Viktor Barzin
e72fd22a17
The Monte Carlo used to compare jurisdictions at a flat London-equivalent spend, which silently overstated the cost-of-living for any move to a cheaper region. Now every cross-jurisdiction simulation auto-scales spending_gbp by the real Numbeo/Expatistan ratio between the user's baseline city and the target city. Architecture: - fire_planner/col/baseline.py — 22 cities with headline Numbeo data (source URLs + snapshot dates embedded) — fallback when scraper fails - col/numbeo.py + col/expatistan.py — httpx async scrapers, regex-parsed, polite 1.1s rate-limit, EUR/USD anchored - col/cache.py — PG-backed cache (col_snapshot table, 1-year TTL) - col/service.py — sync compute_col_ratio() for the simulator; async lookup_city_cached() with source reconciliation for the refresh CronJob - alembic 0005 — col_snapshot table, UNIQUE(city_slug, source_name) Simulator wiring: - SimulateRequest gains col_auto_adjust=True (default), col_baseline_city, col_target_city. Defaults pick the jurisdiction's representative city. - _resolve_col_adjustment scales spending_gbp before path-building. - SimulateResult surfaces col_multiplier_applied + col_adjusted_spending_gbp. CLIs: - python -m fire_planner col-seed — loads BASELINES into col_snapshot (post-migration seed step) - python -m fire_planner col-refresh-stale --within-days 7 — used by the weekly fire-planner-col-refresh CronJob 268 tests pass. Mypy strict + ruff clean. Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
104 lines
4.1 KiB
Python
104 lines
4.1 KiB
Python
"""Tests for the COL module — baseline lookup + ratio + simulator wiring."""
|
|
from __future__ import annotations
|
|
|
|
from decimal import Decimal
|
|
|
|
import pytest
|
|
|
|
from fire_planner.col import (
|
|
JURISDICTION_REPRESENTATIVE_CITY,
|
|
compute_col_ratio,
|
|
lookup_city,
|
|
representative_city_for,
|
|
)
|
|
from fire_planner.col.baseline import BASELINES
|
|
from fire_planner.col.models import CityCostIndex
|
|
|
|
|
|
class TestBaselineCoverage:
|
|
"""Every jurisdiction with a representative city must have a baseline."""
|
|
|
|
def test_all_representative_cities_have_baselines(self) -> None:
|
|
missing = [
|
|
city for city in JURISDICTION_REPRESENTATIVE_CITY.values() if city not in BASELINES
|
|
]
|
|
assert missing == [], (
|
|
f"jurisdiction map points at city(s) without baselines: {missing}"
|
|
)
|
|
|
|
def test_baselines_have_positive_totals(self) -> None:
|
|
for slug, idx in BASELINES.items():
|
|
assert idx.total_single_no_rent_gbp > 0, f"{slug} no_rent must be positive"
|
|
assert idx.total_single_with_rent_gbp > idx.total_single_no_rent_gbp, (
|
|
f"{slug} with_rent must exceed no_rent — rent should be a positive add"
|
|
)
|
|
|
|
def test_baseline_source_provenance_present(self) -> None:
|
|
for slug, idx in BASELINES.items():
|
|
assert idx.source.name in {"numbeo", "expatistan", "baseline", "manual"}
|
|
assert idx.source.url is not None, f"{slug} baseline missing source URL"
|
|
assert idx.source.url.startswith("https://"), f"{slug} URL must be https"
|
|
|
|
|
|
class TestLookup:
|
|
def test_lookup_known_city(self) -> None:
|
|
london = lookup_city("london")
|
|
assert isinstance(london, CityCostIndex)
|
|
assert london.city == "London"
|
|
assert london.country == "United Kingdom"
|
|
|
|
def test_lookup_normalises_input(self) -> None:
|
|
# mixed case, spaces → slug
|
|
assert lookup_city("Kuala Lumpur").city == "Kuala Lumpur"
|
|
assert lookup_city(" Bangkok ").city == "Bangkok"
|
|
|
|
def test_lookup_unknown_raises(self) -> None:
|
|
with pytest.raises(KeyError, match="No COL baseline"):
|
|
lookup_city("atlantis")
|
|
|
|
|
|
class TestColRatio:
|
|
def test_identity_returns_one(self) -> None:
|
|
assert compute_col_ratio("london", "london") == Decimal("1")
|
|
|
|
def test_sofia_cheaper_than_london(self) -> None:
|
|
ratio = compute_col_ratio("london", "sofia")
|
|
assert ratio < Decimal("1"), "Sofia must be cheaper than London"
|
|
assert ratio > Decimal("0.2"), "Sofia ratio looks implausibly low"
|
|
# Real Numbeo number is ~0.41
|
|
assert Decimal("0.35") < ratio < Decimal("0.50")
|
|
|
|
def test_dubai_cheaper_than_london(self) -> None:
|
|
# Dubai is *cheaper* than London on Numbeo's headline because
|
|
# London rent dominates. This was a surprise — flag it in the
|
|
# baseline note for future-us.
|
|
ratio = compute_col_ratio("london", "dubai")
|
|
assert ratio < Decimal("1")
|
|
assert Decimal("0.70") < ratio < Decimal("0.95")
|
|
|
|
def test_bangkok_far_cheaper_than_london(self) -> None:
|
|
ratio = compute_col_ratio("london", "bangkok")
|
|
assert ratio < Decimal("0.40")
|
|
|
|
def test_inverse_consistency(self) -> None:
|
|
# If london→sofia is X, sofia→london should be ~1/X within rounding.
|
|
l2s = compute_col_ratio("london", "sofia")
|
|
s2l = compute_col_ratio("sofia", "london")
|
|
assert abs(l2s * s2l - Decimal("1")) < Decimal("0.001")
|
|
|
|
|
|
class TestRepresentativeCity:
|
|
def test_known_jurisdictions(self) -> None:
|
|
assert representative_city_for("uk") == "london"
|
|
assert representative_city_for("cyprus") == "limassol"
|
|
assert representative_city_for("bulgaria") == "sofia"
|
|
assert representative_city_for("uae") == "dubai"
|
|
assert representative_city_for("malaysia") == "kuala-lumpur"
|
|
assert representative_city_for("thailand") == "bangkok"
|
|
|
|
def test_nomad_returns_none(self) -> None:
|
|
# Nomad mode is COL-invariant by design — auto-adjust skipped.
|
|
assert representative_city_for("nomad") is None
|
|
|
|
def test_unknown_returns_none(self) -> None:
|
|
assert representative_city_for("vulcan") is None
|