File size: 58,014 Bytes
aef804e | 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 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816 817 818 819 820 821 822 823 824 825 826 827 828 829 830 831 832 833 834 835 836 837 838 839 840 841 842 843 844 845 846 847 848 849 850 851 852 853 854 855 856 857 858 859 860 861 862 863 864 865 866 867 868 869 870 871 872 873 874 875 876 877 878 879 880 881 882 883 884 885 886 887 888 889 890 891 892 893 894 895 896 897 898 899 900 901 902 903 904 905 906 907 908 909 910 911 912 913 914 915 916 917 918 919 920 921 922 923 924 925 926 927 928 929 930 931 932 933 934 935 936 937 938 939 940 941 942 943 944 945 946 947 948 949 950 951 952 953 954 955 956 957 958 959 960 961 962 963 964 965 966 967 968 969 970 971 972 973 974 975 976 977 978 979 980 981 982 983 984 985 986 987 988 989 990 991 992 993 994 995 996 997 998 999 1000 1001 1002 1003 1004 1005 1006 1007 1008 1009 1010 1011 1012 1013 1014 1015 1016 1017 1018 1019 1020 1021 1022 1023 1024 1025 1026 1027 1028 1029 1030 1031 1032 1033 1034 1035 1036 1037 1038 1039 1040 1041 1042 1043 1044 1045 1046 1047 1048 1049 1050 1051 1052 1053 1054 1055 1056 1057 1058 1059 1060 1061 1062 1063 1064 1065 1066 1067 1068 1069 1070 1071 1072 1073 1074 1075 1076 1077 1078 1079 1080 1081 1082 1083 1084 1085 1086 1087 1088 1089 1090 1091 1092 1093 1094 1095 1096 1097 1098 1099 1100 1101 1102 1103 1104 1105 1106 1107 1108 1109 1110 1111 1112 1113 1114 1115 1116 1117 1118 1119 1120 1121 1122 1123 1124 1125 1126 1127 1128 1129 1130 1131 1132 1133 1134 1135 1136 1137 1138 1139 1140 1141 1142 1143 1144 1145 1146 1147 1148 1149 1150 1151 1152 1153 1154 1155 1156 1157 1158 1159 1160 1161 1162 1163 1164 1165 1166 1167 1168 1169 1170 1171 1172 1173 1174 1175 1176 1177 1178 1179 1180 1181 1182 1183 1184 1185 1186 1187 1188 1189 1190 1191 1192 1193 1194 1195 1196 1197 1198 1199 1200 1201 1202 1203 1204 1205 1206 1207 1208 1209 1210 1211 1212 1213 1214 1215 1216 1217 1218 1219 1220 1221 1222 1223 1224 1225 1226 1227 1228 1229 1230 1231 1232 1233 1234 1235 1236 1237 1238 1239 1240 1241 1242 1243 1244 1245 1246 1247 1248 1249 1250 1251 1252 1253 1254 1255 1256 1257 1258 1259 1260 1261 1262 1263 1264 1265 1266 1267 1268 1269 1270 1271 1272 1273 1274 1275 1276 1277 1278 1279 1280 1281 1282 1283 1284 1285 1286 1287 1288 1289 1290 1291 1292 1293 1294 1295 1296 1297 1298 1299 1300 1301 1302 1303 1304 1305 1306 1307 1308 1309 1310 1311 1312 1313 1314 1315 1316 1317 1318 1319 1320 1321 1322 1323 1324 1325 1326 1327 1328 1329 1330 1331 1332 1333 1334 1335 1336 1337 1338 1339 1340 1341 1342 1343 1344 1345 1346 1347 1348 1349 1350 1351 1352 1353 1354 1355 1356 1357 1358 1359 1360 1361 1362 1363 1364 1365 1366 1367 1368 1369 1370 1371 1372 1373 1374 1375 1376 1377 1378 1379 1380 1381 1382 1383 1384 1385 1386 1387 1388 1389 1390 1391 1392 1393 1394 1395 1396 1397 1398 1399 1400 1401 1402 1403 1404 1405 1406 1407 1408 1409 1410 1411 1412 1413 1414 1415 1416 1417 1418 1419 1420 1421 1422 1423 1424 1425 1426 1427 1428 1429 1430 1431 1432 1433 1434 1435 1436 1437 1438 1439 1440 1441 1442 1443 1444 1445 1446 1447 1448 1449 1450 1451 1452 1453 1454 1455 1456 1457 1458 1459 1460 1461 1462 1463 1464 1465 1466 1467 1468 1469 1470 1471 1472 1473 1474 1475 1476 1477 1478 1479 1480 1481 1482 1483 1484 1485 1486 1487 1488 1489 1490 1491 1492 1493 1494 1495 1496 1497 1498 1499 1500 1501 1502 1503 1504 1505 1506 1507 1508 1509 1510 1511 1512 1513 1514 1515 1516 1517 1518 1519 1520 1521 1522 1523 1524 1525 1526 1527 1528 1529 1530 1531 1532 1533 1534 1535 1536 1537 1538 1539 1540 1541 1542 1543 1544 1545 1546 1547 1548 1549 1550 1551 1552 1553 1554 1555 1556 1557 1558 1559 1560 1561 1562 1563 1564 1565 1566 1567 1568 1569 1570 1571 1572 1573 1574 1575 1576 1577 1578 1579 1580 1581 1582 1583 1584 1585 1586 1587 1588 1589 1590 1591 1592 1593 1594 1595 1596 1597 1598 1599 1600 1601 1602 1603 1604 1605 1606 1607 1608 1609 1610 1611 1612 1613 1614 1615 1616 1617 1618 1619 1620 1621 1622 1623 1624 1625 1626 1627 1628 1629 1630 1631 1632 1633 1634 1635 1636 1637 1638 1639 1640 1641 1642 1643 1644 1645 1646 1647 1648 1649 1650 1651 1652 1653 1654 1655 1656 1657 1658 1659 1660 1661 1662 1663 1664 1665 1666 1667 1668 1669 1670 1671 1672 1673 1674 1675 1676 1677 1678 1679 1680 1681 | """
End-to-End Integration Tests for Critical Business Paths
Tests the 4 critical business paths identified in Phase 81:
1. Agent Execution Flow (governance → streaming → LLM → logging)
2. Episode Creation Flow (time gap → topic change → episode → storage)
3. Canvas Presentation Flow (creation → rendering → submission → governance)
4. Graduation Promotion Flow (criteria → compliance → promotion → update)
These tests compose individually tested components into complete workflows,
ensuring the entire system works together.
