Export various fields and structures by devrandom · Pull Request #662 · lightningdevkit/rust-lightning

devrandom · 2020-08-02T15:56:36Z

This is for discussion purposes at this time. This is a set of patches we maintain on top of master in order to let Rust Lightning Signer do various things, including:

access to TxCreationKeys~~fields and~~ constructor, so that we can use it to call various InMemoryChannelKeys methods
access to LocalCommitmentTransaction constructor for InMemoryChannelKeys use
export the unsafe methods for testing use, gated by a new feature named "unsafe"
access to node and channel announcement fields so that the contents can be checked by the signer

codecov · 2020-08-02T16:25:08Z

Codecov Report

Merging #662 into master will decrease coverage by 0.00%.
The diff coverage is 75.00%.

@@ Coverage Diff @@## master #662 +/- ##
==========================================
- Coverage 91.36% 91.35% -0.01% 
==========================================
Files 35 35 Lines 21681 21688 +7 ==========================================
+ Hits 19808 19813 +5 - Misses 1873 1875 +2

Impacted Files	Coverage Δ
lightning/src/chain/keysinterface.rs	`95.04% <ø> (ø)`
lightning/src/ln/channelmonitor.rs	`95.70% <ø> (ø)`
lightning/src/ln/onchaintx.rs	`93.94% <ø> (ø)`
lightning/src/util/enforcing_trait_impls.rs	`100.00% <ø> (ø)`
lightning/src/ln/chan_utils.rs	`95.97% <75.00%> (-0.28%)`	⬇️
lightning/src/ln/functional_tests.rs	`97.00% <0.00%> (ø)`

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update 22a0dd5...fd2db40. Read the comment docs.

TheBlueMatt

Mostly questions about use-cases.

ariard

Apart of unsafe feature in discussion, Concept ACK

devrandom · 2020-08-08T17:07:42Z

Removed message field visibility change - will be a separate PR.

Added more detailed commit messaging and removed the draft marker.

TheBlueMatt

I'm fine with this as-is, though I think we need to newtype TxCreationKeys because the APi definitely encourages "unsafe trusting" of the in-RL-derived keys when doing external signing.

I'd prefer a more descriptive name over unsafe, though. Maybe unsafe_revoked_tx_generation or something?

devrandom · 2020-08-09T13:43:38Z

I renamed unsafe to unsafe_revoked_tx_signing.

~~I think the PreCalculatedTxCreationKeys change should go in a separate PR, because the changes (especially involving LocalCommitmentTransaction) will dwarf the other things in this PR.~~ The PreCalculatedTxCreationKeys changes ended up smaller that I predicted, please take a look at #664.

TheBlueMatt

Looks good except for new_missing_local_sig, which could use a bit of clean up if its going to be pub.

Allows unsafe signing in dev code, such as functional testing of justice transactions outside our crate.

Allows calling of InMemoryChannelKeys methods

…_sig

devrandom · 2020-08-02T15:56:36Z

This is for discussion purposes at this time. This is a set of patches we maintain on top of master in order to let Rust Lightning Signer do various things, including:

access to TxCreationKeys~~fields and~~ constructor, so that we can use it to call various InMemoryChannelKeys methods
access to LocalCommitmentTransaction constructor for InMemoryChannelKeys use
export the unsafe methods for testing use, gated by a new feature named "unsafe"
access to node and channel announcement fields so that the contents can be checked by the signer

codecov · 2020-08-02T16:25:08Z

Codecov Report

Merging #662 into master will decrease coverage by 0.00%.
The diff coverage is 75.00%.

@@ Coverage Diff @@## master #662 +/- ##
==========================================
- Coverage 91.36% 91.35% -0.01% 
==========================================
Files 35 35 Lines 21681 21688 +7 ==========================================
+ Hits 19808 19813 +5 - Misses 1873 1875 +2

Impacted Files	Coverage Δ
lightning/src/chain/keysinterface.rs	`95.04% <ø> (ø)`
lightning/src/ln/channelmonitor.rs	`95.70% <ø> (ø)`
lightning/src/ln/onchaintx.rs	`93.94% <ø> (ø)`
lightning/src/util/enforcing_trait_impls.rs	`100.00% <ø> (ø)`
lightning/src/ln/chan_utils.rs	`95.97% <75.00%> (-0.28%)`	⬇️
lightning/src/ln/functional_tests.rs	`97.00% <0.00%> (ø)`

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update 22a0dd5...fd2db40. Read the comment docs.

