thanks!

to git@git.d2evs.net:~ecs/madeline
  0d00d20..7376e8a  madeline -> madeline

thanks!

to git@git.d2evs.net:~ecs/madeline
  17a6c61..0d00d20  madeline -> madeline

thanks!

to git@git.d2evs.net:~ecs/madeline
  2074061..ad5fc67  madeline -> madeline

also pushed some reorganization of the splitter api, you'll need to
update hare-c for that

yeah ik it's not standard but it's the right behavior imo, i frequently
lose messages to that ^y behavior in weechat

we should store the kill ring as a [][]u8, so that ^w^w^y^y is a no-op

On Tue Feb 28, 2023 at 10:49 PM UTC, Sebastian wrote:
> One other option would be to have made::line *only* take in a config
> (maybe config could be renamed to context), and then the kill ring can
> just be stored in the context. I can't think of very many use cases for
> just passing in a prompt without a config, and when that is desired it's
> pretty easy to just, make an empty config.

+1 to that (and +1 to renaming to context). the goal was to make
made::line("prompt") mostly work with minimal effort, but i think
requiring a full context struct makes more sense in light of the kill
ring

thanks!

To git@git.d2evs.net:~ecs/madeline
   04c2052..872fa27  madeline -> madeline

thanks!

To git@git.d2evs.net:~ecs/madeline
   04c2052..872fa27  madeline -> madeline

hm, i think we should keep ^d like it is now (albeit probably fixing
s.hidx like you did with ^g), but i'm open to adding ^g as well

thanks!

To git@git.d2evs.net:~ecs/madeline
   9564bb2..e29d1ee  madeline -> madeline

fish only inserts the next word on ctrl/alt+right, and i'd like to keep
that behavior. can you also merge this with the ^f and ^b patches?