Daniel Mendler

Yes, storing the prefix/suffix would be good to generalize this to arbitrary completion tables using completion-boundaries. Are there any?

> I don't know. Probably the whole scenario of storing a completions buffer long past the closing of its minibuffer is fantasy anyway. Right, crm is an obvious example. Should...

Ah okay, I see. If the action selects it should keep its selection. I am actually unsure which behavior feels more consistent if one considers all the actions. Keeping the...

The workaround you propose is good, but I like it so much that Embark just reuses the plain old stuff we already have :)

I also expected display-buffer to be a function!? But does it always do the right thing? I mean display-buffer is configurable so it could do anything?

> Would a global choice be enough or would it need to finer-grained, per action? Global is good enough since on the per action level we can simply replace the...

I think it is better to prevent this package from being used on Emacs 28. Having both Emacs 28 icomplete-vertical and this package will only lead to conflicts. The Emacs...

> If I followed your suggestion of preventing this package from being installed on Emacs 28, that would have the advantage of not requiring this package to be renamed. Yes,...

> I was thinking of grouping, but I guess that would only affect people simultaneously using Emacs 28, consult and icomplete-vertical. There can't be too many of those. Yes. And...

> As you know, I also prefer Vertico and Selectrum to Icomplete-vertical, but they do have this weird UI that doesn't cycle through candidates and instead "helpfully" gets stuck at...