PowerShellPython
This fork features a modified subprocess.py (subprocess.run) that prefixes code to replace core shell operations from cmd.exe with PowerShell, targeting commands like ninja, cmake, cl/msvc, link, etc. The goal is to bypass cmd.exe issues by using PowerShell for Python execution.
Use / Goals:
Bypass CMD issues causing build/install fails.
- To fix CMD length issues with typical cmd.exe builds and installs.
- If you run into CMD line length exceeded (especially with link.exe), this will likely fix it.
Bring seamless pip ecosystem back to Windows.
- To fix some pip usage with certain large projects, mainly and tested are Xformers (with extensions) and flash-attn.
- If you have pip install issues with these, this may fix it, though pip issues vary.
Make Windows more viable and functionality less dependent on Linux VM-ware or other invasive workarounds.
- Via the above, and give An alternative shell for workarounds, troubleshooting and debugging.
Features
- Portable & Future-Proof – No additional imports; designed for maximum compatibility with minimal invasiveness.
- *Modification Limited to* subprocess.run – The rest of subprocess.py remains untouched.
- Works with Standard Build Toolchains – Requires the usual dependencies: MSVC, C++, CL, VC (Visual Studio), CMake (Windows version, not Python’s), Ninja (Windows version, not Python’s), CUDA, etc.
- Fixes vcvarsall.bat Conflicts – Uses a custom call to initialize MSVC without interference from PowerShell prefixing.
- Supports Different PowerShell Versions – While tested on PowerShell v1.0, it should work with other versions based on pathing/env priority.
- Validated on Both Native & VM Environments – Tested with Python 3.10.6, CUDA 12.1, PyTorch 2.5.1+cu121, Xformers v0.0.29.post1, and Flash-Attn v2.7.4.post1.
- *Works with Plain* pip install – No need for manual builds or Git repo installations.
Recommended Fixes & Enhancements
Optional: PowerShell Helper Script (ver.ps1)
This optional script improves verbose logging compatibility. Place it in system32 or another accessible location and enable execution via:
powershell:
Set-ExecutionPolicy Unrestricted
ver.ps1:
$os = Get-CimInstance -ClassName Win32_OperatingSystem Write-Output "" Write-Output "Microsoft Windows [Version $($os.Version)]"
Highly Recommended: Setuptools Fix (build.py Patch) Issue in build.py 8000 involving "+ plat_specifier" can cause Windows pip install failures (e.g., Xformers). Remove the lines to resolve it:
setuptools/_distutils/command/build.py:
python
if self.build_purelib is None:
self.build_purelib = os.path.join(self.build_base, 'lib') # <<< Remove + plat_specifier
if self.build_platlib is None:
self.build_platlib = os.path.join(self.build_base, 'lib') # <<< Remove + plat_specifier
...
self.build_temp = os.path.join(self.build_base, 'temp') # <<< Remove + plat_specifier
A scripted auto-fix is included in PowerShellPython but is commented out by default.
Installation & Usage
- In this forked version, its all setup and ready to go.
Note: Renamed original subprocess as subprocess_original.py so user may switch between them.
- Apply the build.py patch if needed
- (Optional) Add ver.ps1 to improve log handling
NOTE: You do not actually need to run from powershell at all, you can run cmd like normal, python will just be calling powershell itself from the backend.
Conclusion
PowerShellPython aims to make Windows AI development smoother by replacing cmd.exe with PowerShell where possible. While not a universal fix, it has shown significant improvements in AI package installations and builds, reducing cmd-related failures.
This is a work-in-progress—feedback is welcome!
If you'd like to show appreciation (donate) for this work: https://buymeacoffee.com/leomaxwell
Copyright © 2001 Python Software Foundation. All rights reserved.
See the end of this file for further copyright and license information.
Contents
- Website: https://www.python.org
- Source code: https://github.com/python/cpython
- Issue tracker: https://github.com/python/cpython/issues
- Documentation: https://docs.python.org
- Developer's Guide: https://devguide.python.org/
For more complete instructions on contributing to CPython development, see the Developer Guide.
Installable Python kits, and information about using Python, are available at python.org.
On Unix, Linux, BSD, macOS, and Cygwin:
./configure make make test sudo make install
This will install Python as python3.
You can pass many options to the configure script; run ./configure --help
to find out more. On macOS case-insensitive file systems and on Cygwin,
the executable is called python.exe; elsewhere it's just python.
