-
-
Notifications
You must be signed in to change notification settings - Fork 960
Language servers
- Supported features
- Register custom language servers
-
Example language server configuration
- Ada/SPARK
- Apache Camel
- Arduino
- Bash
- Bitbake
- Blueprint
- C/C++/Objective-C
- CMake
- Clojure
- Common Lisp
- Crystal
- Css/Less/Sass
- Dart
- D
- Deno
- Dhall
- Dockerfile
- Docker Compose
- Elixir
- Elvish
- Erlang
- Elm
- Flow
- Fortran
- FSharp
- Gleam
- Go
- Godot
- GraphQL
- Groovy
- Haskell
- Haxe
- Html
- Java
- Javascript/Typescript
- Json
- Julia
- Kotlin
- LaTeX
- Lua
- Markdown
- Meson
- Nim
- Nix
- OCaml and ReasonML
- Perl
- PHP
- PureScript
- Python
- R
- Racket
- Robot Framework
- Rome
- Ruby
- Ruby RuboCop
- Ruby Sorbet
- Ruby Steep
- Ruby TypeProf
- Ruby Shopify/ruby-lsp
- Rust
- SQL
- Scala
- Solidity
- SystemVerilog
- systemd
- Terraform
- Typst
- V
- Vala
- Vue
- YANG
- Zig
- vim/erb/markdown
Check out the official LSP specification.
Note: different servers can have different capabilities.
User defined language servers are configured in the languageserver
field of the configuration file. See :h coc-config-languageserver
for common language server configuration settings.
There are three types of language servers: module
, executable
, and socket
.
-
module
type language servers are run by node.js and using node IPC for connection. -
executable
type language servers are spawned with an executable command while using stdio for connection. -
socket
language servers are started in a separate process, normally used for debugging purposes.
Different language server types have different configuration schema.
An example of module
language server:
"languageserver": {
"foo": {
"module": "/usr/local/lib/node_modules/foo/index.js",
"args": ["--node-ipc"],
"filetypes": ["foo"],
"trace.server": "verbose",
"rootPatterns": ["root.yml"],
// Used for debugging NodeJS
// "execArgv": ["--nolazy", "--inspect-brk=6045"],
"initializationOptions": {
},
"settings": {
"validate": true
}
}
}
module
and filetypes
are required for module language server.
An example of executable
language server:
"languageserver": {
"bar": {
"command": "bar",
"args": ["--stdio"],
"filetypes": ["bar"],
"cwd": "./src",
"initializationOptions": {
},
"settings": {
}
}
}
command
and filetypes
are required for executable language server.
An example of socket language server:
"languageserver": {
"socketserver": {
"host": "127.0.0.1",
"port": 9527
}
}
port
is required for a socket service and user should start the socket server before coc starts.
Install coc-json for completion and validation support.
Add languageserver
section in your coc-settings.json
for registering custom language servers.
Using Ada Language Server. See installation instructions on the Github homepage of this LSP.
Via coc-als, or:
"languageserver": {
"Ada": {
"command": "path/to/ada_language_server",
"filetypes": ["ada"]
}
}
Use coc-camel
:CocInstall coc-camel
Using Arduino Language Server. See the installation instructions in the README file of this LSP.
"languageserver":{
"arduino":{
"command":"/path/to/arduino-language-server",
"rootPatterns":["*.ino"],
"filetypes":["arduino"],
"args":["-cli", "/path/to/arduino-cli", "-clangd", "/path/to/clangd", "-cli-config", "/path/to/arduino-cli.yaml", "-fqbn", "board:name"],
"disabledFeatures":["semanticTokens"]
}
}
Tips:
- This LSP requires
clangd
andarduino-cli
to work properly. Modify the values according to the locations of the binaries on your machine. - To determine (or generate if there isn't one) the location of
arduino-cli.yaml
, refer to this page.
