddable Package can be downloaded from Python’s official website, it is available for Windows, Linux, and MacOS. To use it, you need to follow these simple steps:

  1. Download and install the appropriate distribution for your platform of the Python Embeddable Package from the official Python website.
  2. Include the Python header files in your C application and use the C APIs to embed Python
  3. Link your application against the Python library.
  4. Create Python scripts that will be executed by the embedded interpreter.
Python Releases for Windows
Download it from here

You can also use some tools and third-party libraries available that help you to simplify the process of embedding Python, such as Boost.Python and PyBind11.

Applications of Python Embeddable Package

The Python Embeddable Package can be used in an array of ways, depending on the type of application you are building. Here are a few examples of how it can be used:

  • Plugins – If you are working on a large project that includes plugins, the Python Embeddable Package can be used to provide an architecture that allows developers to write plugins in Python. This is best for applications that require a lot of customization and flexibility.
  • Automation – Python is very powerful when it comes to automating things and using the PEP you can easily add automation capabilities to your own application. You gain the ability to automate repetitive tasks, execute batch processes, and a variety of other operations to make your life easy.
  • Web Server – You can embed a Python interpreter into a web server to write scripts that can be used to handle requests from clients or generate dynamic web pages and develop server-side applications using frameworks such as the famous Django and Flask.
  • Game Engine – You can use the Python Embeddable Package to embed a Python interpreter into a game engine to create game logic, control game mechanics and modify the behavior of the game engine at runtime. It can be used to create mods and provide an overall more flexible way to develop games.
  • Scientific Application – PEP allows you to write Python scripts that can be used to perform complex calculations and analyze data for statistical analysis and machine learning. You can integrate your application with a lot of scientific libraries, such as NumPY, SciPY, and Pandas, which provide advanced computing capabilities.

Conclusion

We learned about the Python Embeddable Package, what are its advantages, and how to use it, in this blog post. It can be used for working with Plugins, Automation, Web Servers, Game Engines, and Scientific Applications. If you want to learn more about the Python Embeddable Package, be sure to check out the official Python documentation and explore the resources available online.

If you want to learn more about Python, Read more such articles from here.

FAQ’s

Can I distribute my application with a Python Embeddable Package?

Yes, you can distribute your application with a Python Embeddable Package. However, you must comply with the license terms of the Python software.

What is the difference between a Python installer and an embeddable package?

The embeddable package only includes the runtime dependencies while the Python installer provides the full Python developer kit.

Are There Any Common Pitfalls When Using Epoch Time in Python?

Explore potential pitfalls such as time zone considerations and the risk of an integer overflow when working with very large timestamps.

3. Using Python on Windows

This document aims to give an overview of Windows-specific behaviour you should know about when using Python on Microsoft Windows.

Unlike most Unix systems and services, Windows does not include a system supported installation of Python. To make Python available, the CPython team has compiled Windows installers (MSI packages) with every release for many years. These installers are primarily intended to add a per-user installation of Python, with the core interpreter and library being used by a single user. The installer is also able to install for all users of a single machine, and a separate ZIP file is available for application-local distributions.

As specified in PEP 11, a Python release only supports a Windows platform while Microsoft considers the platform under extended support. This means that Python supports Windows Vista and newer. If you require Windows XP support then please install Python 3.4.

There are a number of different installers available for Windows, each with certain benefits and downsides.

The full installer contains all components and is the best option for developers using Python for any kind of project.

The Microsoft Store package is a simple installation of Python that is suitable for running scripts and packages, and using IDLE or other development environments. It requires Windows 10, but can be safely installed without corrupting other programs. It also provides many convenient commands for launching Python and its tools.

The nuget.org packages are lightweight installations intended for continuous integration systems. It can be used to build Python packages or run scripts, but is not updateable and has no user interface tools.

The embeddable package is a minimal package of Python suitable for embedding into a larger application.

3.1. The full installer

3.1.1. Installation steps

Four Python installers are available for download — two each for the 32-bit and 64-bit versions of the interpreter. The web installer is a small initial download, and it will automatically download the required components as necessary. The offline installer includes the components necessary for a default installation and only requires an internet connection for optional features. See Installing Without Downloading for other ways to avoid downloading during installation.

After starting the installer, one of two options may be selected:

../_images/win_installer.png
If you select “Install Now”:

  • You will not need to be an administrator (unless a system update for the C Runtime Library is required or you install the Python Launcher for Windows for all users)
  • Python will be installed into your user directory
  • The Python Launcher for Windows will be installed according to the option at the bottom of the first page
  • The standard library, test suite, launcher and pip will be installed
  • If selected, the install directory will be added to your PATH
  • Shortcuts will only be visible for the current user

Selecting “Customize installation” will allow you to select the features to install, the installation location and other options or post-install actions. To install debugging symbols or binaries, you will need to use this option.

To perform an all-users installation, you should select “Customize installation”. In this case:

  • You may be required to provide administrative credentials or approval
  • Python will be installed into the Program Files directory
  • The Python Launcher for Windows will be installed into the Windows directory
  • Optional features may be selected during installation
  • The standard library can be pre-compiled to bytecode
  • If selected, the install directory will be added to the system PATH
  • Shortcuts are available for all users

3.1.2. Removing the MAX_PATH Limitation

Windows historically has limited path lengths to 260 characters. This meant that paths longer than this would not resolve and errors would result.

In the latest versions of Windows, this limitation can be expanded to approximately 32,000 characters. Your administrator will need to activate the “Enable Win32 long paths” group policy, or set LongPathsEnabled to 1 in the registry key HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\Control\FileSystem.

This allows the open() function, the os module and most other path functionality to accept and return paths longer than 260 characters.

After changing the above option, no further configuration is required.

Changed in version 3.6: Support for long paths was enabled in Python.

3.1.3. Installing Without UI

All of the options available in the installer UI can also be specified from the command line, allowing scripted installers to replicate an installation on many machines without user interaction. These options may also be set without suppressing the UI in order to change some of the defaults.

To completely hide the installer UI and install Python silently, pass the /quiet option. To skip past the user interaction but still display progress and errors, pass the /passive option. The /uninstall option may be passed to immediately begin removing Python — no prompt will be displayed.

All other options are passed as name=value, where the value is usually 0 to disable a feature, 1 to enable a feature, or a path. The full list of available options is shown below.

Name Description Default
InstallAllUsers Perform a system-wide installation. 0
TargetDir The installation directory Selected based on InstallAllUsers
DefaultAllUsersTargetDir The default installation directory for all-user installs %ProgramFiles%\Python X.Y or %ProgramFiles(x86)%\Python X.Y
DefaultJustForMeTargetDir The default install directory for just-for-me installs %LocalAppData%\Programs\PythonXY or %LocalAppData%\Programs\PythonXY-32 or %LocalAppData%\Programs\PythonXY-64
DefaultCustomTargetDir The default custom install directory displayed in the UI (empty)
AssociateFiles Create file associations if the launcher is also installed. 1
CompileAll Compile all .py files to .pyc. 0
PrependPath Add install and Scripts directories to PATH and .PY to PATHEXT 0
Shortcuts Create shortcuts for the interpreter, documentation and IDLE if installed. 1
Include_doc Install Python manual 1
Include_debug Install debug binaries 0
Include_dev Install developer headers and libraries 1
Include_exe Install python.exe and related files 1
Include_launcher Install Python Launcher for Windows. 1
InstallLauncherAllUsers Installs Python Launcher for Windows for all users. 1
Include_lib Install standard library and extension modules 1
Include_pip Install bundled pip and setuptools 1
Include_symbols Install debugging symbols (*.pdb) 0
Include_tcltk Install Tcl/Tk support and IDLE 1
Include_test Install standard library test suite 1
Include_tools Install utility scripts 1
LauncherOnly Only installs the launcher. This will override most other options. 0
SimpleInstall Disable most install UI 0
SimpleInstallDescription A custom message to display when the simplified install UI is used. (empty)

For example, to silently install a default, system-wide Python installation, you could use the following command (from an elevated command prompt):

python-3.8.0.exe /quiet InstallAllUsers=1 PrependPath=1 Include_test=0

To allow users to easily install a personal copy of Python without the test suite, you could provide a shortcut with the following command. This will display a simplified initial page and disallow customization:

python-3.8.0.exe InstallAllUsers=0 Include_launcher=0 Include_test=0
    SimpleInstall=1 SimpleInstallDescription="Just for me, no test suite."

(Note that omitting the launcher also omits file associations, and is only recommended for per-user installs when there is also a system-wide installation that included the launcher.)

