emacs

BuddhiLW 2025-04-05T18:06:14.333099Z

Anyone faced similar mismatch in auto-completion in Emacs? repl vs in-file

oyakushev 2025-04-05T18:08:23.085629Z

Does this happen in other projects too? In other namespaces? Can you try with LSP disabled?

BuddhiLW 2025-04-05T18:10:59.240709Z

LSP disabled: it won't complete anything, aside from copilot/windsurf

BuddhiLW 2025-04-05T18:11:48.597189Z

It happens in every project, currently. I haven't used Clojure in a while, now (1/2 years)

oyakushev 2025-04-05T18:13:50.241229Z

I'm 98% sure this is something with your setup – completion is a base feature and anybody else would notice this too. You having copilot and stuff hints that something may be overriding your regular cider completions.

👀 1
BuddhiLW 2025-04-05T18:13:51.977649Z

tree
.
├── deps.edn
└── src
    └── ugh_core
        └── core.clj
cat deps.edn
{:paths ["src"]

 :deps {org.clojure/clojure {:mvn/version "1.11.1"}}

 :aliases
 {:dev
  {:extra-deps {nrepl/nrepl {:mvn/version "1.3.1"}
                cider/cider-nrepl {:mvn/version "0.54.0"}
                refactor-nrepl/refactor-nrepl {:mvn/version "3.10.0"}}
   :main-opts ["-m" "nrepl.cmdline"
               "--middleware"
               "[refactor-nrepl.middleware/wrap-refactor,cider.nrepl/cider-middleware]"]}}}

oyakushev 2025-04-05T18:14:25.505739Z

What happens if you invoke M-x completion-at-point manully?

BuddhiLW 2025-04-05T18:16:00.668519Z

nothingness. Only message I get in Messages is You can run the command 'completion-at-point' with C-@ [4 times]

oyakushev 2025-04-05T18:18:29.963219Z

Another thing to try out: • M-x nrepl-toggle-message-logging • Try to trigger a completion • Swtich to *nrepl-messages ... buffer and check if the complete op went through

oyakushev 2025-04-05T18:18:48.541569Z

Most likely it won't be there

BuddhiLW 2025-04-05T18:19:20.484009Z

(-->
  id         "24"
  op         "eldoc"
  session    "33198572-497e-47ed-b8fa-a9a336f55ac4"
  time-stamp "2025-04-05 15:18:57.610271040"
  context    "(__prefix__)"
  ns         #("ugh-core.core" 0 13 (face font-lock-type-face ws-butler-chg chg help-echo cider--help-echo fontified t))
  sym        "pp/"
)
(<--
  id         "24"
  session    "33198572-497e-47ed-b8fa-a9a336f55ac4"
  time-stamp "2025-04-05 15:18:57.612717701"
  status     ("done" "no-eldoc")
)

BuddhiLW 2025-04-05T18:19:35.125309Z

Seems like can't fint eldoc

oyakushev 2025-04-05T18:20:01.825119Z

Yeah, that's op "eldoc", you need op "complete"

oyakushev 2025-04-05T18:20:41.920819Z

You can try doing completion from the REPL, then you'll see that op in the messages

oyakushev 2025-04-05T18:21:18.334409Z

BuddhiLW 2025-04-05T18:35:58.315239Z

It's very wierd. When I clojure-mode to refresh the mode, or I start a new folder, I get a non-highlighted text. I think both things are related somehow

BuddhiLW 2025-04-05T18:43:35.664619Z

Ok, I removed anything related to codeium, and now it's working; I removed a lot of extra stuff too; anything related to clojure, expect the vanilla setup from doom

👍 1
BuddhiLW 2025-04-05T18:43:38.695299Z

It's working now

BuddhiLW 2025-04-05T18:45:04.509539Z

I still get the white text highlighting when I clojure-mode though, which is still very weird but unrelated, it seems.

BuddhiLW 2025-04-05T18:45:40.828139Z

BuddhiLW 2025-04-05T19:32:51.653909Z

Found this other culprit

BuddhiLW 2025-04-05T19:33:52.836049Z

indent-bars is incompatible with tree-sitter was causing a crash in hook-time, when I called clojure-mode

BuddhiLW 2025-04-05T19:35:09.448049Z

> If debugging is the process of removing software bugs, then programming must be the process of putting them in

BuddhiLW 2025-04-05T19:36:19.366819Z

Removing 30% of my packages to make emacs more beautiful must solve like 90% of my problems; If I remove AI nonsense it's 99%. 😂

😁 1
BuddhiLW 2025-04-05T19:36:43.691889Z

Thanks @alexyakushev for the help earlier

👍 1