Johannes Altmanninger

> That hook does not do anything for me. The main issue is I'm not sure how I would go about remove exactly what was inserted. yeah I forgot to...

> Toggling autocomplete on/off feels worse than just triggering manual completion whenever I feel like. you can do `set global autocomplete prompt` enable it explicitly with ``, and maybe add...

> I still believe having this available by default would be consistent with the other forms of completion. what behavior exactly? I don't expect that "disabling autocomplete" will be fruitful...

I'm not sure if that makes sense. Note that ` f` will be invalidated as soon as you type anything. The `option=some_completions_opt` completions are more powerful because they have a...

> __fish_complete_subcommand I think that one uses `string escape` already? Is there a problem left? (Well there is probably one for multiline tokens; could probably tweak the `string escape` invocation...

would be interesting to know if this would improve practical cases (esp. cross-shell compatibility). Until then, I'm not sure we need to track this issue

what's the output of fish -c 'type -a apropos' fish -c 'fish_trace=1 __fish_describe_command aa' Your error can happen if you defined a function called "type", shadowing the builtin of the...

On Sat, Mar 30, 2024 at 10:22:13PM -0700, Kovid Goyal wrote: > On Sat, Mar 30, 2024 at 09:47:12PM -0700, Mahmoud Al-Qudsi wrote: > > Fixed (on our end) with...

> Are there any more modes I should bind in? What is the 'default' mode? you want to bind in all of `for mode in (bind --list-modes | string match...

> fish_cursor_end_mode can't use that unfortunately because it's only defined when using Vi key bindings. (Note that `fish_vi_key_bindings` affects future shells which is very surprising I think -- need to...