The options listed above can also be provided in a file named unattend.xml alongside the executable. This file specifies a list of options and values. When a value is provided as an attribute, it will be converted to a number if possible. Values provided as element text are always left as strings. This example file sets the same options as the previous example:

<Options>
    <Option Name="InstallAllUsers" Value="no" />
    <Option Name="Include_launcher" Value="0" />
    <Option Name="Include_test" Value="no" />
    <Option Name="SimpleInstall" Value="yes" />
    <Option Name="SimpleInstallDescription">Just for me, no test suite</Option>
</Options>

3.1.4. Installing Without Downloading

As some features of Python are not included in the initial installer download, selecting those features may require an internet connection. To avoid this need, all possible components may be downloaded on-demand to create a complete layout that will no longer require an internet connection regardless of the selected features. Note that this download may be bigger than required, but where a large number of installations are going to be performed it is very useful to have a locally cached copy.

Execute the following command from Command Prompt to download all possible required files. Remember to substitute python-3.8.0.exe for the actual name of your installer, and to create layouts in their own directories to avoid collisions between files with the same name.

python-3.8.0.exe /layout [optional target directory]

You may also specify the /quiet option to hide the progress display.

3.1.5. Modifying an install

Once Python has been installed, you can add or remove features through the Programs and Features tool that is part of Windows. Select the Python entry and choose “Uninstall/Change” to open the installer in maintenance mode.

“Modify” allows you to add or remove features by modifying the checkboxes — unchanged checkboxes will not install or remove anything. Some options cannot be changed in this mode, such as the install directory; to modify these, you will need to remove and then reinstall Python completely.

“Repair” will verify all the files that should be installed using the current settings and replace any that have been removed or modified.

“Uninstall” will remove Python entirely, with the exception of the Python Launcher for Windows, which has its own entry in Programs and Features.

3.2. The Microsoft Store package

The Microsoft Store package is an easily installable Python interpreter that is intended mainly for interactive use, for example, by students.

To install the package, ensure you have the latest Windows 10 updates and search the Microsoft Store app for “Python “. Ensure that the app you select is published by the Python Software Foundation, and install it.

Warning

Python will always be available for free on the Microsoft Store. If you are asked to pay for it, you have not selected the correct package.

After installation, Python may be launched by finding it in Start. Alternatively, it will be available from any Command Prompt or PowerShell session by typing python. Further, pip and IDLE may be used by typing pip or idle. IDLE can also be found in Start.

All three commands are also available with version number suffixes, for example, as python3.exe and python3.x.exe as well as python.exe (where 3.x is the specific version you want to launch, such as ). Open “Manage App Execution Aliases” through Start to select which version of Python is associated with each command. It is recommended to make sure that pip and idle are consistent with whichever version of python is selected.

Virtual environments can be created with python -m venv and activated and used as normal.

If you have installed another version of Python and added it to your PATH variable, it will be available as python.exe rather than the one from the Microsoft Store. To access the new installation, use python3.exe or python3.x.exe.

The py.exe launcher will detect this Python installation, but will prefer installations from the traditional installer.

To remove Python, open Settings and use Apps and Features, or else find Python in Start and right-click to select Uninstall. Uninstalling will remove all packages you installed directly into this Python installation, but will not remove any virtual environments

3.2.1. Known Issues

Because of restrictions on Microsoft Store apps, Python scripts may not have full write access to shared locations such as TEMP and the registry. Instead, it will write to a private copy. If your scripts must modify the shared locations, you will need to install the full installer.

For more detail on the technical basis for these limitations, please consult Microsoft’s documentation on packaged full-trust apps, currently available at docs.microsoft.com/en-us/windows/msix/desktop/desktop-to-uwp-behind-the-scenes

3.3. The nuget.org packages

The nuget.org package is a reduced size Python environment intended for use on continuous integration and build systems that do not have a system-wide install of Python. While nuget is “the package manager for .NET”, it also works perfectly fine for packages containing build-time tools.

Visit nuget.org for the most up-to-date information on using nuget. What follows is a summary that is sufficient for Python developers.

The nuget.exe command line tool may be downloaded directly from https://aka.ms/nugetclidl, for example, using curl or PowerShell. With the tool, the latest version of Python for 64-bit or 32-bit machines is installed using:

nuget.exe install python -ExcludeVersion -OutputDirectory .
nuget.exe install pythonx86 -ExcludeVersion -OutputDirectory .

To select a particular version, add a -Version 3.x.y. The output directory may be changed from ., and the package will be installed into a subdirectory. By default, the subdirectory is named the same as the package, and without the -ExcludeVersion option this name will include the specific version installed. Inside the subdirectory is a tools directory that contains the Python installation:

# Without -ExcludeVersion
> .\python.3.5.2\tools\python.exe -V
Python 3.5.2

# With -ExcludeVersion
> .\python\tools\python.exe -V
Python 3.5.2

In general, nuget packages are not upgradeable, and newer versions should be installed side-by-side and referenced using the full path. Alternatively, delete the package directory manually and install it again. Many CI systems will do this automatically if they do not preserve files between builds.

Alongside the tools directory is a build\native directory. This contains a MSBuild properties file python.props that can be used in a C++ project to reference the Python install. Including the settings will automatically use the headers and import libraries in your build.

The package information pages on nuget.org are www.nuget.org/packages/python for the 64-bit version and www.nuget.org/packages/pythonx86 for the 32-bit version.

3.4. The embeddable package

The embedded distribution is a ZIP file containing a minimal Python environment. It is intended for acting as part of another application, rather than being directly accessed by end-users.

When extracted, the embedded distribution is (almost) fully isolated from the user’s system, including environment variables, system registry settings, and installed packages. The standard library is included as pre-compiled and optimized .pyc files in a ZIP, and python3.dll, python37.dll, python.exe and pythonw.exe are all provided. Tcl/tk (including all dependants, such as Idle), pip and the Python documentation are not included.

Note

The embedded distribution does not include the Microsoft C Runtime and it is the responsibility of the application installer to provide this. The runtime may have already been installed on a user’s system previously or automatically via Windows Update, and can be detected by finding ucrtbase.dll in the system directory.

Note

When running on Windows 7, Python 3.8 requires the KB2533623 update to be installed. The embeddable distribution does not detect this update, and may fail at runtime. Later versions of Windows include this update.

Third-party packages should be installed by the application installer alongside the embedded distribution. Using pip to manage dependencies as for a regular Python installation is not supported with this distribution, though with some care it may be possible to include and use pip for automatic updates. In general, third-party packages should be treated as part of the application (“vendoring”) so that the developer can ensure compatibility with newer versions before providing updates to users.

The two recommended use cases for this distribution are described below.

3.4.1. Python Application

An application written in Python does not necessarily require users to be aware of that fact. The embedded distribution may be used in this case to include a private version of Python in an install package. Depending on how transparent it should be (or conversely, how professional it should appear), there are two options.

Using a specialized executable as a launcher requires some coding, but provides the most transparent experience for users. With a customized launcher, there are no obvious indications that the program is running on Python: icons can be customized, company and version information can be specified, and file associations behave properly. In most cases, a custom launcher should simply be able to call Py_Main with a hard-coded command line.

The simpler approach is to provide a batch file or generated shortcut that directly calls the python.exe or pythonw.exe with the required command-line arguments. In this case, the application will appear to be Python and not its actual name, and users may have trouble distinguishing it from other running Python processes or file associations.

With the latter approach, packages should be installed as directories alongside the Python executable to ensure they are available on the path. With the specialized launcher, packages can be located in other locations as there is an opportunity to specify the search path before launching the application.

3.4.2. Embedding Python

Applications written in native code often require some form of scripting language, and the embedded Python distribution can be used for this purpose. In general, the majority of the application is in native code, and some part will either invoke python.exe or directly use python3.dll. For either case, extracting the embedded distribution to a subdirectory of the application installation is sufficient to provide a loadable Python interpreter.

As with the application use, packages can be installed to any location as there is an opportunity to specify search paths before initializing the interpreter. Otherwise, there is no fundamental differences between using the embedded distribution and a regular installation.

3.5. Alternative bundles

Besides the standard CPython distribution, there are modified packages including additional functionality. The following is a list of popular versions and their key features:

ActivePython
Installer with multi-platform compatibility, documentation, PyWin32
Anaconda
Popular scientific modules (such as numpy, scipy and pandas) and the conda package manager.
Canopy
A “comprehensive Python analysis environment” with editors and other development tools.
WinPython
Windows-specific distribution with prebuilt scientific packages and tools for building packages.

