friendly-snippets VS cmp-buffer

Those definitely seem to be coming from friendly-snippets, so it seems like it's not being disabled. You can verify this with the :Lazy command to bring up the lazy.nvim menu then checking log or debug to see what is loaded, when, and why.

Here you go :), tested with friendly-snippets lazy loaded (Non lazy-loaded is 500ms)

Snippets for code comments.

I am trying to add snippets for python, i have LuaSnip and friendly-snippets installed, but for some reason it does not load the snippets. This is how i load the plugins:

I just sent [this PR](https://github.com/rafamadriz/friendly-snippets/pull/301) that implements the [full TSdoc specification](https://typedoc.org/guides/overview/) in [friently-snippets](https://github.com/rafamadriz/friendly-snippets). In practice what you get is autocompletion in your typescript comments, which should be a nice of life improvement for most typescript developers.

You mean snippets? If yes, you can try Luasnip and friendly-snippets with nvim-cmp and here's the setup guide. Hope it helps

For example Luasnip with (I guess) friendly-snippets has multi line snippets defined by default. And you can use luasnip with nvim-cmp (read the docs or I guess there’s a YouTube tutorial)

If you copied the config in the readme, then the snippet come from friendly-snippets.

I would recommend lsp-zero, the tsserver lsp provides code actions for auto import. There is auto import when confirming completion as well, I've seen it working with cmp, however I can't remember If that is out of the box behaviour. For snippets, I would go with LuaSnip - It's very versatile snippet engine and it integrates well with many snippet formats. There are specifically react es7 snippet definitions in the friendly-snippets repo which can be used by LuaSnip - https://github.com/rafamadriz/friendly-snippets/blob/main/snippets/javascript/react-es7.json

sources = cmp.config.sources({ { name = "nvim_lsp", priority = 1, group_index = 1 }, { name = "luasnip" }, { group_index = 2, name = "buffer", priority = 5, option = { -- https://github.com/hrsh7th/cmp-buffer get_bufnrs = function() return vim.api.nvim_list_bufs() end, }, }, { name = "nvim_lsp_signature_help" }, { name = "path" }, }),

cmp-buffer: Completion source. It provides suggestions based on the current file.

Then what is the purpose of using cmp-buffer?

Okay I sat down and did exactly what I said in my previous comment. Essentially it's a complete rewrite of my old code that makes things much more generic and flexible. You can take a look at the PR here: https://github.com/hrsh7th/cmp-buffer/pull/60.

Good work but can you tell how is it different from https://github.com/hrsh7th/cmp-buffer

hrsh7th/cmp-buffer - nvim-cmp source for buffer words

cmp-buffer

nvim-cmp is just the autocompletion engine w/ fuzzy matching and other filtering options. To get items to be filled from cmp you will need sources, for example: hrsh7th/cmp-nvim-lsp plugin will provide items from a LSP server to cmp, so you can get items you can fuzzy match in. Another one is items from buffers: hrsh7th/cmp-buffer