Can you try to replace uses of this function with __fish_seen_subcommand_from?

I believe if there are multiple levels of subcommands, it's okay to use __fish_seen_subcommand_from subcmd && __fish_seen_subcommand_from subsubcmd (not always fully correct but good enough)

Do you mean deleting this whole __fish_sops_commands function and use __fish_seen_subcommand_from instead?

Do you mean deleting this whole __fish_sops_commands function and use __fish_seen_subcommand_from instead?

yep

even better, replace it with a function that uses argparse to decide which subcommand is active (similar to __fish_sudo_print_remaining_args)

The reason why I didn't use __fish_seen_subcommand_from subcmd1 && __fish_seen_subcommand_from subcmd2 for multiple levels subcommands is because it accepts anything as long as there are the word subcmd1 and subcmd2 ignoring the placement. For example, if I want group as subcmd1 and add as subcmd2 then even if I write it as sops blablabla add groups it will still return true. While the function that I'm using it right now (I took the code from __btrfs_commands and __btrfs_command_groups) will return false unless the command is sops groups add and nothing else.

I might go with the even better "similar to __fish_sudo_print_remaining_args" approach, but it might need some time because I'm still figuring out how does argparse work. T.T

Ok makes sense. I think copying __fish_sudo_print_remaining_args is the way to go. We should probably add more examples to the argparse docs.

I have modified the way it works, please have another look @krobelus. I don't know if the code is good as I'm not experienced enough for this (not even sure what I did lol), please correct my code if needed. :)