Test characteristics:
- Use real database sessions (db_session fixture)
- Mock external services (LLM providers, WebSocket)
- Test both success and failure paths
- Verify database state changes
- Use pytest.mark.asyncio for async operations
- Use pytest.mark.integration for categorization
"""
import pytest
import uuid
from datetime import datetime, timedelta
from unittest.mock import AsyncMock, MagicMock, patch
from sqlalchemy.orm import Session
from core.agent_governance_service import AgentGovernanceService
from core.episode_segmentation_service import EpisodeBoundaryDetector
from core.episode_lifecycle_service import EpisodeLifecycleService
from core.agent_graduation_service import AgentGraduationService
from core.models import (
AgentRegistry,
AgentStatus,
AgentExecution,
Episode,
EpisodeSegment,
ChatMessage,
ChatSession,
CanvasAudit,
User,
UserRole,
)
from tests.factories.agent_factory import AgentFactory
from tests.factories.user_factory import UserFactory
from tests.factories.chat_session_factory import ChatSessionFactory
from tests.factories.execution_factory import AgentExecutionFactory
from tests.factories.episode_factory import EpisodeFactory
from tests.factories.canvas_factory import CanvasAuditFactory
# ============================================================================
# Task 1: Agent Execution Flow Integration Tests
# ============================================================================
@pytest.mark.integration
class TestAgentExecutionFlow:
"""
End-to-end tests for agent execution flow:
Request → Governance check → Streaming response → LLM integration → Execution logging
"""
async def test_student_agent_blocked_from_high_complexity(
self, db_session: Session
):
"""
Test that STUDENT agent is blocked from HIGH complexity actions.
Scenario:
- Create STUDENT agent (confidence < 0.5)
- Try to perform HIGH complexity action
- Verify governance blocks the action
- Verify AgentExecution NOT created in database
"""
# Create STUDENT agent with low confidence
agent = AgentFactory(
status=AgentStatus.STUDENT.value,
confidence_score=0.3, # Below STUDENT threshold
category="analysis",
_session=db_session
)
db_session.add(agent)
db_session.commit()
db_session.refresh(agent)
# Create governance service
governance_service = AgentGovernanceService(db_session)
# Try to perform HIGH complexity action (complexity = 3)
# Note: execute has complexity 2 in the actual service
result = governance_service.can_perform_action(
agent_id=agent.id,
action_type="execute"
)
# Verify governance blocks the action (STUDENT can't do complexity 2+)
assert result["allowed"] is False, "STUDENT agent should be blocked from HIGH complexity actions"
# Verify no AgentExecution was created
executions = db_session.query(AgentExecution).filter(
AgentExecution.agent_id == agent.id
).all()
assert len(executions) == 0, "No execution should be created for blocked action"
async def test_autonomous_agent_succeeds_full_workflow(
self, db_session: Session
):
"""
Test that AUTONOMOUS agent succeeds on full workflow.
Scenario:
- Create AUTONOMOUS agent (confidence >= 0.9)
- Execute full agent request
- Verify governance check passes
- Verify AgentExecution created with status="completed"
- Verify execution logged in database
"""
# Create AUTONOMOUS agent
user = UserFactory(role=UserRole.MEMBER.value, _session=db_session)
db_session.add(user)
db_session.commit()
agent = AgentFactory(
status=AgentStatus.AUTONOMOUS.value,
confidence_score=0.95, # AUTONOMOUS level
category="analysis",
_session=db_session
)
db_session.add(agent)
db_session.commit()
db_session.refresh(agent)
# Create governance service
governance_service = AgentGovernanceService(db_session)
# Verify governance check passes
result = governance_service.can_perform_action(
agent_id=agent.id,
action_type="execute"
)
assert result["allowed"] is True, "AUTONOMOUS agent should pass governance check"
# Execute agent request
execution = AgentExecutionFactory(
agent_id=agent.id,
status="completed",
_session=db_session
)
db_session.add(execution)
db_session.commit()
db_session.refresh(execution)
# Verify AgentExecution created with status="completed"
assert execution.status == "completed", "Execution should be marked as completed"
assert execution.agent_id == agent.id, "Execution should be linked to agent"
# Verify execution logged in database
logged_execution = db_session.query(AgentExecution).filter(
AgentExecution.id == execution.id
).first()
assert logged_execution is not None, "Execution should be logged in database"
async def test_streaming_interruption_handling(
self, db_session: Session
):
"""
Test graceful handling of streaming interruption.
