select distinct address, label

from algorand.labels

where label_type = 'cex'

and label not like '%group%'

),

transfers as (

select

block_timestamp, tx_id, sender, receiver, amount, label as cex

from algorand.payment_transaction pt

join cex_addresses ca on pt.sender = ca.address

where date(block_timestamp) >= '2022-04-01'

select

date_trunc('day', block_timestamp) as date,

case when date>='2022-05-02' then 'after'

else 'before' end as spliter,

count(tx_id) as transactions,

sum(amount) as volume

from transfers

group by date