TheBlueMatt

Mostly questions about use-cases.

ariard

Apart of unsafe feature in discussion, Concept ACK

devrandom · 2020-08-08T17:07:42Z

Removed message field visibility change - will be a separate PR.

Added more detailed commit messaging and removed the draft marker.

TheBlueMatt

I'm fine with this as-is, though I think we need to newtype TxCreationKeys because the APi definitely encourages "unsafe trusting" of the in-RL-derived keys when doing external signing.

I'd prefer a more descriptive name over unsafe, though. Maybe unsafe_revoked_tx_generation or something?

devrandom · 2020-08-09T13:43:38Z

I renamed unsafe to unsafe_revoked_tx_signing.

~~I think the PreCalculatedTxCreationKeys change should go in a separate PR, because the changes (especially involving LocalCommitmentTransaction) will dwarf the other things in this PR.~~ The PreCalculatedTxCreationKeys changes ended up smaller that I predicted, please take a look at #664.

TheBlueMatt

Looks good except for new_missing_local_sig, which could use a bit of clean up if its going to be pub.

Allows unsafe signing in dev code, such as functional testing of justice transactions outside our crate.

Allows calling of InMemoryChannelKeys methods

…_sig

devrandom · 2020-08-02T15:56:36Z

This is for discussion purposes at this time. This is a set of patches we maintain on top of master in order to let Rust Lightning Signer do various things, including:

access to TxCreationKeys~~fields and~~ constructor, so that we can use it to call various InMemoryChannelKeys methods
access to LocalCommitmentTransaction constructor for InMemoryChannelKeys use
export the unsafe methods for testing use, gated by a new feature named "unsafe"
access to node and channel announcement fields so that the contents can be checked by the signer

codecov · 2020-08-02T16:25:08Z

Codecov Report

Merging #662 into master will decrease coverage by 0.00%.
The diff coverage is 75.00%.

@@ Coverage Diff @@## master #662 +/- ##
==========================================
- Coverage 91.36% 91.35% -0.01% 
==========================================
Files 35 35 Lines 21681 21688 +7 ==========================================
+ Hits 19808 19813 +5 - Misses 1873 1875 +2

Impacted Files	Coverage Δ
lightning/src/chain/keysinterface.rs	`95.04% <ø> (ø)`
lightning/src/ln/channelmonitor.rs	`95.70% <ø> (ø)`
lightning/src/ln/onchaintx.rs	`93.94% <ø> (ø)`
lightning/src/util/enforcing_trait_impls.rs	`100.00% <ø> (ø)`
lightning/src/ln/chan_utils.rs	`95.97% <75.00%> (-0.28%)`	⬇️
lightning/src/ln/functional_tests.rs	`97.00% <0.00%> (ø)`

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update 22a0dd5...fd2db40. Read the comment docs.

TheBlueMatt

Mostly questions about use-cases.

ariard

Apart of unsafe feature in discussion, Concept ACK

devrandom · 2020-08-08T17:07:42Z

Removed message field visibility change - will be a separate PR.

Added more detailed commit messaging and removed the draft marker.

TheBlueMatt

I'm fine with this as-is, though I think we need to newtype TxCreationKeys because the APi definitely encourages "unsafe trusting" of the in-RL-derived keys when doing external signing.

I'd prefer a more descriptive name over unsafe, though. Maybe unsafe_revoked_tx_generation or something?

devrandom · 2020-08-09T13:43:38Z

I renamed unsafe to unsafe_revoked_tx_signing.

~~I think the PreCalculatedTxCreationKeys change should go in a separate PR, because the changes (especially involving LocalCommitmentTransaction) will dwarf the other things in this PR.~~ The PreCalculatedTxCreationKeys changes ended up smaller that I predicted, please take a look at #664.