Note that these packages may not include the latest versions of Python or other libraries, and are not maintained or supported by the core Python team.

3.6. Configuring Python

To run Python conveniently from a command prompt, you might consider changing some default environment variables in Windows. While the installer provides an option to configure the PATH and PATHEXT variables for you, this is only reliable for a single, system-wide installation. If you regularly use multiple versions of Python, consider using the Python Launcher for Windows.

3.6.1. Excursus: Setting environment variables

Windows allows environment variables to be configured permanently at both the User level and the System level, or temporarily in a command prompt.

To temporarily set environment variables, open Command Prompt and use the set command:

C:\>set PATH=C:\Program Files\Python 3.8;%PATH%
C:\>set PYTHONPATH=%PYTHONPATH%;C:\My_python_lib
C:\>python

These changes will apply to any further commands executed in that console, and will be inherited by any applications started from the console.

Including the variable name within percent signs will expand to the existing value, allowing you to add your new value at either the start or the end. Modifying PATH by adding the directory containing python.exe to the start is a common way to ensure the correct version of Python is launched.

To permanently modify the default environment variables, click Start and search for ‘edit environment variables’, or open System properties, Advanced system settings and click the Environment Variables button. In this dialog, you can add or modify User and System variables. To change System variables, you need non-restricted access to your machine (i.e. Administrator rights).

Note

Windows will concatenate User variables after System variables, which may cause unexpected results when modifying PATH.

The PYTHONPATH variable is used by all versions of Python 2 and Python 3, so you should not permanently configure this variable unless it only includes code that is compatible with all of your installed Python versions.

3.6.2. Finding the Python executable

Besides using the automatically created start menu entry for the Python interpreter, you might want to start Python in the command prompt. The installer has an option to set that up for you.

On the first page of the installer, an option labelled “Add Python to PATH” may be selected to have the installer add the install location into the PATH. The location of the Scripts\ folder is also added. This allows you to type python to run the interpreter, and pip for the package installer. Thus, you can also execute your scripts with command line options, see Command line documentation.

If you don’t enable this option at install time, you can always re-run the installer, select Modify, and enable it. Alternatively, you can manually modify the PATH using the directions in Excursus: Setting environment variables. You need to set your PATH environment variable to include the directory of your Python installation, delimited by a semicolon from other entries. An example variable could look like this (assuming the first two entries already existed):

C:\WINDOWS\system32;C:\WINDOWS;C:\Program Files\Python 3.8

3.7. UTF-8 mode

Windows still uses legacy encodings for the system encoding (the ANSI Code Page). Python uses it for the default encoding of text files (e.g. locale.getpreferredencoding()).

This may cause issues because UTF-8 is widely used on the internet and most Unix systems, including WSL (Windows Subsystem for Linux).

You can use UTF-8 mode to change the default text encoding to UTF-8. You can enable UTF-8 mode via the -X utf8 command line option, or the PYTHONUTF8=1 environment variable. See PYTHONUTF8 for enabling UTF-8 mode, and Excursus: Setting environment variables for how to modify environment variables.

When UTF-8 mode is enabled:

  • locale.getpreferredencoding() returns 'UTF-8' instead of the system encoding. This function is used for the default text encoding in many places, including open(), Popen, Path.read_text(), etc.
  • sys.stdin, sys.stdout, and sys.stderr all use UTF-8 as their text encoding.
  • You can still use the system encoding via the “mbcs” codec.

Note that adding PYTHONUTF8=1 to the default environment variables will affect all Python 3.7+ applications on your system. If you have any Python 3.7+ applications which rely on the legacy system encoding, it is recommended to set the environment variable temporarily or use the -X utf8 command line option.

Note

Even when UTF-8 mode is disabled, Python uses UTF-8 by default on Windows for:

  • Console I/O including standard I/O (see PEP 528 for details).
  • The filesystem encoding (see PEP 529 for details).

3.8. Python Launcher for Windows

The Python launcher for Windows is a utility which aids in locating and executing of different Python versions. It allows scripts (or the command-line) to indicate a preference for a specific Python version, and will locate and execute that version.

Unlike the PATH variable, the launcher will correctly select the most appropriate version of Python. It will prefer per-user installations over system-wide ones, and orders by language version rather than using the most recently installed version.

The launcher was originally specified in PEP 397.

3.8.1. Getting started

3.8.1.1. From the command-line

System-wide installations of Python 3.3 and later will put the launcher on your PATH. The launcher is compatible with all available versions of Python, so it does not matter which version is installed. To check that the launcher is available, execute the following command in Command Prompt:

You should find that the latest version of Python you have installed is started — it can be exited as normal, and any additional command-line arguments specified will be sent directly to Python.

If you have multiple versions of Python installed (e.g., 2.7 and ) you will have noticed that Python was started — to launch Python 2.7, try the command:

If you want the latest version of Python 2.x you have installed, try the command:

You should find the latest version of Python 2.x starts.

If you see the following error, you do not have the launcher installed:

'py' is not recognized as an internal or external command,
operable program or batch file.

Per-user installations of Python do not add the launcher to PATH unless the option was selected on installation.

3.8.1.2. Virtual environments

If the launcher is run with no explicit Python version specification, and a virtual environment (created with the standard library venv module or the external virtualenv tool) active, the launcher will run the virtual environment’s interpreter rather than the global one. To run the global interpreter, either deactivate the virtual environment, or explicitly specify the global Python version.

3.8.1.3. From a script

Let’s create a test Python script — create a file called hello.py with the following contents

#! python
import sys
sys.stdout.write("hello from Python %s\n" % (sys.version,))

From the directory in which hello.py lives, execute the command:

You should notice the version number of your latest Python 2.x installation is printed. Now try changing the first line to be:

Re-executing the command should now print the latest Python 3.x information. As with the above command-line examples, you can specify a more explicit version qualifier. Assuming you have Python 2.6 installed, try changing the first line to #! python2.6 and you should find the 2.6 version information printed.

Note that unlike interactive use, a bare “python” will use the latest version of Python 2.x that you have installed. This is for backward compatibility and for compatibility with Unix, where the command python typically refers to Python 2.

3.8.1.4. From file associations

The launcher should have been associated with Python files (i.e. .py, .pyw, .pyc files) when it was installed. This means that when you double-click on one of these files from Windows explorer the launcher will be used, and therefore you can use the same facilities described above to have the script specify the version which should be used.

The key benefit of this is that a single launcher can support multiple Python versions at the same time depending on the contents of the first line.

3.8.2. Shebang Lines

If the first line of a script file starts with #!, it is known as a “shebang” line. Linux and other Unix like operating systems have native support for such lines and they are commonly used on such systems to indicate how a script should be executed. This launcher allows the same facilities to be used with Python scripts on Windows and the examples above demonstrate their use.

To allow shebang lines in Python scripts to be portable between Unix and Windows, this launcher supports a number of ‘virtual’ commands to specify which interpreter to use. The supported virtual commands are:

  • /usr/bin/env python
  • /usr/bin/python
  • /usr/local/bin/python
  • python

For example, if the first line of your script starts with

The default Python will be located and used. As many Python scripts written to work on Unix will already have this line, you should find these scripts can be used by the launcher without modification. If you are writing a new script on Windows which you hope will be useful on Unix, you should use one of the shebang lines starting with /usr.

Any of the above virtual commands can be suffixed with an explicit version (either just the major version, or the major and minor version). Furthermore the 32-bit version can be requested by adding “-32” after the minor version. I.e. /usr/bin/python2.7-32 will request usage of the 32-bit python 2.7.

New in version 3.7: Beginning with python launcher 3.7 it is possible to request 64-bit version by the “-64” suffix. Furthermore it is possible to specify a major and architecture without minor (i.e. /usr/bin/python3-64).

The /usr/bin/env form of shebang line has one further special property. Before looking for installed Python interpreters, this form will search the executable PATH for a Python executable. This corresponds to the behaviour of the Unix env program, which performs a PATH search.

3.8.3. Arguments in shebang lines

The shebang lines can also specify additional options to be passed to the Python interpreter. For example, if you have a shebang line:

Then Python will be started with the -v option

3.8.4. Customization

3.8.4.1. Customization via INI files

Two .ini files will be searched by the launcher — py.ini in the current user’s “application data” directory (i.e. the directory returned by calling the Windows function SHGetFolderPath with CSIDL_LOCAL_APPDATA) and py.ini in the same directory as the launcher. The same .ini files are used for both the ‘console’ version of the launcher (i.e. py.exe) and for the ‘windows’ version (i.e. pyw.exe).

