Author here! I've considered this before, but there are problems such as what does a single atom mean:
Does
test.sh
run test.sh or returns the content of test.sh?
I am a heavy Lisper myself so I end up choosing consistency over one less keystroke (I have paredit, so the difference between (ls) vs ls is just one keystroke), but I could be biased.
That's a fair point. Consider that I have very little experience with any kind of Lisp beyond Scheme, and even that is very limited. I was coming at this recommendation from the perspective of a heavy command line user.
Comments
Author here! I've considered this before, but there are problems such as what does a single atom mean: Does test.sh run test.sh or returns the content of test.sh?
I am a heavy Lisper myself so I end up choosing consistency over one less keystroke (I have paredit, so the difference between (ls) vs ls is just one keystroke), but I could be biased.
That's a fair point. Consider that I have very little experience with any kind of Lisp beyond Scheme, and even that is very limited. I was coming at this recommendation from the perspective of a heavy command line user.
Do you use this as a daily driver?
I got addicted to evil-mode at some point and haven’t squared the circle of modal s-expression editing that this inevitably surfaces.
At least readline respecting programs reward my early emacs years.
Yes I use it daily.
I never squared the circle of modal editing and use mostly vanilla Emacs bindings (and paredit) :)
Would love to hear feedback from evil-mode user. Just in case I port it to vim in the future.
I use these keys every day for just about every sort of balanced delimiter manipulation I do in any language: https://github.com/fiddlerwoaroof/dotfiles/blob/eff889f0b749...
A little below I bind this key map to the “,” prefix and I’ve found my layout of paredit commands pretty ergonomic to use long-term.
would it be hard to modify the readline mechanism to show the "implicit" parens and push them apart as you type?