Interactive Fuzzy-Filter or Keyboard-Driven Prompt

[KyleKing]

I'm new to Rich and I wondered if there was a way to create an interactive prompt. Either keyboard-driven or fuzzy-matching?

---

I saw #2192 where fuzzy-filtering is implemented (which I'm using for now), but I would like to figure out a way to live-filter the choices

I'm looking for something like gum's gum choose "fix" "feat" "docs" "style" "refactor" "test" "chore" "revert"

[KyleKing]

TLDR: not supported in rich (#392 (comment) & #266 (comment)), but possible with textual (#1843 (comment)) or could be done with questionary. This is partially related: #960

However, after looking at a bunch of the rich example apps, it seems like a common workaround is to provide indexing, so you could do something like:

choices = ["fix","feat","docs"]

for idx, choice in enumerate(choices):
    console.print(f"{idx}. {choice}")
selection = Prompt.ask('Which option?', choices=[*map(str, range(len(choices)))], default="0")

Look like:

0. fix
1. feat
2. docs
Which option? [0/1/2] (0):

[FergusFettes]

With iterfzf you can drop into a terminal and use fzf there quite easily, I'm integrating it with rich just now and it works well:

    import iterfzf
    choices = ["apple", "banana", "cherry", "date", "elderberry"]
    selected = iterfzf.iterfzf(choices)
    print(selected)

[KyleKing]

That's a good find! Unfortunately looks like there hasn't been a recent release though and it pins a very old version of fzf: https://github.com/dahlia/iterfzf/blob/d030d71a6748b8507735458e0acadcddc7a6cd01/setup.py#L23-L26

[FergusFettes]

tbh, just using subprocess with fzf also works pretty well :D

import subprocess
choices = ["apple", "banana", "cherry", "date", "elderberry"]
p = subprocess.Popen(["fzf"], stdin=subprocess.PIPE, stdout=subprocess.PIPE)
output, _ = p.communicate("\n".join(choices).encode())
print(output.decode())