Customization specified in the “application directory” will have precedence over the one next to the executable, so a user, who may not have write access to the .ini file next to the launcher, can override commands in that global .ini file.

3.8.4.2. Customizing default Python versions

In some cases, a version qualifier can be included in a command to dictate which version of Python will be used by the command. A version qualifier starts with a major version number and can optionally be followed by a period (‘.’) and a minor version specifier. Furthermore it is possible to specify if a 32 or 64 bit implementation shall be requested by adding “-32” or “-64”.

For example, a shebang line of #!python has no version qualifier, while #!python3 has a version qualifier which specifies only a major version.

If no version qualifiers are found in a command, the environment variable PY_PYTHON can be set to specify the default version qualifier. If it is not set, the default is “3”. The variable can specify any value that may be passed on the command line, such as “3”, “3.7”, “3.7-32” or “3.7-64”. (Note that the “-64” option is only available with the launcher included with Python 3.7 or newer.)

If no minor version qualifiers are found, the environment variable PY_PYTHON{major} (where {major} is the current major version qualifier as determined above) can be set to specify the full version. If no such option is found, the launcher will enumerate the installed Python versions and use the latest minor release found for the major version, which is likely, although not guaranteed, to be the most recently installed version in that family.

On 64-bit Windows with both 32-bit and 64-bit implementations of the same (major.minor) Python version installed, the 64-bit version will always be preferred. This will be true for both 32-bit and 64-bit implementations of the launcher — a 32-bit launcher will prefer to execute a 64-bit Python installation of the specified version if available. This is so the behavior of the launcher can be predicted knowing only what versions are installed on the PC and without regard to the order in which they were installed (i.e., without knowing whether a 32 or 64-bit version of Python and corresponding launcher was installed last). As noted above, an optional “-32” or “-64” suffix can be used on a version specifier to change this behaviour.

Examples:

  • If no relevant options are set, the commands python and python2 will use the latest Python 2.x version installed and the command python3 will use the latest Python 3.x installed.
  • The commands python3.1 and python2.7 will not consult any options at all as the versions are fully specified.
  • If PY_PYTHON=3, the commands python and python3 will both use the latest installed Python 3 version.
  • If PY_PYTHON=3.1-32, the command python will use the 32-bit implementation of 3.1 whereas the command python3 will use the latest installed Python (PY_PYTHON was not considered at all as a major version was specified.)
  • If PY_PYTHON=3 and PY_PYTHON3=3.1, the commands python and python3 will both use specifically 3.1

In addition to environment variables, the same settings can be configured in the .INI file used by the launcher. The section in the INI file is called [defaults] and the key name will be the same as the environment variables without the leading PY_ prefix (and note that the key names in the INI file are case insensitive.) The contents of an environment variable will override things specified in the INI file.

For example:

  • Setting PY_PYTHON=3.1 is equivalent to the INI file containing:
  • Setting PY_PYTHON=3 and PY_PYTHON3=3.1 is equivalent to the INI file containing:
[defaults]
python=3
python3=3.1

3.8.5. Diagnostics

If an environment variable PYLAUNCH_DEBUG is set (to any value), the launcher will print diagnostic information to stderr (i.e. to the console). While this information manages to be simultaneously verbose and terse, it should allow you to see what versions of Python were located, why a particular version was chosen and the exact command-line used to execute the target Python.

3.9. Finding modules

Python usually stores its library (and thereby your site-packages folder) in the installation directory. So, if you had installed Python to C:\Python\, the default library would reside in C:\Python\Lib\ and third-party modules should be stored in C:\Python\Lib\site-packages\.

To completely override sys.path, create a ._pth file with the same name as the DLL (python37._pth) or the executable (python._pth) and specify one line for each path to add to sys.path. The file based on the DLL name overrides the one based on the executable, which allows paths to be restricted for any program loading the runtime if desired.

When the file exists, all registry and environment variables are ignored, isolated mode is enabled, and site is not imported unless one line in the file specifies import site. Blank paths and lines starting with # are ignored. Each path may be absolute or relative to the location of the file. Import statements other than to site are not permitted, and arbitrary code cannot be specified.

Note that .pth files (without leading underscore) will be processed normally by the site module when import site has been specified.

When no ._pth file is found, this is how sys.path is populated on Windows:

  • An empty entry is added at the start, which corresponds to the current directory.
  • If the environment variable PYTHONPATH exists, as described in Environment variables, its entries are added next. Note that on Windows, paths in this variable must be separated by semicolons, to distinguish them from the colon used in drive identifiers (C:\ etc.).
  • Additional “application paths” can be added in the registry as subkeys of \SOFTWARE\Python\PythonCore{version}\PythonPath under both the HKEY_CURRENT_USER and HKEY_LOCAL_MACHINE hives. Subkeys which have semicolon-delimited path strings as their default value will cause each path to be added to sys.path. (Note that all known installers only use HKLM, so HKCU is typically empty.)
  • If the environment variable PYTHONHOME is set, it is assumed as “Python Home”. Otherwise, the path of the main Python executable is used to locate a “landmark file” (either Lib\os.py or pythonXY.zip) to deduce the “Python Home”. If a Python home is found, the relevant sub-directories added to sys.path (Lib, plat-win, etc) are based on that folder. Otherwise, the core Python path is constructed from the PythonPath stored in the registry.
  • If the Python Home cannot be located, no PYTHONPATH is specified in the environment, and no registry entries can be found, a default path with relative entries is used (e.g. .\Lib;.\plat-win, etc).

If a pyvenv.cfg file is found alongside the main executable or in the directory one level above the executable, the following variations apply:

  • If home is an absolute path and PYTHONHOME is not set, this path is used instead of the path to the main executable when deducing the home location.

The end result of all this is:

  • When running python.exe, or any other .exe in the main Python directory (either an installed version, or directly from the PCbuild directory), the core path is deduced, and the core paths in the registry are ignored. Other “application paths” in the registry are always read.
  • When Python is hosted in another .exe (different directory, embedded via COM, etc), the “Python Home” will not be deduced, so the core path from the registry is used. Other “application paths” in the registry are always read.
  • If Python can’t find its home and there are no registry value (frozen .exe, some very strange installation setup) you get a path with some default, but relative, paths.

For those who want to bundle Python into their application or distribution, the following advice will prevent conflicts with other installations:

  • Include a ._pth file alongside your executable containing the directories to include. This will ignore paths listed in the registry and environment variables, and also ignore site unless import site is listed.
  • If you are loading python3.dll or python37.dll in your own executable, explicitly call Py_SetPath() or (at least) Py_SetProgramName() before Py_Initialize().
  • Clear and/or overwrite PYTHONPATH and set PYTHONHOME before launching python.exe from your application.
  • If you cannot use the previous suggestions (for example, you are a distribution that allows people to run python.exe directly), ensure that the landmark file (Lib\os.py) exists in your install directory. (Note that it will not be detected inside a ZIP file, but a correctly named ZIP file will be detected instead.)

These will ensure that the files in a system-wide installation will not take precedence over the copy of the standard library bundled with your application. Otherwise, your users may experience problems using your application. Note that the first suggestion is the best, as the others may still be susceptible to non-standard paths in the registry and user site-packages.

Changed in version 3.6:

  • Adds ._pth file support and removes applocal option from pyvenv.cfg.
  • Adds pythonXX.zip as a potential landmark when directly adjacent to the executable.

Deprecated since version 3.6:

3.10. Additional modules

Even though Python aims to be portable among all platforms, there are features that are unique to Windows. A couple of modules, both in the standard library and external, and snippets exist to use these features.

The Windows-specific standard modules are documented in MS Windows Specific Services.

3.10.1. PyWin32

The PyWin32 module by Mark Hammond is a collection of modules for advanced Windows-specific support. This includes utilities for:

  • Component Object Model (COM)
  • Win32 API calls
  • Registry
  • Event log
  • Microsoft Foundation Classes (MFC) user interfaces

PythonWin is a sample MFC application shipped with PyWin32. It is an embeddable IDE with a built-in debugger.

3.10.2. cx_Freeze

cx_Freeze is a distutils extension (see Extending Distutils) which wraps Python scripts into executable Windows programs (*.exe files). When you have done this, you can distribute your application without requiring your users to install Python.

3.10.3. WConio

Since Python’s advanced terminal handling layer, curses, is restricted to Unix-like systems, there is a library exclusive to Windows as well: Windows Console I/O for Python.