Via coc-sh or
Using mads-hartmann/bash-language-server
"languageserver": {
"bash": {
"command": "bash-language-server",
"args": ["start"],
"filetypes": ["sh"]
}
}
"languageserver": {
"bitbake": {
"command": "bitbake-language-server",
"filetypes": ["bitbake"]
},
}
Using blueprint-compiler
"languageserver": {
"blueprint": {
"command": "blueprint-compiler",
"args": ["lsp"],
"filetypes": ["blueprint", "blp"]
},
}
Using clangd with coc-clangd or:
"languageserver": {
"clangd": {
"command": "clangd",
"rootPatterns": ["compile_flags.txt", "compile_commands.json"],
"filetypes": ["c", "cc", "cpp", "c++", "objc", "objcpp"]
}
}
Like many tools, clangd relies on the presence of a JSON compilation database
Using ccls
"languageserver": {
"ccls": {
"command": "ccls",
"filetypes": ["c", "cc", "cpp", "c++", "objc", "objcpp"],
"rootPatterns": [".ccls", "compile_commands.json", ".git/", ".hg/"],
"initializationOptions": {
"cache": {
"directory": "/tmp/ccls"
}
}
}
}
To make header completion work with clang < 8 on Mac OS X, use "initializationOptions"
like:
"initializationOptions": {
"cache": {
"directory": "/tmp/ccls"
},
"clang": {
// make sure you have installed commandLineTools
"resourceDir": "/Library/Developer/CommandLineTools/usr/lib/clang/11.0.0",
"extraArgs": [
"-isystem",
"/Applications/Xcode.app/Contents/Developer/Toolchains/XcodeDefault.xctoolchain/usr/include/c++/v1",
"-I",
"/Applications/Xcode.app/Contents/Developer/Platforms/MacOSX.platform/Developer/SDKs/MacOSX.sdk/usr/include/"
]
}
},
Using cquery
"languageserver": {
"cquery": {
"command": "cquery",
"args": ["--log-file=/tmp/cq.log"],
"filetypes": ["c", "cc", "cpp", "c++"],
"rootPatterns": ["compile_flags.txt", "compile_commands.json", ".git/", ".hg/"],
"initializationOptions": {
"cacheDirectory": "/tmp/cquery"
}
}
}
Try coc-cmake (not implemented with LSP) or
Using cmake-language-server
"languageserver": {
"cmake": {
"command": "cmake-language-server",
"filetypes": ["cmake"],
"rootPatterns": [
"build/"
],
"initializationOptions": {
"buildDirectory": "build"
}
}
}
Use coc-cl
:CocInstall coc-cl
Use coc-clojure
:CocInstall coc-clojure
Using crystalline
. Follow the instructions to install it.
"languageserver": {
"crystal": {
"command": "crystalline",
"args": [
"--stdio"
],
"filetypes": [
"crystal"
]
}
}
Make sure you have crystalline
available in your PATH
Use coc-css is recommended.
One option is use coc-flutter, that leverages analysis_server from dart-sdk.
Another option is configure analysis_server yourself. Use analysis_server from dart-sdk:
"languageserver": {
"dart": {
"command": "dart",
"args": [
" change this to the path of analysis_server
"dart",
"language-server",
"--lsp",
"--client-id",
"vim",
"--client-version",
"coc.nvim",
],
"filetypes": ["dart"],
"trace.server": "verbose"
},
}
Or use natebosch/dart_language_server
"languageserver": {
"dart": {
"command": "dart_language_server", // in windows is dart_language_server.bat
"args": [],
"filetypes": ["dart"],
"initializationOptions": {},
"settings": {
"dart": {
"validation": {},
"completion": {}
}
}
}
}
Use coc-dlang
:CocInstall coc-dlang
Use coc-deno
:CocInstall coc-deno
Using dhall-lsp-server
. Follow the instructions to install it.
