File size: 9,512 Bytes
fcf8749 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 | """
Unit tests for Final Resolution Agent.
Tests swap logic and fairness validation.
"""
import pytest
from uuid import UUID
from app.services.final_resolution import FinalResolutionAgent
from app.schemas.agent_schemas import (
AllocationItem,
DriverLiaisonDecision,
FairnessMetrics,
RoutePlanResult,
)
# Fixed UUIDs for deterministic tests
D1_UUID = UUID("11111111-1111-1111-1111-111111111111")
D2_UUID = UUID("22222222-2222-2222-2222-222222222222")
R1_UUID = UUID("aaaaaaaa-aaaa-aaaa-aaaa-aaaaaaaaaaaa")
R2_UUID = UUID("bbbbbbbb-bbbb-bbbb-bbbb-bbbbbbbbbbbb")
class TestFinalResolutionSwaps:
"""Tests for swap acceptance and rejection logic."""
def setup_method(self):
"""Set up test fixtures."""
self.agent = FinalResolutionAgent(metric_epsilon=0.05)
def test_swap_applied_improves_fairness(self):
"""Swap should be applied when it improves or maintains fairness."""
# Create a proposal where driver-1 has high effort, driver-2 has low
proposal = RoutePlanResult(
allocation=[
AllocationItem(driver_id=D1_UUID, route_id=R1_UUID, effort=80.0),
AllocationItem(driver_id=D2_UUID, route_id=R2_UUID, effort=40.0),
],
total_effort=120.0,
avg_effort=60.0,
per_driver_effort={str(D1_UUID): 80.0, str(D2_UUID): 40.0},
proposal_number=1,
)
decisions = [
DriverLiaisonDecision(
driver_id=str(D1_UUID),
decision="COUNTER",
preferred_route_id=str(R2_UUID), # Wants the easier route
reason="Too heavy",
),
DriverLiaisonDecision(
driver_id=str(D2_UUID),
decision="ACCEPT",
reason="Within comfort",
),
]
# Effort matrix: d1 can do r2 for 45, d2 can do r1 for 75
effort_matrix = [
[80.0, 45.0], # d1's efforts for r1, r2
[75.0, 40.0], # d2's efforts for r1, r2
]
current_metrics = FairnessMetrics(
avg_effort=60.0,
std_dev=20.0,
max_gap=40.0,
gini_index=0.167,
min_effort=40.0,
max_effort=80.0,
)
result = self.agent.resolve_counters(
approved_proposal=proposal,
decisions=decisions,
effort_matrix=effort_matrix,
driver_ids=[str(D1_UUID), str(D2_UUID)],
route_ids=[str(R1_UUID), str(R2_UUID)],
current_metrics=current_metrics,
)
# Swap should be applied: d1->r2 (45), d2->r1 (75)
assert len(result.swaps_applied) == 1
assert result.swaps_applied[0].driver_a == str(D1_UUID)
assert result.per_driver_effort[str(D1_UUID)] == 45.0
assert result.per_driver_effort[str(D2_UUID)] == 75.0
def test_swap_rejected_worsens_fairness(self):
"""Swap should be rejected when it significantly worsens fairness."""
proposal = RoutePlanResult(
allocation=[
AllocationItem(driver_id=D1_UUID, route_id=R1_UUID, effort=55.0),
AllocationItem(driver_id=D2_UUID, route_id=R2_UUID, effort=45.0),
],
total_effort=100.0,
avg_effort=50.0,
per_driver_effort={str(D1_UUID): 55.0, str(D2_UUID): 45.0},
proposal_number=1,
)
decisions = [
DriverLiaisonDecision(
driver_id=str(D1_UUID),
decision="COUNTER",
preferred_route_id=str(R2_UUID),
reason="Want easier",
),
]
# Effort matrix: swap would make things much worse for d2
effort_matrix = [
[55.0, 40.0], # d1: r1=55, r2=40
[90.0, 45.0], # d2: r1=90, r2=45 - huge penalty for d2
]
current_metrics = FairnessMetrics(
avg_effort=50.0,
std_dev=5.0,
max_gap=10.0,
gini_index=0.05,
min_effort=45.0,
max_effort=55.0,
)
result = self.agent.resolve_counters(
approved_proposal=proposal,
decisions=decisions,
effort_matrix=effort_matrix,
driver_ids=[str(D1_UUID), str(D2_UUID)],
route_ids=[str(R1_UUID), str(R2_UUID)],
current_metrics=current_metrics,
)
# Swap should be rejected: d2 would go from 45 to 90 (100% increase)
assert len(result.swaps_applied) == 0
assert str(D1_UUID) in result.unfulfilled_counters
def test_no_swaps_when_no_counters(self):
"""No swaps should occur when there are no COUNTER decisions."""
