g11tech
g11tech
yea seems like atleast at one place its still missing, can pick it up
i think this should not be an issue anymore as far as i have observed in writing some testcases , may be i can add a test case
I have had discussions on this in past if I remember, yes the fork version depends on attested header, but it doesnt matter if you keep or transmit an updated...
which particular method you are talking about? req/resp? api? gossip?
>.g use capella.LightClientOptimisticUpdate for the update itself but the sync_aggregate should use deneb.SyncAggregate the update is "upgraded" to signature slot's , so the entire update is/SHOULD be one one particular...
whwre don't we do that, can you point out?
> > which particular method you are talking about? req/resp? api? gossip? > > any place where we serialize / derserialize / validate light client updates > > e.g. eventstream...
updated, will continue on this post latest changes are PRed for the deneb ... affects produce/publish api all over
unpinning the issue but keeping it alive for some of the cleanup/optimization items
@hwwhww if you can fix some references and make CI pass, i can probably approve this and get merged in draft status