Nathaniel J. Smith
Nathaniel J. Smith
@Zac-HD That's an interesting idea! I really like Hypothesis's approach, and also I'm really excited to see someone actually trying it for real :-). But I see two big differences...
> If a merged PR becomes to mean "a change worth releasing", does that mean that every single PR is supposed to improve the overall quality of the project? I...
Regarding the whole release automation thing, I think I've convinced myself that the main issues to worry about are: * New maintainers: if we do release-after-merge, that makes it scarier...
I had an idea for this, and started a conversation here: https://forum.bors.tech/t/running-custom-code-before-and-after-the-merge-for-continuous-deployment-or-other-uses/315
The bors-ng maintainer has [accepted](https://bors.tech/rfcs/0322-pre-test-and-pre-merge-hooks.html) the proposal to add hooks for autoformatting and continuous deployment. So the next step is for someone to implement it. The stuff we need to...
@auvipy Basically implementing this spec as a PR for [bors-ng](https://bors.tech/rfcs/0322-pre-test-and-pre-merge-hooks.html): https://bors.tech/rfcs/0322-pre-test-and-pre-merge-hooks.html The alternative plan would be to write our own implementation of bors-ng's features in Python, perhaps in [snekomatic](https://github.com/python-trio/snekomatic). I...
It does sound like an oversight in h11 since we try to preserve header case, but it shouldn't have much practical effect. HTTP header names are case-insensitive, so `Connection` and...
Sure, it wouldn't hurt to fix -- like I said, h11 does try to preserve case. On Fri, Nov 25, 2022 at 1:55 AM Marcelo Trylesinski ***@***.***> wrote: > >...
CI failure is spurious and fixed by b2983aa191eb089b0f23bb813784c7efd0133cdb
SIGTERM doesn't prevent processes from becoming zombies. The Unix process model is counterintuitive, though it has an internal logic. The point of a zombie process is that whenever a process...