Expose ChannelCounterparty and ReserveType in ChannelDetails by enigbe · Pull Request #841 · lightningdevkit/ldk-node

enigbe · 2026-03-24T21:03:24Z

What this PR does

In #305, we needed a way to expose channel type and the channel reserve without leaking implementation details not useful to users. The related discussion in #141 proposed a ReserveType abstraction that bakes in both in a user-relevant way. This PR introduces the said enumeration to ChannelDetails with the following variants:

Legacy: signifying pre-anchor channel types where on-chain fees paying for broadcast transactions following channel closure were pre-determined
TrustedPeersNoReserve: for anchor type channels with trusted peers
Adaptive: indicating anchor channel with adaptive reserve, reflecting dynamic best-effort attempt at fee-bumping.
(see @jkczyz's suggestion)

We modify the ChannelDetails constructor to accept an optional anchor channels config to address the challenge of users cross-referencing the channel details and config to determine if counterparties are trusted.

Additionally, following recommendation in #810 about negotiated features, this PR exposes per-peer InitFeatures, and as a consequence, modifies ChannelDetails by replacing the flattened counterparty_* with ChannelCounterparty that encapsulates them and mirror LDK's.

Issue Addressed

Closes #305.
Builds on discussions and recommendations from #141 and #810.

ldk-reviews-bot · 2026-03-24T21:03:27Z

👋 I see @tnull was un-assigned.
If you'd like another reviewer assignment, please click here.

This fixup moves node feature exposure from freestanding APIs to NodeStatus, as suggested in review. Rather than exposing init_features(), channel_features(), bolt11_invoice_features(), and node_features() as separate public methods on Node, this embeds NodeFeatures in the NodeStatus struct returned by Node::status(). Additionally, channel and invoice features at node level are confusing. Users would expect negotiated per-peer/channel/invoice features, not what the node generally supports. Access to negotiated features are addressed in lightningdevkit#841

tnull

Thanks! Looks pretty good, have a few initial comments.

tnull

One comment, otherwise looks good. Feel free to squash (also the new fixup)

enigbe · 2026-03-30T10:25:22Z

Squashed all fix-ups.

ldk-reviews-bot · 2026-05-13T09:24:42Z

🔔 1st Reminder

Hey @tnull! This PR has been waiting for your review.
Please take a look when you have a chance. If you're unable to review, please let us know so we can find another reviewer.

ldk-reviews-bot · 2026-05-16T00:00:50Z

🔔 2nd Reminder

Hey @tnull! This PR has been waiting for your review.
Please take a look when you have a chance. If you're unable to review, please let us know so we can find another reviewer.

ldk-reviews-bot · 2026-05-18T00:01:50Z

🔔 3rd Reminder

Hey @tnull! This PR has been waiting for your review.
Please take a look when you have a chance. If you're unable to review, please let us know so we can find another reviewer.

tnull

Looks good I think. One nit.

tnull

This needs a minor rebase by now.

enigbe · 2026-06-04T17:58:11Z

Thanks for the review @tnull.

I've rebased and also simplified the documentation on InitFeatures to match the lean style used with NodeFeatures.

tnull · 2026-06-05T08:52:56Z

+/// to better separate parameters.
+#[derive(Clone, Debug, PartialEq)]
+#[cfg_attr(feature = "uniffi", derive(uniffi::Record))]
+pub struct ChannelCounterparty {


This needs to be re-exported via a pub use in src/ffi/types.rs.

Why does this need to be re-exported in src/ffi/types.rs? We don't do this for other uniffi::Records. Do you mean src/lib.rs? If it's the latter, I've re-exported in the fix-up commit.

tnull · 2026-06-16T14:02:25Z

+	pub(crate) inner: LdkInitFeatures,
+}
+
+impl InitFeatures {


This seems to be missing #[uniffi::export]?

This has been added.

No, the one you added is also fine to keep, but we'll also need to add uniffi::export to the impl block, as otherwise the methods won't be available in bindings.

That's right. I've added this and extended the coverage in the python test to check the now-exposed methods. I noticed NodeFeatures had the same issue, so I fixed and covered that as well.

tnull · 2026-06-16T14:04:54Z

+	pub(crate) fn from_ldk(
+		value: LdkChannelDetails, anchor_channels_config: Option<&AnchorChannelsConfig>,
+	) -> Self {
+		let reserve_type =


Hmm, the LDK docs on channel_type say: "None until negotiation completes and the channel type is finalized."

This means we'll erroneously report all channels as Legacy util the negotiation is fully done, no? Probably will need to make this an optional field, too, or have a state for 'not ready' or similar?

Leaning more towards making it optional. Also, I updated the documentation to highlight this.

We previously flattened ChannelCounterparty fields into ChannelDetails as individual counterparty_* fields, and InitFeatures was entirely omitted. This made it impossible for consumers to access per-peer feature flags, and awkward to access counterparty forwarding information without navigating the flattened field names. This commit replaces the flattened fields with a structured ChannelCounterparty type that mirrors LDK's ChannelCounterparty, exposing InitFeatures and CounterpartyForwardingInfo that were previously inaccessible. We keep outbound_htlc_minimum_msat optional because it is unavailable before receiving OpenChannel or AcceptChannel, and re-export ChannelCounterparty so Rust consumers can name the type.