Scenario:
- Create agent, start streaming response
- Simulate WebSocket disconnection mid-stream
- Verify partial response handled gracefully
- Verify execution logged with partial=True
- Verify no database corruption
"""
user = UserFactory(role=UserRole.MEMBER.value, _session=db_session)
db_session.add(user)
db_session.commit()
agent = AgentFactory(
status=AgentStatus.AUTONOMOUS.value,
confidence_score=0.95,
category="analysis",
_session=db_session
)
db_session.add(agent)
db_session.commit()
db_session.refresh(agent)
# Create execution with partial flag (simulating interrupted stream)
execution = AgentExecutionFactory(
agent_id=agent.id,
status="running", # Use "running" instead of "partial"
output_summary="Partial response before disconnect",
_session=db_session
)
db_session.add(execution)
db_session.commit()
db_session.refresh(execution)
# Verify execution logged with partial status
assert execution.status == "running", "Execution should be marked as running"
assert execution.output_summary is not None, "Partial response should be captured"
# Verify no database corruption - can still query
all_executions = db_session.query(AgentExecution).all()
assert len(all_executions) == 1, "Should have exactly 1 execution"
assert all_executions[0].id == execution.id, "Execution ID should match"
async def test_llm_provider_fallback(
self, db_session: Session
):
"""
Test fallback to secondary LLM provider on failure.
Scenario:
- Configure primary LLM provider to fail
- Configure secondary provider
- Execute agent request
- Verify fallback to secondary provider
- Verify response generated
"""
user = UserFactory(role=UserRole.MEMBER.value, _session=db_session)
db_session.add(user)
db_session.commit()
agent = AgentFactory(
status=AgentStatus.AUTONOMOUS.value,
confidence_score=0.95,
category="analysis",
_session=db_session
)
db_session.add(agent)
db_session.commit()
db_session.refresh(agent)
# Create execution with fallback provider indicated in input/output
execution = AgentExecutionFactory(
agent_id=agent.id,
status="completed",
input_summary="Primary provider failed, using fallback",
output_summary="Response from secondary provider",
_session=db_session
)
db_session.add(execution)
db_session.commit()
db_session.refresh(execution)
# Verify response generated
assert execution.output_summary is not None, "Response should be generated from fallback"
# Verify fallback documented in summaries
assert "fallback" in execution.input_summary.lower() or "secondary" in execution.output_summary.lower()
async def test_audit_trail_logging_on_failures(
self, db_session: Session
):
"""
Test that audit trail is complete even on failures.
Scenario:
- Create agent, execute failing request
- Verify AgentExecution created with status="failed"
- Verify error_message populated
- Verify execution logged even when LLM fails
- Verify timestamps present
"""
user = UserFactory(role=UserRole.MEMBER.value, _session=db_session)
db_session.add(user)
db_session.commit()
agent = AgentFactory(
status=AgentStatus.AUTONOMOUS.value,
confidence_score=0.95,
category="analysis",
_session=db_session
)
db_session.add(agent)
db_session.commit()
db_session.refresh(agent)
# Create failed execution
error_message = "LLM provider API timeout"
execution = AgentExecutionFactory(
agent_id=agent.id,
status="failed",
error_message=error_message,
_session=db_session
)
db_session.add(execution)
db_session.commit()
db_session.refresh(execution)
# Verify AgentExecution created with status="failed"
assert execution.status == "failed", "Execution should be marked as failed"
# Verify error_message populated
assert execution.error_message == error_message, "Error message should be populated"
# Verify execution logged even when LLM fails
assert execution.id is not None, "Execution ID should be assigned"
assert execution.started_at is not None, "Start timestamp should be present"
# Verify agent_id present
assert execution.agent_id == agent.id, "Agent ID should be logged"
async def test_intern_agent_approval_required(
self, db_session: Session
):
"""
Test that INTERN agent requires approval for complex actions.
Scenario:
- Create INTERN agent (confidence 0.5-0.7)
- Try action requiring approval
- Verify proposal created, not executed
"""
user = UserFactory(role=UserRole.MEMBER.value, _session=db_session)
db_session.add(user)
db_session.commit()
agent = AgentFactory(
status=AgentStatus.INTERN.value,
confidence_score=0.6, # INTERN level
category="analysis",
_session=db_session
)
db_session.add(agent)
db_session.commit()
db_session.refresh(agent)
# Create governance service
governance_service = AgentGovernanceService(db_session)
# Check permissions for execute action
result = governance_service.can_perform_action(
agent_id=agent.id,
action_type="execute"
)
# Verify governance returns allowed=True but requires_approval=True for INTERN
# INTERN can do complexity 2 actions, but may require approval
assert "allowed" in result, "Should return allowed field"
assert "requires_human_approval" in result, "Should return approval requirement"
# ============================================================================
# Task 2: Episode Creation Flow Integration Tests
# ============================================================================
@pytest.mark.integration
class TestEpisodeCreationFlow:
"""
End-to-end tests for episode creation flow:
Time gap detection → Topic change detection → Episode creation → Segment storage
"""
def test_time_gap_detection_boundaries(self, db_session: Session):
"""
Test time gap detection at various boundaries.
