openlawlibrary/pygls

Add some project info to docs

Closed this issue · 5 comments

Hi.
I'd like to request some background info added to the docs. Specifically, about the pygls inception and development goals.
I'm considering options for developing a language server, and honestly, I'd much rather deal with Python than JS, but I'm not quite clear how's pygls compared to MS vscode-languageserver.

I assume it's not a core project for OpenLawLibrary, but rather something to support your actual project, whatever that is?

I see a list of implemented features, but is there a list on unimplemented ones, maybe even a roadmap? What's missing in pygls compared to vscode-languageserver? And vice versa?

Any performance, compatiblity considerations?
Python 3.7 should still run on windows 7, I think. Any plans to drop that?

Just to be clear, I'm not trying to be imposing/demanding. I understand that open source is open source. Simply want to get a clearer picture of the project, and I think that would be helpful to other people, too.

tombh commented

This is a really good point. Thank you. I'll research and respond soon.

augb commented

@tombh I'll reach out to you via email to provide some background information on the origination of the project.

tombh commented

So I've made a PR, #315, that addresses some of the questions here and more.

Pygls' inception is now addressed in a HISTORY.md file, thanks to @augb.

There's now also an Alternatives section that should address the pros/cons of Pygls compared to its competitors/alternatives.

@burner1024 Those implemented features that you linked to are merely for the example VSCode extension. In actual fact, with Pygls' recent v1 release that depends on Microsoft's official lsprotocol types, we should in theory both support all of LSP's current and potential future features.

I'm not very familiar with Python's performance characteristics, other than it's a scripting language, so compared to Language Servers written in compiled languages, its performance is, like NodeJS, going to be relatively slower, but certainly not impractical. I don't know of any reasons to drop Python 3.7 support in the near future.

Cool, thanks!

tombh commented

Thanks for asking ✨