The publicly available dataset is not properly formatted for importing into db or further analysis. In order to clean up the data for analysis, I did the following:

• Added headers based on data descriptions for Atlantic and Pacific datasets (major difference seem to be in the possible values for basins, record identifiers, and hurricane status)
• Normalize by adding in header row information (alternatively, this could also be moved to different table)
• Convert date and time fields to ISO8601 format, which is easier for most db and software to interpret
• Added helper tables that link basins, record identifiers, and statuses to their meanings
• Normalize longitude/latitude to numbers rather than using E/W, N/S suffixes
• Mark maximum_wind_knots as null -99
• Mark -999 wind and pressure values as null so it doesn't interfere with calculations
• Give option to combine Atlantic and Pacific datasets (Though it it worth noting that the Atlantic dataset has data form 1851, and the Pacific one only has data from 1949)

• SQLite version
• CSV version

python3 hurricane/process_hurricane.py -h: See help

python3 hurricane/process_hurricane.py: Create new hurricane.db SQLite file in main

python3 hurricane/process_hurricane.py <<path_to_sqlite_file>>: Add all data to SQLite file

python3 hurricane/process_hurricane.py -b atlantic <<path_to_sqlite_file>>: Add atlantic hurricane data to SQLite file

python3 hurricane/process_hurricane.py -c <<path_to_csv_file>>: Add all data to csv file