Zooce

Just learned about this on Twitter now and will be trying this out today: https://twitter.com/DappForce/status/1183111484911566848?s=09

Looking at the code, in `UnixFileLock._acquire` the call to `os.open` is not wrapped in a try-except, but `WindowsFileLock._acquire` does wrap the `os.open` call in a try-except.

Any progress on this? I've had to add so many words to the folder dictionaries that it's becoming kind of annoying.

I'm not an expert in Qt applications, but launching LibreCAD from the terminal like this seems to work: ``` $ QT_AUTO_SCREEN_SCALE_FACTOR=1 librecad & ``` I got this from: https://doc.qt.io/qt-5/highdpi.html. And...

I'm having similar problems but with GitBash. I see the fnm_multishells in my path but for some reason it's not picking it up. One thing that is different about that...

Might I suggest making this configurable as well. Some workflows may not favor this approach.

We use `effect` to do this fetch-some-new-data-when-a-signal-changes-and-store-it-in-another-signal pattern in several places with the use of `untracked` and it works great. We do all of this in services and testing is...

BTW: This does not happen on my Pop!_OS 20.04 machine. Maybe a Windows-only issue? Maybe that will help narrow things down a bit...

I'm seeing this as well, but I noticed it with `bun outdated` which printed no output: ```bash $ bun oudated bun outdated v1.1.28 (cf4e9cb6) $ ```

This seems to be broken again. Even with a trailing slash, I'm getting 404s. ```toml # .bunfig.toml [install] registry = "https://my.registry.com/npm/" ``` ```bash $ bun install bun install v1.1.34 (5e5e7c60)...