notes.txt

- order of entry in history is important for an accurate EMA.
- What this implies is that if there is an issue with the oracles(s) no inflation calulations will occur until they are fixed
  and any missing days are backfilled.

- Normalize all data collection to happen at 00:00:00 UTC
- setttotal validations:
     - timestamp is 00:00:00 for given day
     - timestamp doesn't already exist
     - timestamp is in the past
     - timestamp is the next day after max hisory.timestamp (to preserve order)

- adddistrib validations: 
    - timestamp doesn't already exist
    - timestamp exists in historys table
    - UtilX locked details exist for timestamp
    - total distribution < 100

onblock: 
    - each day at 00:00:00 we collect currency stats and utilityX details

Notes:
Look into using blocks per day instead of time to validate oracles.
Instead of timestamp they provide the last blocknumber in their sample set.
Do they use lib or head?
The problem is there is will be a time skew due to missed blocks.

TODO: 
Split out distribution collection from adding distrivution to distribution.
In case there is an error with the oracle

allow multiple leasing platforms



Sequence:

onblock:
   - update metrics_table once a day
   - if inflation_record
     then
        issue currency and transfer:
           - bp
           - locking
           - utilityX