"languageserver": {
"dhall": {
"command": "dhall-lsp-server",
"filetypes": [
"dhall"
]
}
}
Using rcjsuen/dockerfile-language-server-nodejs
"languageserver": {
"dockerfile": {
"command": "docker-langserver",
"filetypes": ["dockerfile"],
"args": ["--stdio"]
}
}
Using microsoft/compose-language-service
First, set the filetype of docker-compose.yml
and compose.yml
to yaml.docker-compose
in vimrc
or init.vim
. Also set g:coc_filetype_map
to map yaml.docker-compose
to dockercompose
.
Set the .vimrc or init.vim:
au FileType yaml if bufname("%") =~# "docker-compose.yml" | set ft=yaml.docker-compose | endif
au FileType yaml if bufname("%") =~# "compose.yml" | set ft=yaml.docker-compose | endif
let g:coc_filetype_map = {
\ 'yaml.docker-compose': 'dockercompose',
\ }
Installing a Language Server:
npm i -g @microsoft/compose-language-service
Set the coc-settings.json:
{
"languageserver": {
"dockercompose": {
"command": "docker-compose-langserver",
"args": ["--stdio"],
"filetypes": ["dockercompose"],
"rootPatterns": [".git", ".env", "docker-compose.yml", "compose.yml"]
}
}
}
Using elixir-ls
If you want to use @spec suggestions you have to enable codelens.
"codeLens.enable": true,
"languageserver": {
"elixirLS": {
"command": "/absolute/path/to/elixir-ls/release/language_server.sh",
"filetypes": ["elixir", "eelixir"]
}
}
Using the builtin language server
{
"languageserver": {
"elvish": {
"command": "elvish",
"args": ["-lsp"],
"filetypes": ["elvish"]
}
}
}
Using erlang_ls
Assuming that erlang_ls
is in the execution path
"languageserver": {
"erlang": {
"command": "erlang_ls",
"filetypes": ["erlang"]
}
}
Using elm-tooling/elm-language-server
"languageserver": {
"elmLS": {
"command": "elm-language-server",
"filetypes": ["elm"],
"rootPatterns": ["elm.json"],
"initializationOptions": {
"elmPath": "elm", // optional
"elmFormatPath": "elm-format", // optional
"elmTestPath": "elm-test", // optional
"elmAnalyseTrigger": "change" // optional
}
}
}
elm-language-server
needs elm
, elm-format
and elm-test
- If a path like
elmPath
is defined, it'll be used. - If a path is missing,
elm-language-server
will try to load it from a local npm installation folder. Otherwise a global installation provided viaPATH
is used. -
elmAnalyseTrigger
:elm-analyse
(linting) is executed on 'change', 'save' or 'never' (default: 'change')
Check out github page for more information.
Using flow-language-server (Note: project no longer maintained)
// disable tsserver for javascript
"tsserver.enableJavascript": false,
"languageserver": {
"flow": {
"command": "flow-language-server",
"args": ["--stdio"],
"filetypes": ["javascript", "javascriptreact"],
"rootPatterns": [".flowconfig"]
},
}
Using flow lsp
"languageserver": {
"flow": {
"command": "flow",
"args": ["lsp"],
"filetypes": ["javascript", "javascriptreact"],
"initializationOptions": {},
"requireRootPattern": true,
"settings": {},
"rootPatterns": [".flowconfig"]
}
},
Using fortran-language-server
Make sure the fortls executable is available on your $PATH.
