Remove exclusive reference and generic from `CommitmentTransaction` API by tankyleo · Pull Request #3689 · lightningdevkit/rust-lightning

tankyleo · 2025-03-29T00:49:22Z

 Remove exclusive reference and generic from `CommitmentTransaction` API
This commit reworks how channel is told about which HTLCs were assigned
which index upon the build of a `CommitmentTransaction`.
Previously, `CommitmentTransaction` was given an exclusive reference to
channel's HTLC-source table, and `CommitmentTransaction` populated the
transaction output indices in that table via this exclusive reference.
As a result, the public API of `CommitmentTransaction` included a
generic parameter, and an exclusive reference.
We remove both of these in preparation for the upcoming `TxBuilder`
trait. This cleans up the API, and makes it more bindings-friendly.
Henceforth, channel populates the HTLC-source table via a brute-force
search of each htlc in `CommitmentTransaction`. This is an O(n^2)
operation, but n is small enough that we ignore the performance hit.
We also take this opportunity to cleanup how we build and sort the
commitment transaction outputs together with the htlc output data
in `CommitmentTransaction`. The goal is to keep the number of vector
allocations to a minimum; we now only allocate a single vector to hold
the transaction outputs, and do all the sorting in-place.

ldk-reviews-bot · 2025-03-29T00:49:24Z

👋 Thanks for assigning @wpaulino as a reviewer!
I'll wait for their review and will help manage the review process.
Once they submit their review, I'll check if a second reviewer would be helpful.

codecov · 2025-04-10T23:26:43Z

Codecov Report

Attention: Patch coverage is 99.15254% with 2 lines in your changes missing coverage. Please review.

Project coverage is 89.21%. Comparing base (ef0fcab) to head (75f81b5).
Report is 19 commits behind head on main.

Files with missing lines	Patch %	Lines
lightning/src/ln/chan_utils.rs	98.81%	2 Missing ⚠️

Additional details and impacted files

@@ Coverage Diff @@## main #3689 +/- ##
==========================================
+ Coverage 89.07% 89.21% +0.13% 
==========================================
Files 156 156 Lines 123507 124130 +623 Branches 123507 124130 +623 ==========================================
+ Hits 110017 110742 +725 + Misses 10801 10724 -77 + Partials 2689 2664 -25

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

TheBlueMatt

One last real comment

TheBlueMatt

New commit basically LTGM.

TheBlueMatt

Basically LGTM, Probably needs another look from @wpaulino otherwise I'm happy.

TheBlueMatt · 2025-04-14T20:11:25Z

 	///
-	/// This is not exported to bindings users due to the generic though we likely should expose a version without
-	pub fn new_with_auxiliary_htlc_data<T>(commitment_number: u64, per_commitment_point: &PublicKey, to_broadcaster_value_sat: u64, to_countersignatory_value_sat: u64, feerate_per_kw: u32, htlcs_with_aux: &mut Vec<(HTLCOutputInCommitment, T)>, channel_parameters: &DirectedChannelTransactionParameters, secp_ctx: &Secp256k1<secp256k1::All>) -> CommitmentTransaction {
+	/// All HTLCs MUST be above the dust limit for the channel.\


stray , I think? note that if you dont leave an extra line between paragraphs, rustdoc assumes you meant for them to be in the same paragraph and ignores newlines in the source.

rustdoc actually turns \ into a single newline character.

Heh, TIL, but it also makes the source unreadable :)

TheBlueMatt

Probably worth adding the sort in the rebuild case, but gonna go ahead and land it.

TheBlueMatt · 2025-04-16T01:02:38Z

+ let (obscured_commitment_transaction_number, txins) = Self::build_inputs(self.commitment_number, channel_parameters);
+
+ // First rebuild the htlc outputs, note that these are already sorted
+ let mut outputs = Self::build_htlc_outputs(keys, &self.nondust_htlcs, channel_parameters.channel_type_features());


It would be nice to resort the HTLCs here before we return. The point of rebuild_transaction is to avoid trusting the pre-built commitment transaction entirely and build it from the HTLC list.

Should I get this done now ?

Can we have fn verify take a &mut self ? To highlight that this call will resort the HTLCs if necessary. Then fn rebuild_transaction can call build_outputs_and_htlcs just like fn new without any new memory allocations.

Ah looks like this upsets the validation APIs of the signer that take a &CommitmentTransaction

Maybe we make it return a Result? Can happen in a followup but up to you if you want to do it now.

Meaning rebuild_transaction returns an Err if the HTLCs are not sorted ? Yes

TheBlueMatt · 2025-04-16T01:08:10Z

Oops, sorry lol, didn't mean for the nit on docs to be blocking.

tankyleo · 2025-04-16T01:09:39Z

Ah all good I do agree with the aesthetics.

This commit reworks how channel is told about which HTLCs were assigned which index upon the build of a `CommitmentTransaction`. Previously, `CommitmentTransaction` was given an exclusive reference to channel's HTLC-source table, and `CommitmentTransaction` populated the transaction output indices in that table via this exclusive reference. As a result, the public API of `CommitmentTransaction` included a generic parameter, and an exclusive reference. We remove both of these in preparation for the upcoming `TxBuilder` trait. This cleans up the API, and makes it more bindings-friendly. Henceforth, channel populates the HTLC-source table via a brute-force search of each htlc in `CommitmentTransaction`. This is an O(n^2) operation, but n is small enough that we ignore the performance hit. We also take this opportunity to cleanup how we build and sort the commitment transaction outputs together with the htlc output data in `CommitmentTransaction`. The goal is to keep the number of vector allocations to a minimum; we now only allocate a single vector to hold the transaction outputs, and do all the sorting in-place. Finally, we add test coverage to assert that `CommitmentTransaction::verify` actually checks that the HTLCs in the `nondust_htlcs` member are actually sorted.

tankyleo · 2025-04-16T20:00:48Z

@wpaulino I previously mistakenly started trusting the order of HTLCs in the nondust_htlcs member in the verify call - I correct this here.

tankyleo · 2025-03-29T00:49:22Z

 Remove exclusive reference and generic from `CommitmentTransaction` API
This commit reworks how channel is told about which HTLCs were assigned
which index upon the build of a `CommitmentTransaction`.
Previously, `CommitmentTransaction` was given an exclusive reference to
channel's HTLC-source table, and `CommitmentTransaction` populated the
transaction output indices in that table via this exclusive reference.
As a result, the public API of `CommitmentTransaction` included a
generic parameter, and an exclusive reference.
We remove both of these in preparation for the upcoming `TxBuilder`
trait. This cleans up the API, and makes it more bindings-friendly.
Henceforth, channel populates the HTLC-source table via a brute-force
search of each htlc in `CommitmentTransaction`. This is an O(n^2)
operation, but n is small enough that we ignore the performance hit.
We also take this opportunity to cleanup how we build and sort the
commitment transaction outputs together with the htlc output data
in `CommitmentTransaction`. The goal is to keep the number of vector
allocations to a minimum; we now only allocate a single vector to hold
the transaction outputs, and do all the sorting in-place.

ldk-reviews-bot · 2025-03-29T00:49:24Z

👋 Thanks for assigning @wpaulino as a reviewer!
I'll wait for their review and will help manage the review process.
Once they submit their review, I'll check if a second reviewer would be helpful.

codecov · 2025-04-10T23:26:43Z

Codecov Report

Attention: Patch coverage is 99.15254% with 2 lines in your changes missing coverage. Please review.

