Spaces:
Sleeping
Sleeping
File size: 30,261 Bytes
67e153c |
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 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 |
#!/usr/bin/env python3
"""
Comprehensive tests for LLM Fallback System in VoucherBot
This test suite is designed to challenge the LLM fallback system's ability
to handle complex, ambiguous, and edge case queries that the regex-based
system cannot process effectively.
"""
import unittest
import sys
import os
import json
import time
from unittest.mock import Mock, patch
# Add the parent directory to the path to import modules
sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
from llm_fallback_router import (
LLMFallbackRouter,
InvalidInputError,
LLMProcessingError,
InvalidLLMResponseError,
IntentType,
RouterResponse
)
class MockLLMClient:
"""Enhanced mock LLM client for comprehensive testing"""
def __init__(self, response_mode="normal", fail_mode=None, delay=0):
self.response_mode = response_mode
self.fail_mode = fail_mode
self.delay = delay
self.call_count = 0
self.call_history = []
def generate(self, prompt: str) -> str:
"""Generate mock responses based on test configuration"""
self.call_count += 1
self.call_history.append(prompt)
if self.delay > 0:
time.sleep(self.delay)
if self.fail_mode == "exception":
raise Exception("Mock LLM client failure")
elif self.fail_mode == "invalid_json":
return "This is not valid JSON"
elif self.fail_mode == "malformed_response":
return '{"intent": "INVALID_INTENT", "parameters": "not_a_dict"}'
elif self.fail_mode == "timeout":
time.sleep(10) # Simulate timeout
return self._generate_normal_response(prompt)
elif self.fail_mode == "partial_response":
return '{"intent": "SEARCH_LISTINGS"' # Incomplete JSON
return self._generate_normal_response(prompt)
def _generate_normal_response(self, prompt: str) -> str:
"""Generate realistic mock responses based on prompt content"""
# Extract message from prompt
message_start = prompt.find('Message: "') + 10
message_end = prompt.find('"', message_start)
message = prompt[message_start:message_end] if message_start > 9 else ""
message_lower = message.lower()
# Sophisticated pattern matching for realistic responses
if any(phrase in message_lower for phrase in ["find", "search", "looking for", "need apartment"]):
return json.dumps({
"intent": "SEARCH_LISTINGS",
"parameters": {
"borough": self._extract_borough(message_lower),
"bedrooms": self._extract_bedrooms(message_lower),
"max_rent": self._extract_rent(message_lower),
"voucher_type": self._extract_voucher(message_lower)
},
"reasoning": "User is requesting to search for apartment listings with specific criteria"
})
elif any(phrase in message_lower for phrase in ["what if", "try", "instead", "change to"]):
return json.dumps({
"intent": "REFINE_SEARCH",
"parameters": {
"borough": self._extract_borough(message_lower),
"bedrooms": self._extract_bedrooms(message_lower),
"max_rent": self._extract_rent(message_lower),
"voucher_type": self._extract_voucher(message_lower)
},
"reasoning": "User wants to modify their existing search parameters"
})
elif any(phrase in message_lower for phrase in ["violation", "safe", "building", "inspect"]):
return json.dumps({
"intent": "CHECK_VIOLATIONS",
"parameters": {},
"reasoning": "User wants to check building safety violations"
})
elif any(phrase in message_lower for phrase in ["help", "assist", "what can", "how do"]):
return json.dumps({
"intent": "HELP_REQUEST",
"parameters": {},
"reasoning": "User is requesting help or information"
})
elif any(phrase in message_lower for phrase in ["what is", "explain", "tell me about"]) and \
any(voucher in message_lower for voucher in ["section 8", "hasa", "cityfheps", "voucher"]):
return json.dumps({
"intent": "ASK_VOUCHER_SUPPORT",
"parameters": {
"voucher_type": self._extract_voucher(message_lower)
},
"reasoning": "User is asking for information about voucher programs"
})
else:
return json.dumps({
"intent": "UNKNOWN",
"parameters": {},
"reasoning": "Unable to determine clear intent from the message"
})
def _extract_borough(self, message: str) -> str:
"""Extract borough from message"""
boroughs = {
"brooklyn": "Brooklyn", "bk": "Brooklyn",
"manhattan": "Manhattan", "mnh": "Manhattan",
