John Hawthorn
John Hawthorn
@catwell we are no longer using this gem. Ruby has had a built-in and efficient `Process.spawn` for a long time so I don't think there's any reason to use this...
Good idea. I would accept a PR which added `--only-exact` (or something similar) but I'd like to keep the default behaviour.
I'd definitely consider adding this. What behaviour you expect in this mode if one of the choices has a newline?
They're ignored, but I'm not sure that's the best behaviour either
I'm no familiar with AppImages, would that provide any advantages over just a static binary of fzy? (if that's even necessary... fzy only links against `libc` and `pthread`)
This seems to hurt the benchmark from #44469 a little bit... still faster than before that PR, but measurable. Maybe due to the extra filtering? Before: ``` $ be ruby...
One option might be creating a new group_klass for these `silencable?` subscribers. They're all the same type, so it should just be one more type of group, that way we'd...
Yep! That's correct. We can't use available records for a scoped association as there's no way to be sure that they match, so this is working as intended. Please note...
Updated to just call `find_by_sql`. This is a slight behaviour change in that we'll now print out ex. "User Load" instead of "User Load via GitHub::SQL". This move us to...
> `Array#max` which creates an Array instance which must be later garbage collected. Actually for very simple cases like this where `max` is called on an new array literal there...