WConio is a wrapper for Turbo-C’s CONIO.H, used to create text user interfaces.

3.11. Compiling Python on Windows

If you want to compile CPython yourself, first thing you should do is get the source. You can download either the latest release’s source or just grab a fresh checkout.

The source tree contains a build solution and project files for Microsoft Visual Studio 2015, which is the compiler used to build the official Python releases. These files are in the PCbuild directory.

Check PCbuild/readme.txt for general information on the build process.

For extension modules, consult Building C and C++ Extensions on Windows.

See also

Python + Windows + distutils + SWIG + gcc MinGW
or “Creating Python extensions in C/C++ with SWIG and compiling them with MinGW gcc under Windows” or “Installing Python extension with distutils and without Microsoft Visual C++” by Sébastien Sauvage, 2003

3.12. Other Platforms

With ongoing development of Python, some platforms that used to be supported earlier are no longer supported (due to the lack of users or developers). Check PEP 11 for details on all unsupported platforms.

  • Windows CE is still supported.
  • The Cygwin installer offers to install the Python interpreter as well (cf. Cygwin package source, Maintainer releases)

See Python for Windows for detailed information about platforms with pre-compiled installers.

This document aims to give an overview of Windows-specific behaviour you should know about when using Python on Microsoft Windows.

Unlike most Unix systems and services, Windows does not include a system supported installation of Python. To make Python available, the CPython team has compiled Windows installers (MSI packages) with every release for many years. These installers are primarily intended to add a per-user installation of Python, with the core interpreter and library being used by a single user. The installer is also able to install for all users of a single machine, and a separate ZIP file is available for application-local distributions.

As specified in PEP 11 , a Python release only supports a Windows platform while Microsoft considers the platform under extended support. This means that Python 3.8 supports Windows Vista and newer. If you require Windows XP support then please install Python 3.4.

There are a number of different installers available for Windows, each with certain benefits and downsides.

The full installer contains all components and is the best option for developers using Python for any kind of project.

The Microsoft Store package is a simple installation of Python that is suitable for running scripts and packages, and using IDLE or other development environments. It requires Windows 10, but can be safely installed without corrupting other programs. It also provides many convenient commands for launching Python and its tools.

The nuget.org packages are lightweight installations intended for continuous integration systems. It can be used to build Python packages or run scripts, but is not updateable and has no user interface tools.

The embeddable package is a minimal package of Python suitable for embedding into a larger application.

3.1. The full installer

3.1.1. Installation steps

Four Python 3.8 installers are available for download — two each for the 32-bit and 64-bit versions of the interpreter. The web installer is a small initial download, and it will automatically download the required components as necessary. The offline installer includes the components necessary for a default installation and only requires an internet connection for optional features. See Installing Without Downloading for other ways to avoid downloading during installation.

After starting the installer, one of two options may be selected:

../_images/win_installer.png

If you select “Install Now”:

  • You will not need to be an administrator (unless a system update for the C Runtime Library is required or you install the Python Launcher for Windows for all users)

  • Python will be installed into your user directory

  • The Python Launcher for Windows will be installed according to the option at the bottom of the first page

  • The standard library, test suite, launcher and pip will be installed

  • If selected, the install directory will be added to your PATH

  • Shortcuts will only be visible for the current user

Selecting “Customize installation” will allow you to select the features to install, the installation location and other options or post-install actions. To install debugging symbols or binaries, you will need to use this option.

To perform an all-users installation, you should select “Customize installation”. In this case:

  • You may be required to provide administrative credentials or approval

  • Python will be installed into the Program Files directory

  • The Python Launcher for Windows will be installed into the Windows directory

  • Optional features may be selected during installation

  • The standard library can be pre-compiled to bytecode

  • If selected, the install directory will be added to the system PATH

  • Shortcuts are available for all users

3.1.2. Removing the MAX_PATH Limitation

Windows historically has limited path lengths to 260 characters. This meant that paths longer than this would not resolve and errors would result.

In the latest versions of Windows, this limitation can be expanded to approximately 32,000 characters. Your administrator will need to activate the “Enable Win32 long paths” group policy, or set the registry value HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\Control\FileSystem@LongPathsEnabled to 1.

This allows the open() function, the os module and most other path functionality to accept and return paths longer than 260 characters.

After changing the above option, no further configuration is required.

Changed in version 3.6: Support for long paths was enabled in Python.

3.1.3. Installing Without UI

All of the options available in the installer UI can also be specified from the command line, allowing scripted installers to replicate an installation on many machines without user interaction. These options may also be set without suppressing the UI in order to change some of the defaults.

To completely hide the installer UI and install Python silently, pass the /quiet option. To skip past the user interaction but still display progress and errors, pass the /passive option. The /uninstall option may be passed to immediately begin removing Python — no prompt will be displayed.

All other options are passed as name=value, where the value is usually 0 to disable a feature, 1 to enable a feature, or a path. The full list of available options is shown below.

Name

Description

Default

InstallAllUsers

Perform a system-wide installation.

0

TargetDir

The installation directory

Selected based on InstallAllUsers

DefaultAllUsersTargetDir

The default installation directory for all-user installs

%ProgramFiles%\Python X.Y or %ProgramFiles(x86)%\Python X.Y

DefaultJustForMeTargetDir

The default install directory for just-for-me installs

%LocalAppData%\Programs\PythonXY or %LocalAppData%\Programs\PythonXY-32 or %LocalAppData%\Programs\PythonXY-64

DefaultCustomTargetDir

The default custom install directory displayed in the UI

(empty)

AssociateFiles

Create file associations if the launcher is also installed.

1

CompileAll

Compile all .py files to .pyc.

0

PrependPath

Add install and Scripts directories to PATH and .PY to PATHEXT

0

Shortcuts

Create shortcuts for the interpreter, documentation and IDLE if installed.

1

Include_doc

Install Python manual

1

Include_debug

Install debug binaries

0

Include_dev

Install developer headers and libraries

1

Include_exe

Install python.exe and related files

1

Include_launcher

Install Python Launcher for Windows.

1

InstallLauncherAllUsers

Installs Python Launcher for Windows for all users.

1

Include_lib

Install standard library and extension modules

1

Include_pip

Install bundled pip and setuptools

1

Include_symbols

Install debugging symbols (*.pdb)

0

Include_tcltk

Install Tcl/Tk support and IDLE

1

Include_test

Install standard library test suite

1

Include_tools

Install utility scripts

1

LauncherOnly

Only installs the launcher. This will override most other options.

0

SimpleInstall

Disable most install UI

0

SimpleInstallDescription

A custom message to display when the simplified install UI is used.

(empty)

For example, to silently install a default, system-wide Python installation, you could use the following command (from an elevated command prompt):

python-3.8.0.exe /quiet InstallAllUsers=1 PrependPath=1 Include_test=0

To allow users to easily install a personal copy of Python without the test suite, you could provide a shortcut with the following command. This will display a simplified initial page and disallow customization:

python-3.8.0.exe InstallAllUsers=0 Include_launcher=0 Include_test=0
    SimpleInstall=1 SimpleInstallDescription="Just for me, no test suite."

(Note that omitting the launcher also omits file associations, and is only recommended for per-user installs when there is also a system-wide installation that included the launcher.)

The options listed above can also be provided in a file named unattend.xml alongside the executable. This file specifies a list of options and values. When a value is provided as an attribute, it will be converted to a number if possible. Values provided as element text are always left as strings. This example file sets the same options as the previous example:

<Options>
    <Option Name="InstallAllUsers" Value="no" />
    <Option Name="Include_launcher" Value="0" />
    <Option Name="Include_test" Value="no" />
    <Option Name="SimpleInstall" Value="yes" />
    <Option Name="SimpleInstallDescription">Just for me, no test suite</Option>
</Options>

3.1.4. Installing Without Downloading

As some features of Python are not included in the initial installer download, selecting those features may require an internet connection. To avoid this need, all possible components may be downloaded on-demand to create a complete layout that will no longer require an internet connection regardless of the selected features. Note that this download may be bigger than required, but where a large number of installations are going to be performed it is very useful to have a locally cached copy.

Execute the following command from Command Prompt to download all possible required files. Remember to substitute python-3.8.0.exe for the actual name of your installer, and to create layouts in their own directories to avoid collisions between files with the same name.

python-3.8.0.exe /layout [optional target directory]

You may also specify the /quiet option to hide the progress display.