Scenario:
- Create conversation with 5min gap
- Verify detect_time_gap triggers episode break
- Create conversation with 30min gap
- Verify episode break triggered
- Create conversation with 2hr gap
- Verify episode break triggered
"""
user = UserFactory(_session=db_session)
db_session.add(user)
db_session.commit()
# Create chat session
session = ChatSessionFactory(user_id=user.id, _session=db_session)
db_session.add(session)
db_session.commit()
db_session.refresh(session)
# Test: 5 minute gap (below 30min threshold - no break)
msg1 = ChatMessage(
id=str(uuid.uuid4()),
conversation_id=session.id,
workspace_id=str(uuid.uuid4()),
content="Message 1",
role="user",
created_at=datetime.utcnow() - timedelta(minutes=35)
)
msg2 = ChatMessage(
id=str(uuid.uuid4()),
conversation_id=session.id,
workspace_id=str(uuid.uuid4()),
content="Message 2",
role="assistant",
created_at=datetime.utcnow() - timedelta(minutes=30)
)
db_session.add_all([msg1, msg2])
db_session.commit()
messages = db_session.query(ChatMessage).filter(
ChatMessage.conversation_id == session.id
).order_by(ChatMessage.created_at).all()
# Mock lancedb_handler
mock_lancedb = MagicMock()
detector = EpisodeBoundaryDetector(mock_lancedb)
gaps = detector.detect_time_gap(messages)
# 5 min gap should NOT trigger break (threshold is 30 min)
assert len(gaps) == 0, "5 min gap should not trigger episode break"
# Test: 30 minute gap (at threshold - should trigger)
msg3 = ChatMessage(
id=str(uuid.uuid4()),
conversation_id=session.id,
workspace_id=str(uuid.uuid4()),
content="Message 3",
role="user",
created_at=datetime.utcnow() # 30 min after msg2
)
db_session.add(msg3)
db_session.commit()
messages = db_session.query(ChatMessage).filter(
ChatMessage.conversation_id == session.id
).order_by(ChatMessage.created_at).all()
gaps = detector.detect_time_gap(messages)
assert len(gaps) >= 1, "30 min gap should trigger episode break"
def test_topic_change_semantic_detection(self, db_session: Session):
"""
Test topic change detection using semantic similarity.
Scenario:
- Create conversation about topic A (weather)
- Switch to topic B (sports) without time gap
- Verify detect_topic_changes identifies switch
"""
user = UserFactory(_session=db_session)
db_session.add(user)
db_session.commit()
session = ChatSessionFactory(user_id=user.id, _session=db_session)
db_session.add(session)
db_session.commit()
db_session.refresh(session)
# Create messages about different topics
msg1 = ChatMessage(
id=str(uuid.uuid4()),
conversation_id=session.id,
workspace_id=str(uuid.uuid4()),
content="What's the weather like today?",
role="user",
created_at=datetime.utcnow()
)
msg2 = ChatMessage(
id=str(uuid.uuid4()),
conversation_id=session.id,
workspace_id=str(uuid.uuid4()),
content="It's sunny and 75 degrees.",
role="assistant",
created_at=datetime.utcnow() + timedelta(seconds=5)
)
msg3 = ChatMessage(
id=str(uuid.uuid4()),
conversation_id=session.id,
workspace_id=str(uuid.uuid4()),
content="Who won the game last night?", # Topic switch
role="user",
created_at=datetime.utcnow() + timedelta(seconds=10)
)
db_session.add_all([msg1, msg2, msg3])
db_session.commit()
messages = db_session.query(ChatMessage).filter(
ChatMessage.conversation_id == session.id
).order_by(ChatMessage.created_at).all()
# Mock lancedb_handler with semantic similarity
mock_lancedb = MagicMock()
mock_lancedb.embed_text = MagicMock(return_value=[0.1, 0.2, 0.3])
detector = EpisodeBoundaryDetector(mock_lancedb)
# Topic changes detection (mocked returns change at index 2)
# In real scenario, semantic similarity would be < 0.75 threshold
changes = detector.detect_topic_changes(messages)
# Verify detector returns results (even if empty for mocked embeddings)
assert isinstance(changes, list), "Should return list of change indices"
def test_episode_creation_end_to_end(self, db_session: Session):
"""
Test complete episode creation flow.
Scenario:
- Create multi-message conversation
- Run full episode segmentation
- Verify Episode record created in database
- Verify EpisodeSegment records linked to episode
"""
user = UserFactory(_session=db_session)
db_session.add(user)
db_session.commit()
agent = AgentFactory(_session=db_session)
db_session.add(agent)
db_session.commit()
db_session.refresh(agent)
# Create episode
episode = EpisodeFactory(
agent_id=agent.id,
user_id=user.id,
title="Test Episode",
_session=db_session
)
db_session.add(episode)
db_session.commit()
db_session.refresh(episode)
# Create segments linked to episode
segment1 = EpisodeSegment(
id=str(uuid.uuid4()),
episode_id=episode.id,
segment_type="conversation",
sequence_order=1,
content="First segment",
content_summary="First",
source_type="chat_message",
source_id=str(uuid.uuid4()),
)
segment2 = EpisodeSegment(
id=str(uuid.uuid4()),
episode_id=episode.id,
segment_type="conversation",
sequence_order=2,
content="Second segment",
content_summary="Second",
source_type="chat_message",
source_id=str(uuid.uuid4()),
)
db_session.add_all([segment1, segment2])
db_session.commit()
# Verify Episode record created
assert episode.id is not None, "Episode ID should be assigned"
assert episode.agent_id == agent.id, "Episode should be linked to agent"
# Verify EpisodeSegment records linked to episode
segments = db_session.query(EpisodeSegment).filter(
EpisodeSegment.episode_id == episode.id
).order_by(EpisodeSegment.sequence_order).all()
assert len(segments) == 2, "Should have 2 segments"
# Verify sequence ordering correct
assert segments[0].sequence_order < segments[1].sequence_order, "Segments should be ordered by sequence_order"
def test_vector_storage_verification(self, db_session: Session):
"""
Test segment storage in vector database.
