chat VS rust-analyzer

TUIs over ssh/telnet can be a lot of fun. Especially in cases where multiple people can interact with each other on the server. It simplifies the programming model as you only have one state on the backend that you render to multiple connections. Syncing up everyone becomes trivial. You can also use some React concepts, like rendering a virtual TUI and sending just the right set of minimal escape sequences back to the user to bring their display up to date.

A few months ago I implemented a telnet chat server[0] for fun and it was surprisingly easy to do so. Even by using a wasm vm that I was still working on at the same time.

[0]: https://github.com/lunatic-solutions/chat

We are investing a lot of effort into making Lunatic feel native to the particular language and ecosystem. If you look at the Rust chat server we built in Lunatic (https://github.com/lunatic-solutions/chat), it fully integrates with cargo. You just run your typical “cargo run” command, it will compile the app to wasm and use lunatic to run it. If you want to run your test, you can just do “cargo test”.

wasm-bindgen is necessary only because it’s really hard right now to merge the wasm world and the JS one in the browser. We have the advantage here of staying out of the browser.

It took me around a week to build it with Rust + Lunatic and you can check out the code here. If you would like to try it out you can connect to it with:

Recently I implemented a command line chat server in Rust using an actor framework. I model each TCP connection as an actor.

Hi HN,

I'm working on an Erlang inspired WebAssembly runtime for the backend[0]. Recently I added TCP support and was looking for apps I could build with it. I ended up building a telnet line chat app. It was a great dogfooding experience and has a nice retro feel to it.

You can access the US server with:

> telnet lunatic.chat

or the EU one with:

> telnet eu.lunatic.chat

Pick the one closer to you, as all the rendering is done on the backend and lower latency will mean better UX.

The server is open source[1] and written in Rust. The Rust code is then compiled to WebAssembly and runs on top of Lunatic. Each connection runs in a separate (lightweight) process, has it's own state and sends just a diff of esc-sequences back to the terminal to bring it up to date with the current render buffer. Everything is deployed to an ARM Linux box.

[0]: https://github.com/lunatic-solutions/lunatic

[1]: https://github.com/lunatic-solutions/chat

The server is open source and written in Rust. The Rust code is then compiled to WebAssembly and runs on top of Lunatic. Each connection runs in a separate (lightweight) process, has it's own state and sends just a diff of esc-sequences back to the terminal to bring it up to date with the current render buffer. Everything is deployed to an ARM Linux box.

#14561 map tokens from include! expansion to the included file

return { tools = { -- autoSetHints = false, on_initialized = function() vim.api.nvim_create_autocmd({ "BufWritePost", "BufEnter", "CursorHold", "InsertLeave" }, { pattern = { "*.rs" }, callback = function() vim.lsp.codelens.refresh() end, }) end, auto = false, inlay_hints = { -- Only show inlay hints for the current line only_current_line = false, auto = false, -- Event which triggers a refersh of the inlay hints. -- You can make this "CursorMoved" or "CursorMoved,CursorMovedI" but -- not that this may cause higher CPU usage. -- This option is only respected when only_current_line and -- autoSetHints both are true. only_current_line_autocmd = "CursorHold", -- whether to show parameter hints with the inlay hints or not -- default: true show_parameter_hints = false, -- whether to show variable name before type hints with the inlay hints or not -- default: false show_variable_name = false, -- prefix for parameter hints -- default: "<-" -- parameter_hints_prefix = "<- ", parameter_hints_prefix = " ", -- prefix for all the other hints (type, chaining) -- default: "=>" -- other_hints_prefix = "=> ", other_hints_prefix = " ", -- whether to align to the lenght of the longest line in the file max_len_align = false, -- padding from the left if max_len_align is true max_len_align_padding = 1, -- whether to align to the extreme right or not right_align = false, -- padding from the right if right_align is true right_align_padding = 7, -- The color of the hints highlight = "Comment", }, hover_actions = { auto_focus = false, border = "rounded", width = 60, -- height = 30, }, }, server = { --[[ $ mkdir -p ~/.local/bin $ curl -L https://github.com/rust-analyzer/rust-analyzer/releases/latest/download/rust-analyzer-x86_64-unknown-linux-gnu.gz | gunzip -c - > ~/.local/bin/rust-analyzer $ chmod +x ~/.local/bin/rust-analyzer --]] -- cmd = { os.getenv "HOME" .. "/.local/bin/rust-analyzer" }, cmd = { os.getenv "HOME" .. "~/.cargo/bin/rust-analyzer" }, on_attach = require("user.lsp.handlers").on_attach, capabilities = require("user.lsp.handlers").capabilities, settings = { ["rust-analyzer"] = { lens = { enable = true, }, checkOnSave = { command = "clippy", }, }, }, }, }

I would like changes like https://github.com/rust-analyzer/rust-analyzer/pull/13799 to be listed in 'Breaking Changes' category, to приманка draw the users' attention.

For those of you who haven’t heard of Mun before, Mun is an embeddable programming language empowering creation through iteration. The idea to create Mun originated out of frustration with the Lua dynamic scripting language and a desire to have similar hot reloading functionality available in Rust. As such, it’s not a direct competitor with Rust, but instead is intended to be used with Rust (or C/C++) as a host/embedded language pairing. Actually, Mun is completely written in Rust, building on similar crates as rust-analyzer and rustc. Its key features include:

#13728 upgrade chalk to make solver fuel work again (works around most trait solving hangs).

#13221 (first contribution) add option to move lenses above doc comments (rust-analyzer.lens.location):

Rust-analyzer has a good overview: https://github.com/rust-analyzer/rust-analyzer/blob/master/docs/dev/syntax.md

#12517 (first contribution) fix completion for methods in trait generated by macro.

-- all the opts to send to nvim-lspconfig -- these override the defaults set by rust-tools.nvim -- see https://github.com/neovim/nvim-lspconfig/blob/master/doc/server_configurations.md#rust_analyzer server = { -- on_attach is a callback called when the language server attachs to the buffer -- on_attach = on_attach, settings = { -- to enable rust-analyzer settings visit: -- https://github.com/rust-analyzer/rust-analyzer/blob/master/docs/user/generated_config.adoc ["rust-analyzer"] = { -- enable clippy on save checkOnSave = { command = "clippy" }, assist = { importGranularity = "module", importPrefix = "self", }, cargo = { loadOutDirsFromCheck = true }, procMacro = { enable = true }, } } },

#12349 publish universal VSIX to make Code happy.