Refactor ConstructedTransaction to contain Transaction
#4097
Conversation
|
👋 Thanks for assigning @wpaulino as a reviewer! |
Codecov Report❌ Patch coverage is
Additional details and impacted files@@ Coverage Diff @@
## main #4097 +/- ##
==========================================
+ Coverage 87.81% 88.71% +0.89%
==========================================
Files 176 177 +1
Lines 131770 133237 +1467
Branches 131770 133237 +1467
==========================================
+ Hits 115719 118196 +2477
+ Misses 13416 12326 -1090
- Partials 2635 2715 +80
Flags with carried forward coverage won't be shown. Click here to find out more. ☔ View full report in Codecov by Sentry. 🚀 New features to boost your workflow:
|
lightning/src/ln/interactivetxs.rs
Outdated
| fn add_local_witnesses(&mut self, witnesses: Vec<Witness>) { | ||
| self.inputs | ||
| self.tx | ||
| .input | ||
| .iter_mut() | ||
| .zip(self.inputs.iter()) | ||
| .enumerate() | ||
| .filter(|(_, input)| input.is_local(self.holder_is_initiator)) | ||
| .filter(|(_, (_, input))| input.is_local(self.holder_is_initiator)) | ||
| .filter(|(index, _)| { | ||
| self.shared_input_index | ||
| .map(|shared_index| *index != shared_index as usize) | ||
| .unwrap_or(true) | ||
| }) | ||
| .map(|(_, input)| &mut input.txin) | ||
| .map(|(_, (txin, _))| txin) | ||
| .zip(witnesses) | ||
| .for_each(|(input, witness)| input.witness = witness); | ||
| } |
There was a problem hiding this comment.
Wondering if we should have InteractiveTxSigningSession store both local and remote signatures until finalize_funding_tx is called. That way, ConstructedTransaction would always contain an unsigned transaction instead of one that may be partially signed.
There was a problem hiding this comment.
SGTM, we already store the local ones separately
InteractiveTxConstructor::new converts inputs given in channel.rs to InputOwned. Instead of converting FundingTxInput in channel.rs to the type needed by that constructor, pass in FundingTxInput and have it converted directly to InputOwned there. This saves one conversion and Vec allocation. It also lets us use the satisfaction_weight in an upcoming refactor.
Instead of estimating the input weight, track the satisfaction_weight in InteractiveTxInput. This then can be used compute the transaction weight without storing individual weights in NegotiatedTxInput in an upcoming commit.
The weight is no longer needed once a ConstructedTransaction is created, so it doesn't need to be persisted.
Instead of defining a custom function for calculating a transaction's weight, have ConstructedTransaction hold a Transaction so that its weight method can be re-used. As a result NegotiatedTxInput no longer needs to store the transaction inputs.
jkczyz
force-pushed
the
2025-09-interactive-tx-refactor
branch
from
f96f304 to
9eeaf52
Compare
|
Pushed |
|
🔔 1st Reminder Hey @wpaulino! This PR has been waiting for your review. |
lightning/src/ln/interactivetxs.rs
Outdated
| fn add_local_witnesses(&mut self, witnesses: Vec<Witness>) { | ||
| self.inputs | ||
| self.tx | ||
| .input | ||
| .iter_mut() | ||
| .zip(self.inputs.iter()) | ||
| .enumerate() | ||
| .filter(|(_, input)| input.is_local(self.holder_is_initiator)) | ||
| .filter(|(_, (_, input))| input.is_local(self.holder_is_initiator)) | ||
| .filter(|(index, _)| { | ||
| self.shared_input_index | ||
| .map(|shared_index| *index != shared_index as usize) | ||
| .unwrap_or(true) | ||
| }) | ||
| .map(|(_, input)| &mut input.txin) | ||
| .map(|(_, (txin, _))| txin) | ||
| .zip(witnesses) | ||
| .for_each(|(input, witness)| input.witness = witness); | ||
| } |
There was a problem hiding this comment.
SGTM, we already store the local ones separately
|
👋 The first review has been submitted! Do you think this PR is ready for a second reviewer? If so, click here to assign a second reviewer. |
Now that NegotiatedTxInput no longer holds a TxIn, rename it accordingly.
Now that ConstructedTransaction holds the actual Transaction, there's no need to keep track of and persist the outputs separately. However, the serial IDs are still needed to later reconstruct which outputs were contributed.
jkczyz
force-pushed
the
2025-09-interactive-tx-refactor
branch
from
9eeaf52 to
7958dd9
Compare
The output values are never used so there's no need to persist them. If needed, they can be re-computed though the shared output's local and remote amounts would be lost.
The local and remote input values are used to determine which node sends tx_signatures first. Instead of persisting these values, compute them only when needed from the input metadata. The spec states that the entire shared input value is to be included for the node sending the corresponding tx_add_input, so it isn't necessary to know the local and remote balances which the metadata does not contain.
InteractiveTxSigningSession currently persists holder witnesses directly, but persists counterparty witnesses as part of its unsigned ConstructedTransaction. This makes the ConstructedTransaction actually partially signed even though it is held in a field named unsigned_tx. Instead, persists the counterparty witnesses alongside the holder witnesses directly in InteractiveTxSigningSession, leaving the transaction it holds unsigned.
jkczyz
force-pushed
the
2025-09-interactive-tx-refactor
branch
from
7958dd9 to
86ed012
Compare
| .as_ref() | ||
| .map(|shared_input| SharedInputSignature { | ||
| holder_signature_first: shared_input.holder_sig_first, | ||
| counterparty_signature: None, |
There was a problem hiding this comment.
Should probably rename SharedInputSignature since it doesn't contain a signature anymore
| if let Some(shared_input_index) = self.shared_input_index { | ||
| let holder_shared_input_sig = | ||
| holder_tx_signatures.shared_input_signature.or_else(|| { | ||
| debug_assert!(false); |
There was a problem hiding this comment.
I think it'd be better to debug_assert_eq!(shared_input_index.is_some(), holder_tx_signatures.shared_input_signature.is_some()) to catch the other possible cases as well
ConstructedTransactiontracks the result of theInteractiveTxConstructionand is used during anInteractiveTxSigningSession. This PR refactors it such that it holds an actualTransactionrather than the contributed parts, though some metadata in the form ofNegotiatedTxInputandNegotiatedTxOutputneed to be maintained.This allows us to re-use the weight calculation provided by
Transaction. It also facilitates reconstructing contributed inputs and outputs to produce aSpliceFailedevent in #4077.