Data cleaning: duplicates, missing values, messy formats and a cleaning checklist

raw = [
    {"id": "1", "county": " Nairobi", "phone": "0712 345 678", "amount": "KSh 1,200"},
    {"id": "2", "county": "nairobi", "phone": "+254712345679", "amount": "850"},
    {"id": "2", "county": "nairobi", "phone": "+254712345679", "amount": "850"},
    {"id": "3", "county": "KIAMBU", "phone": "712345680", "amount": ""},
    {"id": "4", "county": "Murang'a", "phone": "0722-000-111", "amount": "2,500"},
]

def clean_phone(p):
    digits = "".join(ch for ch in p if ch.isdigit())
    if digits.startswith("0"):
        digits = "254" + digits[1:]
    elif len(digits) == 9:
        digits = "254" + digits
    return digits

def clean_amount(a):
    digits = "".join(ch for ch in a if ch.isdigit() or ch == ".")
    return float(digits) if digits else None

seen, clean = set(), []
for row in raw:
    if row["id"] in seen:          # remove duplicates
        continue
    seen.add(row["id"])
    clean.append({
        "id": int(row["id"]),
        "county": row["county"].strip().capitalize(),
        "phone": clean_phone(row["phone"]),
        "amount": clean_amount(row["amount"]),
    })

for row in clean:
    print(row)

missing = [r["id"] for r in clean if r["amount"] is None]
print("Rows with a missing amount:", missing)