Closed jp1ac4 closed 8 months ago
I still think it's wrong after this. Why are we adding the lower bound thing to it. In the case of no drain the metric should just be: (cs.selected_value() - self.target.value) as f32
i.e. the fee. Then with drain it should be (input_weight + base_weight + drain_weight) * feerate
.
cc @evanlinjin
Just noting that I've talked with @evanlinjin and I'm now in charge of fixing it.
Thanks for the update. Feel free to close this PR if you'll be opening a new one.
@evanlinjin can you close this PR and give me permission to manage this repo.
This is to fix the part of #6 relating to https://github.com/bitcoindevkit/bdk/pull/1072#discussion_r1393687511.
For the
LowestFee
metric, it adds the excess to changeless solutions instead of those with change.