integrate votemarket vebal/vlaura data into allocator automation - #302
Conversation
| data = _fetch_json(round_url_template.format(round_id=rid)) | ||
| for gauge in data["analytics"]: | ||
| addr = gauge["gauge"].lower() | ||
| deposited[addr] = deposited.get(addr, 0) + gauge["totalDeposited"] |
There was a problem hiding this comment.
did a deep dive and this is a big bug:
get_aura_share_gauge should be based on the votes, not the deposit
i think this is also why we are seeing unexpected low numbers for the backfilled auravebalShare (<=.5), whereas i would expect the usual ~.65
There was a problem hiding this comment.
good catch. yes, should be using nonBlacklistedVotes. fixed and updated backfilled data too: 76e038
There was a problem hiding this comment.
Agreed! Great find and fix looks good
| bal_round_ids = _find_matching_rounds(bal_metadata, period_start, period_end) | ||
| aura_round_ids = _find_matching_rounds(aura_metadata, period_start, period_end) |
There was a problem hiding this comment.
should we check if there is actually data? not sure on the cron timing and the data availability on stakedao's side
There was a problem hiding this comment.
good point. added to check to handle this scenario better. if no data, it will just exit early and new data will be caught the following week: 76e038
… run on thursday weekly
3 participants
solves #295 - problem is that data cant be queried for a round until that round ends. solution is new automation, backfill_aura_split.yaml, that runs on after a new rounds ends and backfills the data automatically. backfilled data starting from fee runs after
1767225600have been added here tooalso fixes #300