Spaces:
Sleeping
Sleeping
File size: 12,854 Bytes
e86dfae | 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 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 | """
Unit Tests: Agents
Tests for all 4 agents and the AgentOrchestrator.
Uses mocked Gemini API to test without real API calls.
"""
import json
from unittest.mock import AsyncMock, MagicMock, patch
import pytest
MOCK_CLASSIFICATION_JSON = json.dumps(
{
"category": "safety_protocol",
"subcategory": "ventilation",
"confidence": 0.93,
"reasoning": "Document contains MSHA ventilation requirements and PPE protocols.",
}
)
MOCK_SAFETY_JSON = json.dumps(
{
"score": 82.0,
"status": "compliant",
"confidence": 0.88,
"hazards": [
{
"type": "atmospheric",
"severity": "medium",
"description": "Potential methane buildup in Section C",
"regulation": "30 CFR 75.323",
}
],
"recommendations": ["Install additional methane detectors in Section C"],
"compliance_details": {
"msha_compliant": True,
"osha_compliant": True,
"dgms_compliant": True,
"missing_elements": [],
},
"summary": "Generally compliant. One atmospheric hazard identified.",
}
)
MOCK_ENTITIES_JSON = json.dumps(
{
"equipment": ["Caterpillar D11", "Joy 12CM15 Continuous Miner"],
"chemicals": ["methane (CH4)", "carbon monoxide (CO)"],
"locations": ["Section C", "Main Heading 3"],
"personnel": ["Safety Officer", "Mine Foreman"],
"dates": ["2024-01-15", "quarterly"],
"regulations": ["30 CFR 75.321", "30 CFR 75.323", "MSHA 1910.134"],
}
)
MOCK_SUMMARY_JSON = json.dumps(
{
"summary": "This document outlines safety procedures for underground coal mining operations, covering ventilation standards, PPE requirements, and emergency evacuation protocols.",
"key_points": [
"Methane levels must stay below 1% in all working areas",
"PPE inspection required before each shift",
"Emergency drills must occur quarterly",
],
"action_items": [
"Schedule quarterly emergency drill",
"Replace methane detectors in Section C",
],
"document_purpose": "Define safety standards for underground coal mine personnel.",
"confidence": 0.91,
}
)
def make_mock_model(json_response: str):
"""Create a mock Gemini model that returns json_response."""
mock_response = MagicMock()
mock_response.text = json_response
mock_model = MagicMock()
mock_model.generate_content = MagicMock(return_value=mock_response)
return mock_model
def make_mock_client(json_response: str):
"""Create a mock OpenAI client that returns json_response."""
mock_choice = MagicMock()
mock_choice.message.content = json_response
mock_completion = MagicMock()
mock_completion.choices = [mock_choice]
mock_chat = MagicMock()
mock_chat.completions.create = AsyncMock(return_value=mock_completion)
mock_client = MagicMock()
mock_client.chat = mock_chat
return mock_client
class TestClassifierAgent:
"""Tests for ClassifierAgent."""
@pytest.mark.unit
@pytest.mark.asyncio
async def test_analyze_returns_expected_keys(self, sample_mining_text):
"""ClassifierAgent.analyze() returns dict with required keys."""
from app.agents.classifier import ClassifierAgent
agent = ClassifierAgent()
agent.client = make_mock_client(MOCK_CLASSIFICATION_JSON)
result = await agent.analyze(sample_mining_text)
assert "category" in result
assert "subcategory" in result
assert "confidence" in result
assert "reasoning" in result
@pytest.mark.unit
@pytest.mark.asyncio
async def test_analyze_maps_category_correctly(self, sample_mining_text):
"""ClassifierAgent maps 'safety_protocol' string to correct value."""
from app.agents.classifier import ClassifierAgent
agent = ClassifierAgent()
agent.client = make_mock_client(MOCK_CLASSIFICATION_JSON)
result = await agent.analyze(sample_mining_text)
assert result["category"] == "safety_protocol"
@pytest.mark.unit
@pytest.mark.asyncio
async def test_analyze_confidence_is_float_in_range(self, sample_mining_text):
"""Confidence value is a float between 0 and 1."""
