In-Memory Transaction Authorizer with MCC Blacklists and Rolling Velocity Limit

Problem: In-Memory Transaction Authorizer (MCC Blacklists + User Overrides + Rolling Velocity Limit) Implement a transaction authorizer TransactionAuthorizer that decides whether to approve or decline a transaction request from a card network. Assume everything is in-memory (no persistence). Transaction fields Each transaction has: user_id (string) transaction_id (string) mcc (string, merchant category code) timestamp (integer, in seconds) amount (decimal, or integer cents) Example: { "user_id": "1111", "transaction_id": "0001", "mcc": "9999", "timestamp": 0, "amount": 10.00 } Part 1: Single MCC rejection rule Implement approve(transaction) -> boolean: Rule 1: decline (false) if mcc == "9999", otherwise approve (true). Part 2: Global + per-user MCC rejection lists Extend Part 1: Globally rejected MCCs: { "9999", "0000" } User-specific rejected MCCs: user_id == "1234": reject { "1111", "2222" } user_id == "5678": reject { "5555", "6666" } Decision order: If the transaction mcc is in the global reject set, decline. Else if the user has a user-specific reject set and mcc is in it, decline. Otherwise approve. Part 3: Rolling 1-hour velocity limit Add a spending velocity limit on top of Part 2: Rule 3: For each user, if the sum of attempted transaction amounts within a rolling 1-hour window exceeds $5000, decline further transactions. Notes: Timestamp unit is seconds. For a transaction at time t, the rolling window is the last 3600 seconds (choose (t-3600, t] or [t-3600, t] and be consistent; explain your choice). “Attempted transactions” means both approved and declined transactions contribute to the running window total. Example For a single user: | timestamp | amount | result | |---:|---:|---| | 0 | 500 | approved | | 60 | 2000 | approved | | 3660 | 3000 | approved | | 7260 | 2001 | rejected | | 7261 | 3000 | rejected | Requirements Implement TransactionAuthorizer (any language; Java class stub is acceptable). Support the three rules (can be implemented incrementally or via feature toggles). Explain your data structure choices and time/space complexity. Example Input 5 1111 0001 1234 0 500.00 1111 0002 1234 60 2000.00 1111 0003 1234 3660 3000.00 1111 0004 1234 7260 2001.00 1111 0005 1234 7261 3000.00 Output approved approved approved rejected rejected