Project coverage is 89.21%. Comparing base (ef0fcab) to head (75f81b5).
Report is 19 commits behind head on main.

Files with missing lines	Patch %	Lines
lightning/src/ln/chan_utils.rs	98.81%	2 Missing ⚠️

Additional details and impacted files

@@ Coverage Diff @@## main #3689 +/- ##
==========================================
+ Coverage 89.07% 89.21% +0.13% 
==========================================
Files 156 156 Lines 123507 124130 +623 Branches 123507 124130 +623 ==========================================
+ Hits 110017 110742 +725 + Misses 10801 10724 -77 + Partials 2689 2664 -25

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

TheBlueMatt

One last real comment

TheBlueMatt

New commit basically LTGM.

TheBlueMatt

Basically LGTM, Probably needs another look from @wpaulino otherwise I'm happy.

TheBlueMatt · 2025-04-14T20:11:25Z

 	///
-	/// This is not exported to bindings users due to the generic though we likely should expose a version without
-	pub fn new_with_auxiliary_htlc_data<T>(commitment_number: u64, per_commitment_point: &PublicKey, to_broadcaster_value_sat: u64, to_countersignatory_value_sat: u64, feerate_per_kw: u32, htlcs_with_aux: &mut Vec<(HTLCOutputInCommitment, T)>, channel_parameters: &DirectedChannelTransactionParameters, secp_ctx: &Secp256k1<secp256k1::All>) -> CommitmentTransaction {
+	/// All HTLCs MUST be above the dust limit for the channel.\


stray , I think? note that if you dont leave an extra line between paragraphs, rustdoc assumes you meant for them to be in the same paragraph and ignores newlines in the source.

rustdoc actually turns \ into a single newline character.

Heh, TIL, but it also makes the source unreadable :)

TheBlueMatt

Probably worth adding the sort in the rebuild case, but gonna go ahead and land it.

TheBlueMatt · 2025-04-16T01:02:38Z

+ let (obscured_commitment_transaction_number, txins) = Self::build_inputs(self.commitment_number, channel_parameters);
+
+ // First rebuild the htlc outputs, note that these are already sorted
+ let mut outputs = Self::build_htlc_outputs(keys, &self.nondust_htlcs, channel_parameters.channel_type_features());


It would be nice to resort the HTLCs here before we return. The point of rebuild_transaction is to avoid trusting the pre-built commitment transaction entirely and build it from the HTLC list.

Should I get this done now ?

Can we have fn verify take a &mut self ? To highlight that this call will resort the HTLCs if necessary. Then fn rebuild_transaction can call build_outputs_and_htlcs just like fn new without any new memory allocations.

Ah looks like this upsets the validation APIs of the signer that take a &CommitmentTransaction

Maybe we make it return a Result? Can happen in a followup but up to you if you want to do it now.

Meaning rebuild_transaction returns an Err if the HTLCs are not sorted ? Yes

TheBlueMatt · 2025-04-16T01:08:10Z

Oops, sorry lol, didn't mean for the nit on docs to be blocking.

tankyleo · 2025-04-16T01:09:39Z

Ah all good I do agree with the aesthetics.

This commit reworks how channel is told about which HTLCs were assigned which index upon the build of a `CommitmentTransaction`. Previously, `CommitmentTransaction` was given an exclusive reference to channel's HTLC-source table, and `CommitmentTransaction` populated the transaction output indices in that table via this exclusive reference. As a result, the public API of `CommitmentTransaction` included a generic parameter, and an exclusive reference. We remove both of these in preparation for the upcoming `TxBuilder` trait. This cleans up the API, and makes it more bindings-friendly. Henceforth, channel populates the HTLC-source table via a brute-force search of each htlc in `CommitmentTransaction`. This is an O(n^2) operation, but n is small enough that we ignore the performance hit. We also take this opportunity to cleanup how we build and sort the commitment transaction outputs together with the htlc output data in `CommitmentTransaction`. The goal is to keep the number of vector allocations to a minimum; we now only allocate a single vector to hold the transaction outputs, and do all the sorting in-place. Finally, we add test coverage to assert that `CommitmentTransaction::verify` actually checks that the HTLCs in the `nondust_htlcs` member are actually sorted.

tankyleo · 2025-04-16T20:00:48Z

@wpaulino I previously mistakenly started trusting the order of HTLCs in the nondust_htlcs member in the verify call - I correct this here.

tankyleo · 2025-03-29T00:49:22Z

 Remove exclusive reference and generic from `CommitmentTransaction` API
This commit reworks how channel is told about which HTLCs were assigned
which index upon the build of a `CommitmentTransaction`.
Previously, `CommitmentTransaction` was given an exclusive reference to
channel's HTLC-source table, and `CommitmentTransaction` populated the
transaction output indices in that table via this exclusive reference.
As a result, the public API of `CommitmentTransaction` included a
generic parameter, and an exclusive reference.
We remove both of these in preparation for the upcoming `TxBuilder`
trait. This cleans up the API, and makes it more bindings-friendly.
Henceforth, channel populates the HTLC-source table via a brute-force
search of each htlc in `CommitmentTransaction`. This is an O(n^2)
operation, but n is small enough that we ignore the performance hit.
We also take this opportunity to cleanup how we build and sort the
commitment transaction outputs together with the htlc output data
in `CommitmentTransaction`. The goal is to keep the number of vector
allocations to a minimum; we now only allocate a single vector to hold
the transaction outputs, and do all the sorting in-place.

ldk-reviews-bot · 2025-03-29T00:49:24Z

👋 Thanks for assigning @wpaulino as a reviewer!
I'll wait for their review and will help manage the review process.
Once they submit their review, I'll check if a second reviewer would be helpful.

codecov · 2025-04-10T23:26:43Z

Codecov Report

Attention: Patch coverage is 99.15254% with 2 lines in your changes missing coverage. Please review.