TheBlueMatt

Looks good except for new_missing_local_sig, which could use a bit of clean up if its going to be pub.

Allows unsafe signing in dev code, such as functional testing of justice transactions outside our crate.

Allows calling of InMemoryChannelKeys methods

…_sig

devrandom · 2020-08-02T15:56:36Z

This is for discussion purposes at this time. This is a set of patches we maintain on top of master in order to let Rust Lightning Signer do various things, including:

access to TxCreationKeys~~fields and~~ constructor, so that we can use it to call various InMemoryChannelKeys methods
access to LocalCommitmentTransaction constructor for InMemoryChannelKeys use
export the unsafe methods for testing use, gated by a new feature named "unsafe"
access to node and channel announcement fields so that the contents can be checked by the signer

codecov · 2020-08-02T16:25:08Z

Codecov Report

Merging #662 into master will decrease coverage by 0.00%.
The diff coverage is 75.00%.

@@ Coverage Diff @@## master #662 +/- ##
==========================================
- Coverage 91.36% 91.35% -0.01% 
==========================================
Files 35 35 Lines 21681 21688 +7 ==========================================
+ Hits 19808 19813 +5 - Misses 1873 1875 +2

Impacted Files	Coverage Δ
lightning/src/chain/keysinterface.rs	`95.04% <ø> (ø)`
lightning/src/ln/channelmonitor.rs	`95.70% <ø> (ø)`
lightning/src/ln/onchaintx.rs	`93.94% <ø> (ø)`
lightning/src/util/enforcing_trait_impls.rs	`100.00% <ø> (ø)`
lightning/src/ln/chan_utils.rs	`95.97% <75.00%> (-0.28%)`	⬇️
lightning/src/ln/functional_tests.rs	`97.00% <0.00%> (ø)`

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update 22a0dd5...fd2db40. Read the comment docs.

TheBlueMatt

Mostly questions about use-cases.

ariard

Apart of unsafe feature in discussion, Concept ACK

devrandom · 2020-08-08T17:07:42Z

Removed message field visibility change - will be a separate PR.

Added more detailed commit messaging and removed the draft marker.

TheBlueMatt

I'm fine with this as-is, though I think we need to newtype TxCreationKeys because the APi definitely encourages "unsafe trusting" of the in-RL-derived keys when doing external signing.

I'd prefer a more descriptive name over unsafe, though. Maybe unsafe_revoked_tx_generation or something?

devrandom · 2020-08-09T13:43:38Z

I renamed unsafe to unsafe_revoked_tx_signing.

~~I think the PreCalculatedTxCreationKeys change should go in a separate PR, because the changes (especially involving LocalCommitmentTransaction) will dwarf the other things in this PR.~~ The PreCalculatedTxCreationKeys changes ended up smaller that I predicted, please take a look at #664.

TheBlueMatt

Looks good except for new_missing_local_sig, which could use a bit of clean up if its going to be pub.

Allows unsafe signing in dev code, such as functional testing of justice transactions outside our crate.

Allows calling of InMemoryChannelKeys methods

…_sig

devrandom · 2020-08-02T15:56:36Z

This is for discussion purposes at this time. This is a set of patches we maintain on top of master in order to let Rust Lightning Signer do various things, including:

access to TxCreationKeys~~fields and~~ constructor, so that we can use it to call various InMemoryChannelKeys methods
access to LocalCommitmentTransaction constructor for InMemoryChannelKeys use
export the unsafe methods for testing use, gated by a new feature named "unsafe"
access to node and channel announcement fields so that the contents can be checked by the signer

codecov · 2020-08-02T16:25:08Z

Codecov Report

Merging #662 into master will decrease coverage by 0.00%.
The diff coverage is 75.00%.

