Conversation
|
see if a grab of #2250 can make it. no strong opinion. |
|
That PR is still pending someone doing an analysis of what the impact is on node selection in others' routing algorithms. |
Codecov ReportAll modified and coverable lines are covered by tests ✅
❗ Your organization needs to install the Codecov GitHub app to enable full functionality. Additional details and impacted files@@ Coverage Diff @@
## main #2678 +/- ##
==========================================
- Coverage 88.76% 88.76% -0.01%
==========================================
Files 112 112
Lines 88474 88474
Branches 88474 88474
==========================================
- Hits 78537 78533 -4
- Misses 7702 7706 +4
Partials 2235 2235 ☔ View full report in Codecov by Sentry. |
valentinewallace
left a comment
There was a problem hiding this comment.
LGTM after squash
e2575e7 to
cc14449
Compare
CHANGELOG.md
Outdated
| ## API Updates | ||
| * BOLT12 sending and receiving is now supported as an alpha feature. You may | ||
| run into unexpected issues and will need to have a direct connection with | ||
| onion message peer targets in order to exchange messages. We are seeking |
There was a problem hiding this comment.
Maybe say "payment recipient" instead of "onion message peer targets in order to exchange messages"? Simpler and also includes the payment portion of the flow, which also requires a direct connection.
There was a problem hiding this comment.
But its not the "payment recipient" in the case of blinded paths. I agree we should make it less wordy, though.
There was a problem hiding this comment.
I guess it is something like "payment recipient, which must also be the introduction node in blinded paths".
There was a problem hiding this comment.
I'm confused, we definitely dont need a direct connection to the payment recipient, right?
There was a problem hiding this comment.
Right, we can't forward, and we currently only generate one-hop, but none of those require direct-connection. We need to be able to get a message to our peer via direct-connection-to-blinded-intro, but the responses from our peer can all come routed.
There was a problem hiding this comment.
Yeah, that's right. I was confusing myself about support for blinded payment paths, actually. We can find a payment path to them, though, so it's "the offer's recipient or the introduction node of its blinded paths".
There was a problem hiding this comment.
Okay, pushed a potential change, lmk if its reasonable.
24c98d2 to
25f7196
Compare
|
Also added missing backwards compat notes. Correct me if the second one is wrong but I believe we've changed the serialization of the pending waiting-for-invoice payment timeout, which will fail deserialization. |
25f7196 to
6c901b9
Compare
|
Squashed, added diff stats, ready to go 🎉 |
That's accurate, though I wasn't sure if it was necessary to have a release note since I think the |
|
True, but calling a method then trying to downgrade and immediately failing is generally something we document. |
6c901b9 to
b664875
Compare
Will need updates for the last three prs still.