Problem Solving — CS & DS

Lab 9 — Dictionaries as Decisions (Lecture 13 follow-up)

Machine lab · 2 hours · Pairs. You will replace a wall of if-branches with a mapping, then discover what the mapping version makes easier (and one thing it makes harder). Submit: the checkpoints at the end.

Setup

The task

A campus printer charges by user category: student 0.05/page, staff 0.08/page, visitor 0.15/page, honorary free. New categories appear every term, and one category (visitor) also has a per-month cap.

Part A — The branch version (20 min)

A1. Write cost_branches(category, pages) using an if/elif chain. Handle an unknown category by returning None (and say why None beats guessing a price). A2. Count the places in your function that must change when the business adds a category. Write the number down.

Part B — The mapping version (25 min)

B1. Rewrite as cost_map(category, pages) with a module-level dictionary RATES and no category branches. The unknown-category case must behave identically to A1. B2. Count the change-places again. Compare with A2 and explain the difference in one sentence (the word "data" should appear in it). B3. The visitor cap: a visitor's monthly total cannot exceed 5.00. Your function sees only (category, pages) — so the cap cannot live inside it honestly. Write the wrapper visitor_month_cost(pages_list) that enforces the cap, and state which rule decision you had to make (truncate the last job? refuse it? charge partially?).

Part C — Inverting the map (30 min)

C1. Build the inverse map who_pays(price) from price back to category. What breaks? (Hint: two categories can share a price — and one price is 0 for honorary but also for zero pages of anyone.) C2. Make who_pays total: decide and document a rule for ambiguous and missing prices. Your rule must be checkable by a one-line test table. C3. Swap test tables with another pair; adjudicate their who_pays against their stated rule, not yours.

Part D — Aggregation (15 min)

D1. Given the day's log [(category, pages), ...], produce revenue per category in one pass using the mapping version. Two sentences: which accumulator pattern from lecture 09 this is, and where the mapping earned its keep.

Checkpoints (submit these as your report)

  1. cost_branches with its A2 change-place count.
  2. cost_map with its B2 count and the comparison sentence.
  3. Your B3 cap decision, stated as a rule.
  4. The C1 breakage story and your C2 totality rule with its test table.
  5. Your C3 adjudication notes.
  6. The D1 one-pass aggregation and its pattern name.

What completion looks like

Completion-graded as in Lab 1. The graded insight is the A2/B2 contrast: the branch version hides its policy in control flow, the mapping version exposes it as data. Code that "works both ways" without the sentence earns less than broken code with the sentence.