Data Cleaning

074_data_cleaning.cell

// ============================================================================
// 074 — Data Cleaning
// ============================================================================
// Clean and normalize messy data.
//
// Run:   soma run 074_data_cleaning.cell run
// Expect:
//   Raw: 5 records
//   Cleaned:
//     alice -> [email protected]
//     bob -> [email protected]
//     carol -> [email protected]
//   Removed 2 invalid records

cell DataCleaning {

    on run() {
        // Messy data with inconsistent formatting
        let raw = list(
            map("name", "  Alice  ", "email", "  [email protected]  "),
            map("name", "BOB", "email", "[email protected]"),
            map("name", "", "email", "[email protected]"),
            map("name", "Carol", "email", "[email protected]"),
            map("name", "Dave", "email", "")
        )

        let raw_count = raw.length
        print("Raw: {raw_count} records")

        // Clean and validate
        let cleaned = list()
        let removed = 0
        for row in raw {
            let name = lowercase(trim(row.name))
            let email = lowercase(trim(row.email))

            // Skip if name or email is empty
            if name == "" {
                removed = removed + 1
            } else {
                if email == "" {
                    removed = removed + 1
                } else {
                    let clean_row = map("name", name, "email", email)
                    cleaned = list(cleaned, clean_row)
                }
            }
        }

        print("Cleaned:")
        for row in cleaned {
            print("  {row.name} -> {row.email}")
        }
        print("Removed {removed} invalid records")

        return "done"
    }
}