Building a complete Python installation requires the use of various additional third-party libraries, depending on your build platform and configure options. Not all standard library modules are buildable or usable on all platforms. Refer to the Install dependencies section of the Developer Guide for current detailed information on dependencies for various Linux distributions and macOS.
On macOS, there are additional configure and build options related to macOS framework and universal builds. Refer to Mac/README.rst.
On Windows, see PCbuild/readme.txt.
To build Windows installer, see Tools/msi/README.txt.
If you wish, you can create a subdirectory and invoke configure from there. For example:
mkdir debug cd debug ../configure --with-pydebug make make test
(This will fail if you also built at the top-level directory. You should do
a make clean at the top-level first.)
To get an optimized build of Python, configure --enable-optimizations
before you run make. This sets the default make targets up to enable
Profile Guided Optimization (PGO) and may be used to auto-enable Link Time
Optimization (LTO) on some platforms. For more details, see the sections
below.
PGO takes advantage of recent versions of the GCC or Clang compilers. If used,
either via configure --enable-optimizations or by manually running
make profile-opt regardless of configure flags, the optimized build
process will perform the following steps:
The entire Python directory is cleaned of temporary files that may have resulted from a previous compilation.
An instrumented version of the interpreter is built, using suitable compiler flags for each flavor. Note that this is just an intermediary step. The binary resulting from this step is not good for real-life workloads as it has profiling instructions embedded inside.
After the instrumented interpreter is built, the Makefile will run a training workload. This is necessary in order to profile the interpreter's execution. Note also that any output, both stdout and stderr, that may appear at this step is suppressed.
The final step is to build the actual interpreter, using the information collected from the instrumented one. The end result will be a Python binary that is optimized; suitable for distribution or production installation.
Enabled via configure's --with-lto flag. LTO takes advantage of the
ability of recent compiler toolchains to optimize across the otherwise
arbitrary .o file boundary when building final executables or shared
libraries for additional performance gains.
We have a comprehensive overview of the changes in the What's New in Python 3.14 document. For a more detailed change log, read Misc/NEWS, but a full accounting of changes can only be gleaned from the commit history.
If you want to install multiple versions of Python, see the section below entitled "Installing multiple versions".
Documentation for Python 3.14 is online, updated daily.
It can also be downloaded in many formats for faster access. The documentation is downloadable in HTML, PDF, and reStructuredText formats; the latter version is primarily for documentation authors, translators, and people with special formatting requirements.
For information about building Python's documentation, refer to Doc/README.rst.
To test the interpreter, type make test in the top-level directory. The
test set produces some output. You can generally ignore the messages about
skipped tests due to optional features which can't be imported. If a message
is printed about a failed test or a traceback or core dump is produced,
something is wrong.
By default, tests are prevented from overusing resources like disk space and
memory. To enable these tests, run make buildbottest.
If any tests fail, you can re-run the failing test(s) in verbose mode. For
example, if test_os and test_gdb failed, you can run:
make test TESTOPTS="-v test_os test_gdb"
If the failure persists and appears to be a problem with Python rather than your environment, you can file a bug report and include relevant output from that command to show the issue.
See Running & Writing Tests for more on running tests.
On Unix and Mac systems if you intend to install multiple versions of Python
using the same installation prefix (--prefix argument to the configure
script) you must take care that your primary python executable is not
overwritten by the installation of a different version. All files and
directories installed using make altinstall contain the major and minor
version and can thus live side-by-side. make install also creates
${prefix}/bin/python3 which refers to ${prefix}/bin/python3.X. If you
intend to install multiple versions using the same prefix you must decide which
version (if any) is your "primary" version. Install that version using
make install. Install all other versions using make altinstall.
For example, if you want to install Python 2.7, 3.6, and 3.14 with 3.14 being the
primary version, you would execute make install in your 3.14 build directory
and make altinstall in the others.
See PEP 745 for Python 3.14 release details.
Copyright © 2001 Python Software Foundation. All rights reserved.
Copyright © 2000 BeOpen.com. All rights reserved.
Copyright © 1995-2001 Corporation for National Research Initiatives. All rights reserved.
Copyright © 1991-1995 Stichting Mathematisch Centrum. All rights reserved.
See the LICENSE for information on the history of this software, terms & conditions for usage, and a DISCLAIMER OF ALL WARRANTIES.
This Python distribution contains no GNU General Public License (GPL) code, so it may be used in proprietary projects. There are interfaces to some GNU code but these are entirely optional.
All trademarks referenced herein are property of their respective holders.