"languageserver": {
"fortran": {
"command": "fortls",
"filetypes": ["fortran"],
"rootPatterns": [".fortls", ".git/"]
}
}
Install FsAutoComplete
dotnet tool install --global fsautocomplete
Associate F# file extensions with fsharp
autocmd BufNewFile,BufRead *.fs,*.fsx,*.fsi set filetype=fsharp
"languageserver": {
"fsharp": {
"command": "fsautocomplete",
"args": [],
"filetypes": ["fsharp"],
"trace.server": "verbose",
"initializationOptions": {
"AutomaticWorkspaceInit": true
},
"settings": {
"FSharp.keywordsAutocomplete": true,
"FSharp.ExternalAutocomplete": false,
"FSharp.Linter": true,
"FSharp.UnionCaseStubGeneration": true,
"FSharp.UnionCaseStubGenerationBody": "failwith \"Not Implemented\"",
"FSharp.RecordStubGeneration": true,
"FSharp.RecordStubGenerationBody": "failwith \"Not Implemented\"",
"FSharp.InterfaceStubGeneration": true,
"FSharp.InterfaceStubGenerationObjectIdentifier": "this",
"FSharp.InterfaceStubGenerationMethodBody": "failwith \"Not Implemented\"",
"FSharp.UnusedOpensAnalyzer": true,
"FSharp.UnusedDeclarationsAnalyzer": true,
"FSharp.UseSdkScripts": true,
"FSharp.SimplifyNameAnalyzer": false,
"FSharp.ResolveNamespaces": true,
"FSharp.EnableReferenceCodeLens": true
}
}
}
Using the builtin language server
{
"languageserver": {
"gleam": {
"command": "gleam",
"args": ["lsp"],
"filetypes": ["gleam"]
}
}
}
Via coc-go or
Using gopls
"languageserver": {
"golang": {
"command": "gopls",
"rootPatterns": ["go.mod"],
"filetypes": ["go"]
}
}
If want more debug,add "args": ["serve", "-debug", "0.0.0.0:8484", "-rpc.trace"],
below command
.
Using sourcegraph/go-langserver
"languageserver": {
"golang": {
"command": "go-langserver",
"filetypes": ["go"],
"initializationOptions": {
"gocodeCompletionEnabled": true,
"diagnosticsEnabled": true,
"lintTool": "golint"
}
}
}
"languageserver": {
"godot": {
"host": "127.0.0.1",
"filetypes": ["gd", "gdscript", "gdscript3"],
"port": 6008
}
}
Using graphql-language-service-cli
"languageserver": {
"graphql": {
"command": "graphql-lsp",
"args": ["server", "-m", "stream"],
// customize filetypes to your needs
"filetypes": ["typescript", "typescriptreact", "graphql"]
}
}
Using Language server for Groovy
. Check out sources, build it and place groovy-language-server-all.jar
to any folder.
"languageserver": {
"groovy": {
"command": "java",
"args" : ["-jar", "/path/to/groovy-language-server-all.jar"],
"filetypes": ["groovy"]
}
}
Using Haskell Language Server
"languageserver": {
"haskell": {
"command": "haskell-language-server-wrapper",
"args": ["--lsp"],
"rootPatterns": ["*.cabal", "stack.yaml", "cabal.project", "package.yaml", "hie.yaml"],
"filetypes": ["haskell", "lhaskell"],
// Settings are optional, here are some example values
"settings": {
"haskell": {
"checkParents": "CheckOnSave",
"checkProject": true,
"maxCompletions": 40,
"formattingProvider": "ormolu",
"plugin": {
"stan": { "globalOn": true }
}
}
}
}
}
- You may also opt to use
ghcup
to easily install latestghc
,cabal
andhls
binaries to yourPATH
. - Check HLS README about global cabal configuration to enable documentation on hover.
- Run
haskell-language-server-wrapper generate-default-config
in order to see an example for all configuration and default values.
With this you can avoid building anything from scratch and can start coding Haskell files right away.
Install the coc-haxe plugin, then run :CocInstall coc-haxe
.
Otherwise, you can use the language server that comes with VS Code, or manually build it yourself: see vshaxe/haxe-language-server, and then use the following json-settings.