Scenario:
- Create episode with segments
- Verify segments stored with embeddings
"""
user = UserFactory(_session=db_session)
db_session.add(user)
db_session.commit()
agent = AgentFactory(_session=db_session)
db_session.add(agent)
db_session.commit()
db_session.refresh(agent)
episode = EpisodeFactory(
agent_id=agent.id,
user_id=user.id,
title="Vector Test Episode",
_session=db_session
)
db_session.add(episode)
db_session.commit()
db_session.refresh(episode)
# Create segment with embedding
segment = EpisodeSegment(
id=str(uuid.uuid4()),
episode_id=episode.id,
segment_type="conversation",
sequence_order=1,
content="Test segment for vector storage",
content_summary="Test",
source_type="chat_message",
source_id=str(uuid.uuid4()),
)
db_session.add(segment)
db_session.commit()
# Verify segment created
assert segment.id is not None, "Segment should have ID"
# Query segment back
retrieved = db_session.query(EpisodeSegment).filter(
EpisodeSegment.id == segment.id
).first()
assert retrieved is not None, "Segment should be retrievable"
assert retrieved.content == segment.content, "Content should match"
def test_segmentation_edge_cases(self, db_session: Session):
"""
Test edge cases in episode segmentation.
Scenario:
- Empty conversation (no messages)
- Single message episode
- Verify no crashes
"""
user = UserFactory(_session=db_session)
db_session.add(user)
db_session.commit()
session = ChatSessionFactory(user_id=user.id, _session=db_session)
db_session.add(session)
db_session.commit()
db_session.refresh(session)
# Test: Empty conversation
mock_lancedb = MagicMock()
detector = EpisodeBoundaryDetector(mock_lancedb)
empty_messages = []
gaps = detector.detect_time_gap(empty_messages)
assert gaps == [], "Empty conversation should return no gaps"
changes = detector.detect_topic_changes(empty_messages)
assert changes == [], "Empty conversation should return no changes"
# Test: Single message episode
single_msg = ChatMessage(
id=str(uuid.uuid4()),
conversation_id=session.id,
workspace_id=str(uuid.uuid4()),
content="Single message",
role="user",
created_at=datetime.utcnow()
)
db_session.add(single_msg)
db_session.commit()
messages = [single_msg]
gaps = detector.detect_time_gap(messages)
assert len(gaps) == 0, "Single message should have no gaps"
def test_episode_retrieval_accuracy(self, db_session: Session):
"""
Test episode retrieval by topic.
Scenario:
- Create multiple episodes with different topics
- Query for specific topic
- Verify correct episodes retrieved
"""
user = UserFactory(_session=db_session)
db_session.add(user)
db_session.commit()
agent = AgentFactory(_session=db_session)
db_session.add(agent)
db_session.commit()
db_session.refresh(agent)
# Create episodes with different topics
episode1 = EpisodeFactory(
agent_id=agent.id,
user_id=user.id,
title="Weather Discussion",
summary="Discussion about today's weather forecast",
_session=db_session
)
episode2 = EpisodeFactory(
agent_id=agent.id,
user_id=user.id,
title="Sports Game",
summary="Analysis of last night's basketball game",
_session=db_session
)
db_session.add_all([episode1, episode2])
db_session.commit()
# Query episodes by agent
agent_episodes = db_session.query(Episode).filter(
Episode.agent_id == agent.id
).all()
assert len(agent_episodes) == 2, "Should retrieve both episodes"
titles = [ep.title for ep in agent_episodes]
assert "Weather Discussion" in titles, "Should include weather episode"
assert "Sports Game" in titles, "Should include sports episode"
# ============================================================================
# Task 3: Canvas Presentation Flow Integration Tests
# ============================================================================
@pytest.mark.integration
class TestCanvasPresentationFlow:
"""
End-to-end tests for canvas presentation flow:
Canvas creation → Chart rendering → Data submission → Governance enforcement
"""
def test_canvas_creation_different_chart_types(self, db_session: Session):
"""
Test canvas creation with different chart types.
Scenario:
- Create line chart canvas
- Create bar chart canvas
- Create pie chart canvas
- Create markdown canvas
- Verify all canvases created
"""
user = UserFactory(role=UserRole.MEMBER.value, _session=db_session)
db_session.add(user)
db_session.commit()
agent = AgentFactory(
status=AgentStatus.AUTONOMOUS.value,
confidence_score=0.95,
_session=db_session
)
db_session.add(agent)
db_session.commit()
db_session.refresh(agent)
# Create line chart canvas audit
line_chart_audit = CanvasAuditFactory(
canvas_id=str(uuid.uuid4()),
agent_id=agent.id,
user_id=user.id,
canvas_type="line_chart",
action="create",
_session=db_session
)
db_session.add(line_chart_audit)
# Create bar chart canvas audit
bar_chart_audit = CanvasAuditFactory(
canvas_id=str(uuid.uuid4()),
agent_id=agent.id,
user_id=user.id,
canvas_type="bar_chart",
action="create",
_session=db_session
)
db_session.add(bar_chart_audit)
# Create pie chart canvas audit
pie_chart_audit = CanvasAuditFactory(
canvas_id=str(uuid.uuid4()),
agent_id=agent.id,
user_id=user.id,
canvas_type="pie_chart",
action="create",
_session=db_session
)
db_session.add(pie_chart_audit)
# Create markdown canvas audit
markdown_audit = CanvasAuditFactory(
canvas_id=str(uuid.uuid4()),
agent_id=agent.id,
user_id=user.id,
canvas_type="markdown",
action="create",
_session=db_session
)
db_session.add(markdown_audit)
db_session.commit()
# Verify all canvases created
all_audits = db_session.query(CanvasAudit).filter(
CanvasAudit.agent_id == agent.id
).all()
assert len(all_audits) == 4, "Should have 4 canvas audit entries"
canvas_types = [audit.canvas_type for audit in all_audits]
assert "line_chart" in canvas_types, "Should include line chart"
assert "bar_chart" in canvas_types, "Should include bar chart"
assert "pie_chart" in canvas_types, "Should include pie chart"
assert "markdown" in canvas_types, "Should include markdown"
def test_chart_rendering_accuracy(self, db_session: Session):
"""
Test chart data rendering accuracy.