@@ Coverage Diff @@## master #662 +/- ##
==========================================
- Coverage 91.36% 91.35% -0.01% 
==========================================
Files 35 35 Lines 21681 21688 +7 ==========================================
+ Hits 19808 19813 +5 - Misses 1873 1875 +2

Impacted Files	Coverage Δ
lightning/src/chain/keysinterface.rs	`95.04% <ø> (ø)`
lightning/src/ln/channelmonitor.rs	`95.70% <ø> (ø)`
lightning/src/ln/onchaintx.rs	`93.94% <ø> (ø)`
lightning/src/util/enforcing_trait_impls.rs	`100.00% <ø> (ø)`
lightning/src/ln/chan_utils.rs	`95.97% <75.00%> (-0.28%)`	⬇️
lightning/src/ln/functional_tests.rs	`97.00% <0.00%> (ø)`

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update 22a0dd5...fd2db40. Read the comment docs.

TheBlueMatt

Mostly questions about use-cases.

ariard

Apart of unsafe feature in discussion, Concept ACK

devrandom · 2020-08-08T17:07:42Z

Removed message field visibility change - will be a separate PR.

Added more detailed commit messaging and removed the draft marker.

TheBlueMatt

I'm fine with this as-is, though I think we need to newtype TxCreationKeys because the APi definitely encourages "unsafe trusting" of the in-RL-derived keys when doing external signing.

I'd prefer a more descriptive name over unsafe, though. Maybe unsafe_revoked_tx_generation or something?

devrandom · 2020-08-09T13:43:38Z

I renamed unsafe to unsafe_revoked_tx_signing.

~~I think the PreCalculatedTxCreationKeys change should go in a separate PR, because the changes (especially involving LocalCommitmentTransaction) will dwarf the other things in this PR.~~ The PreCalculatedTxCreationKeys changes ended up smaller that I predicted, please take a look at #664.

TheBlueMatt

Looks good except for new_missing_local_sig, which could use a bit of clean up if its going to be pub.

Allows unsafe signing in dev code, such as functional testing of justice transactions outside our crate.

Allows calling of InMemoryChannelKeys methods

…_sig

devrandom · 2020-08-02T15:56:36Z

This is for discussion purposes at this time. This is a set of patches we maintain on top of master in order to let Rust Lightning Signer do various things, including:

access to TxCreationKeys~~fields and~~ constructor, so that we can use it to call various InMemoryChannelKeys methods
access to LocalCommitmentTransaction constructor for InMemoryChannelKeys use
export the unsafe methods for testing use, gated by a new feature named "unsafe"
access to node and channel announcement fields so that the contents can be checked by the signer

codecov · 2020-08-02T16:25:08Z

Codecov Report

Merging #662 into master will decrease coverage by 0.00%.
The diff coverage is 75.00%.

@@ Coverage Diff @@## master #662 +/- ##
==========================================
- Coverage 91.36% 91.35% -0.01% 
==========================================
Files 35 35 Lines 21681 21688 +7 ==========================================
+ Hits 19808 19813 +5 - Misses 1873 1875 +2

Impacted Files	Coverage Δ
lightning/src/chain/keysinterface.rs	`95.04% <ø> (ø)`
lightning/src/ln/channelmonitor.rs	`95.70% <ø> (ø)`
lightning/src/ln/onchaintx.rs	`93.94% <ø> (ø)`
lightning/src/util/enforcing_trait_impls.rs	`100.00% <ø> (ø)`
lightning/src/ln/chan_utils.rs	`95.97% <75.00%> (-0.28%)`	⬇️
lightning/src/ln/functional_tests.rs	`97.00% <0.00%> (ø)`

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update 22a0dd5...fd2db40. Read the comment docs.

TheBlueMatt

Mostly questions about use-cases.

ariard

Apart of unsafe feature in discussion, Concept ACK

devrandom · 2020-08-08T17:07:42Z

Removed message field visibility change - will be a separate PR.

Added more detailed commit messaging and removed the draft marker.

TheBlueMatt

I'm fine with this as-is, though I think we need to newtype TxCreationKeys because the APi definitely encourages "unsafe trusting" of the in-RL-derived keys when doing external signing.

I'd prefer a more descriptive name over unsafe, though. Maybe unsafe_revoked_tx_generation or something?

devrandom · 2020-08-09T13:43:38Z

I renamed unsafe to unsafe_revoked_tx_signing.

~~I think the PreCalculatedTxCreationKeys change should go in a separate PR, because the changes (especially involving LocalCommitmentTransaction) will dwarf the other things in this PR.~~ The PreCalculatedTxCreationKeys changes ended up smaller that I predicted, please take a look at #664.