"queens": "Queens", "qns": "Queens",
"bronx": "Bronx", "bx": "Bronx",
"staten island": "Staten Island", "si": "Staten Island"
}
for key, value in boroughs.items():
if key in message:
return value
return None
def _extract_bedrooms(self, message: str) -> int:
"""Extract bedroom count from message"""
import re
bedroom_match = re.search(r'(\d+)\s*(?:bed|br|bedroom)', message)
if bedroom_match:
return int(bedroom_match.group(1))
elif "studio" in message:
return 0
return None
def _extract_rent(self, message: str) -> int:
"""Extract rent amount from message"""
import re
rent_match = re.search(r'\$(\d+(?:,\d{3})*)', message)
if rent_match:
return int(rent_match.group(1).replace(',', ''))
return None
def _extract_voucher(self, message: str) -> str:
"""Extract voucher type from message"""
if "section 8" in message or "section-8" in message:
return "Section 8"
elif "cityfheps" in message or "city fheps" in message:
return "CityFHEPS"
elif "hasa" in message:
return "HASA"
elif "voucher" in message:
return "Housing Voucher"
return None
class TestLLMFallbackSystem(unittest.TestCase):
"""Comprehensive test suite for LLM fallback system"""
def setUp(self):
"""Set up test components"""
self.mock_llm = MockLLMClient()
self.router = LLMFallbackRouter(self.mock_llm, debug=True)
def test_basic_functionality(self):
"""Test basic LLM fallback functionality"""
test_cases = [
{
"message": "I need to find a 2-bedroom apartment in Brooklyn under $2500",
"expected_intent": "SEARCH_LISTINGS",
"expected_params": ["borough", "bedrooms", "max_rent"]
},
{
"message": "What if I try Queens instead?",
"expected_intent": "REFINE_SEARCH",
"expected_params": ["borough"]
},
{
"message": "Can you help me understand Section 8?",
"expected_intent": "ASK_VOUCHER_SUPPORT",
"expected_params": ["voucher_type"]
},
{
"message": "I need help with this system",
"expected_intent": "HELP_REQUEST",
"expected_params": []
}
]
print("\n🧠 Testing Basic LLM Fallback Functionality")
print("=" * 60)
for test_case in test_cases:
with self.subTest(message=test_case["message"]):
result = self.router.route(test_case["message"])
print(f"Message: '{test_case['message']}'")
print(f" Intent: {result['intent']}")
print(f" Parameters: {result['parameters']}")
print(f" Reasoning: {result['reasoning']}")
self.assertEqual(result["intent"], test_case["expected_intent"])
# Check that expected parameters are present (if any)
for param in test_case["expected_params"]:
self.assertIn(param, result["parameters"])
self.assertIsNotNone(result["parameters"][param])
print(" ✅ Test passed")
print()
def test_complex_ambiguous_queries(self):
"""Test LLM's ability to handle complex and ambiguous queries"""
complex_queries = [
{
"message": "I'm not sure what I'm looking for but I need somewhere to live",
"description": "Vague housing request"
},
{
"message": "My current situation is complicated and I need housing assistance",
"description": "Complex personal situation"
},
{
"message": "Can you help me figure out what kind of apartment I can afford with my voucher?",
"description": "Multi-part question with implicit search intent"
},
{
"message": "I've been looking everywhere but nothing seems right, maybe you can suggest something different",
"description": "Frustration with search refinement request"
},
{
"message": "The landlord said something about my voucher not being accepted, what should I do?",
"description": "Problem-solving request with voucher context"
},
{
"message": "I heard Brooklyn is good but expensive, what about other places that might work?",
"description": "Comparative analysis request"
},
{
"message": "My caseworker mentioned some options but I'm confused about the differences",
"description": "Information clarification request"
},
{
"message": "Is it worth looking in Manhattan or should I focus on outer boroughs?",
"description": "Strategic advice request"
}
]
print("\n🌀 Testing Complex and Ambiguous Queries")
print("=" * 60)
for query_info in complex_queries:
with self.subTest(message=query_info["message"]):
result = self.router.route(query_info["message"])
print(f"Query: '{query_info['message']}'")
print(f"Description: {query_info['description']}")
print(f" Intent: {result['intent']}")
print(f" Parameters: {result['parameters']}")
