Clojurians Log v2

Where do you position it in relation to the recently introduced core cli repl?

It's a nREPL specific client, that will emulate some of CIDER’s features out-of-the-box. I was mostly using REPLy for ages, but now that it's kind of abandoned I decided to go down this route. (And I did not want to make the built-in nREPl client more complex than it currently is)

Admittedly I haven't even heard of the core cli repl. I will have to check it out. 😁

How does it relate to bb repl --connect (which is a also a TUI REPL connecting to nREPL)

oh I see neorepl is also written in bb

Indeed. 😁 Seemed like a very good fit after your recent work. Well, it works a bit like CIDER - jack-in with the middleware, built-in support for it, etc. I think the built-in client is purer, right?

But in a nutshell - my main idea for neorepl is “cider in your terminal for humans and agents alike”.

The built-in also uses the built-in nREPL stuff (nrepl.core and some built-in CIDER middeware (like the inspector, the debugger is coming soon). :) I'm curious what middleware you emulated and added?

I’m not sure I understand the question. I meant that the client looks for the cider-nrepl ops and uses them if available (it can also inject cider-nrepl like CIDER can). Right now it already has support for things like cider-nrepl’s complete, info (doc, eldoc, source, ,where), apropos, macroexpand, analyze-last-stacktrace and refresh. I’ll add support for more of those along the way. On non-Clojure projects some of those won’t be available, of course, but neorepl will gracefully degrade as most nrepl clients do.

bb has had complete, info, eldoc for a long time