TheBlueMatt

Looks good except for new_missing_local_sig, which could use a bit of clean up if its going to be pub.

Allows unsafe signing in dev code, such as functional testing of justice transactions outside our crate.

Allows calling of InMemoryChannelKeys methods

…_sig

devrandom · 2020-08-02T15:56:36Z

This is for discussion purposes at this time. This is a set of patches we maintain on top of master in order to let Rust Lightning Signer do various things, including:

access to TxCreationKeys~~fields and~~ constructor, so that we can use it to call various InMemoryChannelKeys methods
access to LocalCommitmentTransaction constructor for InMemoryChannelKeys use
export the unsafe methods for testing use, gated by a new feature named "unsafe"
access to node and channel announcement fields so that the contents can be checked by the signer

codecov · 2020-08-02T16:25:08Z

Codecov Report

Merging #662 into master will decrease coverage by 0.00%.
The diff coverage is 75.00%.

@@ Coverage Diff @@## master #662 +/- ##
==========================================
- Coverage 91.36% 91.35% -0.01% 
==========================================
Files 35 35 Lines 21681 21688 +7 ==========================================
+ Hits 19808 19813 +5 - Misses 1873 1875 +2

Impacted Files	Coverage Δ
lightning/src/chain/keysinterface.rs	`95.04% <ø> (ø)`
lightning/src/ln/channelmonitor.rs	`95.70% <ø> (ø)`
lightning/src/ln/onchaintx.rs	`93.94% <ø> (ø)`
lightning/src/util/enforcing_trait_impls.rs	`100.00% <ø> (ø)`
lightning/src/ln/chan_utils.rs	`95.97% <75.00%> (-0.28%)`	⬇️
lightning/src/ln/functional_tests.rs	`97.00% <0.00%> (ø)`

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update 22a0dd5...fd2db40. Read the comment docs.

TheBlueMatt

Mostly questions about use-cases.

ariard

Apart of unsafe feature in discussion, Concept ACK

devrandom · 2020-08-08T17:07:42Z

Removed message field visibility change - will be a separate PR.

Added more detailed commit messaging and removed the draft marker.

TheBlueMatt

I'm fine with this as-is, though I think we need to newtype TxCreationKeys because the APi definitely encourages "unsafe trusting" of the in-RL-derived keys when doing external signing.

I'd prefer a more descriptive name over unsafe, though. Maybe unsafe_revoked_tx_generation or something?

devrandom · 2020-08-09T13:43:38Z

I renamed unsafe to unsafe_revoked_tx_signing.

~~I think the PreCalculatedTxCreationKeys change should go in a separate PR, because the changes (especially involving LocalCommitmentTransaction) will dwarf the other things in this PR.~~ The PreCalculatedTxCreationKeys changes ended up smaller that I predicted, please take a look at #664.

TheBlueMatt

Looks good except for new_missing_local_sig, which could use a bit of clean up if its going to be pub.

Allows unsafe signing in dev code, such as functional testing of justice transactions outside our crate.

Allows calling of InMemoryChannelKeys methods

…_sig

devrandom · 2020-08-02T15:56:36Z

This is for discussion purposes at this time. This is a set of patches we maintain on top of master in order to let Rust Lightning Signer do various things, including:

access to TxCreationKeys~~fields and~~ constructor, so that we can use it to call various InMemoryChannelKeys methods
access to LocalCommitmentTransaction constructor for InMemoryChannelKeys use
export the unsafe methods for testing use, gated by a new feature named "unsafe"
access to node and channel announcement fields so that the contents can be checked by the signer

codecov · 2020-08-02T16:25:08Z

Codecov Report

Merging #662 into master will decrease coverage by 0.00%.
The diff coverage is 75.00%.

