Felicián Németh

I think there is a bug in Eglot, so this issue shouldn't be closed. Even if smart-jump is a good workaround. At least it wouldn't hurt to left it open...

I haven't followed this discussion, but https://github.com/emacs-helm/helm/wiki says "helm-mode enables Helm completion globally for any Emacs command using completing-read or read-file-name." So, if something wrong with integrating Helm with Eglot,...

Sure, the language server protocol was designed differently than Emacs' completion system, but there are clients that found a way to cope with the protocol. So I do not think...

> Exactly how did those clients manage to "cope with the protocol"? It's hard for me to say because I've only seen animgifs of them. They are probably better at...

I think we can enhance Eglot with the following compromise. Introduce two project-local variables with names like `eglot-completion-bound-function` and `elgot-completion-bound-error-behavior`. The bound-function is `bounds-of-thing-at-point` by default, but there's another choice...

Does it print anything if you start it from the command line?

I'm on testing and it (obviously) works for me. Can you debug the code? (For example, by adding a couple of "print" lines to ldoce5viewer/qtgui/utils/soundplayer.py) Which backend it decides to...

> @nemethf What do you think about changing/fixing the behavior to close the doc buffer at the end even when `company-auto-update-doc` is non-nil? The documentation is helpful to find the...

So `C-u ` would cycle through three options instead of the current behavior when it toggles between on and off? I can try to prepare a pull request. (But I...

In the end, I took a different approach. In the pull request, auto-update-doc always restores the window configuration when company finishes. But I also added a new command that shows...