print(f" Reasoning: {result['reasoning']}")
# These should not be UNKNOWN if LLM is working properly
self.assertNotEqual(result["intent"], "UNKNOWN",
f"LLM failed to classify complex query: {query_info['message']}")
# Reasoning should be provided
self.assertIsNotNone(result["reasoning"])
self.assertNotEqual(result["reasoning"].strip(), "")
print(" ✅ Successfully handled complex query")
print()
def test_multilingual_support(self):
"""Test LLM's multilingual capabilities"""
multilingual_queries = [
{
"message": "Necesito encontrar un apartamento en Brooklyn",
"language": "es",
"expected_intent": "SEARCH_LISTINGS",
"description": "Spanish apartment search"
},
{
"message": "¿Qué es Section 8?",
"language": "es",
"expected_intent": "ASK_VOUCHER_SUPPORT",
"description": "Spanish voucher information request"
},
{
"message": "我需要在布鲁克林找房子",
"language": "zh",
"expected_intent": "SEARCH_LISTINGS",
"description": "Chinese apartment search"
},
{
"message": "আমার ব্রুকলিনে একটি অ্যাপার্টমেন্ট দরকার",
"language": "bn",
"expected_intent": "SEARCH_LISTINGS",
"description": "Bengali apartment search"
},
{
"message": "Help me find housing - ayuda por favor",
"language": "mixed",
"expected_intent": "SEARCH_LISTINGS",
"description": "Mixed language request"
}
]
print("\n🌍 Testing Multilingual Support")
print("=" * 60)
for query_info in multilingual_queries:
with self.subTest(message=query_info["message"]):
result = self.router.route(
query_info["message"],
language=query_info["language"]
)
print(f"Query: '{query_info['message']}'")
print(f"Language: {query_info['language']}")
print(f"Description: {query_info['description']}")
print(f" Intent: {result['intent']}")
print(f" Parameters: {result['parameters']}")
print(f" Reasoning: {result['reasoning']}")
# Should handle multilingual queries appropriately
self.assertNotEqual(result["intent"], "UNKNOWN",
f"LLM failed to handle multilingual query: {query_info['message']}")
print(" ✅ Successfully handled multilingual query")
print()
def test_context_awareness(self):
"""Test LLM's ability to use context for better classification"""
context_tests = [
{
"message": "try something else",
"context": '{"borough": "Brooklyn", "bedrooms": 2, "max_rent": 2500}',
"expected_intent": "REFINE_SEARCH",
"description": "Vague refinement with search context"
},
{
"message": "what about Manhattan?",
"context": '{"last_search": "Brooklyn apartments", "results": 5}',
"expected_intent": "REFINE_SEARCH",
"description": "Borough change with search history"
},
{
"message": "show me more",
"context": '{"current_listings": 3, "total_available": 15}',
"expected_intent": "SEARCH_LISTINGS",
"description": "Continuation request with listings context"
},
{
"message": "that's too expensive",
"context": '{"last_shown_rent": 3000, "user_budget": 2500}',
"expected_intent": "REFINE_SEARCH",
"description": "Budget feedback with price context"
}
]
print("\n🧠 Testing Context Awareness")
print("=" * 60)
for test_case in context_tests:
with self.subTest(message=test_case["message"]):
result = self.router.route(test_case["message"], test_case["context"])
print(f"Message: '{test_case['message']}'")
print(f"Context: {test_case['context']}")
print(f"Description: {test_case['description']}")
print(f" Intent: {result['intent']}")
print(f" Parameters: {result['parameters']}")
print(f" Reasoning: {result['reasoning']}")
# Context should improve classification
self.assertNotEqual(result["intent"], "UNKNOWN",
f"LLM failed to use context for: {test_case['message']}")
print(" ✅ Successfully used context")
print()
def test_error_handling_and_recovery(self):
"""Test LLM fallback system's error handling and recovery"""
error_scenarios = [
{
"fail_mode": "invalid_json",
"description": "Invalid JSON response from LLM"
},
{
"fail_mode": "malformed_response",
"description": "Malformed response structure"
},
{
"fail_mode": "partial_response",
"description": "Incomplete JSON response"
},
{
"fail_mode": "exception",
"description": "LLM client exception"
}
]
print("\n🚨 Testing Error Handling and Recovery")
print("=" * 60)
for scenario in error_scenarios:
with self.subTest(fail_mode=scenario["fail_mode"]):
# Create router with failing mock LLM
failing_llm = MockLLMClient(fail_mode=scenario["fail_mode"])