{
"languageserver": {
"haxe": {
"command": "node",
"args": ["<path-to-server.js>"],
"filetypes": ["haxe"],
"trace.server": "verbose",
"initializationOptions": {
"displayArguments": ["build.hxml"]
},
"settings": {
"haxe.executable": "haxe"
}
}
}
}
Where <path-to-server.js> can either be a server.js you built from source or simply downloaded as part of the Haxe Visual Studio Code extension ("//.vscode/extensions/nadako.vshaxe-/bin/server.js").
Use coc-html is recommended.
Use coc-java is recommended.
Use coc-tsserver is recommended.
use coc-json is recommended.
Using LanguageServer.jl
The LanguageServer
, SymbolServer
and StaticLint
packages must be installed in Julia (1.x), i.e.
julia> using Pkg
julia> Pkg.add("LanguageServer")
julia> Pkg.add("SymbolServer")
julia> Pkg.add("StaticLint")
Install coc-julia, or register the server in coc-settings.json
:
"languageserver": {
"julia": {
"command": "/usr/bin/julia",
"args" : ["--startup-file=no", "--history-file=no", "-e",
"using LanguageServer;\n using Pkg;\n import StaticLint;\n import SymbolServer;\n env_path = dirname(Pkg.Types.Context().env.project_file);\n server = LanguageServer.LanguageServerInstance(stdin, stdout, env_path, \"\");\n server.runlinter = true;\n run(server);" ],
"filetypes": ["julia"]
}
}
Check out JuliaEditorSupport/LanguageServer.jl for more information.
In case you hit issue #836, the language server needs to be installed from master to avoid the problem:
julia> using Pkg
julia> Pkg.add(url="https://github.com/julia-vscode/LanguageServer.jl")
julia> Pkg.add("SymbolServer")
julia> Pkg.add("StaticLint")
Using kotlin-language-server
- Download server.zip from the releases page.
- Unzip the file in a convenient directory, for example inside
~/lsp/kotlin/
.
"languageserver": {
"kotlin": {
"command": "~/lsp/kotlin/server/bin/kotlin-language-server",
"filetypes": ["kotlin"]
}
}
Using astoff/digestif
Make sure the digestif executable is available on your $PATH or use absolute path as command. Installation instructions can be found here.
"languageserver": {
"digestif": {
"command": "digestif",
"filetypes": ["tex", "plaintex", "context"]
}
}
For Texlab, use coc-texlab or:
"languageserver": {
"latex": {
"command": "/PATH/TO/texlab",
"filetypes": ["tex", "bib", "plaintex", "context"]
}
}
- May need to add
let g:tex_flavor = "latex"
to correct buffer filetype, check it by:echo &filetype
. - Adjust the path to textlab accordingly, or simply use as command name, from
PATH
. - For bibTeX integration, you should use package
biblatex
, check the gif on https://texlab.netlify.com/
Using Alloyed/lua-lsp
"languageserver": {
"lua": {
"command": "lua-lsp",
"filetypes": ["lua"]
}
}
Using sumneko/lua-language-server. It's a little difficult to use sumneko/lua-language-server
in coc.nvim directly, it's recommend to use coc-sumneko-lua that can download and setup for Lua. If you want to setup sumneko/lua-language-server
in coc-settings.json, follow the steps: (tests on macOS, change the path if you're using Windows)
- install
sumneko.lua
in VSCode cd ~/.vscode/extensions/sumneko.lua-2.3.7
-
chmod +x ./server/bin/macOS/lua-language-server
, have no idea why this is needed, because in my VSCode extension, the server is not executable. - setup in your coc-settings.json:
{
"languageserver": {
"lua": {
"command": "~/.vscode/extensions/sumneko.lua-2.3.7/server/bin/macOS/lua-language-server",
"args": ["-E", "~/.vscode/extensions/sumneko.lua-2.3.7/server/main.lua"],
"rootPatterns": [".git"],
"filetypes": ["lua"]
}
}
}
Using EmmyLua-LanguageServer
Make sure your Java environment variables are rights and change the path in the args field according to your installation.
