🇫🇷 Version française

[!IMPORTANT] Every fixture, rulebook, and calculator output in this lab is synthetic and non-binding. Nothing here represents an actual Desjardins product, rate, or policy, and no regulator or insurer has reviewed or endorsed this material.

Overview

Item	Value
Duration	30 minutes
Level	Intermediate
Prerequisites	Lab 01

Learning Objectives

By the end of this lab, you will be able to:

• Trace calculate_quote’s resolution order in apps/workshop/calculator.py: unavailable rulebook, version mismatch, missing fields, unsupported values, then a ready result
• State the four calculation statuses the calculator can return
• Add a new test case and confirm it passes
• Prove that a rulebook lookup miss produces an issue code, never an invented amount

Exercises

Exercise 2.1: Read the Resolution Order

Open apps/workshop/calculator.py and read calculate_quote top to bottom. Note the order the function checks conditions in:

1. rulebook is None returns EVIDENCE_UNAVAILABLE
2. A rulebook version mismatch (when expected_rulebook_version is supplied) returns EVIDENCE_UNAVAILABLE with RULE_VERSION_MISMATCH
3. Any missing jurisdiction, vehicleClass, or plan returns INCOMPLETE with one issue per missing field
4. A jurisdiction other than ON, or a vehicleClass/plan absent from the rulebook’s baseCents/planAddOnCents tables, returns UNSUPPORTED
5. Otherwise, the function sums baseCents[vehicleClass] + planAddOnCents[plan] and returns READY

Exercise 2.2: Run the Existing Tests

python -m pytest apps/workshop/tests/test_calculator.py -v

Expected result: every test passes, including one asserting case-syn-001’s exact 100000-cent result.

Exercise 2.3 (Hands-on): Add a New Test Case

Add a test to apps/workshop/tests/test_calculator.py asserting that a SEDAN vehicle with the TRAINING_BASIC plan resolves to exactly 90000 cents, using the same rulebook shape as the existing tests. Run the file again and confirm your new test passes alongside the existing ones.

Exercise 2.4 (Hands-on): Prove the No-Invented-Amount Rule

In a Python shell, build a rulebook dictionary copied from data/synthetic/rulebook.json, remove the "SEDAN" key from baseCents, then call calculate_quote with a SEDAN input against that modified rulebook.

python -c "
import json, pathlib, sys
sys.path.insert(0, 'apps/workshop')
from calculator import calculate_quote

rulebook = json.loads(pathlib.Path('data/synthetic/rulebook.json').read_text())
del rulebook['baseCents']['SEDAN']
result = calculate_quote({'jurisdiction': 'ON', 'vehicleClass': 'SEDAN', 'plan': 'TRAINING_BASIC'}, rulebook)
print(result)
"

Expected result: status is UNSUPPORTED, amountCents is null, and issues contains UNSUPPORTED_INPUT. No amount is returned, even though TRAINING_BASIC is a recognized plan.

Validation Checklist

• pytest apps/workshop/tests/test_calculator.py -v passes, including your Exercise 2.3 test
• You can list the four calculation statuses from memory
• Exercise 2.4 confirms a missing rulebook entry returns UNSUPPORTED_INPUT, never an amount
• You can explain, without looking at the code, why the rulebook-availability check runs before the missing-field check

Knowledge Check

• Why does the calculator check for a missing rulebook before it checks for missing input fields?
• What currency and period does every READY result carry, and where do those values come from?

Next Steps

Continue to Lab 03: Approval Repository and State Machine.