Closed viniabussafi closed 1 month ago
Analysing the outliers (results where the number of BPTs staked > BPT supply), me and Fábio identified an error on the bpt_supply spell, which was addressed on this PR.
Previous code:
`
joins_and_exits AS (
SELECT
j.block_date,
j.tokenOut AS bpt,
SUM(COALESCE(ajoins, 0) - COALESCE(aexits, 0)) OVER (PARTITION BY j.tokenOut ORDER BY j.block_date ASC) AS adelta
FROM joins j
FULL OUTER JOIN exits e ON j.block_date = e.block_date AND e.tokenIn = j.tokenOut
),
Now:
joins_and_exits_1 AS (
SELECT
*
FROM joins
UNION ALL
SELECT
*
FROM exits
),
joins_and_exits AS (
SELECT
block_date,
token AS bpt,
LEAD(block_date, 1, NOW()) OVER (PARTITION BY token ORDER BY block_date) AS day_of_next_change,
SUM(amount) OVER (PARTITION BY token ORDER BY block_date ROWS BETWEEN UNBOUNDED PRECEDING AND CURRENT ROW) AS adelta
FROM joins_and_exits_1
),
A request came up from the maxis in order to check what's the % of BPTs staked per pool on mainnet and, in turn, to what % of our TVL it amounts to.