"languageserver": {
"lua": {
"command": "java",
"args": ["-cp", "/your/path/to/EmmyLua-LanguageServer/EmmyLua-LS/build/libs/EmmyLua-LS-all.jar", "com.tang.vscode.MainKt"],
"filetypes": ["lua"],
"rootPatterns": [".git/"]
}
}
Using coc-markdownlint, which has a codeAction support to autofix errs.
:CocInstall coc-markdownlint
Using mesonlsp
"languageserver": {
"meson": {
"command": "mesonlsp",
"rootPatterns": ["meson.build", "meson_options.txt"],
"args": ["--lsp"],
"filetypes": ["meson"]
}
}
Using nimlsp
. Follow the instructions in the README of the repository. (Make sure that your Nimble bin directory is in your path)
"languageserver": {
"nim": {
"command": "nimlsp",
"filetypes": ["nim"]
}
}
Using nixd
"languageserver": {
"nixd": {
"command": "nixd",
"rootPatterns": [".nixd.json"],
"filetypes": ["nix"]
}
}
Using rnix-lsp
"languageserver": {
"nix": {
"command": "rnix-lsp",
"filetypes": ["nix"]
}
}
Using ocaml-language-server
"languageserver": {
"ocaml": {
"command": "ocaml-language-server",
"args": ["--stdio"],
"filetypes": ["ocaml", "reason"]
}
}
If you installed merlin with opam
"languageserver": {
"ocaml": {
"command": "opam",
"args": ["config", "exec", "--", "ocaml-language-server", "--stdio"],
"filetypes": ["ocaml", "reason"]
}
}
Using ocaml-lsp and opam
"languageserver": {
"ocaml-lsp": {
"command": "opam",
"args": ["config", "exec", "--", "ocamllsp"],
"filetypes": ["ocaml", "reason"]
}
}
Using ocaml-lsp and esy
"languageserver": {
"ocaml-lsp": {
"command": "esy",
"args": ["sh", "-c", "ocamllsp"],
"filetypes": ["ocaml", "reason"]
}
}
Using reason-language-server
"languageserver": {
"reason": {
"command": "reason-language-server",
"filetypes": ["reason"]
}
}
Use coc-perl is recommended.
Try marlonfan/coc-phpls or one of the following methods:
Using bmewburn/intelephense-docs
Recommended (way faster than php-language-server)
"languageserver": {
"intelephense": {
"command": "intelephense",
"args": ["--stdio"],
"filetypes": ["php"],
"initializationOptions": {
"storagePath": "/tmp/intelephense"
}
}
}
Using felixfbecker/php-language-server
"languageserver": {
"phplang": {
"command": "php",
"args": ["/path/to/vendor/felixfbecker/language-server/bin/php-language-server.php"],
"filetypes": ["php"]
}
}
Note: make sure you can start the server by use command and args.
Via coc-psalm or
Using vimeo/psalm (psalm-language-server)
"languageserver": {
"psalmls": {
"command": "vendor/bin/psalm-language-server",
"filetypes": ["php"],
"rootPatterns": ["psalm.xml", "psalm.xml.dist"],
"requireRootPattern": true
}
}
Using purescript-language-server (Configuration)
"languageserver": {
"purescript": {
"command": "purescript-language-server",
"args": ["--stdio"],
"filetypes": ["purescript"],
"rootPatterns": ["bower.json", "psc-package.json", "spago.dhall"]
}
}
Use one of them below:
- coc-pyright for Pyright
- coc-pylsp for python-lsp-server
- coc-jedi for jedi-language-server
- coc-ruff for ruff-lsp
First install language server from R
:
install.packages("languageserver")
Use coc-r-lsp extension.
