Matt Corallo
Matt Corallo
cc https://github.com/lightning/bolts/issues/964 but we need to update our shutdown sending/handling - we should accept a shutdown even if there's pending updates, and we also need to consider update_fail/update_fail_malformed and update_fee...
It should be checking if `low == usize::MAX as u64` not `low == 0`. Should be harmless cause our uses of it shouldn't care about the value, only that its...
After handling #2706 locally, 44% of my allocations locally are from path building when using the MonitorUpdatingPersister! That's obviously nuts, and luckily we can cut this down with careful buffer...
We currently panic in spend_spendable_outputs if the script doesn't match, which is a bit strange given we tell users explicitly they should override the coop close scripts. Instead we should...
....but this makes no sense for MPP. Its still interesting to have, but we should either share the vec of inbounds or we should at *least* update the docs to...
If we have an inbound channel where funding timed out and we have zero balance, the monitor can be deleted, so `get_claimable_balances` should return an empty set.
See https://github.com/lightningdevkit/rust-lightning/pull/1351#discussion_r828514699
Some LSPs running LDK want to give users scorer data directly from their node. Sadly, even if they're probing, this data will not include the channels the LSP has, which...
We should build a second `get_route`-like method that, instead of seeking to send funds from one node to another, seeks to balance the funds in a set of `ChannelDetails` objects...
The `Cannot generate an open_channel after we've moved forward` panic is now reachable as of ddf75afd16. This isn't on 0.0.121 so we just need to fix it before the next...