proposal = RoutePlanResult(
allocation=[
AllocationItem(driver_id=D1_UUID, route_id=R1_UUID, effort=50.0),
AllocationItem(driver_id=D2_UUID, route_id=R2_UUID, effort=50.0),
],
total_effort=100.0,
avg_effort=50.0,
per_driver_effort={str(D1_UUID): 50.0, str(D2_UUID): 50.0},
proposal_number=1,
)
decisions = [
DriverLiaisonDecision(driver_id=str(D1_UUID), decision="ACCEPT", reason="OK"),
DriverLiaisonDecision(driver_id=str(D2_UUID), decision="ACCEPT", reason="OK"),
]
effort_matrix = [[50.0, 60.0], [60.0, 50.0]]
current_metrics = FairnessMetrics(
avg_effort=50.0,
std_dev=0.0,
max_gap=0.0,
gini_index=0.0,
min_effort=50.0,
max_effort=50.0,
)
result = self.agent.resolve_counters(
approved_proposal=proposal,
decisions=decisions,
effort_matrix=effort_matrix,
driver_ids=[str(D1_UUID), str(D2_UUID)],
route_ids=[str(R1_UUID), str(R2_UUID)],
current_metrics=current_metrics,
)
assert len(result.swaps_applied) == 0
assert len(result.unfulfilled_counters) == 0
def test_unfulfilled_counter_recorded(self):
"""Unfulfilled counters should be recorded in result."""
proposal = RoutePlanResult(
allocation=[
AllocationItem(driver_id=D1_UUID, route_id=R1_UUID, effort=70.0),
AllocationItem(driver_id=D2_UUID, route_id=R2_UUID, effort=50.0),
],
total_effort=120.0,
avg_effort=60.0,
per_driver_effort={str(D1_UUID): 70.0, str(D2_UUID): 50.0},
proposal_number=1,
)
decisions = [
DriverLiaisonDecision(
driver_id=str(D1_UUID),
decision="COUNTER",
preferred_route_id="non-existent-route", # Non-existent route
reason="Want different route",
),
]
effort_matrix = [[70.0, 55.0], [65.0, 50.0]]
current_metrics = FairnessMetrics(
avg_effort=60.0,
std_dev=10.0,
max_gap=20.0,
gini_index=0.083,
min_effort=50.0,
max_effort=70.0,
)
result = self.agent.resolve_counters(
approved_proposal=proposal,
decisions=decisions,
effort_matrix=effort_matrix,
driver_ids=[str(D1_UUID), str(D2_UUID)],
route_ids=[str(R1_UUID), str(R2_UUID)],
current_metrics=current_metrics,
)
# Counter for non-existent route should be unfulfilled
assert str(D1_UUID) in result.unfulfilled_counters
class TestFinalResolutionMetrics:
"""Tests for metric computation."""
def setup_method(self):
"""Set up test fixtures."""
self.agent = FinalResolutionAgent()
def test_compute_metrics(self):
"""Test metric computation."""
efforts = [40.0, 50.0, 60.0]
metrics = self.agent._compute_metrics(efforts)
assert metrics["avg_effort"] == 50.0
assert metrics["min_effort"] == 40.0
assert metrics["max_effort"] == 60.0
assert metrics["max_gap"] == 20.0
assert metrics["gini_index"] > 0
def test_compute_gini_perfect_equality(self):
"""Gini should be 0 for perfect equality."""
values = [50.0, 50.0, 50.0]
gini = self.agent._compute_gini(values)
assert gini == 0.0
def test_snapshot_generation(self):
"""Test input/output snapshot generation."""
metrics = FairnessMetrics(
avg_effort=50.0,
std_dev=10.0,
max_gap=20.0,
gini_index=0.1,
min_effort=40.0,
max_effort=60.0,
)
input_snap = self.agent.get_input_snapshot(3, metrics, 50.0)
assert input_snap["num_counters"] == 3
assert input_snap["original_gini"] == 0.1
from app.schemas.agent_schemas import FinalResolutionResult
result = FinalResolutionResult(
allocation=[],
per_driver_effort={},
metrics={"gini_index": 0.08},
swaps_applied=[],
unfulfilled_counters=[],
)
output_snap = self.agent.get_output_snapshot(result)
assert output_snap["num_swaps_applied"] == 0
|