@@ Coverage Diff @@## master #662 +/- ##
==========================================
- Coverage 91.36% 91.35% -0.01% 
==========================================
Files 35 35 Lines 21681 21688 +7 ==========================================
+ Hits 19808 19813 +5 - Misses 1873 1875 +2

Impacted Files	Coverage Δ
lightning/src/chain/keysinterface.rs	`95.04% <ø> (ø)`
lightning/src/ln/channelmonitor.rs	`95.70% <ø> (ø)`
lightning/src/ln/onchaintx.rs	`93.94% <ø> (ø)`
lightning/src/util/enforcing_trait_impls.rs	`100.00% <ø> (ø)`
lightning/src/ln/chan_utils.rs	`95.97% <75.00%> (-0.28%)`	⬇️
lightning/src/ln/functional_tests.rs	`97.00% <0.00%> (ø)`

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update 22a0dd5...fd2db40. Read the comment docs.

TheBlueMatt

Mostly questions about use-cases.

ariard

Apart of unsafe feature in discussion, Concept ACK

devrandom · 2020-08-08T17:07:42Z

Removed message field visibility change - will be a separate PR.

Added more detailed commit messaging and removed the draft marker.

TheBlueMatt

I'm fine with this as-is, though I think we need to newtype TxCreationKeys because the APi definitely encourages "unsafe trusting" of the in-RL-derived keys when doing external signing.

I'd prefer a more descriptive name over unsafe, though. Maybe unsafe_revoked_tx_generation or something?

devrandom · 2020-08-09T13:43:38Z

I renamed unsafe to unsafe_revoked_tx_signing.

~~I think the PreCalculatedTxCreationKeys change should go in a separate PR, because the changes (especially involving LocalCommitmentTransaction) will dwarf the other things in this PR.~~ The PreCalculatedTxCreationKeys changes ended up smaller that I predicted, please take a look at #664.

TheBlueMatt

Looks good except for new_missing_local_sig, which could use a bit of clean up if its going to be pub.

Allows unsafe signing in dev code, such as functional testing of justice transactions outside our crate.

Allows calling of InMemoryChannelKeys methods

…_sig

TheBlueMatt reviewed Aug 3, 2020
View reviewed changes

Comment threadlightning/src/chain/keysinterface.rs Outdated
Comment threadlightning/src/ln/chan_utils.rs
Comment threadlightning/src/ln/chan_utils.rs
Comment threadlightning/src/ln/msgs.rs Outdated

ariard reviewed Aug 5, 2020
View reviewed changes

devrandom force-pushed the compat5 branch from 4136c0e to 1c3cb78Compare August 5, 2020 06:44

devrandom marked this pull request as draft August 5, 2020 06:47

devrandom changed the title ~~WIP: export various fields and structures~~Draft: export various fields and structuresAug 5, 2020

devrandom force-pushed the compat5 branch from 1c3cb78 to 50a2e26Compare August 8, 2020 17:06

devrandom marked this pull request as ready for review August 8, 2020 17:06

devrandom changed the title ~~Draft: export various fields and structures~~Export various fields and structuresAug 8, 2020

TheBlueMatt previously approved these changes Aug 8, 2020
View reviewed changes

devrandom dismissed TheBlueMatt’s stale review via f61ce28August 9, 2020 10:55

devrandom force-pushed the compat5 branch from 50a2e26 to f61ce28Compare August 9, 2020 10:55

devrandom mentioned this pull request Aug 9, 2020
Wrap transaction creation keys #664
Merged

TheBlueMatt approved these changes Aug 10, 2020
View reviewed changes

Comment threadlightning/src/ln/chan_utils.rs

devrandom added 3 commits August 11, 2020 09:23

export "unsafe_revoked_tx_signing" feature
ce0cecd
Allows unsafe signing in dev code, such as functional testing of justice transactions outside our crate.

make SimpleManyChannelMonitor.monitors public
31b4d89

make LocalCommitmentTransaction constructor public
f600263
Allows calling of InMemoryChannelKeys methods

devrandom force-pushed the compat5 branch 2 times, most recently from f05076d to 8bcdc87Compare August 11, 2020 08:48

check the input shape in LocalCommitmentTransaction.new_missing_local…
fd2db40
…_sig

devrandom force-pushed the compat5 branch from 8bcdc87 to fd2db40Compare August 11, 2020 09:00

TheBlueMatt merged commit 1d2d393 into lightningdevkit:masterAug 11, 2020