We expose the reserve type of each channel through a new ReserveType enum on ChannelDetails. This tells users whether a channel uses adaptive anchor reserves, has no reserve due to a trusted peer, or is a legacy pre-anchor channel. The reserve type is derived at query time in list_channels by checking the channel's type features against trusted_peers_no_reserve. We replace the From<LdkChannelDetails> implementation with an explicit from_ldk method that takes the anchor channels config. Additionally, we document the rationale behind selecting adaptive reserve type in the unlikely event the anchor channels config was previously set and then later removed.

We add requires_* counterparts for every supports_* method on InitFeatures, completing the BOLT 9 feature flag coverage for FFI consumers. We export the existing feature helper methods for both InitFeatures and NodeFeatures through UniFFI. NodeFeatures had the same missing export path as InitFeatures and was noticed while addressing the InitFeatures FFI exposure. Additionally, we extend the Python full-cycle test to cover node features and init features on their real runtime paths.

ldk-reviews-bot requested a review from tnull March 24, 2026 21:13

enigbe mentioned this pull request Mar 24, 2026

Expose node-invariant features via NodeStatus #810

Merged

tnull reviewed Mar 25, 2026

View reviewed changes

Comment thread src/types.rs Outdated

Comment thread src/types.rs Outdated

Comment thread src/types.rs Outdated

Comment thread src/types.rs Outdated

enigbe force-pushed the 2026-03-channel-reserve-type-exposure branch 4 times, most recently from 33a36bd to 63f32ab Compare March 30, 2026 06:34

tnull reviewed Mar 30, 2026

View reviewed changes

Comment thread src/types.rs Outdated

enigbe force-pushed the 2026-03-channel-reserve-type-exposure branch from 63f32ab to 56d86f7 Compare March 30, 2026 09:57

tnull reviewed Mar 30, 2026

View reviewed changes

Comment thread src/types.rs

enigbe force-pushed the 2026-03-channel-reserve-type-exposure branch from 56d86f7 to 77dcbc4 Compare April 3, 2026 09:24

tnull reviewed Apr 3, 2026

View reviewed changes

Comment thread src/ffi/types.rs

tnull reviewed Apr 6, 2026

View reviewed changes

Comment thread src/ffi/types.rs

enigbe force-pushed the 2026-03-channel-reserve-type-exposure branch from fe4597a to eb7f021 Compare April 9, 2026 08:01

enigbe force-pushed the 2026-03-channel-reserve-type-exposure branch 2 times, most recently from 86e9936 to 863d5e7 Compare April 22, 2026 09:04

enigbe force-pushed the 2026-03-channel-reserve-type-exposure branch from 863d5e7 to b5736f1 Compare May 11, 2026 08:14

enigbe requested a review from tnull May 11, 2026 09:24

tnull reviewed May 18, 2026

View reviewed changes

Comment thread src/ffi/types.rs Outdated

tnull reviewed Jun 1, 2026

View reviewed changes

enigbe force-pushed the 2026-03-channel-reserve-type-exposure branch 4 times, most recently from d67aaa8 to fdbd4ff Compare June 4, 2026 17:31

tnull reviewed Jun 5, 2026

View reviewed changes

enigbe force-pushed the 2026-03-channel-reserve-type-exposure branch 4 times, most recently from aa631a4 to b4dcc99 Compare June 12, 2026 11:00

enigbe force-pushed the 2026-03-channel-reserve-type-exposure branch from b4dcc99 to 3b06d8e Compare June 16, 2026 08:35

tnull reviewed Jun 16, 2026

View reviewed changes

enigbe requested a review from tnull June 17, 2026 08:07

tnull removed their request for review June 17, 2026 08:09

enigbe force-pushed the 2026-03-channel-reserve-type-exposure branch from 84095d9 to 2613151 Compare June 19, 2026 09:08

enigbe added 3 commits June 19, 2026 11:35

enigbe force-pushed the 2026-03-channel-reserve-type-exposure branch from 2613151 to fee7a73 Compare June 19, 2026 18:53

Conversation

enigbe commented Mar 24, 2026

What this PR does

Issue Addressed

Uh oh!

ldk-reviews-bot commented Mar 24, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

tnull left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

tnull left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

enigbe commented Mar 30, 2026

Uh oh!

Uh oh!

Uh oh!

Uh oh!

ldk-reviews-bot commented May 13, 2026

Uh oh!

ldk-reviews-bot commented May 16, 2026

Uh oh!

ldk-reviews-bot commented May 18, 2026

Uh oh!

tnull left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

tnull left a comment

Choose a reason for hiding this comment

Uh oh!

enigbe commented Jun 4, 2026

Uh oh!

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

3 participants

ldk-reviews-bot commented Mar 24, 2026 •

edited

Loading