Workshop: a full analysis

Everything at once

This is the shape of a real job: read the file, clean it, join it, group it, and hand back a summary someone can act on.

The data is a month of orders. Some rows have a missing quantity. One order references a customer who is not in the customer list.

Your turn: write analyse(orders_csv, customers_csv) that:

1. reads both CSV strings, treating "" and n/a as missing
2. drops orders with a missing quantity
3. adds a revenue column (quantity * unit_price)
4. joins the customers on, keeping every remaining order
5. returns a dictionary:

{
    "orders": 4,                  # rows after cleaning
    "revenue": 235.0,             # total, rounded to 2dp
    "by_region": {"north": 135.0, "south": 100.0},
    "unknown_customer_orders": 1, # orders whose customer was not found
}

Regions come from the customer table; an order with no matching customer has no region and is counted separately rather than being dropped.