WIP: Htlc v3 taproot #471
WIP: Htlc v3 taproot #471
Conversation
carlaKC
left a comment
carlaKC
left a comment
There was a problem hiding this comment.
Tapscript changes looking good! However the structure of this PR makes it quite difficult to review in depth, needs some cleanup/ refactors before next review please.
| } | ||
| } | ||
|
|
||
| func (h *HtlcScriptV3) IsSuccessWitness(witness wire.TxWitness) bool { |
swap/htlc_test.go
Outdated
| <reciever_key> OP_CHECKSIGVERIFY OP_SIZE 20 OP_EQUALVERIFY OP_RIPEMD160 <hash> OP_EQUALVERIFY 1 OP_CHECKSEQUENCEVERIFY | ||
| */ | ||
| func createClaimPathLeaf(t *testing.T, recieverHtlcKey [32]byte, swapHash lntypes.Hash) (txscript.TapLeaf, []byte) { |
There was a problem hiding this comment.
Rather than storing claimScript/timeoutScript in HtlcScriptV3, move these functions into htlc.go and then we can just recreate them using our receiver/sender key and cltv when we need these scripts?
| return nil, err | ||
| } | ||
|
|
||
| randomPub, _ := hex.DecodeString( |
There was a problem hiding this comment.
If we're going to deploy these chagnes incrementally (ie, do tapscript then musig keyspend) this is going to need to be deterministic / communicated between client/server so that each party knows the other doesn't know the privkey (and can just claim the htlc).
Probably not worth the effort, so would just add a clarifying comment that this is a dummy value that should not be used in prod.
swap/htlc.go
Outdated
|
|
||
| // TODO: Unsilence errors | ||
| controlBlockBytes, _ := h.genControlBlock(h.claimScript) | ||
| controlBlockBytes, _ := h.genControlBlock(h.timeoutScript) |
swap/htlc.go
Outdated
|
|
||
| // TODO: Unsilence errors | ||
| controlBlockBytes, _ := h.genControlBlock(h.timeoutScript) | ||
| controlBlockBytes, _ := h.genControlBlock(h.claimScript) |
There was a problem hiding this comment.
likewise, seems like these have switched?
| // SenderKey is the key of the sender that will be used in the on-chain | ||
| // HTLC. | ||
| SenderKey [33]byte | ||
| SenderKey []byte |
There was a problem hiding this comment.
move this commit to before we add HtlcScriptV3 + don't intermingle refactors with logical changes (eg SuccessSequence returns 10 now?)
|
|
||
| // A human-readable message received from the loop server. | ||
| string server_message = 6; | ||
| string server_message = 7; |
There was a problem hiding this comment.
changing the proto number of server_message will break the API for old clients.
| HtlcV2 | ||
|
|
||
| HtlcV3 | ||
| HtlcV3 = 10 |
There was a problem hiding this comment.
why not use the existing iota?
|
Closing as #477 has taken over |
3 participants
In this PR we introduce V3 loop out and loop in submarine swaps via taproot. The general approach has been to implement v3 htlcs in
htlc.goand patch the plumbing from loop -> loopserver to accommodate the extra values. In some cases we default to creating taproot htlcs instead of traditional p2sh, which may be something a keen reviewer might want to look into, should we decide not to go all in on taproot for the next release.TODOs
Known Issues: