Implement readline's yank-last-arg

[pkgw]

The implementation of readline's yank-last-arg command is currently a TODO. It would be good for it to actually get implemented!

[haridsv]

How about creating a patch based on this gist ? I doesn't currently cycle back to the beginning, but it should be easy to modify it. It is also not identical to the readline's behavior, specifically on how readline recalls arguments from (n-1)th item if you are currently at nth item, but honestly, I prefer always starting from -1 irrespective of where you are in the history.

@jonathanslenders

[haridsv]

The gist also takes care of yank-nth-arg, though I noticed that the issue #359 makes it less useful.

[adamchainz]

Thanks for the gist! I adapted it into my implementation in #407 which uses function 'local' variables rather than a separate state and also deduplicates on repeat presses like readline does.

[jonathanslenders]

Thanks everyone for the pull requests! So far, I did not had any time yet to review/merge this functionality, but I hope to do it soon. It's very useful!

[jonathanslenders]

Hi everyone, this functionality has been mergen in 0c09bf0

If you have time, could you have a look and see whether it works?

Jonathan

[amitkot]

Can this useful feature be used with vi binding as well?

I'm using Xonsh and for the time being I added this binding:

@events.on_ptk_create
def custom_keybindings(bindings, **kw):

    @bindings.add('escape', '.')
    def yank_last(event):
        event.current_buffer.yank_last_arg()

[adamchainz]

Please don’t comment on eight year old issues. Instead, open a new one referencing this one...