Clangd remote server

[ghost]

The feature is described here:

https://clangd.llvm.org/design/remote-index

I have a weird idea. I will setup the llvm-mingw cross compiler on Linux and configure CMake to cross compile to Windows using this llvm-mingw cross compiler. With the compile_commands.json generated by CMake, Clangd running on Linux will generate the index for Windows headers!

I then setup the Clangd remote server in Linux and set the local Clangd on my Windows to use the remote index. This way, my entrance price eMMC laptop can now do C++ development with much faster code completion!

In VS Code, you can configure Clangd settings in your project's .vscode/settings.json file. What about Ecode?

[SpartanJ]

ecode LSP currently does not allow per-project configuration of LSP servers (I'll add it, this can be quite useful) but you can configure the LSP globally in the ecode configuration file. Documentation is here.

You'll just need to add some command parameters, so you can just add those commands to "cland", this is for example my configuration:

{
  "config": {
    // ...
  },
  "keybindings": {
    // ...
  },
  "servers": [
    {
      "command_parameters": "--malloc-trim --background-index-priority=background --use-dirty-headers",
      "name": "clangd"
    }
  ]
}