period_tail in get_eth1_vote

[paulhauner]

The change I have suggested targets get_eth1_vote() and is the result of the following 3 (potentially incorrect) observations:

1. slot seems a misleading variable name for indexing state.eth1_data_votes.
2. slot % SLOTS_PER_ETH1_VOTING_PERIOD when len(state.eth1_data_votes) <= SLOTS_PER_ETH1_VOTING_PERIOD is a no-op.
3. I assume the period_tail is designed to validate earlier eth1_data_votes if they are in all_eth1_data (but not in new_eth1_data). However, when period_tail = slot >= integer_squareroot(...), only later eth1_data_votes are tested against all_eth1_data.

[paulhauner]

It's probably worth considering this alongside #1464

[djrtwo]

1. agreed
2. agreed
3. You have this flipped. Essentially, for earlier in the period, we just want people to attempt to vote on new data, so in the beginning of the period, only pile on your vote to votes that are for newer data. But once we've entered the tail of the period, we just want consensus so we consider all votes

---

The previous method was making the assumption that there was one eth1 vote per slot. I am considering that maybe it should just be the following to remove this assumption (uses state.slot instead of counting the votes in state for figuring out if in period tail)

def get_eth1_vote(state: BeaconState, previous_eth1_distance: uint64) -> Eth1Data:
    new_eth1_data = [get_eth1_data(distance) for distance in range(ETH1_FOLLOW_DISTANCE, 2 * ETH1_FOLLOW_DISTANCE)]
    all_eth1_data = [get_eth1_data(distance) for distance in range(ETH1_FOLLOW_DISTANCE, previous_eth1_distance)]

    period_tail = state.slot % SLOTS_PER_ETH1_VOTING_PERIOD >= integer_squareroot(SLOTS_PER_ETH1_VOTING_PERIOD)
    if period_tail:
        votes_to_consider = all_eth1_data
    else:
        votes_to_consider = new_eth1_data

    valid_votes = [vote for vote in state.eth1_data_votes if vote in votes_to_consider]

    return max(
        valid_votes,
        key=lambda v: (valid_votes.count(v), -all_eth1_data.index(v)),  # Tiebreak by smallest distance
        default=get_eth1_data(ETH1_FOLLOW_DISTANCE),
    )

[djrtwo]

closing in favor of #1468