ALE 3.0 - Four Years of ALE!

Just a moment ago, I released version 3.0.0 of ALE. This version of ALE has been released on the four year anniversary of ALE! Four years ago, on this very same day, was when I made the very first git commit for ALE. ALE was originally released in the same week that Vim 8.0 was released. (Search the page for “w0rp.”) In that time, I have gone from being a regular-old developer to a team leader, I am very close to getting married, and the entire world has gone to Hell in a repeat of what happened in 1918.

It’s been difficult to find the time to work on ALE in the past year, but I have made a huge push in the past couple of months to get a lot of important work done. The changelog for version 3.0.0 of ALE contains many important changes, from many weeks of work from both myself and others. The full list of changes is long, but here are some important changes which I personally like a lot.

• ALE supports the way project configuration works with newer ESLint versions.
• Pylint checking now checks files as you type if the version of Pylint is new enough, and works just like before if you have to stick with an older version.
• Pyright should work without any configuration, or very little configuration at the very least.
• You can use a new ALEImport command to just import something at your cursor if there happens to be a completion result you can get that will import it.
• You can configure ALE to automatically fix your code after inserting completions.
• gcc and clang linters have been merged into a uniform linter, which automatically gets compiler flags from compile_commands.json files in generally good way.
• It’s now possible to run linters in Docker, in a virtual machine, or even on some remote machine.

There are even more goodies in the changelog to look for. I would like to once again thank everyone who keeps contributing to the project. I was able to get a lot more pull requests merged recently thanks to the Stale bot. I firmly believe that issues are never “stale,” and should only be closed if they are resolved in same way, but pull requests certainly go stale, and getting the bot to nudge people or myself has really helped a lot.

I’d like to thank the “Dense Analysis” teams for all of their help. “Dense Analysis” is the GitHub organisation I created just so I could configure permission levels for ALE. If anyone feels like they can help manage ALE issues, pull requests, or possibly help with ALE maintenance, please don’t hesitate to shoot me an email at dev@w0rp.com. It takes a while to vet people before they can be given enough permissions, but any help is appreciated.

Here’s to another four years of ALE. Cheers!