failing_router = LLMFallbackRouter(failing_llm, debug=True)
print(f"Scenario: {scenario['description']}")
with self.assertRaises((LLMProcessingError, InvalidLLMResponseError)):
failing_router.route("Find apartments in Brooklyn")
print(" ✅ Properly raised expected exception")
print()
def test_retry_mechanism(self):
"""Test the retry mechanism for failed LLM calls"""
print("\n🔄 Testing Retry Mechanism")
print("=" * 60)
# Create a mock that fails twice then succeeds
class RetryTestLLM:
def __init__(self):
self.attempt_count = 0
def generate(self, prompt):
self.attempt_count += 1
if self.attempt_count <= 2:
raise Exception(f"Attempt {self.attempt_count} failed")
return json.dumps({
"intent": "SEARCH_LISTINGS",
"parameters": {"borough": "Brooklyn"},
"reasoning": f"Succeeded on attempt {self.attempt_count}"
})
retry_llm = RetryTestLLM()
retry_router = LLMFallbackRouter(retry_llm, debug=True, max_retries=3)
result = retry_router.route("Find apartments in Brooklyn")
print(f"Total attempts made: {retry_llm.attempt_count}")
print(f"Final result: {result}")
self.assertEqual(retry_llm.attempt_count, 3)
self.assertEqual(result["intent"], "SEARCH_LISTINGS")
print(" ✅ Retry mechanism worked correctly")
print()
def test_performance_under_load(self):
"""Test LLM fallback performance under various load conditions"""
print("\n⚡ Testing Performance Under Load")
print("=" * 60)
# Test with different response delays
delay_tests = [
{"delay": 0, "description": "Instant response"},
{"delay": 0.1, "description": "Fast response (100ms)"},
{"delay": 0.5, "description": "Moderate response (500ms)"},
{"delay": 1.0, "description": "Slow response (1s)"}
]
for delay_test in delay_tests:
with self.subTest(delay=delay_test["delay"]):
delayed_llm = MockLLMClient(delay=delay_test["delay"])
delayed_router = LLMFallbackRouter(delayed_llm, debug=False)
start_time = time.time()
result = delayed_router.route("Find apartments in Brooklyn")
end_time = time.time()
actual_time = end_time - start_time
expected_time = delay_test["delay"]
print(f"Test: {delay_test['description']}")
print(f" Expected delay: {expected_time}s")
print(f" Actual time: {actual_time:.3f}s")
print(f" Result: {result['intent']}")
# Allow for some overhead but should be close to expected
self.assertGreaterEqual(actual_time, expected_time)
self.assertLess(actual_time, expected_time + 0.5) # Max 500ms overhead
print(" ✅ Performance within acceptable range")
print()
def test_parameter_normalization(self):
"""Test parameter normalization and validation"""
normalization_tests = [
{
"input_params": {"borough": "bk", "bedrooms": "2", "max_rent": "2,500"},
"expected_borough": "Brooklyn",
"expected_bedrooms": 2,
"expected_rent": 2500,
"description": "Abbreviation and string normalization"
},
{
"input_params": {"borough": "staten island", "voucher_type": "section 8"},
"expected_borough": "Staten Island",
"expected_voucher": "Section 8",
"description": "Multi-word and voucher normalization"
},
{
"input_params": {"borough": "manhattan", "bedrooms": 0},
"expected_borough": "Manhattan",
"expected_bedrooms": 0,
"description": "Studio apartment (0 bedrooms)"
}
]
print("\n🔧 Testing Parameter Normalization")
print("=" * 60)
for test_case in normalization_tests:
with self.subTest(description=test_case["description"]):
# Create a custom mock response with test parameters
class CustomMockLLM:
def generate(self, prompt):
return json.dumps({
"intent": "SEARCH_LISTINGS",
"parameters": test_case["input_params"],
"reasoning": "Test normalization"
})
custom_router = LLMFallbackRouter(CustomMockLLM())
result = custom_router.route("Test message")
print(f"Test: {test_case['description']}")
print(f" Input params: {test_case['input_params']}")
print(f" Normalized params: {result['parameters']}")
# Check normalization
if "expected_borough" in test_case:
self.assertEqual(result["parameters"]["borough"], test_case["expected_borough"])
if "expected_bedrooms" in test_case:
self.assertEqual(result["parameters"]["bedrooms"], test_case["expected_bedrooms"])
if "expected_rent" in test_case:
self.assertEqual(result["parameters"]["max_rent"], test_case["expected_rent"])
if "expected_voucher" in test_case:
self.assertEqual(result["parameters"]["voucher_type"], test_case["expected_voucher"])
print(" ✅ Parameters normalized correctly")
print()
def test_edge_cases_and_boundary_conditions(self):
"""Test edge cases and boundary conditions"""
edge_cases = [
{
"message": "",
"description": "Empty message"
},
{
"message": " ",
"description": "Whitespace only"
},
{
"message": "a" * 10000,
"description": "Very long message"
},
{
"message": "🏠🏡🏘️🏚️🏗️",
"description": "Emoji only"
},
{
"message": "!@#$%^&*()_+{}[]|\\:;\"'<>?,./",
"description": "Special characters only"
},
{
"message": "find apartments" + "\n" * 100,
"description": "Message with many newlines"
}
]
print("\n🔧 Testing Edge Cases and Boundary Conditions")
print("=" * 60)
for edge_case in edge_cases:
with self.subTest(message=edge_case["description"]):
try:
if edge_case["message"] == "":
# Empty message should raise InvalidInputError
with self.assertRaises(InvalidInputError):
self.router.route(edge_case["message"])
print(f"Test: {edge_case['description']}")
print(" ✅ Correctly raised InvalidInputError for empty message")
else:
result = self.router.route(edge_case["message"])
print(f"Test: {edge_case['description']}")
print(f" Intent: {result['intent']}")
print(f" Parameters: {result['parameters']}")
print(" ✅ Handled edge case without error")
except Exception as e:
print(f"Test: {edge_case['description']}")
print(f" ❌ Unexpected error: {e}")
self.fail(f"Edge case caused unexpected error: {e}")
print()
def test_regression_scenarios(self):
"""Test known regression scenarios and previously problematic queries"""
regression_tests = [
{
"message": "I live in Brooklyn but work in Manhattan",
"description": "Multiple borough mentions",
"expected_behavior": "Should not extract both boroughs"
},
{
"message": "My 3 kids need a place to live",
"description": "Family size vs bedroom count",
"expected_behavior": "Should not extract '3' as bedrooms"
},
{
"message": "I make $50,000 per year",
"description": "Annual income vs monthly rent",
"expected_behavior": "Should not extract as max_rent"
},
{
"message": "Section 8 is a good program",
"description": "Informational statement vs request",
"expected_behavior": "Should not be SEARCH_LISTINGS"
},
{
"message": "What if I told you I need help?",
"description": "Hypothetical vs what-if scenario",
"expected_behavior": "Should be HELP_REQUEST, not REFINE_SEARCH"
}
]
print("\n🔍 Testing Regression Scenarios")
print("=" * 60)
for test_case in regression_tests:
with self.subTest(message=test_case["message"]):
result = self.router.route(test_case["message"])
print(f"Message: '{test_case['message']}'")
print(f"Description: {test_case['description']}")
print(f"Expected behavior: {test_case['expected_behavior']}")
print(f" Intent: {result['intent']}")
print(f" Parameters: {result['parameters']}")
print(f" Reasoning: {result['reasoning']}")
# Verify the specific expected behavior
if "Multiple borough mentions" in test_case["description"]:
# Should not extract both Brooklyn and Manhattan
borough = result["parameters"].get("borough")
if borough:
self.assertIn(borough, ["Brooklyn", "Manhattan"])
print(f" ✅ Correctly extracted single borough: {borough}")
else:
print(" ✅ Correctly extracted no borough")
elif "Family size vs bedroom count" in test_case["description"]:
# Should not extract 3 as bedrooms
bedrooms = result["parameters"].get("bedrooms")
self.assertNotEqual(bedrooms, 3)
print(" ✅ Correctly did not extract family size as bedrooms")
elif "Annual income vs monthly rent" in test_case["description"]:
# Should not extract 50000 as max_rent
max_rent = result["parameters"].get("max_rent")
self.assertNotEqual(max_rent, 50000)
print(" ✅ Correctly did not extract annual income as rent")
elif "Informational statement vs request" in test_case["description"]:
# Should not be SEARCH_LISTINGS
self.assertNotEqual(result["intent"], "SEARCH_LISTINGS")
print(" ✅ Correctly did not classify as search request")
elif "Hypothetical vs what-if scenario" in test_case["description"]:
# Should be HELP_REQUEST, not REFINE_SEARCH
self.assertEqual(result["intent"], "HELP_REQUEST")
print(" ✅ Correctly classified as help request")
print()
if __name__ == "__main__":
# Run the tests with verbose output
unittest.main(verbosity=2, buffer=True) |