Clerk was just open sourced today and I am very excited. Nextjournal has been doing amazing things and I've taken much inspiration from their work, we seem to have the same taste in tooling. They provide a next-generation coding notebook experience, and have previously open-sourced clojure-mode for Codemirror 6 which I'd love to integrate into our web editor as well... but that's another issue.

The reason I'm so excited about Clerk is that it makes wonderful use of tools.analyzer, and I tend to work best when given good examples to follow. The techniques they use to analyze Clojure files and gather comprehensive data to provide their live notebook environment will also apply to our purpose for analyzing user submissions.

Seems we have a run script, which calls the main function with the right args, and the Dockerfile. Details are still left to work out but the pieces are in place. I'm very slow when it comes to docker stuff because it often causes my machines to explode.

Probably ought to complete #20 because it will avoid the JVM and make analysis time nearly instantaneous.

The next thing that I know of is to create the template file in https://github.com/exercism/website-copy/tree/main/analyzer-comments to display the analysis results. Since we are just displaying the results of the script (and not creating individual analyzer comments), it will just be one file called generic.md. For example, the bash track uses results from ShellCheck: https://github.com/exercism/website-copy/blob/main/analyzer-comments/bash/shellcheck/generic.md

In line with our new org-wide policy, the master branch of this repo will be renamed to main. All open PRs will be automatically repointed.

GitHub will show you a notification about this when you look at this repo after renaming:

In case it doesn't, this is the command it suggests:

git branch -m master main
git fetch origin
git branch -u origin/main main

You may like to update the primary branch on your forks too, which you can do under Settings->Branches and clicking the pencil icon on the right-hand-side under Default Branch:

We will post a comment below when this is done. We expect it to happen within the next 12 hours.

The analyzer currently uses clj-kondo as a Clojure library to extract analysis data which is then processed into JSON. This is not ideal for production, because clj-kondo is already shipped as a native binary that runs on Alpine.

I think it would certainly be preferable to instead use an image with clj-kondo in it and call it as a CLI, then have another script that handles those results.

Doc for getting analysis data via CLI: https://github.com/clj-kondo/clj-kondo/tree/master/analysis