Project coverage is 89.21%. Comparing base (ef0fcab) to head (75f81b5).
Report is 19 commits behind head on main.

Files with missing lines	Patch %	Lines
lightning/src/ln/chan_utils.rs	98.81%	2 Missing ⚠️

Additional details and impacted files

@@ Coverage Diff @@## main #3689 +/- ##
==========================================
+ Coverage 89.07% 89.21% +0.13% 
==========================================
Files 156 156 Lines 123507 124130 +623 Branches 123507 124130 +623 ==========================================
+ Hits 110017 110742 +725 + Misses 10801 10724 -77 + Partials 2689 2664 -25

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

TheBlueMatt

One last real comment

TheBlueMatt

New commit basically LTGM.

TheBlueMatt

Basically LGTM, Probably needs another look from @wpaulino otherwise I'm happy.

TheBlueMatt · 2025-04-14T20:11:25Z

 	///
-	/// This is not exported to bindings users due to the generic though we likely should expose a version without
-	pub fn new_with_auxiliary_htlc_data<T>(commitment_number: u64, per_commitment_point: &PublicKey, to_broadcaster_value_sat: u64, to_countersignatory_value_sat: u64, feerate_per_kw: u32, htlcs_with_aux: &mut Vec<(HTLCOutputInCommitment, T)>, channel_parameters: &DirectedChannelTransactionParameters, secp_ctx: &Secp256k1<secp256k1::All>) -> CommitmentTransaction {
+	/// All HTLCs MUST be above the dust limit for the channel.\


stray , I think? note that if you dont leave an extra line between paragraphs, rustdoc assumes you meant for them to be in the same paragraph and ignores newlines in the source.

rustdoc actually turns \ into a single newline character.

Heh, TIL, but it also makes the source unreadable :)

TheBlueMatt

Probably worth adding the sort in the rebuild case, but gonna go ahead and land it.

TheBlueMatt · 2025-04-16T01:02:38Z

+ let (obscured_commitment_transaction_number, txins) = Self::build_inputs(self.commitment_number, channel_parameters);
+
+ // First rebuild the htlc outputs, note that these are already sorted
+ let mut outputs = Self::build_htlc_outputs(keys, &self.nondust_htlcs, channel_parameters.channel_type_features());


It would be nice to resort the HTLCs here before we return. The point of rebuild_transaction is to avoid trusting the pre-built commitment transaction entirely and build it from the HTLC list.

Should I get this done now ?

Can we have fn verify take a &mut self ? To highlight that this call will resort the HTLCs if necessary. Then fn rebuild_transaction can call build_outputs_and_htlcs just like fn new without any new memory allocations.

Ah looks like this upsets the validation APIs of the signer that take a &CommitmentTransaction

Maybe we make it return a Result? Can happen in a followup but up to you if you want to do it now.

Meaning rebuild_transaction returns an Err if the HTLCs are not sorted ? Yes

TheBlueMatt · 2025-04-16T01:08:10Z

Oops, sorry lol, didn't mean for the nit on docs to be blocking.

tankyleo · 2025-04-16T01:09:39Z

Ah all good I do agree with the aesthetics.

This commit reworks how channel is told about which HTLCs were assigned which index upon the build of a `CommitmentTransaction`. Previously, `CommitmentTransaction` was given an exclusive reference to channel's HTLC-source table, and `CommitmentTransaction` populated the transaction output indices in that table via this exclusive reference. As a result, the public API of `CommitmentTransaction` included a generic parameter, and an exclusive reference. We remove both of these in preparation for the upcoming `TxBuilder` trait. This cleans up the API, and makes it more bindings-friendly. Henceforth, channel populates the HTLC-source table via a brute-force search of each htlc in `CommitmentTransaction`. This is an O(n^2) operation, but n is small enough that we ignore the performance hit. We also take this opportunity to cleanup how we build and sort the commitment transaction outputs together with the htlc output data in `CommitmentTransaction`. The goal is to keep the number of vector allocations to a minimum; we now only allocate a single vector to hold the transaction outputs, and do all the sorting in-place. Finally, we add test coverage to assert that `CommitmentTransaction::verify` actually checks that the HTLCs in the `nondust_htlcs` member are actually sorted.

tankyleo · 2025-04-16T20:00:48Z

@wpaulino I previously mistakenly started trusting the order of HTLCs in the nondust_htlcs member in the verify call - I correct this here.

tankyleo · 2025-03-29T00:49:22Z

 Remove exclusive reference and generic from `CommitmentTransaction` API
This commit reworks how channel is told about which HTLCs were assigned
which index upon the build of a `CommitmentTransaction`.
Previously, `CommitmentTransaction` was given an exclusive reference to
channel's HTLC-source table, and `CommitmentTransaction` populated the
transaction output indices in that table via this exclusive reference.
As a result, the public API of `CommitmentTransaction` included a
generic parameter, and an exclusive reference.
We remove both of these in preparation for the upcoming `TxBuilder`
trait. This cleans up the API, and makes it more bindings-friendly.
Henceforth, channel populates the HTLC-source table via a brute-force
search of each htlc in `CommitmentTransaction`. This is an O(n^2)
operation, but n is small enough that we ignore the performance hit.
We also take this opportunity to cleanup how we build and sort the
commitment transaction outputs together with the htlc output data
in `CommitmentTransaction`. The goal is to keep the number of vector
allocations to a minimum; we now only allocate a single vector to hold
the transaction outputs, and do all the sorting in-place.

ldk-reviews-bot · 2025-03-29T00:49:24Z

👋 Thanks for assigning @wpaulino as a reviewer!
I'll wait for their review and will help manage the review process.
Once they submit their review, I'll check if a second reviewer would be helpful.

codecov · 2025-04-10T23:26:43Z

Codecov Report

Attention: Patch coverage is 99.15254% with 2 lines in your changes missing coverage. Please review.

Project coverage is 89.21%. Comparing base (ef0fcab) to head (75f81b5).
Report is 19 commits behind head on main.