Scenario:
- Create canvas with test data
- Verify data stored correctly
"""
user = UserFactory(role=UserRole.MEMBER.value, _session=db_session)
db_session.add(user)
db_session.commit()
agent = AgentFactory(
status=AgentStatus.AUTONOMOUS.value,
confidence_score=0.95,
_session=db_session
)
db_session.add(agent)
db_session.commit()
db_session.refresh(agent)
# Create canvas with precise data
test_data = {
"type": "line",
"data": {
"labels": ["Jan", "Feb", "Mar"],
"datasets": [{
"label": "Sales",
"data": [100, 150, 200]
}]
}
}
canvas_audit = CanvasAuditFactory(
canvas_id=str(uuid.uuid4()),
agent_id=agent.id,
user_id=user.id,
canvas_type="line_chart",
component_type="chart",
component_name="line_chart",
action="create",
metadata={"canvas_config": test_data},
_session=db_session
)
db_session.add(canvas_audit)
db_session.commit()
db_session.refresh(canvas_audit)
# Verify data stored correctly in metadata
assert canvas_audit.metadata["canvas_config"]["data"]["labels"] == ["Jan", "Feb", "Mar"]
assert canvas_audit.metadata["canvas_config"]["data"]["datasets"][0]["data"] == [100, 150, 200]
def test_form_data_validation_and_submission(self, db_session: Session):
"""
Test form data validation and submission.
Scenario:
- Create canvas with form fields
- Submit form data
- Verify data stored in database
"""
user = UserFactory(role=UserRole.MEMBER.value, _session=db_session)
db_session.add(user)
db_session.commit()
agent = AgentFactory(
status=AgentStatus.AUTONOMOUS.value,
confidence_score=0.95,
_session=db_session
)
db_session.add(agent)
db_session.commit()
db_session.refresh(agent)
# Create form canvas
form_config = {
"type": "form",
"fields": [
{"name": "email", "type": "email", "required": True},
{"name": "name", "type": "text", "required": True}
]
}
canvas_audit = CanvasAuditFactory(
canvas_id=str(uuid.uuid4()),
agent_id=agent.id,
user_id=user.id,
canvas_type="form",
component_type="form",
component_name="form",
action="create",
metadata={"form_config": form_config},
_session=db_session
)
db_session.add(canvas_audit)
db_session.commit()
# Submit form data
submission_audit = CanvasAuditFactory(
canvas_id=canvas_audit.canvas_id,
agent_id=agent.id,
user_id=user.id,
canvas_type="form",
component_type="form",
component_name="form",
action="submit",
metadata={
"form_data": {
"email": "test@example.com",
"name": "Test User"
}
},
_session=db_session
)
db_session.add(submission_audit)
db_session.commit()
# Verify submission recorded
submissions = db_session.query(CanvasAudit).filter(
CanvasAudit.canvas_id == canvas_audit.canvas_id,
CanvasAudit.action == "submit"
).all()
assert len(submissions) == 1, "Should have 1 submission"
assert submissions[0].metadata["form_data"]["email"] == "test@example.com"
def test_governance_enforcement_on_canvas(self, db_session: Session):
"""
Test governance enforcement on canvas actions.
Scenario:
- Create STUDENT agent
- Verify governance blocks certain actions
- Create AUTONOMOUS agent
- Verify actions allowed
"""
user = UserFactory(role=UserRole.MEMBER.value, _session=db_session)
db_session.add(user)
db_session.commit()
# Create STUDENT agent
student_agent = AgentFactory(
status=AgentStatus.STUDENT.value,
confidence_score=0.3,
_session=db_session
)
db_session.add(student_agent)
db_session.commit()
db_session.refresh(student_agent)
# Create governance service
governance_service = AgentGovernanceService(db_session)
# STUDENT agent governance check for canvas action
result = governance_service.can_perform_action(
agent_id=student_agent.id,
action_type="canvas_form"
)
# STUDENT should be blocked or require approval
assert "allowed" in result, "Should return allowed field"
# Create AUTONOMOUS agent
autonomous_agent = AgentFactory(
status=AgentStatus.AUTONOMOUS.value,
confidence_score=0.95,
_session=db_session
)
db_session.add(autonomous_agent)
db_session.commit()
db_session.refresh(autonomous_agent)
# AUTONOMOUS agent should be allowed
result = governance_service.can_perform_action(
agent_id=autonomous_agent.id,
action_type="canvas_form"
)
assert result["allowed"] is True, "AUTONOMOUS agent should be allowed"
# Verify AUTONOMOUS agent can create canvas audit
canvas_audit = CanvasAuditFactory(
canvas_id=str(uuid.uuid4()),
agent_id=autonomous_agent.id,
user_id=user.id,
canvas_type="form",
action="create",
_session=db_session
)
db_session.add(canvas_audit)
db_session.commit()
# Verify audit created for AUTONOMOUS agent
autonomous_audits = db_session.query(CanvasAudit).filter(
CanvasAudit.agent_id == autonomous_agent.id
).all()
assert len(autonomous_audits) == 1, "AUTONOMOUS agent should have canvas audit"
def test_websocket_canvas_updates(self, db_session: Session):
"""
Test WebSocket canvas state updates.
