Bump ldk-node dependency to 00dba456a10cb60172eaa74e0a8b11c5e0473dcc#230
Bump ldk-node dependency to 00dba456a10cb60172eaa74e0a8b11c5e0473dcc#230enigbe wants to merge 3 commits into
00dba456a10cb60172eaa74e0a8b11c5e0473dcc#230Conversation
|
👋 Thanks for assigning @benthecarman as a reviewer! |
|
Can you bump to |
| reserved 3; | ||
| reserved "bolt11_jit"; |
There was a problem hiding this comment.
we aren't worrying about backwards compat yet so you can remove this
There was a problem hiding this comment.
Noted. I'll remove all code that considered backwards compatibility.
| // See more: https://docs.rs/ldk-node/latest/ldk_node/enum.Event.html#variant.PaymentForwarded | ||
| message ForwardedPayment{ | ||
| // The channel id of the incoming channel between the previous node and us. | ||
| // Deprecated: use `prev_htlcs` instead. This is populated from `prev_htlcs[0]` |
There was a problem hiding this comment.
if these fields are deprecated it's probably best if we drop them all
| // | ||
| // Will only be `Some` once we received the payment. | ||
| optional uint64 counterparty_skimmed_fee_msat = 5; | ||
| // Replaces the former `PaymentKind.bolt11_jit.counterparty_skimmed_fee_msat` |
There was a problem hiding this comment.
we dont need to mention the old format, we haven't released yet so no one knows the former one
There was a problem hiding this comment.
this doc still talks about old format
Sure thing. Working on this now. |
enigbe
force-pushed
the
2026-06-bump-ldk-node
branch
3 times, most recently
from
94e233f to
f936ac6
Compare
|
Hi @benthecarman |
benthecarman
changed the title
47dad6d909af0fbb53e76d07740c04df5abdde3b00dba456a10cb60172eaa74e0a8b11c5e0473dcc
| // | ||
| // Will only be `Some` once we received the payment. | ||
| optional uint64 counterparty_skimmed_fee_msat = 5; | ||
| // Replaces the former `PaymentKind.bolt11_jit.counterparty_skimmed_fee_msat` |
There was a problem hiding this comment.
this doc still talks about old format
enigbe
force-pushed
the
2026-06-bump-ldk-node
branch
from
f936ac6 to
30e04a1
Compare
Cleaned up the documentation still referencing |
- Update ldk-node to a newer revision and adapt ldk-server to the changed payment event surface. PaymentForwarded now preserves the full prev_htlcs and next_htlcs vectors in the gRPC model. - Remove the Bolt11Jit payment kind as a breaking API change. JIT BOLT 11 payments are now represented as regular Bolt11 payments with counterparty_skimmed_fee_msat set when applicable. - Regenerate protobuf bindings and drop the obsolete Bolt11Jit serde attribute. AI-Assisted-By: OpenAI Codex
enigbe
force-pushed
the
2026-06-bump-ldk-node
branch
from
30e04a1 to
d41f6fd
Compare
LDK may render invoice and offer descriptions through safe display wrappers before CLI output sanitization sees them. In that case, bidi controls are replaced instead of escaped as \uXXXX. Update the tests to check the behavior that matters: decoded CLI output must not contain raw bidi override characters.
Build the test TLS server config with rustls's ring provider explicitly instead of relying on process-global provider auto-detection. This avoids panics when rustls cannot infer a single CryptoProvider from the compiled feature set.
enigbe
force-pushed
the
2026-06-bump-ldk-node
branch
from
e407a43 to
1020c46
Compare
|
Added a couple more commits to address failing tests. If you consider them out of scope for this PR, I am happy to move them to another. |
3 participants
What this PR does:
Bumps the
ldk-nodedependency to revision 47dad6d, which, among other changes, contains the prerequisite for exposingNodeFeaturestoldk-servercallers via the nodestatusendpoint. We need this for callers interested in querying the features supported by the node.Notable upstream changes adapted here
Bolt11Jitmerged into Bolt11: ldk-node folded the JIT payment variant intoBolt11, adding acounterparty_skimmed_fee_msatfield. TheBolt11Jitproto message, its serde attribute inbuild.rs, and the corresponding proto_adapter arm are all removed. The original proto tag numbers for Bolt12Offer (4), Bolt12Refund (5), and Spontaneous (6) are preserved — tag 3 is reserved to avoid wire-compatibility issues with previously-serializedBolt11Jitdata.PaymentForwardedevent restructured aroundHTLCLocator: The six flat channel/node/user-channel-id fields are replaced byprev_htlcsandnext_htlcs(Vec<HTLCLocator>), supporting multi-HTLC forwards. The legacy scalar fields are retained in the proto as deprecated, populated from the first element of each vec for backward compatibility (not sure if this is the correct way to handle this).