3.1.5. Modifying an install

Once Python has been installed, you can add or remove features through the Programs and Features tool that is part of Windows. Select the Python entry and choose “Uninstall/Change” to open the installer in maintenance mode.

“Modify” allows you to add or remove features by modifying the checkboxes — unchanged checkboxes will not install or remove anything. Some options cannot be changed in this mode, such as the install directory; to modify these, you will need to remove and then reinstall Python completely.

“Repair” will verify all the files that should be installed using the current settings and replace any that have been removed or modified.

“Uninstall” will remove Python entirely, with the exception of the Python Launcher for Windows, which has its own entry in Programs and Features.

3.2. The Microsoft Store package

New in version 3.7.2.

The Microsoft Store package is an easily installable Python interpreter that is intended mainly for interactive use, for example, by students.

To install the package, ensure you have the latest Windows 10 updates and search the Microsoft Store app for “Python 3.8”. Ensure that the app you select is published by the Python Software Foundation, and install it.

Warning

Python will always be available for free on the Microsoft Store. If you are asked to pay for it, you have not selected the correct package.

After installation, Python may be launched by finding it in Start. Alternatively, it will be available from any Command Prompt or PowerShell session by typing python. Further, pip and IDLE may be used by typing pip or idle. IDLE can also be found in Start.

All three commands are also available with version number suffixes, for example, as python3.exe and python3.x.exe as well as python.exe (where 3.x is the specific version you want to launch, such as 3.8). Open “Manage App Execution Aliases” through Start to select which version of Python is associated with each command. It is recommended to make sure that pip and idle are consistent with whichever version of python is selected.

Virtual environments can be created with python -m venv and activated and used as normal.

If you have installed another version of Python and added it to your PATH variable, it will be available as python.exe rather than the one from the Microsoft Store. To access the new installation, use python3.exe or python3.x.exe.

The py.exe launcher will detect this Python installation, but will prefer installations from the traditional installer.

To remove Python, open Settings and use Apps and Features, or else find Python in Start and right-click to select Uninstall. Uninstalling will remove all packages you installed directly into this Python installation, but will not remove any virtual environments

3.2.1. Known Issues

Because of restrictions on Microsoft Store apps, Python scripts may not have full write access to shared locations such as TEMP and the registry. Instead, it will write to a private copy. If your scripts must modify the shared locations, you will need to install the full installer.

3.3. The nuget.org packages

New in version 3.5.2.

The nuget.org package is a reduced size Python environment intended for use on continuous integration and build systems that do not have a system-wide install of Python. While nuget is “the package manager for .NET”, it also works perfectly fine for packages containing build-time tools.

Visit nuget.org for the most up-to-date information on using nuget. What follows is a summary that is sufficient for Python developers.

The nuget.exe command line tool may be downloaded directly from https://aka.ms/nugetclidl, for example, using curl or PowerShell. With the tool, the latest version of Python for 64-bit or 32-bit machines is installed using:

nuget.exe install python -ExcludeVersion -OutputDirectory .
nuget.exe install pythonx86 -ExcludeVersion -OutputDirectory .

To select a particular version, add a -Version 3.x.y. The output directory may be changed from ., and the package will be installed into a subdirectory. By default, the subdirectory is named the same as the package, and without the -ExcludeVersion option this name will include the specific version installed. Inside the subdirectory is a tools directory that contains the Python installation:

# Without -ExcludeVersion
> .\python.3.5.2\tools\python.exe -V
Python 3.5.2

# With -ExcludeVersion
> .\python\tools\python.exe -V
Python 3.5.2

In general, nuget packages are not upgradeable, and newer versions should be installed side-by-side and referenced using the full path. Alternatively, delete the package directory manually and install it again. Many CI systems will do this automatically if they do not preserve files between builds.

Alongside the tools directory is a build\native directory. This contains a MSBuild properties file python.props that can be used in a C++ project to reference the Python install. Including the settings will automatically use the headers and import libraries in your build.

The package information pages on nuget.org are www.nuget.org/packages/python for the 64-bit version and www.nuget.org/packages/pythonx86 for the 32-bit version.

3.4. The embeddable package

New in version 3.5.

The embedded distribution is a ZIP file containing a minimal Python environment. It is intended for acting as part of another application, rather than being directly accessed by end-users.

When extracted, the embedded distribution is (almost) fully isolated from the user’s system, including environment variables, system registry settings, and installed packages. The standard library is included as pre-compiled and optimized .pyc files in a ZIP, and python3.dll, python37.dll, python.exe and pythonw.exe are all provided. Tcl/tk (including all dependants, such as Idle), pip and the Python documentation are not included.

Note

The embedded distribution does not include the Microsoft C Runtime and it is the responsibility of the application installer to provide this. The runtime may have already been installed on a user’s system previously or automatically via Windows Update, and can be detected by finding ucrtbase.dll in the system directory.

Third-party packages should be installed by the application installer alongside the embedded distribution. Using pip to manage dependencies as for a regular Python installation is not supported with this distribution, though with some care it may be possible to include and use pip for automatic updates. In general, third-party packages should be treated as part of the application (“vendoring”) so that the developer can ensure compatibility with newer versions before providing updates to users.

The two recommended use cases for this distribution are described below.

3.4.1. Python Application

An application written in Python does not necessarily require users to be aware of that fact. The embedded distribution may be used in this case to include a private version of Python in an install package. Depending on how transparent it should be (or conversely, how professional it should appear), there are two options.

Using a specialized executable as a launcher requires some coding, but provides the most transparent experience for users. With a customized launcher, there are no obvious indications that the program is running on Python: icons can be customized, company and version information can be specified, and file associations behave properly. In most cases, a custom launcher should simply be able to call Py_Main with a hard-coded command line.

The simpler approach is to provide a batch file or generated shortcut that directly calls the python.exe or pythonw.exe with the required command-line arguments. In this case, the application will appear to be Python and not its actual name, and users may have trouble distinguishing it from other running Python processes or file associations.

With the latter approach, packages should be installed as directories alongside the Python executable to ensure they are available on the path. With the specialized launcher, packages can be located in other locations as there is an opportunity to specify the search path before launching the application.

3.4.2. Embedding Python

Applications written in native code often require some form of scripting language, and the embedded Python distribution can be used for this purpose. In general, the majority of the application is in native code, and some part will either invoke python.exe or directly use python3.dll. For either case, extracting the embedded distribution to a subdirectory of the application installation is sufficient to provide a loadable Python interpreter.

As with the application use, packages can be installed to any location as there is an opportunity to specify search paths before initializing the interpreter. Otherwise, there is no fundamental differences between using the embedded distribution and a regular installation.

3.5. Alternative bundles

Besides the standard CPython distribution, there are modified packages including additional functionality. The following is a list of popular versions and their key features:

ActivePython

Installer with multi-platform compatibility, documentation, PyWin32

Anaconda

Popular scientific modules (such as numpy, scipy and pandas) and the conda package manager.

Canopy

A “comprehensive Python analysis environment” with editors and other development tools.

WinPython

Windows-specific distribution with prebuilt scientific packages and tools for building packages.

Note that these packages may not include the latest versions of Python or other libraries, and are not maintained or supported by the core Python team.

3.6. Configuring Python

To run Python conveniently from a command prompt, you might consider changing some default environment variables in Windows. While the installer provides an option to configure the PATH and PATHEXT variables for you, this is only reliable for a single, system-wide installation. If you regularly use multiple versions of Python, consider using the Python Launcher for Windows.

3.6.1. Excursus: Setting environment variables

Windows allows environment variables to be configured permanently at both the User level and the System level, or temporarily in a command prompt.

To temporarily set environment variables, open Command Prompt and use the set command:

C:\>set PATH=C:\Program Files\Python 3.8;%PATH%
C:\>set PYTHONPATH=%PYTHONPATH%;C:\My_python_lib
C:\>python

These changes will apply to any further commands executed in that console, and will be inherited by any applications started from the console.

Including the variable name within percent signs will expand to the existing value, allowing you to add your new value at either the start or the end. Modifying PATH by adding the directory containing python.exe to the start is a common way to ensure the correct version of Python is launched.

To permanently modify the default environment variables, click Start and search for ‘edit environment variables’, or open System properties, Advanced system settings and click the Environment Variables button. In this dialog, you can add or modify User and System variables. To change System variables, you need non-restricted access to your machine (i.e. Administrator rights).

Note

Windows will concatenate User variables after System variables, which may cause unexpected results when modifying PATH.