Files with missing lines	Patch %	Lines
lightning/src/ln/chan_utils.rs	98.81%	2 Missing ⚠️

Additional details and impacted files

@@ Coverage Diff @@## main #3689 +/- ##
==========================================
+ Coverage 89.07% 89.21% +0.13% 
==========================================
Files 156 156 Lines 123507 124130 +623 Branches 123507 124130 +623 ==========================================
+ Hits 110017 110742 +725 + Misses 10801 10724 -77 + Partials 2689 2664 -25

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

TheBlueMatt

One last real comment

TheBlueMatt

New commit basically LTGM.

TheBlueMatt

Basically LGTM, Probably needs another look from @wpaulino otherwise I'm happy.

TheBlueMatt · 2025-04-14T20:11:25Z

 	///
-	/// This is not exported to bindings users due to the generic though we likely should expose a version without
-	pub fn new_with_auxiliary_htlc_data<T>(commitment_number: u64, per_commitment_point: &PublicKey, to_broadcaster_value_sat: u64, to_countersignatory_value_sat: u64, feerate_per_kw: u32, htlcs_with_aux: &mut Vec<(HTLCOutputInCommitment, T)>, channel_parameters: &DirectedChannelTransactionParameters, secp_ctx: &Secp256k1<secp256k1::All>) -> CommitmentTransaction {
+	/// All HTLCs MUST be above the dust limit for the channel.\


stray , I think? note that if you dont leave an extra line between paragraphs, rustdoc assumes you meant for them to be in the same paragraph and ignores newlines in the source.

rustdoc actually turns \ into a single newline character.

Heh, TIL, but it also makes the source unreadable :)

TheBlueMatt

Probably worth adding the sort in the rebuild case, but gonna go ahead and land it.

TheBlueMatt · 2025-04-16T01:02:38Z

+ let (obscured_commitment_transaction_number, txins) = Self::build_inputs(self.commitment_number, channel_parameters);
+
+ // First rebuild the htlc outputs, note that these are already sorted
+ let mut outputs = Self::build_htlc_outputs(keys, &self.nondust_htlcs, channel_parameters.channel_type_features());


It would be nice to resort the HTLCs here before we return. The point of rebuild_transaction is to avoid trusting the pre-built commitment transaction entirely and build it from the HTLC list.

Should I get this done now ?

Can we have fn verify take a &mut self ? To highlight that this call will resort the HTLCs if necessary. Then fn rebuild_transaction can call build_outputs_and_htlcs just like fn new without any new memory allocations.

Ah looks like this upsets the validation APIs of the signer that take a &CommitmentTransaction

Maybe we make it return a Result? Can happen in a followup but up to you if you want to do it now.

Meaning rebuild_transaction returns an Err if the HTLCs are not sorted ? Yes

TheBlueMatt · 2025-04-16T01:08:10Z

Oops, sorry lol, didn't mean for the nit on docs to be blocking.

tankyleo · 2025-04-16T01:09:39Z

Ah all good I do agree with the aesthetics.

This commit reworks how channel is told about which HTLCs were assigned which index upon the build of a `CommitmentTransaction`. Previously, `CommitmentTransaction` was given an exclusive reference to channel's HTLC-source table, and `CommitmentTransaction` populated the transaction output indices in that table via this exclusive reference. As a result, the public API of `CommitmentTransaction` included a generic parameter, and an exclusive reference. We remove both of these in preparation for the upcoming `TxBuilder` trait. This cleans up the API, and makes it more bindings-friendly. Henceforth, channel populates the HTLC-source table via a brute-force search of each htlc in `CommitmentTransaction`. This is an O(n^2) operation, but n is small enough that we ignore the performance hit. We also take this opportunity to cleanup how we build and sort the commitment transaction outputs together with the htlc output data in `CommitmentTransaction`. The goal is to keep the number of vector allocations to a minimum; we now only allocate a single vector to hold the transaction outputs, and do all the sorting in-place. Finally, we add test coverage to assert that `CommitmentTransaction::verify` actually checks that the HTLCs in the `nondust_htlcs` member are actually sorted.

tankyleo · 2025-04-16T20:00:48Z

@wpaulino I previously mistakenly started trusting the order of HTLCs in the nondust_htlcs member in the verify call - I correct this here.

tankyleo · 2025-03-29T00:49:22Z

 Remove exclusive reference and generic from `CommitmentTransaction` API
This commit reworks how channel is told about which HTLCs were assigned
which index upon the build of a `CommitmentTransaction`.
Previously, `CommitmentTransaction` was given an exclusive reference to
channel's HTLC-source table, and `CommitmentTransaction` populated the
transaction output indices in that table via this exclusive reference.
As a result, the public API of `CommitmentTransaction` included a
generic parameter, and an exclusive reference.
We remove both of these in preparation for the upcoming `TxBuilder`
trait. This cleans up the API, and makes it more bindings-friendly.
Henceforth, channel populates the HTLC-source table via a brute-force
search of each htlc in `CommitmentTransaction`. This is an O(n^2)
operation, but n is small enough that we ignore the performance hit.
We also take this opportunity to cleanup how we build and sort the
commitment transaction outputs together with the htlc output data
in `CommitmentTransaction`. The goal is to keep the number of vector
allocations to a minimum; we now only allocate a single vector to hold
the transaction outputs, and do all the sorting in-place.

ldk-reviews-bot · 2025-03-29T00:49:24Z

👋 Thanks for assigning @wpaulino as a reviewer!
I'll wait for their review and will help manage the review process.
Once they submit their review, I'll check if a second reviewer would be helpful.

codecov · 2025-04-10T23:26:43Z

Codecov Report

Attention: Patch coverage is 99.15254% with 2 lines in your changes missing coverage. Please review.

Project coverage is 89.21%. Comparing base (ef0fcab) to head (75f81b5).
Report is 19 commits behind head on main.

Files with missing lines	Patch %	Lines
lightning/src/ln/chan_utils.rs	98.81%	2 Missing ⚠️

Additional details and impacted files

@@ Coverage Diff @@## main #3689 +/- ##
==========================================
+ Coverage 89.07% 89.21% +0.13% 
==========================================
Files 156 156 Lines 123507 124130 +623 Branches 123507 124130 +623 ==========================================
+ Hits 110017 110742 +725 + Misses 10801 10724 -77 + Partials 2689 2664 -25

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

TheBlueMatt

One last real comment

TheBlueMatt

New commit basically LTGM.

TheBlueMatt

Basically LGTM, Probably needs another look from @wpaulino otherwise I'm happy.

