Updated docs on PeerManager::process_events.
#3395
Conversation
lightning/src/ln/peer_handler.rs
Outdated
| /// or one of the other clients provided in our language bindings, as it will be called | ||
| /// automatically when a message is received from a peer. However, it can in some cases delay | ||
| /// messages from going out if peers are not sending messages, so it is still recommended to | ||
| /// call it routinely from [`lightning-background-processor`] or a similar background task. |
There was a problem hiding this comment.
The logic of when its called in lightning-background-processor is that that BP gets woken up ~every time there's a state change in the channel logic (implying we likely have a message to send). So its not really just "calling it routinely" but rather calling it when we have a message to send because we changed the channel state by sending/claiming/forwarding a payment. Would be nice to capture that here somehow.
There was a problem hiding this comment.
Tried to capture that, but LMK if you have any better wording in mind!
mhrheaume
force-pushed
the
mhr/process_events_docs
branch
from
575d2b1
to
b375d71
Compare
lightning/src/ln/peer_handler.rs
Outdated
| /// automatically when a message is received from a peer. However, it can in some cases delay | ||
| /// messages from going out if peers are not sending messages, so it is still recommended to | ||
| /// call it when messages may need to be sent, e.g. after processing channel events that may | ||
| /// have resulted in an update to the channel state. |
There was a problem hiding this comment.
Now it seems to imply that even if you're using the BP you have to call it manually? Maybe we should just flip this on its head and say that this needs to be called any time we may have messages to send, and that its automatically called after processing messages by net-tokio and by the BP when the channel state has changed, but for users not using both of those they may need to call it manually?
Codecov ReportAll modified and coverable lines are covered by tests ✅
Additional details and impacted files@@ Coverage Diff @@
## main #3395 +/- ##
==========================================
+ Coverage 89.56% 89.64% +0.08%
==========================================
Files 127 128 +1
Lines 103547 104881 +1334
Branches 103547 104881 +1334
==========================================
+ Hits 92744 94025 +1281
- Misses 8094 8139 +45
- Partials 2709 2717 +8 ☔ View full report in Codecov by Sentry. |
Try to make it a bit more clear that there are downsides to solely relying on `lightning-net-tokio`, and it's still recommended to occasionally call this function in a separate loop.
mhrheaume
force-pushed
the
mhr/process_events_docs
branch
from
b375d71
to
9d07106
Compare
Try to make it a bit more clear that there are downsides to solely relying on
lightning-net-tokio, and it's still recommended to occasionally call this function in a separate loop.