Testing Plan and Continuous Integration

Testing Strategy

Tests cover the complete core logic of the project. Python code is tested with pytest, with Hypothesis for property tests. TypeScript code is tested with Jest. Mocking isolates components from Supabase and the network.

Test type	What it checks	Example
Unit	One function or class against hand-calculated expected values	test_events_scheduled_out_of_order_are_processed_in_time_order
Property (Hypothesis)	Rules that must hold for any input	test_processing_order_never_goes_backwards
Determinism	Same scenario and seed give byte-identical output	test_same_scenario_and_seed_give_byte_identical_output
Integration	Components working together	test_smoke_run_with_no_control_strategy_writes_the_expected_log
Data processing	GTFS extraction and walking-time fit	Planned
Front end	UI behaviour (Jest, Playwright)	Planned

PR Review Stategy

All pull requests to the main branch will by policy be required to be reviewed and approved by one of team members (one who was not also an author on said PR).

Continuous integration

We will set up a CI/CD pipeline using GitHub Actions. It will automatically run our tests and lint our code on each push and opened PR. Any failing test or failing the lint check will prevent merging.

Pipeline steps

• ruff check . (lint, complexity and statement limits)
• ruff format --check .
• pytest, including the determinism test
• Workflow file: planned in #11