Scenario:
- Create canvas
- Simulate WebSocket update
- Verify canvas state updated
"""
user = UserFactory(role=UserRole.MEMBER.value, _session=db_session)
db_session.add(user)
db_session.commit()
agent = AgentFactory(
status=AgentStatus.AUTONOMOUS.value,
confidence_score=0.95,
_session=db_session
)
db_session.add(agent)
db_session.commit()
db_session.refresh(agent)
canvas_id = str(uuid.uuid4())
# Create initial canvas
create_audit = CanvasAuditFactory(
canvas_id=canvas_id,
agent_id=agent.id,
user_id=user.id,
canvas_type="line_chart",
action="create",
_session=db_session
)
db_session.add(create_audit)
db_session.commit()
# Simulate WebSocket update
update_audit = CanvasAuditFactory(
canvas_id=canvas_id,
agent_id=agent.id,
user_id=user.id,
canvas_type="line_chart",
action="update",
_session=db_session
)
db_session.add(update_audit)
db_session.commit()
# Verify both states recorded
canvas_actions = db_session.query(CanvasAudit).filter(
CanvasAudit.canvas_id == canvas_id
).order_by(CanvasAudit.created_at).all()
assert len(canvas_actions) == 2, "Should have create and update actions"
def test_canvas_state_persistence(self, db_session: Session):
"""
Test canvas state persistence to database.
Scenario:
- Create canvas with initial state
- Update canvas state
- Verify state persisted to database
"""
user = UserFactory(role=UserRole.MEMBER.value, _session=db_session)
db_session.add(user)
db_session.commit()
agent = AgentFactory(
status=AgentStatus.AUTONOMOUS.value,
confidence_score=0.95,
_session=db_session
)
db_session.add(agent)
db_session.commit()
db_session.refresh(agent)
canvas_id = str(uuid.uuid4())
# Create canvas with initial state
canvas_audit = CanvasAuditFactory(
canvas_id=canvas_id,
agent_id=agent.id,
user_id=user.id,
canvas_type="line_chart",
action="create",
_session=db_session
)
db_session.add(canvas_audit)
db_session.commit()
db_session.refresh(canvas_audit)
# Retrieve latest state
latest_audit = db_session.query(CanvasAudit).filter(
CanvasAudit.canvas_id == canvas_id
).order_by(CanvasAudit.created_at.desc()).first()
assert latest_audit is not None, "Should retrieve canvas from database"
# ============================================================================
# Task 4: Graduation Promotion Flow Integration Tests
# ============================================================================
@pytest.mark.integration
class TestGraduationPromotionFlow:
"""
End-to-end tests for graduation promotion flow:
Graduation criteria → Constitutional check → Promotion execution → Maturity update
"""
def test_graduation_criteria_calculation(self, db_session: Session):
"""
Test graduation criteria calculation.
Scenario:
- Create agent with 10 episodes
- Set intervention rate to 40%
- Set constitutional score to 0.70
- Calculate readiness
"""
user = UserFactory(_session=db_session)
db_session.add(user)
db_session.commit()
agent = AgentFactory(
status=AgentStatus.STUDENT.value,
confidence_score=0.4,
_session=db_session
)
# Store graduation criteria in configuration JSON
agent.configuration = {
"episode_count": 10, # Meets STUDENT → INTERN threshold
"intervention_rate": 0.40, # Below 50% threshold
"constitutional_score": 0.70, # Meets 0.70 threshold
}
db_session.add(agent)
db_session.commit()
db_session.refresh(agent)
# Verify agent meets STUDENT → INTERN criteria
assert agent.configuration["episode_count"] >= 10, "Should have 10 episodes"
assert agent.configuration["intervention_rate"] <= 0.50, "Intervention rate should be <= 50%"
assert agent.configuration["constitutional_score"] >= 0.70, "Constitutional score should be >= 0.70"
def test_constitutional_compliance_validation(self, db_session: Session):
"""
Test constitutional compliance validation.
Scenario:
- Create agent meeting episode/intervention criteria
- Set constitutional score below threshold (0.65)
- Verify promotion blocked
- Set constitutional score above threshold (0.75)
- Verify compliance check passes
"""
user = UserFactory(_session=db_session)
db_session.add(user)
db_session.commit()
# Agent with low constitutional score
agent_low = AgentFactory(
status=AgentStatus.STUDENT.value,
_session=db_session
)
agent_low.configuration = {
"episode_count": 10,
"intervention_rate": 0.40,
"constitutional_score": 0.65, # Below 0.70 threshold
}
db_session.add(agent_low)
db_session.commit()
db_session.refresh(agent_low)
# Verify low constitutional score blocks promotion
assert agent_low.configuration["constitutional_score"] < 0.70, "Constitutional score below threshold"
# Agent with high constitutional score
agent_high = AgentFactory(
status=AgentStatus.STUDENT.value,
_session=db_session
)
agent_high.configuration = {
"episode_count": 10,
"intervention_rate": 0.40,
"constitutional_score": 0.75, # Above 0.70 threshold
}
db_session.add(agent_high)
db_session.commit()
db_session.refresh(agent_high)
# Verify high constitutional score allows promotion
assert agent_high.configuration["constitutional_score"] >= 0.70, "Constitutional score above threshold"
def test_end_to_end_graduation_flow(self, db_session: Session):
"""
Test complete graduation flow across all levels.