from app.agents.classifier import ClassifierAgent
agent = ClassifierAgent()
agent.client = make_mock_client(MOCK_CLASSIFICATION_JSON)
result = await agent.analyze(sample_mining_text)
assert isinstance(result["confidence"], float)
assert 0.0 <= result["confidence"] <= 1.0
@pytest.mark.unit
@pytest.mark.asyncio
async def test_analyze_unknown_category_maps_to_other(self, sample_mining_text):
"""Unknown category value falls back to 'other'."""
from app.agents.classifier import ClassifierAgent
agent = ClassifierAgent()
agent.client = make_mock_client(
json.dumps(
{
"category": "completely_unknown_xyz",
"confidence": 0.5,
"reasoning": "unknown",
}
)
)
result = await agent.analyze(sample_mining_text)
assert result["category"] == "other"
@pytest.mark.unit
@pytest.mark.asyncio
async def test_analyze_handles_empty_response(self, sample_mining_text):
"""Empty JSON response returns 'other' without crashing."""
from app.agents.classifier import ClassifierAgent
agent = ClassifierAgent()
agent.client = make_mock_client("{}")
result = await agent.analyze(sample_mining_text)
assert "category" in result
assert result["confidence"] == 0.5 # default
class TestSafetyAnalyzerAgent:
"""Tests for SafetyAnalyzerAgent."""
@pytest.mark.unit
@pytest.mark.asyncio
async def test_analyze_returns_expected_keys(self, sample_mining_text):
"""SafetyAnalyzerAgent.analyze() returns dict with required keys."""
from app.agents.safety_analyzer import SafetyAnalyzerAgent
agent = SafetyAnalyzerAgent()
agent.client = make_mock_client(MOCK_SAFETY_JSON)
result = await agent.analyze(
sample_mining_text, {"category": "safety_protocol"}
)
for key in (
"score",
"status",
"hazards",
"recommendations",
"compliance_details",
):
assert key in result
@pytest.mark.unit
@pytest.mark.asyncio
async def test_score_is_numeric(self, sample_mining_text):
"""Safety score must be a numeric type."""
from app.agents.safety_analyzer import SafetyAnalyzerAgent
agent = SafetyAnalyzerAgent()
agent.client = make_mock_client(MOCK_SAFETY_JSON)
result = await agent.analyze(sample_mining_text)
assert isinstance(result["score"], (int, float))
@pytest.mark.unit
@pytest.mark.asyncio
async def test_hazards_is_list(self, sample_mining_text):
"""Hazards must be a list."""
from app.agents.safety_analyzer import SafetyAnalyzerAgent
agent = SafetyAnalyzerAgent()
agent.client = make_mock_client(MOCK_SAFETY_JSON)
result = await agent.analyze(sample_mining_text)
assert isinstance(result["hazards"], list)
class TestEntityExtractorAgent:
"""Tests for EntityExtractorAgent."""
@pytest.mark.unit
@pytest.mark.asyncio
async def test_analyze_returns_all_entity_types(self, sample_mining_text):
"""EntityExtractorAgent returns all 6 entity categories."""
from app.agents.entity_extractor import EntityExtractorAgent
agent = EntityExtractorAgent()
agent.client = make_mock_client(MOCK_ENTITIES_JSON)
result = await agent.analyze(sample_mining_text)
for key in (
"equipment",
"chemicals",
"locations",
"personnel",
"dates",
"regulations",
):
assert key in result
assert isinstance(result[key], list)
@pytest.mark.unit
@pytest.mark.asyncio
async def test_entity_count_is_correct(self, sample_mining_text):
"""entity_count equals sum of all entity lists."""
from app.agents.entity_extractor import EntityExtractorAgent
agent = EntityExtractorAgent()
agent.client = make_mock_client(MOCK_ENTITIES_JSON)
result = await agent.analyze(sample_mining_text)
expected = sum(
len(result[k])
for k in (
"equipment",
"chemicals",
"locations",
"personnel",
"dates",
"regulations",
)
)
assert result["entity_count"] == expected
class TestSummarizerAgent:
"""Tests for SummarizerAgent."""