If coc-r-lsp does not work properly, according to instructions from REditorSupport/languageserver, add below languageserver object to configuration file:
"languageserver": {
"R": {
"enable": true,
"command": "/usr/bin/R",
"args": [
"--slave",
"-e",
"languageserver::run()"
],
"filetypes": [
"r"
]
}
}
- install robotframework-lsp
- Configure robot filetype detection in neovim configuration file
autocmd BufNewFile,BufRead *.robot setlocal filetype=robot
"languageserver": {
"robotframework_ls": {
"command": "robotframework_ls",
"filetypes": ["robot"],
"settings": {
// here the ls configuration
}
}
}
Using the racket-langserver
- Install racket-langserver
raco pkg install racket-langserver
- Add the below snippet to your coc config
"languageserver": {
"racket": {
"command": "racket",
"args": [
"-l",
"racket-langserver"
],
"filetypes": [
"scheme", "rkt"
]
}
}
"languageserver": {
"rome-lsp": {
"command": "rome",
"args": ["lsp"],
"filetypes": [
"javascript",
"javascriptreact",
"typescript",
"typescriptreact",
"json"
],
"rootPatterns": [".config"],
"requireRootPattern": true
}
}
Using coc-solargraph
Make sure solargraph is in your $PATH (sudo gem install solargraph) or use solargraph.commandPath
to configure executable path of solargraph.
Using RuboCop:
"languageserver": {
"rubocop": {
"command": "rubocop",
"args": ["--lsp"],
"filetypes": ["ruby"]
}
}
Using Sorbet:
"languageserver": {
"sorbet": {
"command": "srb",
"args": ["tc", "--typed", "true", "--enable-all-experimental-lsp-features", "--lsp", "--disable-watchman"],
"filetypes": ["ruby"],
"rootPatterns": ["sorbet/config"],
"initializationOptions": {},
"settings": {}
}
}
Using Steep:
"languageserver": {
"steep": {
"command": "steep",
"args" : ["langserver"],
"filetypes": ["ruby"]
}
}
Using TypeProf:
"languageserver": {
"typeprof": {
"command": "typeprof",
"args": ["--lsp", "--stdio"],
"filetypes": ["ruby"]
}
}
Using Shopify/ruby-lsp:
"languageserver": {
"ruby-lsp": {
"command": "ruby-lsp",
"initializationOptions": {
"experimentalFeaturesEnabled": false,
"enabledFeatures": [
"codeActions",
"diagnostics",
"documentHighlights",
"documentLink",
"documentSymbols",
"foldingRanges",
"formatting",
"hover",
"inlayHint",
"onTypeFormatting",
"selectionRanges",
"semanticHighlighting",
"completion"
],
"formatter": "auto"
},
"filetypes": ["ruby"]
}
}
- install coc-rust-analyzer which uses
rust-analyzer
- or install coc-rls which uses
rls
- or try rust-analyzer without an extension:
"languageserver": {
"rust": {
"command": "rust-analyzer",
"filetypes": ["rust"],
"rootPatterns": ["Cargo.toml"]
}
}
It's necessary to rustup component add rust-src
and build rust-analyzer
from sources, follow rust-analyzer User Manual.
For coc-rls do not add above configuration in coc-settings.json
file just use ( rustup component add rls rust-analysis rust-src
)
Using sql-language-server
"languageserver": {
"sql": {
"module": "/path/to/node_modules/sql-language-server/dist/bin/cli.js",
"args": ["up", "--method", "node-ipc"],
"filetypes": ["sql", "mysql"]
}
}
The sdtio option is broken by it's console.log
.
Using scalameta/metals:
Install coc-metals, which will automate the Metals installation and also provide extra helpers.
If you'd like to use Metals without the coc-metals extension, make sure the generated metals-vim binary is available on your $PATH and follow the instructions on the Metals Website.
"languageserver": {
"metals": {
"command": "metals-vim",
"rootPatterns": ["build.sbt"],
"filetypes": ["scala", "sbt"]
}
}
Note that the Dotty Language server is no longer recommended. Instead, it is recommended to use Metals for Dotty/Scala 3.
