neovim-ui VS pylance-release

mjlbach/neovim-ui

I understand the frustration (making neovim more cohesive is definitely a goal). It sounds like coc provides the UI you are used to, so there's no harm in sticking with that!

Many of our users explicitly don't want automatically called functions that would slow down the editor (autocommands that map signature requests to the language server, for example), so by nature neovim's core implementation is extremely conservative.

One thing I would like to do, is make the automatic pop-ups for signature easier to implement with our current handler, which means a plugin like signature-x could use our upcoming lsp.config option to configure it's borders (https://github.com/neovim/neovim/pull/14681), and match the rest of the UI.

I also have another project I was working on before the 0.5 stabilization phase (https://github.com/mjlbach/neovim-ui). The goal with this is to have composable/overridable UI elements built into core (which we would use for our internal lsp functions), that can be used (or overridden) by UI plugins.

In summary, I think the likelihood of autocompletion (and generally auto-anything) being built-into core is very small, but providing the APIs in neovim core to make snippets - autocompletion - automated UI elements easier for plugin authors is a high priority.

Yep, I started working on this awhile ago https://github.com/mjlbach/neovim-ui but put it on hold for 0.5 stabilization.

I am working on this, but it will probably have to wait until after the 0.5 release (I don't know when this will be). This is part of the larger UI effort in neovim-ui which is currently glorified vaporware. As far as I know, TJ, Smolck, and I are the only ones who have expressed interest in fixing this.

One of the things that comes to mind here is the fact that the default Python extension for VS Code is, perhaps surprisingly to many, not open source. https://github.com/microsoft/pylance-release

While it's possible to fork VS Code, it is not possible to fork VS Code and provide a seamless onramp towards a Python editing experience that is fully open source, because users are used to the nuances of the closed-source Pylance experience in VS Code proper. You could use the minified/compiled Pylance plugin in your fork, but you'd have no way to expand its capabilities to new hooks your fork provides. Microsoft's development process would always be able to move faster than a fork, because it could coordinate VS Code internal API development with its internal Pylance team, and could become incompatible with forks at any time.

It's worth re-reading the quote from J Allard in https://en.wikipedia.org/wiki/Embrace,_extend,_and_extinguis... with this modern example in mind.

(Also worth mentioning https://github.com/detachhead/basedpyright?tab=readme-ov-fil... which is a heroic effort to derisk this, but it's an uphill battle for sure!)

Anyone know how can we fix this issue if we build the vscode locally

Do any of these issue tickets explain the behaviour you're seeing? https://github.com/microsoft/pylance-release/issues/4341, https://github.com/microsoft/pylance-release/issues/4071

Maybe they "are the same thing" in terms of behavior (I don't know), but "A uses B" doesn't mean that "A is B".

One important difference in this case is that while "Pylance leverages Microsoft's open-source static type checking tool, Pyright" [1], Pylance itself is not open source. In fact, the license [2] restricts you to "use [...] the software only with [...] Microsoft products and services", which means that you are not allowed to use it with a non-Microsoft open source fork of VS Code, for example.

The license terms also say that by accepting the license, you agree that "The software may collect information about you and your use of the software, and send that to Microsoft" and that "You may opt-out of many of these scenarios, but not all".

[1] https://github.com/microsoft/pylance-release

[2] https://marketplace.visualstudio.com/items/ms-python.vscode-...

There's this one: https://github.com/microsoft/pylance-release/issues/4174 (rules don't apply properly, and ovverrides don't work even after being set, this is especially for the more generic ones like )

The example is not .NET in general, but that specific event when Microsoft reneged on open development tooling[1]. For some people, that was the moment they stopped trusting "new Microsoft" to keep their word (though for me, it was when the Python language server was replaced with a DRM-locked, LSP-noncompliant one[2] a bit before that; unlike .NET hot reload, they didn't backtrack there). I can think the company makes great open .NET tools and at the same time not trust them to close it down on a whim.

Does anyone know where the open xlang reimplementation of MIDL went[3], by the way? (Unlike 1990s MIDL, you can't reimplement this one from the language grammar in the docs, because there is no language grammar in the docs.)

[1] https://dusted.codes/can-we-trust-microsoft-with-open-source and links there

[2] https://github.com/microsoft/pylance-release/issues

[3] https://github.com/microsoft/xlang/pull/529