The PYTHONPATH variable is used by all versions of Python 2 and Python 3, so you should not permanently configure this variable unless it only includes code that is compatible with all of your installed Python versions.

3.6.2. Finding the Python executable

Changed in version 3.5.

Besides using the automatically created start menu entry for the Python interpreter, you might want to start Python in the command prompt. The installer has an option to set that up for you.

On the first page of the installer, an option labelled “Add Python to PATH” may be selected to have the installer add the install location into the PATH. The location of the Scripts\ folder is also added. This allows you to type python to run the interpreter, and pip for the package installer. Thus, you can also execute your scripts with command line options, see Command line documentation.

If you don’t enable this option at install time, you can always re-run the installer, select Modify, and enable it. Alternatively, you can manually modify the PATH using the directions in Excursus: Setting environment variables. You need to set your PATH environment variable to include the directory of your Python installation, delimited by a semicolon from other entries. An example variable could look like this (assuming the first two entries already existed):

C:\WINDOWS\system32;C:\WINDOWS;C:\Program Files\Python 3.8

3.7. UTF-8 mode

New in version 3.7.

Windows still uses legacy encodings for the system encoding (the ANSI Code Page). Python uses it for the default encoding of text files (e.g. locale.getpreferredencoding()).

This may cause issues because UTF-8 is widely used on the internet and most Unix systems, including WSL (Windows Subsystem for Linux).

You can use UTF-8 mode to change the default text encoding to UTF-8. You can enable UTF-8 mode via the -X utf8 command line option, or the PYTHONUTF8=1 environment variable. See PYTHONUTF8 for enabling UTF-8 mode, and Excursus: Setting environment variables for how to modify environment variables.

When UTF-8 mode is enabled:

  • locale.getpreferredencoding() returns 'UTF-8' instead of the system encoding. This function is used for the default text encoding in many places, including open(), Popen, Path.read_text(), etc.

  • sys.stdin, sys.stdout, and sys.stderr all use UTF-8 as their text encoding.

  • You can still use the system encoding via the “mbcs” codec.

Note that adding PYTHONUTF8=1 to the default environment variables will affect all Python 3.7+ applications on your system. If you have any Python 3.7+ applications which rely on the legacy system encoding, it is recommended to set the environment variable temporarily or use the -X utf8 command line option.

Note

Even when UTF-8 mode is disabled, Python uses UTF-8 by default on Windows for:

  • Console I/O including standard I/O (see PEP 528 for details).

  • The filesystem encoding (see PEP 529 for details).

3.8. Python Launcher for Windows

New in version 3.3.

The Python launcher for Windows is a utility which aids in locating and executing of different Python versions. It allows scripts (or the command-line) to indicate a preference for a specific Python version, and will locate and execute that version.

Unlike the PATH variable, the launcher will correctly select the most appropriate version of Python. It will prefer per-user installations over system-wide ones, and orders by language version rather than using the most recently installed version.

The launcher was originally specified in PEP 397 .

3.8.1. Getting started

3.8.1.1. From the command-line

Changed in version 3.6.

System-wide installations of Python 3.3 and later will put the launcher on your PATH. The launcher is compatible with all available versions of Python, so it does not matter which version is installed. To check that the launcher is available, execute the following command in Command Prompt:

You should find that the latest version of Python you have installed is started — it can be exited as normal, and any additional command-line arguments specified will be sent directly to Python.

If you have multiple versions of Python installed (e.g., 2.7 and 3.8) you will have noticed that Python 3.8 was started — to launch Python 2.7, try the command:

If you want the latest version of Python 2.x you have installed, try the command:

You should find the latest version of Python 2.x starts.

If you see the following error, you do not have the launcher installed:

'py' is not recognized as an internal or external command,
operable program or batch file.

Per-user installations of Python do not add the launcher to PATH unless the option was selected on installation.

3.8.1.2. Virtual environments

New in version 3.5.

If the launcher is run with no explicit Python version specification, and a virtual environment (created with the standard library venv module or the external virtualenv tool) active, the launcher will run the virtual environment’s interpreter rather than the global one. To run the global interpreter, either deactivate the virtual environment, or explicitly specify the global Python version.

3.8.1.3. From a script

Let’s create a test Python script — create a file called hello.py with the following contents

#! python
import sys
sys.stdout.write("hello from Python %s\n" % (sys.version,))

From the directory in which hello.py lives, execute the command:

You should notice the version number of your latest Python 2.x installation is printed. Now try changing the first line to be:

Re-executing the command should now print the latest Python 3.x information. As with the above command-line examples, you can specify a more explicit version qualifier. Assuming you have Python 2.6 installed, try changing the first line to #! python2.6 and you should find the 2.6 version information printed.

Note that unlike interactive use, a bare “python” will use the latest version of Python 2.x that you have installed. This is for backward compatibility and for compatibility with Unix, where the command python typically refers to Python 2.

3.8.1.4. From file associations

The launcher should have been associated with Python files (i.e. .py, .pyw, .pyc files) when it was installed. This means that when you double-click on one of these files from Windows explorer the launcher will be used, and therefore you can use the same facilities described above to have the script specify the version which should be used.

The key benefit of this is that a single launcher can support multiple Python versions at the same time depending on the contents of the first line.

3.8.2. Shebang Lines

If the first line of a script file starts with #!, it is known as a “shebang” line. Linux and other Unix like operating systems have native support for such lines and they are commonly used on such systems to indicate how a script should be executed. This launcher allows the same facilities to be used with Python scripts on Windows and the examples above demonstrate their use.

To allow shebang lines in Python scripts to be portable between Unix and Windows, this launcher supports a number of ‘virtual’ commands to specify which interpreter to use. The supported virtual commands are:

  • /usr/bin/env python

  • /usr/bin/python

  • /usr/local/bin/python

  • python

For example, if the first line of your script starts with

The default Python will be located and used. As many Python scripts written to work on Unix will already have this line, you should find these scripts can be used by the launcher without modification. If you are writing a new script on Windows which you hope will be useful on Unix, you should use one of the shebang lines starting with /usr.

Any of the above virtual commands can be suffixed with an explicit version (either just the major version, or the major and minor version). Furthermore the 32-bit version can be requested by adding “-32” after the minor version. I.e. /usr/bin/python2.7-32 will request usage of the 32-bit python 2.7.

New in version 3.7: Beginning with python launcher 3.7 it is possible to request 64-bit version by the “-64” suffix. Furthermore it is possible to specify a major and architecture without minor (i.e. /usr/bin/python3-64).

The /usr/bin/env form of shebang line has one further special property. Before looking for installed Python interpreters, this form will search the executable PATH for a Python executable. This corresponds to the behaviour of the Unix env program, which performs a PATH search.

3.8.3. Arguments in shebang lines

The shebang lines can also specify additional options to be passed to the Python interpreter. For example, if you have a shebang line:

Then Python will be started with the -v option

3.8.4. Customization

3.8.4.1. Customization via INI files

Two .ini files will be searched by the launcher — py.ini in the current user’s “application data” directory (i.e. the directory returned by calling the Windows function SHGetFolderPath with CSIDL_LOCAL_APPDATA) and py.ini in the same directory as the launcher. The same .ini files are used for both the ‘console’ version of the launcher (i.e. py.exe) and for the ‘windows’ version (i.e. pyw.exe).

Customization specified in the “application directory” will have precedence over the one next to the executable, so a user, who may not have write access to the .ini file next to the launcher, can override commands in that global .ini file.

3.8.4.2. Customizing default Python versions

In some cases, a version qualifier can be included in a command to dictate which version of Python will be used by the command. A version qualifier starts with a major version number and can optionally be followed by a period (‘.’) and a minor version specifier. Furthermore it is possible to specify if a 32 or 64 bit implementation shall be requested by adding “-32” or “-64”.

For example, a shebang line of #!python has no version qualifier, while #!python3 has a version qualifier which specifies only a major version.

If no version qualifiers are found in a command, the environment variable PY_PYTHON can be set to specify the default version qualifier. If it is not set, the default is “3”. The variable can specify any value that may be passed on the command line, such as “3”, “3.7”, “3.7-32” or “3.7-64”. (Note that the “-64” option is only available with the launcher included with Python 3.7 or newer.)

If no minor version qualifiers are found, the environment variable PY_PYTHON{major} (where {major} is the current major version qualifier as determined above) can be set to specify the full version. If no such option is found, the launcher will enumerate the installed Python versions and use the latest minor release found for the major version, which is likely, although not guaranteed, to be the most recently installed version in that family.