TheBlueMatt · 2025-04-14T20:11:25Z

 	///
-	/// This is not exported to bindings users due to the generic though we likely should expose a version without
-	pub fn new_with_auxiliary_htlc_data<T>(commitment_number: u64, per_commitment_point: &PublicKey, to_broadcaster_value_sat: u64, to_countersignatory_value_sat: u64, feerate_per_kw: u32, htlcs_with_aux: &mut Vec<(HTLCOutputInCommitment, T)>, channel_parameters: &DirectedChannelTransactionParameters, secp_ctx: &Secp256k1<secp256k1::All>) -> CommitmentTransaction {
+	/// All HTLCs MUST be above the dust limit for the channel.\


stray , I think? note that if you dont leave an extra line between paragraphs, rustdoc assumes you meant for them to be in the same paragraph and ignores newlines in the source.

rustdoc actually turns \ into a single newline character.

Heh, TIL, but it also makes the source unreadable :)

TheBlueMatt

Probably worth adding the sort in the rebuild case, but gonna go ahead and land it.

TheBlueMatt · 2025-04-16T01:02:38Z

+ let (obscured_commitment_transaction_number, txins) = Self::build_inputs(self.commitment_number, channel_parameters);
+
+ // First rebuild the htlc outputs, note that these are already sorted
+ let mut outputs = Self::build_htlc_outputs(keys, &self.nondust_htlcs, channel_parameters.channel_type_features());


It would be nice to resort the HTLCs here before we return. The point of rebuild_transaction is to avoid trusting the pre-built commitment transaction entirely and build it from the HTLC list.

Should I get this done now ?

Can we have fn verify take a &mut self ? To highlight that this call will resort the HTLCs if necessary. Then fn rebuild_transaction can call build_outputs_and_htlcs just like fn new without any new memory allocations.

Ah looks like this upsets the validation APIs of the signer that take a &CommitmentTransaction

Maybe we make it return a Result? Can happen in a followup but up to you if you want to do it now.

Meaning rebuild_transaction returns an Err if the HTLCs are not sorted ? Yes

TheBlueMatt · 2025-04-16T01:08:10Z

Oops, sorry lol, didn't mean for the nit on docs to be blocking.

tankyleo · 2025-04-16T01:09:39Z

Ah all good I do agree with the aesthetics.

This commit reworks how channel is told about which HTLCs were assigned which index upon the build of a `CommitmentTransaction`. Previously, `CommitmentTransaction` was given an exclusive reference to channel's HTLC-source table, and `CommitmentTransaction` populated the transaction output indices in that table via this exclusive reference. As a result, the public API of `CommitmentTransaction` included a generic parameter, and an exclusive reference. We remove both of these in preparation for the upcoming `TxBuilder` trait. This cleans up the API, and makes it more bindings-friendly. Henceforth, channel populates the HTLC-source table via a brute-force search of each htlc in `CommitmentTransaction`. This is an O(n^2) operation, but n is small enough that we ignore the performance hit. We also take this opportunity to cleanup how we build and sort the commitment transaction outputs together with the htlc output data in `CommitmentTransaction`. The goal is to keep the number of vector allocations to a minimum; we now only allocate a single vector to hold the transaction outputs, and do all the sorting in-place. Finally, we add test coverage to assert that `CommitmentTransaction::verify` actually checks that the HTLCs in the `nondust_htlcs` member are actually sorted.

tankyleo · 2025-04-16T20:00:48Z

@wpaulino I previously mistakenly started trusting the order of HTLCs in the nondust_htlcs member in the verify call - I correct this here.

tankyleo · 2025-03-29T00:49:22Z

 Remove exclusive reference and generic from `CommitmentTransaction` API
This commit reworks how channel is told about which HTLCs were assigned
which index upon the build of a `CommitmentTransaction`.
Previously, `CommitmentTransaction` was given an exclusive reference to
channel's HTLC-source table, and `CommitmentTransaction` populated the
transaction output indices in that table via this exclusive reference.
As a result, the public API of `CommitmentTransaction` included a
generic parameter, and an exclusive reference.
We remove both of these in preparation for the upcoming `TxBuilder`
trait. This cleans up the API, and makes it more bindings-friendly.
Henceforth, channel populates the HTLC-source table via a brute-force
search of each htlc in `CommitmentTransaction`. This is an O(n^2)
operation, but n is small enough that we ignore the performance hit.
We also take this opportunity to cleanup how we build and sort the
commitment transaction outputs together with the htlc output data
in `CommitmentTransaction`. The goal is to keep the number of vector
allocations to a minimum; we now only allocate a single vector to hold
the transaction outputs, and do all the sorting in-place.

ldk-reviews-bot · 2025-03-29T00:49:24Z

👋 Thanks for assigning @wpaulino as a reviewer!
I'll wait for their review and will help manage the review process.
Once they submit their review, I'll check if a second reviewer would be helpful.

codecov · 2025-04-10T23:26:43Z

Codecov Report

Attention: Patch coverage is 99.15254% with 2 lines in your changes missing coverage. Please review.

Project coverage is 89.21%. Comparing base (ef0fcab) to head (75f81b5).
Report is 19 commits behind head on main.

Files with missing lines	Patch %	Lines
lightning/src/ln/chan_utils.rs	98.81%	2 Missing ⚠️

Additional details and impacted files

@@ Coverage Diff @@## main #3689 +/- ##
==========================================
+ Coverage 89.07% 89.21% +0.13% 
==========================================
Files 156 156 Lines 123507 124130 +623 Branches 123507 124130 +623 ==========================================
+ Hits 110017 110742 +725 + Misses 10801 10724 -77 + Partials 2689 2664 -25

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

TheBlueMatt

One last real comment

TheBlueMatt

New commit basically LTGM.

TheBlueMatt

Basically LGTM, Probably needs another look from @wpaulino otherwise I'm happy.

TheBlueMatt · 2025-04-14T20:11:25Z

 	///
-	/// This is not exported to bindings users due to the generic though we likely should expose a version without
-	pub fn new_with_auxiliary_htlc_data<T>(commitment_number: u64, per_commitment_point: &PublicKey, to_broadcaster_value_sat: u64, to_countersignatory_value_sat: u64, feerate_per_kw: u32, htlcs_with_aux: &mut Vec<(HTLCOutputInCommitment, T)>, channel_parameters: &DirectedChannelTransactionParameters, secp_ctx: &Secp256k1<secp256k1::All>) -> CommitmentTransaction {
+	/// All HTLCs MUST be above the dust limit for the channel.\


stray , I think? note that if you dont leave an extra line between paragraphs, rustdoc assumes you meant for them to be in the same paragraph and ignores newlines in the source.

rustdoc actually turns \ into a single newline character.

