Here is a list of events: https://microsoft.github.io/debug-adapter-protocol/specification#Events. We currently don't handle all of them; they should be added to the dap--on-event.

The message "No message handler for ..." indicates that we don't implement some event.

At the moment the user has to press Ctrl (or a different modifier key if changed in settings) to bring up the hover tooltip. We could allow specifying a delay as an alternative/additional activation method. This is how it is implemented in VScode I think.

Sketch for setting chema:

"delay": {
  "title": "Hover delay",
  "type": ["number", "null"],
  "default": 300,
  "

Add a configuration flag that disables the Java-to-Kotlin converter in external sources. This would also require declaring the option in the vscode-kotlin client.

I'm seeing duplicate diagnostics in the textDocument/publishDiagnostics notification, one from clojure-lsp and one from clj-kondo, for unused private vars

From clojure-lsp

        {
            "range": {
                "start": {
                    "line": 13,
                    "character": 7
                },
                "end": {
                    "line": 13,
  

Current Version

0.5.3

Use-cases

Currently we only expose rootModulePaths in "LSP native" configuration:

https://github.com/hashicorp/terraform-ls/blob/a1ff20f9f5378eaf2b224f84c8bc05b2028da18f/internal/settings/settings.go#L11-L13

Some runtime settings which we don't strictly need to know before the server is actually initialized is still exposed via CLI flags onl

Often, something goes wrong in my setup of kak-lsp (a language server is installed incorrectly, the command I specified for that LSP is wrong, the plugin just isn't enabled for the window, I forgot to start the LSP, my kak-lsp.toml is invalid or in the wrong place, my code project is setup wrong, etc., etc.) When this happens there are two problems that make it more difficult to solve:

1. **T

Describe the bug

If I set

otp_path: "/packages/erlang/erlang-22.3.2/"

then OTP packages are not indexed. If I remove the trailing path separator, then they are.

Our path expansion algorithms shouldn't care.

Expected Behavior

I would like to be warned that an "in port" would be stripped if no subscription is implemented.

Current Behavior

port module Main exposing (..)

-- Define an inPort (but no implementation yet)
port inPort : (String -> msg) -> Sub msg

subscriptions : Model -> Sub Msg
subscriptions _ =
    Sub.none -- inPort is never subscribed to

Current

For some reason FCS reports FSharpFindDeclResult.DeclFound for symbols declared in external F# libraries - we then handle it at normal go-to and attempt to return the result to the client, which ends up in error as such file probably doesn't exist on user machine. Example (note it reports path from @baronfel machine, when I try to do go-to-def on my machine):

![image](https://user-images.gith