- install coc-solidity
- or install @nomicfoundation/coc-solidity
- or try the LSP of the native binary
solc
without an extension:
"languageserver": {
"solidity": {
"command": "/path/to/solc",
"args": [ "--lsp" ],
"rootPatterns": [".git/"],
"filetypes": ["solidity"]
}
}
- or try the language server from vscode-solidity plugin. install the language server with
npm install -g @juanfranblanco/vscode-solidity-server
, then configure coc:
"languageserver": {
"solidity": {
"command": "vscode-solidity-server",
"args": [ "--stdio" ],
"rootPatterns": [".git/"],
"filetypes": ["solidity"]
}
}
Using svlangserver.
For installation, please check this section in the readme
Example settings file
{
"languageserver": {
"svlangserver": {
"module": "/INSTALLATION/PATH/lib/svlangserver.js",
"filetypes": ["systemverilog"],
"settings": {
"systemverilog.includeIndexing": ["**/*.{sv,svh}"],
"systemverilog.excludeIndexing": ["test/**/*.sv*"],
"systemverilog.defines" : [],
"systemverilog.launchConfiguration": "/TOOL/PATH/verilator -sv -Wall --lint-only",
"systemverilog.formatCommand": "/TOOL/PATH/verible-verilog-format"
}
}
}
}
Uses systemd-language-server.
Installation: pip install systemd-language-server
Settings file:
{
"languageserver": {
"systemd-language-server": {
"command": "systemd-language-server",
"filetypes": ["systemd"]
}
}
}
Using Terraform-LSP
. Build it and place it into any folder.
"languageserver": {
"terraform": {
"command": "terraform-lsp",
"filetypes": ["terraform"],
"initializationOptions": {}
}
}
Using terraform-ls
.
"languageserver": {
"terraform": {
"command": "terraform-ls",
"args": ["serve"],
"filetypes": ["terraform", "tf"],
"initializationOptions": {}
}
}
Using typst-lsp
.
"languageserver": {
"typst": {
"command": "typst-lsp",
"filetypes": ["typst"]
}
}
Using vls
. v ls --install
to install it.
"languageserver": {
"vlang": {
"command": "v",
"args": ["ls"],
"rootPatterns": ["v.mod", ".git"],
"filetypes": ["v", "vlang"],
}
}
Using vala-language-server
"languageserver": {
"vala": {
"command": "vala-language-server",
"filetypes": ["vala"],
}
}
Using coc-vetur
:CocInstall coc-vetur
Using yang-lsp
Make sure yang-language-server
is findable in your PATH
variable, otherwise specify the full path to the yang-language-server
executable
"languageserver": {
"yls": {
"command": "yang-language-server",
"filetypes": ["yang"],
"rootPatterns": ["yang.settings", ".git/"],
}
}
Using coc-zig
:CocInstall coc-zig
Using zls
Make sure zls
is findable in your PATH
variable, otherwise specify the full path to the zls
executable
"zig": {
"command": "zls",
"filetypes": [
"zig"
]
},
Using efm-langserver
Location of efm-langserver config.yaml is:
- UNIX:
$HOME/.config/efm-langserver/config.yaml
- Windows:
%APPDATA%\efm-langserver\config.yaml
efm-langserver config:
languages:
eruby:
lint-command: 'erb -x -T - | ruby -c'
lint-stdin: true
lint-offset: 1
format-command: 'htmlbeautifier'
vim:
lint-command: 'vint -'
lint-stdin: true
markdown:
lint-command: 'markdownlint -s'
lint-stdin: true
lint-formats:
- '%f:%l %m'
- '%f:%l:%c %m'
- '%f: %l: %m'
coc-settings.json:
"languageserver": {
"efm": {
"command": "efm-langserver",
"args": [],
// custom config path
// "args": ["-c", "/path/to/your/config.yaml"],
"filetypes": ["vim", "eruby", "markdown"]
}
}