Heh, TIL, but it also makes the source unreadable :)

TheBlueMatt

Probably worth adding the sort in the rebuild case, but gonna go ahead and land it.

TheBlueMatt · 2025-04-16T01:02:38Z

+ let (obscured_commitment_transaction_number, txins) = Self::build_inputs(self.commitment_number, channel_parameters);
+
+ // First rebuild the htlc outputs, note that these are already sorted
+ let mut outputs = Self::build_htlc_outputs(keys, &self.nondust_htlcs, channel_parameters.channel_type_features());


It would be nice to resort the HTLCs here before we return. The point of rebuild_transaction is to avoid trusting the pre-built commitment transaction entirely and build it from the HTLC list.

Should I get this done now ?

Can we have fn verify take a &mut self ? To highlight that this call will resort the HTLCs if necessary. Then fn rebuild_transaction can call build_outputs_and_htlcs just like fn new without any new memory allocations.

Ah looks like this upsets the validation APIs of the signer that take a &CommitmentTransaction

Maybe we make it return a Result? Can happen in a followup but up to you if you want to do it now.

Meaning rebuild_transaction returns an Err if the HTLCs are not sorted ? Yes

TheBlueMatt · 2025-04-16T01:08:10Z

Oops, sorry lol, didn't mean for the nit on docs to be blocking.

tankyleo · 2025-04-16T01:09:39Z

Ah all good I do agree with the aesthetics.

This commit reworks how channel is told about which HTLCs were assigned which index upon the build of a `CommitmentTransaction`. Previously, `CommitmentTransaction` was given an exclusive reference to channel's HTLC-source table, and `CommitmentTransaction` populated the transaction output indices in that table via this exclusive reference. As a result, the public API of `CommitmentTransaction` included a generic parameter, and an exclusive reference. We remove both of these in preparation for the upcoming `TxBuilder` trait. This cleans up the API, and makes it more bindings-friendly. Henceforth, channel populates the HTLC-source table via a brute-force search of each htlc in `CommitmentTransaction`. This is an O(n^2) operation, but n is small enough that we ignore the performance hit. We also take this opportunity to cleanup how we build and sort the commitment transaction outputs together with the htlc output data in `CommitmentTransaction`. The goal is to keep the number of vector allocations to a minimum; we now only allocate a single vector to hold the transaction outputs, and do all the sorting in-place. Finally, we add test coverage to assert that `CommitmentTransaction::verify` actually checks that the HTLCs in the `nondust_htlcs` member are actually sorted.

tankyleo · 2025-04-16T20:00:48Z

@wpaulino I previously mistakenly started trusting the order of HTLCs in the nondust_htlcs member in the verify call - I correct this here.

tankyleo · 2025-03-29T00:49:22Z

 Remove exclusive reference and generic from `CommitmentTransaction` API
This commit reworks how channel is told about which HTLCs were assigned
which index upon the build of a `CommitmentTransaction`.
Previously, `CommitmentTransaction` was given an exclusive reference to
channel's HTLC-source table, and `CommitmentTransaction` populated the
transaction output indices in that table via this exclusive reference.
As a result, the public API of `CommitmentTransaction` included a
generic parameter, and an exclusive reference.
We remove both of these in preparation for the upcoming `TxBuilder`
trait. This cleans up the API, and makes it more bindings-friendly.
Henceforth, channel populates the HTLC-source table via a brute-force
search of each htlc in `CommitmentTransaction`. This is an O(n^2)
operation, but n is small enough that we ignore the performance hit.
We also take this opportunity to cleanup how we build and sort the
commitment transaction outputs together with the htlc output data
in `CommitmentTransaction`. The goal is to keep the number of vector
allocations to a minimum; we now only allocate a single vector to hold
the transaction outputs, and do all the sorting in-place.

ldk-reviews-bot · 2025-03-29T00:49:24Z

👋 Thanks for assigning @wpaulino as a reviewer!
I'll wait for their review and will help manage the review process.
Once they submit their review, I'll check if a second reviewer would be helpful.

codecov · 2025-04-10T23:26:43Z

Codecov Report

Attention: Patch coverage is 99.15254% with 2 lines in your changes missing coverage. Please review.

Project coverage is 89.21%. Comparing base (ef0fcab) to head (75f81b5).
Report is 19 commits behind head on main.

Files with missing lines	Patch %	Lines
lightning/src/ln/chan_utils.rs	98.81%	2 Missing ⚠️

Additional details and impacted files

@@ Coverage Diff @@## main #3689 +/- ##
==========================================
+ Coverage 89.07% 89.21% +0.13% 
==========================================
Files 156 156 Lines 123507 124130 +623 Branches 123507 124130 +623 ==========================================
+ Hits 110017 110742 +725 + Misses 10801 10724 -77 + Partials 2689 2664 -25

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

TheBlueMatt

One last real comment

TheBlueMatt

New commit basically LTGM.

TheBlueMatt

Basically LGTM, Probably needs another look from @wpaulino otherwise I'm happy.

TheBlueMatt · 2025-04-14T20:11:25Z

 	///
-	/// This is not exported to bindings users due to the generic though we likely should expose a version without
-	pub fn new_with_auxiliary_htlc_data<T>(commitment_number: u64, per_commitment_point: &PublicKey, to_broadcaster_value_sat: u64, to_countersignatory_value_sat: u64, feerate_per_kw: u32, htlcs_with_aux: &mut Vec<(HTLCOutputInCommitment, T)>, channel_parameters: &DirectedChannelTransactionParameters, secp_ctx: &Secp256k1<secp256k1::All>) -> CommitmentTransaction {
+	/// All HTLCs MUST be above the dust limit for the channel.\


stray , I think? note that if you dont leave an extra line between paragraphs, rustdoc assumes you meant for them to be in the same paragraph and ignores newlines in the source.

rustdoc actually turns \ into a single newline character.

Heh, TIL, but it also makes the source unreadable :)

TheBlueMatt

Probably worth adding the sort in the rebuild case, but gonna go ahead and land it.

TheBlueMatt · 2025-04-16T01:02:38Z

+ let (obscured_commitment_transaction_number, txins) = Self::build_inputs(self.commitment_number, channel_parameters);
+
+ // First rebuild the htlc outputs, note that these are already sorted
+ let mut outputs = Self::build_htlc_outputs(keys, &self.nondust_htlcs, channel_parameters.channel_type_features());


It would be nice to resort the HTLCs here before we return. The point of rebuild_transaction is to avoid trusting the pre-built commitment transaction entirely and build it from the HTLC list.

Should I get this done now ?

