pandemiclatte
a month ago
The original blog post of the author is: "Compiling a Python extension in C, with the Zig toolchain – Increased number of ziglang downloads" but honestly should have been called "How not to deploy breaking changes on NYE 2026".
The subject of the blog post is not truly reflecting the fact that a series of (bad) decisions to release a breaking change on NYE could have side effects on many of the 1000s or 10,000s of users of ruamel.yaml that didn't pin the version in their Python projects.
I appreciate the time the author of ruamel.yaml detailing the reasoning for his changes, but again we see how much responsibility a single developer has for libraries that we take for granted and the lack of guardrails around those can cause unforeseen issues.
pandemiclatte
a month ago
There is some more discussion from the author of the package here https://ziggit.dev/t/using-the-ziglang-and-setuptools-zig-py...
I still don't understand the logic to deploy this on the last day of the year. ¯\_(ツ)_/¯
pandemiclatte
a month ago
It was kind of expected after Jan 1st for people to start to see the CI pipelines to fail or any other random side effects: https://sourceforge.net/p/ruamel-yaml/activity/?page=0&limit...
pandemiclatte
a month ago
and it's fixed. https://sourceforge.net/p/ruamel-yaml/code/ci/6846cf136e775e...