-- Run from the package root with DuckDB.
SELECT
  contract_symbol,
  MIN(timestamp_chicago_wall) AS first_bar,
  MAX(timestamp_chicago_wall) AS last_bar,
  COUNT(*) AS observations,
  SUM(volume) AS reported_volume
FROM read_parquet('data/minute/*/*/*.parquet', hive_partitioning = false, filename = true)
GROUP BY contract_symbol
ORDER BY contract_symbol;