event_inputs:_to as buyer_address,

event_inputs:_id as tokenid

from polygon.core.fact_event_logs

where tx_status = 'SUCCESS' and block_timestamp >= '2022-08-26' and event_name = 'TransferSingle' and event_inputs:_from != '0x0000000000000000000000000000000000000000'

and contract_address = lower('0x8d0501d85becda92b89e56177ddfcea5fc1f0af2') and buyer_address != '0x83bfa75c9391dbe19e0a1d2c53c9b6c4baef5ab7' and buyer_address != '0x0000a26b00c1f0df003000390027140000faa719'),

df_sold as (select block_timestamp as sold_date,

event_inputs:_from as seller_address,

b.event_inputs:_id as tokenid

from df_purchase a

join polygon.core.fact_event_logs b on a.buyer_address = b.event_inputs:_from and a.tokenid = b.event_inputs:_id

where b.block_timestamp > purchase_date),

df as (select datediff(day, purchase_date, sold_date) as hold_time,

1 as count

from df_purchase a

join df_sold b on a.buyer_address = b.seller_address and a.tokenid = b.tokenid)

select case

when hold_time < 1 then 'Less than 1 day'

when hold_time > 1 and hold_time <= 7 then '1-7 days'

when hold_time > 7 and hold_time <= 14 then '7-14 days'

when hold_time > 14 and hold_time <= 30 then '14-30 days'

else 'More then 30 days'

end as duration,

sum(count) as count

from df

group by 1

order by 1