You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
convert.RunnerRole seems like a bit of a hack, as if we're adapting Alan to the existing code rather than the other way around.
For example in QBFT storage: is there a good reason to store QBFT instances by BeaconRole at post-fork? At post-fork we perform consensus per RunnerRole not per BeaconRole, so intuitively they should be stored by RunnerRole.
This issue exists to discuss this and come up with a PR to fix if necessary.
The text was updated successfully, but these errors were encountered:
@moshe-blox, in exporter, we want to save consensus results (qbft storage..) in the 'old' format, so convert.RunnerRole basically enables using RunnerRole but also saving Attester and Sync committee duties. Since exporter doesn't share anything about committee consensus now and only about post-consensus.
@y0sher i see, but still i think the QBFT storage should use the spectypes role, and only in exporter we should see usage of the convert function (and not in other packages)
convert.RunnerRole
seems like a bit of a hack, as if we're adapting Alan to the existing code rather than the other way around.For example in QBFT storage: is there a good reason to store QBFT instances by BeaconRole at post-fork? At post-fork we perform consensus per RunnerRole not per BeaconRole, so intuitively they should be stored by RunnerRole.
This issue exists to discuss this and come up with a PR to fix if necessary.
The text was updated successfully, but these errors were encountered: