[$] Accessing new kernel features from Python
Every release of the Linux kernel has lots of new features, many of which are accessible from user space. Usually, though, the GNU C Library (glibc) and tools that access the Linu [...]
| Web | https://ldpreload.com |
| GitHub | https://github.com/geofft |
| https://twitter.com/geofft | |
| Bluesky | https://bsky.app/profile/ldpreload.com |
@villares @feoh @paco Yes, please give the latest build a try when you get a chance! See https://github.com/astral-sh/python-build-standalone/issues/129#issuecomment-3169282440 for upgrade instructions (the Python startup message should say it was compiled today).
I tested matplotlib and PIL.ImageTk myself and they work now. Haven't tested other things but I'm optimistic.
Basic matplotlib usage such as import matplotlib.pyplot as plt; plt.plot(list(range(10)), list(range(10))) crashes with module '_tkinter' has no attribute '__file__'. On system python, _tkinter is ...
[$] Accessing new kernel features from Python
Every release of the Linux kernel has lots of new features, many of which are accessible from user space. Usually, though, the GNU C Library (glibc) and tools that access the Linu [...]
ulexecve is a userland execve() implementation which helps you execute arbitrary ELF binaries on Linux from userland without the binaries ever having to touch storage. This is useful for red-teamin...
Don't forget that SPDX also gives you convenient syntax for letting people opt into being blessed:
# SPDX-License-Identifier: 0BSD OR blessing
or combine a blessing with another license:
# SPDX-License-Identifier: GPL-2.0-or-later WITH Autoconf-exception-2.0 AND blessing
Software licensing fun fact: you can just write
# SPDX-License-Identifier: blessing
at the top of your file and it is spiritually equivalent to this full blessing: https://spdx.org/licenses/blessing.html
@glyph ah, yeah, I got sufficiently distracted by trying to make this work at all that I forgot about that goal, and indeed it completely does not work.
I feel like mostly I want @ dataclass class Foo(Enum) to work like this out of the box (right now it is documented to be broken but does not throw which seems bad), and I feel like getting type checkers to handle it should not be hard - they already know how to handle both enums and dataclasses.