@pytest.mark.unit
@pytest.mark.asyncio
async def test_analyze_returns_expected_keys(self, sample_mining_text):
"""SummarizerAgent returns summary, key_points, action_items."""
from app.agents.summarizer import SummarizerAgent
agent = SummarizerAgent()
agent.model = make_mock_model(MOCK_SUMMARY_JSON)
result = await agent.analyze(sample_mining_text)
assert "summary" in result
assert "key_points" in result
assert isinstance(result["summary"], str)
assert len(result["summary"]) > 0
@pytest.mark.unit
@pytest.mark.asyncio
async def test_key_points_is_list(self, sample_mining_text):
"""key_points must be a list of strings."""
from app.agents.summarizer import SummarizerAgent
agent = SummarizerAgent()
agent.model = make_mock_model(MOCK_SUMMARY_JSON)
result = await agent.analyze(sample_mining_text)
assert isinstance(result["key_points"], list)
class TestAgentOrchestrator:
"""Tests for AgentOrchestrator."""
@pytest.mark.unit
@pytest.mark.asyncio
async def test_orchestrator_returns_all_sections(self, sample_mining_text):
"""Orchestrator returns classification, safety, entities, summary, metadata."""
from app.agents.orchestrator import AgentOrchestrator
orchestrator = AgentOrchestrator()
orchestrator.classifier.client = make_mock_client(MOCK_CLASSIFICATION_JSON)
orchestrator.safety_analyzer.client = make_mock_client(MOCK_SAFETY_JSON)
orchestrator.entity_extractor.client = make_mock_client(MOCK_ENTITIES_JSON)
orchestrator.summarizer.model = make_mock_model(MOCK_SUMMARY_JSON)
result = await orchestrator.analyze_document(sample_mining_text)
for section in ("classification", "safety", "entities", "summary", "metadata"):
assert section in result
@pytest.mark.unit
@pytest.mark.asyncio
async def test_orchestrator_includes_processing_time(self, sample_mining_text):
"""Metadata includes processing_time_ms > 0."""
from app.agents.orchestrator import AgentOrchestrator
orchestrator = AgentOrchestrator()
orchestrator.classifier.client = make_mock_client(MOCK_CLASSIFICATION_JSON)
orchestrator.safety_analyzer.client = make_mock_client(MOCK_SAFETY_JSON)
orchestrator.entity_extractor.client = make_mock_client(MOCK_ENTITIES_JSON)
orchestrator.summarizer.model = make_mock_model(MOCK_SUMMARY_JSON)
result = await orchestrator.analyze_document(sample_mining_text)
assert "processing_time_ms" in result["metadata"]
assert isinstance(result["metadata"]["processing_time_ms"], int)
assert result["metadata"]["processing_time_ms"] >= 0
@pytest.mark.unit
@pytest.mark.asyncio
async def test_orchestrator_handles_agent_failure_gracefully(
self, sample_mining_text
):
"""If one agent fails, orchestrator still returns results from other agents."""
from app.agents.orchestrator import AgentOrchestrator
orchestrator = AgentOrchestrator()
orchestrator.classifier.client = make_mock_client(MOCK_CLASSIFICATION_JSON)
orchestrator.safety_analyzer.client = make_mock_client(MOCK_SAFETY_JSON)
orchestrator.entity_extractor.client = make_mock_client(MOCK_ENTITIES_JSON)
# Make summarizer fail
fail_model = MagicMock()
fail_model.generate_content.side_effect = Exception("API rate limit exceeded")
orchestrator.summarizer.model = fail_model
result = await orchestrator.analyze_document(sample_mining_text)
# Should still return — summary will have "error" key
assert "summary" in result
assert "error" in result["summary"] or "summary" in result["summary"]
|