Fork me on GitHub

No special setup is need for company when using CIDER.


I know, working fine for me but I'm curious about it as I want to customize it. Which of the backends in company-backends is it? I think that'd give me a starting point


capf (completion-at-point-functions)


Thank you!


In case it rings a bell, what I'm trying to accomplish is to complete blindly if there's a single candidate which hasn't been compiled by clojure yet (i.e. I introduced blargh, haven't saved yet, now I type bla<TAB>). As opposed to having to hit TAB twice.


It seems cider-connection-browser disappeared in the latest version, not sure many people were using it but I used it all the time, so I hacked together a replacement.


I was just telling @vspinu that now it’s hard to figure out what connections are mapped to which session.


@bozhidar what's the story with sesman? was this created for/spun off from CIDER, or did it exist already? is anyone else using it?


@plexus It was created for CIDER, but with the aim to generic enough to be used with other projects. I know ESS (Emacs Speaks Statistics) is also going to use it. The aim is to adopt this in other similar projects that have to deal with multiple connections/sessions.


I was against the idea to have this as something generic and outside of CIDER, but @vspinu was really convinced that sesman has a potential to help simplify many projects (and bring consistency to them), so I agreed in the end.


I definitely agree that sesman’s approach is significantly better from how multiple connections were handled in the past, but the project is still very young and pretty rough around the edges (especially in the UI department).


I know prefixing variables/functions with the package name is common, but is there a way in code to easily reference only variables/functions that belong to a certain package, mainly when using autocomplete? is there a way to do the same with "core" emacs functions and vars?