Error handling api design

with that proposal it would look like something this:

await lh.pick_up_tips(
  tip_rack["A1"],
  use_channels=[1],
  error_handler=sequential([
    retry(),
    retry(),
    retry(),
    z_probe(),
  ])
)

(this is just a sketch, not actually implemented in the PR)

Sweet. Ok one last thing. So on the Venus side, horrible check box aside, grouping the errors by type then allowing for a different sequential approach is rather handy.

I would never want to retry if a hardware error was returned, but would if missing tip. How do we get around this without writing out each check box.

in the proposal, have a different function wrapped around retry() that checks the error and conditionally calls the handler

1 Like