nrepl-sync-request:eval is breaking

[looselytyped]

I am not sure quite what the issue here is, so allow me to post what I have found so far.

• I posted an issue on lispy.el - lispy-eval reports "No buffer named ... " abo-abo/lispy#115. The issue is that lispy-eval is breaking with a No buffer named com.mycompany.job error.
• @abo-abo responded with lispy-eval reports "No buffer named ... " abo-abo/lispy#115 (comment) (reposting here)

This is a breakage in CIDER and it should be fixed there, I think. Even the most basic eval request fails now:

(nrepl-sync-request:eval
 "(+ 2 2)"
 (cider-current-ns)
 (nrepl-current-session))

I am not quite sure if the the issue lies with CIDER or lispy, however, the issue seems to permeate from the fact that my buffer's name is actually job.clj - but the error reports the name as the full ns.

My apologies if this seems vague - I am trying to debug this on my machine, but my emacs-lisp-fu isn't all that great.

Here is what I am running

Currently running GNU Emacs 24.5.1 with

cider 20150901.1605
lispy 20150901.550

I will keep looking into this and see if I can come up with more info.

On a final note, thank you for all your hard work on CIDER, and all other emacs/clojure tooling :)

[Malabarba]

This is because the CIDER internals are undergoing some breaking changes. If you look at the docstring for that function you'll see that its signature has changed to

(nrepl-sync-request:eval INPUT CONNECTION SESSION &optional NS)

So that the correct way to call it is now

(nrepl-sync-request:eval
 "(+ 2 2)"
 (cider-current-connection)
 (cider-current-session)
 (cider-current-ns))

I'm not sure if the refactoring is finished or not (@bbatsov?), but when it is finished we should probably make a list of all the breaking changes.

[bbatsov]

It's a work in progress, but I have a pretty busy week at work and it's on hold for now. Looking at my schedule for the month I might not be able to finish everything until the beginning of October.

[looselytyped]

@Malabarba @bbatsov - Thank you for the prompt response. I can easily work around this if need be. And again, thank you for all your hard work.

Would you rather I close this issue?

[bbatsov]

Guess that would be best.