On 64-bit Windows with both 32-bit and 64-bit implementations of the same (major.minor) Python version installed, the 64-bit version will always be preferred. This will be true for both 32-bit and 64-bit implementations of the launcher — a 32-bit launcher will prefer to execute a 64-bit Python installation of the specified version if available. This is so the behavior of the launcher can be predicted knowing only what versions are installed on the PC and without regard to the order in which they were installed (i.e., without knowing whether a 32 or 64-bit version of Python and corresponding launcher was installed last). As noted above, an optional “-32” or “-64” suffix can be used on a version specifier to change this behaviour.

Examples:

  • If no relevant options are set, the commands python and python2 will use the latest Python 2.x version installed and the command python3 will use the latest Python 3.x installed.

  • The commands python3.1 and python2.7 will not consult any options at all as the versions are fully specified.

  • If PY_PYTHON=3, the commands python and python3 will both use the latest installed Python 3 version.

  • If PY_PYTHON=3.1-32, the command python will use the 32-bit implementation of 3.1 whereas the command python3 will use the latest installed Python (PY_PYTHON was not considered at all as a major version was specified.)

  • If PY_PYTHON=3 and PY_PYTHON3=3.1, the commands python and python3 will both use specifically 3.1

In addition to environment variables, the same settings can be configured in the .INI file used by the launcher. The section in the INI file is called [defaults] and the key name will be the same as the environment variables without the leading PY_ prefix (and note that the key names in the INI file are case insensitive.) The contents of an environment variable will override things specified in the INI file.

For example:

  • Setting PY_PYTHON=3.1 is equivalent to the INI file containing:

  • Setting PY_PYTHON=3 and PY_PYTHON3=3.1 is equivalent to the INI file containing:

[defaults]
python=3
python3=3.1

3.8.5. Diagnostics

If an environment variable PYLAUNCH_DEBUG is set (to any value), the launcher will print diagnostic information to stderr (i.e. to the console). While this information manages to be simultaneously verbose and terse, it should allow you to see what versions of Python were located, why a particular version was chosen and the exact command-line used to execute the target Python.

3.9. Finding modules

Python usually stores its library (and thereby your site-packages folder) in the installation directory. So, if you had installed Python to C:\Python\, the default library would reside in C:\Python\Lib\ and third-party modules should be stored in C:\Python\Lib\site-packages\.

To completely override sys.path, create a ._pth file with the same name as the DLL (python37._pth) or the executable (python._pth) and specify one line for each path to add to sys.path. The file based on the DLL name overrides the one based on the executable, which allows paths to be restricted for any program loading the runtime if desired.

When the file exists, all registry and environment variables are ignored, isolated mode is enabled, and site is not imported unless one line in the file specifies import site. Blank paths and lines starting with # are ignored. Each path may be absolute or relative to the location of the file. Import statements other than to site are not permitted, and arbitrary code cannot be specified.

Note that .pth files (without leading underscore) will be processed normally by the site module when import site has been specified.

When no ._pth file is found, this is how sys.path is populated on Windows:

  • An empty entry is added at the start, which corresponds to the current directory.

  • If the environment variable PYTHONPATH exists, as described in Environment variables, its entries are added next. Note that on Windows, paths in this variable must be separated by semicolons, to distinguish them from the colon used in drive identifiers (C:\ etc.).

  • Additional “application paths” can be added in the registry as subkeys of \SOFTWARE\Python\PythonCore{version}\PythonPath under both the HKEY_CURRENT_USER and HKEY_LOCAL_MACHINE hives. Subkeys which have semicolon-delimited path strings as their default value will cause each path to be added to sys.path. (Note that all known installers only use HKLM, so HKCU is typically empty.)

  • If the environment variable PYTHONHOME is set, it is assumed as “Python Home”. Otherwise, the path of the main Python executable is used to locate a “landmark file” (either Lib\os.py or pythonXY.zip) to deduce the “Python Home”. If a Python home is found, the relevant sub-directories added to sys.path (Lib, plat-win, etc) are based on that folder. Otherwise, the core Python path is constructed from the PythonPath stored in the registry.

  • If the Python Home cannot be located, no PYTHONPATH is specified in the environment, and no registry entries can be found, a default path with relative entries is used (e.g. .\Lib;.\plat-win, etc).

If a pyvenv.cfg file is found alongside the main executable or in the directory one level above the executable, the following variations apply:

  • If home is an absolute path and PYTHONHOME is not set, this path is used instead of the path to the main executable when deducing the home location.

The end result of all this is:

  • When running python.exe, or any other .exe in the main Python directory (either an installed version, or directly from the PCbuild directory), the core path is deduced, and the core paths in the registry are ignored. Other “application paths” in the registry are always read.

  • When Python is hosted in another .exe (different directory, embedded via COM, etc), the “Python Home” will not be deduced, so the core path from the registry is used. Other “application paths” in the registry are always read.

  • If Python can’t find its home and there are no registry value (frozen .exe, some very strange installation setup) you get a path with some default, but relative, paths.

For those who want to bundle Python into their application or distribution, the following advice will prevent conflicts with other installations:

  • Include a ._pth file alongside your executable containing the directories to include. This will ignore paths listed in the registry and environment variables, and also ignore site unless import site is listed.

  • If you are loading python3.dll or python37.dll in your own executable, explicitly call Py_SetPath() or (at least) Py_SetProgramName() before Py_Initialize().

  • Clear and/or overwrite PYTHONPATH and set PYTHONHOME before launching python.exe from your application.

  • If you cannot use the previous suggestions (for example, you are a distribution that allows people to run python.exe directly), ensure that the landmark file (Lib\os.py) exists in your install directory. (Note that it will not be detected inside a ZIP file, but a correctly named ZIP file will be detected instead.)

These will ensure that the files in a system-wide installation will not take precedence over the copy of the standard library bundled with your application. Otherwise, your users may experience problems using your application. Note that the first suggestion is the best, as the others may still be susceptible to non-standard paths in the registry and user site-packages.

Changed in version 3.6:

  • Adds ._pth file support and removes applocal option from pyvenv.cfg.

  • Adds pythonXX.zip as a potential landmark when directly adjacent to the executable.

Deprecated since version 3.6:

Modules specified in the registry under Modules (not PythonPath) may be imported by importlib.machinery.WindowsRegistryFinder. This finder is enabled on Windows in 3.6.0 and earlier, but may need to be explicitly added to sys.meta_path in the future.

3.10. Additional modules

Even though Python aims to be portable among all platforms, there are features that are unique to Windows. A couple of modules, both in the standard library and external, and snippets exist to use these features.

The Windows-specific standard modules are documented in MS Windows Specific Services.

3.10.1. PyWin32

The PyWin32 module by Mark Hammond is a collection of modules for advanced Windows-specific support. This includes utilities for:

  • Component Object Model (COM)

  • Win32 API calls

  • Registry

  • Event log

  • Microsoft Foundation Classes (MFC) user interfaces

PythonWin is a sample MFC application shipped with PyWin32. It is an embeddable IDE with a built-in debugger.

3.10.2. cx_Freeze

cx_Freeze is a distutils extension (see Extending Distutils) which wraps Python scripts into executable Windows programs (*.exe files). When you have done this, you can distribute your application without requiring your users to install Python.

3.10.3. WConio

Since Python’s advanced terminal handling layer, curses, is restricted to Unix-like systems, there is a library exclusive to Windows as well: Windows Console I/O for Python.

WConio is a wrapper for Turbo-C’s CONIO.H, used to create text user interfaces.

3.11. Compiling Python on Windows

If you want to compile CPython yourself, first thing you should do is get the source . You can download either the latest release’s source or just grab a fresh checkout .

The source tree contains a build solution and project files for Microsoft Visual Studio 2015, which is the compiler used to build the official Python releases. These files are in the PCbuild directory.

Check PCbuild/readme.txt for general information on the build process.

For extension modules, consult Building C and C++ Extensions on Windows.

3.12. Other Platforms

With ongoing development of Python, some platforms that used to be supported earlier are no longer supported (due to the lack of users or developers). Check PEP 11 for details on all unsupported platforms.

  • Windows CE is still supported.

  • The Cygwin installer offers to install the Python interpreter as well (cf. Cygwin package source , Maintainer releases )

See Python for Windows for detailed information about platforms with pre-compiled installers.

  • Windows embedded standard 7 32 torrent
  • Windows embedded standard 2010 sp1
  • Windows driver package asus что это
  • Windows driver kit visual studio
  • Windows driver kit for windows 7 скачать