Can we have fn verify take a &mut self ? To highlight that this call will resort the HTLCs if necessary. Then fn rebuild_transaction can call build_outputs_and_htlcs just like fn new without any new memory allocations.

Ah looks like this upsets the validation APIs of the signer that take a &CommitmentTransaction

Maybe we make it return a Result? Can happen in a followup but up to you if you want to do it now.

Meaning rebuild_transaction returns an Err if the HTLCs are not sorted ? Yes

TheBlueMatt · 2025-04-16T01:08:10Z

Oops, sorry lol, didn't mean for the nit on docs to be blocking.

tankyleo · 2025-04-16T01:09:39Z

Ah all good I do agree with the aesthetics.

This commit reworks how channel is told about which HTLCs were assigned which index upon the build of a `CommitmentTransaction`. Previously, `CommitmentTransaction` was given an exclusive reference to channel's HTLC-source table, and `CommitmentTransaction` populated the transaction output indices in that table via this exclusive reference. As a result, the public API of `CommitmentTransaction` included a generic parameter, and an exclusive reference. We remove both of these in preparation for the upcoming `TxBuilder` trait. This cleans up the API, and makes it more bindings-friendly. Henceforth, channel populates the HTLC-source table via a brute-force search of each htlc in `CommitmentTransaction`. This is an O(n^2) operation, but n is small enough that we ignore the performance hit. We also take this opportunity to cleanup how we build and sort the commitment transaction outputs together with the htlc output data in `CommitmentTransaction`. The goal is to keep the number of vector allocations to a minimum; we now only allocate a single vector to hold the transaction outputs, and do all the sorting in-place. Finally, we add test coverage to assert that `CommitmentTransaction::verify` actually checks that the HTLCs in the `nondust_htlcs` member are actually sorted.

tankyleo · 2025-04-16T20:00:48Z

@wpaulino I previously mistakenly started trusting the order of HTLCs in the nondust_htlcs member in the verify call - I correct this here.

tankyleo · 2025-03-29T00:49:22Z

 Remove exclusive reference and generic from `CommitmentTransaction` API
This commit reworks how channel is told about which HTLCs were assigned
which index upon the build of a `CommitmentTransaction`.
Previously, `CommitmentTransaction` was given an exclusive reference to
channel's HTLC-source table, and `CommitmentTransaction` populated the
transaction output indices in that table via this exclusive reference.
As a result, the public API of `CommitmentTransaction` included a
generic parameter, and an exclusive reference.
We remove both of these in preparation for the upcoming `TxBuilder`
trait. This cleans up the API, and makes it more bindings-friendly.
Henceforth, channel populates the HTLC-source table via a brute-force
search of each htlc in `CommitmentTransaction`. This is an O(n^2)
operation, but n is small enough that we ignore the performance hit.
We also take this opportunity to cleanup how we build and sort the
commitment transaction outputs together with the htlc output data
in `CommitmentTransaction`. The goal is to keep the number of vector
allocations to a minimum; we now only allocate a single vector to hold
the transaction outputs, and do all the sorting in-place.

ldk-reviews-bot · 2025-03-29T00:49:24Z

👋 Thanks for assigning @wpaulino as a reviewer!
I'll wait for their review and will help manage the review process.
Once they submit their review, I'll check if a second reviewer would be helpful.

codecov · 2025-04-10T23:26:43Z

Codecov Report

Attention: Patch coverage is 99.15254% with 2 lines in your changes missing coverage. Please review.

Project coverage is 89.21%. Comparing base (ef0fcab) to head (75f81b5).
Report is 19 commits behind head on main.

Files with missing lines	Patch %	Lines
lightning/src/ln/chan_utils.rs	98.81%	2 Missing ⚠️

Additional details and impacted files

@@ Coverage Diff @@## main #3689 +/- ##
==========================================
+ Coverage 89.07% 89.21% +0.13% 
==========================================
Files 156 156 Lines 123507 124130 +623 Branches 123507 124130 +623 ==========================================
+ Hits 110017 110742 +725 + Misses 10801 10724 -77 + Partials 2689 2664 -25

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

TheBlueMatt

One last real comment

TheBlueMatt

New commit basically LTGM.

TheBlueMatt

Basically LGTM, Probably needs another look from @wpaulino otherwise I'm happy.

TheBlueMatt · 2025-04-14T20:11:25Z

 	///
-	/// This is not exported to bindings users due to the generic though we likely should expose a version without
-	pub fn new_with_auxiliary_htlc_data<T>(commitment_number: u64, per_commitment_point: &PublicKey, to_broadcaster_value_sat: u64, to_countersignatory_value_sat: u64, feerate_per_kw: u32, htlcs_with_aux: &mut Vec<(HTLCOutputInCommitment, T)>, channel_parameters: &DirectedChannelTransactionParameters, secp_ctx: &Secp256k1<secp256k1::All>) -> CommitmentTransaction {
+	/// All HTLCs MUST be above the dust limit for the channel.\


stray , I think? note that if you dont leave an extra line between paragraphs, rustdoc assumes you meant for them to be in the same paragraph and ignores newlines in the source.

rustdoc actually turns \ into a single newline character.

Heh, TIL, but it also makes the source unreadable :)

TheBlueMatt

Probably worth adding the sort in the rebuild case, but gonna go ahead and land it.

TheBlueMatt · 2025-04-16T01:02:38Z

+ let (obscured_commitment_transaction_number, txins) = Self::build_inputs(self.commitment_number, channel_parameters);
+
+ // First rebuild the htlc outputs, note that these are already sorted
+ let mut outputs = Self::build_htlc_outputs(keys, &self.nondust_htlcs, channel_parameters.channel_type_features());


It would be nice to resort the HTLCs here before we return. The point of rebuild_transaction is to avoid trusting the pre-built commitment transaction entirely and build it from the HTLC list.

Should I get this done now ?

Can we have fn verify take a &mut self ? To highlight that this call will resort the HTLCs if necessary. Then fn rebuild_transaction can call build_outputs_and_htlcs just like fn new without any new memory allocations.

Ah looks like this upsets the validation APIs of the signer that take a &CommitmentTransaction

Maybe we make it return a Result? Can happen in a followup but up to you if you want to do it now.

Meaning rebuild_transaction returns an Err if the HTLCs are not sorted ? Yes

TheBlueMatt · 2025-04-16T01:08:10Z

Oops, sorry lol, didn't mean for the nit on docs to be blocking.

tankyleo · 2025-04-16T01:09:39Z

Ah all good I do agree with the aesthetics.

