Move justice transactions signature behind signer interface #560
Conversation
|
Rebased on top of #598 |
Codecov Report
@@ Coverage Diff @@
## master #560 +/- ##
==========================================
- Coverage 91.10% 91.08% -0.03%
==========================================
Files 34 34
Lines 20447 20520 +73
==========================================
+ Hits 18628 18690 +62
- Misses 1819 1830 +11
Continue to review full report at Codecov.
|
|
I'm really not a fan of the direction this is going in terms of OnchainTxHandler. More caching in OnchainTxHandler doesn't feel like the solution - its bad enough with the local_commitment and prev_local_commitment having almost entirely duplicated data but doing the same for remote commitment txn seems wrong. Its not that expensive to copy the data, and we shouldn't ever be in a state where we're tracking 10 different copies of remote HTLCs. |
…TxCache Used in next commits to avoid passing script between ChannelMonitor and OnchainTxHandler. ChannelMonitor duplicata will be removed in future commits.
As we can't predict if any and which revoked commitment tx is going to appear onchain we have by design to cache all htlc information to regenerate htlc script if needed.
As we cache more and more transaction elements in OnchainTxHandler we should dry up completly InputMaterial until them being replaced directly by InputDescriptor
By moving script generation inside OnchainTxHandler, we may dry-up further ChannelMonitor in next commits.
|
Closed by #610 |
1 similar comment
|
Closed by #610 |
2 participants
Build on top of #559.