Justin M. Keyes

Mainly want to avoid footnotes where possible. How can we document this as "options support Lua functions", instead of "options support functions if you use `vim.o`"? Maybe that means that...

See also https://github.com/neovim/neovim/issues/5054 # Proposal A short-term, high-leverage, low-cost improvement would be to add `:read :` which opens any `:` output into a buffer. Bram also was [favorable to that...

The door is open for those questions, but adding `:read :cmd` in "vimscript" should be rather low-risk, and makes sense as a first step before the other stuff. > do...

> The point of my comment is to make sure whatever is added also applies to `:=`. I would expect `:read :=...` to be supported since any `:read :` should...

> We could upstream a form of this without introducing another ex command in C. That's the idea. And this doesn't introduce any new Ex command, it only enhances `:read`...

Also, I mentioned that https://github.com/neovim/neovim/issues/5054 tracks more elaborate "builtin pager" changes. That is why I renamed this issue to track the more near-term solution of enhancing `:read`.

Hashing the filepath is indirection which makes troubleshooting/debugging more cumbersome. Is that the only reason we don't hash the filepaths? Maybe it's worth the tradeoff.

should be easy to modify loader.lua and try it out, for people having this issue.

I mean to modify vim.loader to use hashed paths (use `VIMRUNTIME=... nvim --luamod-dev` to test). Please, don't leave comments unless you have implemented that. This issue remains open until someone...

@mochsner or use `vim.loader.disable()` ?