This commit reworks how channel is told about which HTLCs were assigned which index upon the build of a `CommitmentTransaction`. Previously, `CommitmentTransaction` was given an exclusive reference to channel's HTLC-source table, and `CommitmentTransaction` populated the transaction output indices in that table via this exclusive reference. As a result, the public API of `CommitmentTransaction` included a generic parameter, and an exclusive reference. We remove both of these in preparation for the upcoming `TxBuilder` trait. This cleans up the API, and makes it more bindings-friendly. Henceforth, channel populates the HTLC-source table via a brute-force search of each htlc in `CommitmentTransaction`. This is an O(n^2) operation, but n is small enough that we ignore the performance hit. We also take this opportunity to cleanup how we build and sort the commitment transaction outputs together with the htlc output data in `CommitmentTransaction`. The goal is to keep the number of vector allocations to a minimum; we now only allocate a single vector to hold the transaction outputs, and do all the sorting in-place. Finally, we add test coverage to assert that `CommitmentTransaction::verify` actually checks that the HTLCs in the `nondust_htlcs` member are actually sorted.

tankyleo · 2025-04-16T20:00:48Z

@wpaulino I previously mistakenly started trusting the order of HTLCs in the nondust_htlcs member in the verify call - I correct this here.

tankyleo force-pushed the small-n-squared branch from 1cc3c50 to fd57c35Compare March 29, 2025 00:50

tankyleo changed the title ~~Remove generic and mutable references from commitment transaction API~~Remove generic and mutable references from CommitmentTransaction APIMar 29, 2025

tankyleo force-pushed the small-n-squared branch from fd57c35 to ecc7dcfCompare March 29, 2025 00:58

tankyleo changed the title ~~Remove generic and mutable references from CommitmentTransaction API~~Remove generic and exclusive references from CommitmentTransaction APIMar 29, 2025

tankyleo changed the title ~~Remove generic and exclusive references from CommitmentTransaction API~~Remove generic and exclusive reference from CommitmentTransaction APIMar 29, 2025

tankyleo force-pushed the small-n-squared branch 2 times, most recently from c8a753e to b04cc76Compare March 29, 2025 01:07

tankyleo changed the title ~~Remove generic and exclusive reference from CommitmentTransaction API~~Remove exclusive reference and generic from CommitmentTransaction APIMar 29, 2025

tankyleo mentioned this pull request Mar 10, 2025
Custom Transactions Tracking Issue #3603
Open
24 tasks

tankyleo force-pushed the small-n-squared branch 4 times, most recently from f4a3cc3 to acad6d8Compare April 8, 2025 18:34

tankyleo force-pushed the small-n-squared branch from acad6d8 to c86b528Compare April 9, 2025 18:00

tankyleo marked this pull request as ready for review April 9, 2025 18:25

ldk-reviews-bot requested a review from wpaulino April 9, 2025 18:26

tankyleo commented Apr 9, 2025
View reviewed changes

Comment threadlightning/src/ln/chan_utils.rs Outdated

wpaulino reviewed Apr 10, 2025
View reviewed changes

Comment threadlightning/src/ln/channel.rs Outdated

tankyleo force-pushed the small-n-squared branch from c86b528 to fcdacc9Compare April 10, 2025 23:17

tankyleo requested review from TheBlueMatt and wpaulino April 10, 2025 23:31

TheBlueMatt reviewed Apr 11, 2025
View reviewed changes

Comment threadlightning/src/ln/chan_utils.rs
Comment threadlightning/src/ln/chan_utils.rs
Comment threadlightning/src/ln/chan_utils.rs Outdated

tankyleo force-pushed the small-n-squared branch from fcdacc9 to 7e3a4e1Compare April 11, 2025 21:17

tankyleo requested a review from TheBlueMatt April 11, 2025 21:23

tankyleo force-pushed the small-n-squared branch 4 times, most recently from 1d7a1d8 to 262351cCompare April 12, 2025 03:47

tankyleo force-pushed the small-n-squared branch from f021734 to 1621c12Compare April 14, 2025 06:16

TheBlueMatt reviewed Apr 14, 2025
View reviewed changes

Comment threadlightning/src/ln/chan_utils.rs Outdated
Comment threadlightning/src/ln/chan_utils.rs Outdated
Comment threadlightning/src/ln/chan_utils.rs Outdated
Comment threadlightning/src/ln/chan_utils.rs Outdated

tankyleo force-pushed the small-n-squared branch from 1621c12 to 258b277Compare April 14, 2025 19:17

tankyleo requested a review from TheBlueMatt April 14, 2025 19:22

tankyleo force-pushed the small-n-squared branch from 258b277 to ee42e2aCompare April 14, 2025 20:11

TheBlueMatt reviewed Apr 14, 2025
View reviewed changes

tankyleo force-pushed the small-n-squared branch 2 times, most recently from 8bc75b9 to 5252437Compare April 15, 2025 02:10

wpaulino reviewed Apr 15, 2025
View reviewed changes

Comment threadlightning/src/ln/channel.rs
Comment threadlightning/src/ln/chan_utils.rs Outdated
Comment threadlightning/src/ln/chan_utils.rs Outdated

tankyleo force-pushed the small-n-squared branch 2 times, most recently from b82a9d5 to b44f272Compare April 15, 2025 18:45

tankyleo requested a review from wpaulino April 15, 2025 18:45

tankyleo force-pushed the small-n-squared branch 2 times, most recently from 2f7567d to 5e67a1cCompare April 15, 2025 19:01

wpaulino previously approved these changes Apr 15, 2025
View reviewed changes

tankyleo requested a review from TheBlueMatt April 15, 2025 21:08

tankyleo dismissed wpaulino’s stale review via 75f81b5April 16, 2025 00:47

tankyleo force-pushed the small-n-squared branch from 5e67a1c to 75f81b5Compare April 16, 2025 00:47

TheBlueMatt previously approved these changes Apr 16, 2025
View reviewed changes

tankyleo dismissed TheBlueMatt’s stale review via 6460db0April 16, 2025 19:45

tankyleo force-pushed the small-n-squared branch from 75f81b5 to 6460db0Compare April 16, 2025 19:45

tankyleo requested a review from TheBlueMatt April 16, 2025 19:58

tankyleo requested a review from wpaulino April 16, 2025 20:01

TheBlueMatt approved these changes Apr 16, 2025
View reviewed changes

wpaulino approved these changes Apr 16, 2025
View reviewed changes

TheBlueMatt merged commit 22a18a8 into lightningdevkit:mainApr 16, 2025