Normalization

Normalization is the process of structuring tables to eliminate redundancy and prevent certain classes of update bugs. First Normal Form (1NF) is the starting requirement: every column must hold a single, atomic value — no repeating groups (like a "phone numbers" column holding a whole list).

def has_repeating_group(row):
    return any(isinstance(v, list) for v in row.values())

bad_row = {"id": 1, "name": "Alice", "phones": ["555-1234", "555-5678"]}
print(has_repeating_group(bad_row))   # True -- phones should be its own table

The fix: a separate phone_numbers table with a foreign key back to id, one row per phone number — exactly the kind of table-splitting normalization is all about.

Second Normal Form (2NF) goes further, but only matters for tables with a COMPOSITE primary key (more than one column). It requires every non-key column to depend on the WHOLE key, not just part of it:

# order_items table, primary key = (order_id, product_id)
dependencies = [
    (("order_id", "product_id"), "quantity"),      # depends on the FULL key -- fine
    (("product_id",), "product_name"),                # depends on ONLY product_id -- 2NF violation!
]

product_name doesn't actually depend on WHICH ORDER it's in — it only depends on the product. Storing it in order_items means it's needlessly repeated for every order line, AND creates an update anomaly: change a product's name, and now you have to update every single order row that references it, or the data goes inconsistent. The fix: move product_name into its own products table, keyed by product_id alone.