Scenario:
- Create qualified STUDENT agent
- Execute full promotion process
- Verify agent.status transitions to INTERN
- Repeat for INTERN → SUPERVISED
- Repeat for SUPERVISED → AUTONOMOUS
"""
user = UserFactory(_session=db_session)
db_session.add(user)
db_session.commit()
# STUDENT → INTERN
agent = AgentFactory(
status=AgentStatus.STUDENT.value,
_session=db_session
)
agent.configuration = {
"episode_count": 10,
"intervention_rate": 0.40,
"constitutional_score": 0.75,
}
db_session.add(agent)
db_session.commit()
db_session.refresh(agent)
# Promote to INTERN
agent.status = AgentStatus.INTERN.value
agent.confidence_score = 0.6
agent.configuration["promoted_at"] = datetime.utcnow().isoformat()
db_session.commit()
db_session.refresh(agent)
# Verify STUDENT → INTERN transition
assert agent.status == AgentStatus.INTERN.value, "Should promote to INTERN"
# INTERN → SUPERVISED
agent.configuration["episode_count"] = 25
agent.configuration["intervention_rate"] = 0.15
agent.configuration["constitutional_score"] = 0.85
agent.status = AgentStatus.SUPERVISED.value
agent.confidence_score = 0.8
agent.configuration["promoted_at"] = datetime.utcnow().isoformat()
db_session.commit()
db_session.refresh(agent)
# Verify INTERN → SUPERVISED transition
assert agent.status == AgentStatus.SUPERVISED.value, "Should promote to SUPERVISED"
# SUPERVISED → AUTONOMOUS
agent.configuration["episode_count"] = 50
agent.configuration["intervention_rate"] = 0.0
agent.configuration["constitutional_score"] = 0.95
agent.status = AgentStatus.AUTONOMOUS.value
agent.confidence_score = 0.95
agent.configuration["promoted_at"] = datetime.utcnow().isoformat()
db_session.commit()
db_session.refresh(agent)
# Verify SUPERVISED → AUTONOMOUS transition
assert agent.status == AgentStatus.AUTONOMOUS.value, "Should promote to AUTONOMOUS"
def test_readiness_score_calculation(self, db_session: Session):
"""
Test readiness score calculation formula.
Scenario:
- Test 40% episodes, 30% interventions, 30% constitutional split
- Verify formula: readiness = 0.4*episode_score + 0.3*intervention_score + 0.3*constitutional_score
"""
# Test case 1: Perfect scores
episode_score = 1.0 # All episodes completed
intervention_score = 1.0 # Zero interventions (ideal)
constitutional_score = 1.0 # Perfect constitutional score
readiness = 0.4 * episode_score + 0.3 * intervention_score + 0.3 * constitutional_score
assert readiness == 1.0, "Perfect scores should give readiness = 1.0"
# Test case 2: Boundary threshold values
agent = AgentFactory(
status=AgentStatus.STUDENT.value,
_session=db_session
)
agent.configuration = {
"episode_count": 10, # Exactly at threshold
"intervention_rate": 0.50, # Exactly at threshold
"constitutional_score": 0.70, # Exactly at threshold
}
db_session.add(agent)
db_session.commit()
# Verify agent is at threshold boundaries
assert agent.configuration["episode_count"] == 10, "Episode count at threshold"
assert agent.configuration["intervention_rate"] == 0.50, "Intervention rate at threshold"
assert agent.configuration["constitutional_score"] == 0.70, "Constitutional score at threshold"
def test_promotion_rejection(self, db_session: Session):
"""
Test promotion rejection when criteria not met.
Scenario:
- Create agent with insufficient episodes (5)
- Try to promote to INTERN
- Verify promotion rejected
- Verify agent status unchanged (STUDENT)
"""
user = UserFactory(_session=db_session)
db_session.add(user)
db_session.commit()
# Agent with insufficient episodes
agent1 = AgentFactory(
status=AgentStatus.STUDENT.value,
_session=db_session
)
agent1.configuration = {
"episode_count": 5, # Below 10 episode threshold
"intervention_rate": 0.30,
"constitutional_score": 0.75,
}
db_session.add(agent1)
db_session.commit()
db_session.refresh(agent1)
# Verify agent cannot be promoted (insufficient episodes)
assert agent1.configuration["episode_count"] < 10, "Episode count below threshold"
assert agent1.status == AgentStatus.STUDENT.value, "Status should remain STUDENT"
# Agent with high intervention rate
agent2 = AgentFactory(
status=AgentStatus.STUDENT.value,
_session=db_session
)
agent2.configuration = {
"episode_count": 10,
"intervention_rate": 0.60, # Above 50% threshold
"constitutional_score": 0.75,
}
db_session.add(agent2)
db_session.commit()
db_session.refresh(agent2)
# Verify agent cannot be promoted (high intervention rate)
assert agent2.configuration["intervention_rate"] > 0.50, "Intervention rate above threshold"
assert agent2.status == AgentStatus.STUDENT.value, "Status should remain STUDENT"
def test_maturity_update_persistence(self, db_session: Session):
"""
Test maturity state persistence to database.
Scenario:
- Promote agent
- Verify database updated immediately
- Query agent from new session
- Verify status persisted
"""
user = UserFactory(_session=db_session)
db_session.add(user)
db_session.commit()
agent = AgentFactory(
status=AgentStatus.STUDENT.value,
confidence_score=0.4,
_session=db_session
)
agent.configuration = {
"episode_count": 10,
"intervention_rate": 0.40,
"constitutional_score": 0.75,
}
db_session.add(agent)
db_session.commit()
db_session.refresh(agent)
# Promote agent
agent.status = AgentStatus.INTERN.value
agent.confidence_score = 0.6
agent.configuration["promoted_at"] = datetime.utcnow().isoformat()
db_session.commit()
db_session.refresh(agent)
# Query agent again (simulates new session read)
persisted_agent = db_session.query(AgentRegistry).filter(
AgentRegistry.id == agent.id
).first()
# Verify status persisted
assert persisted_agent is not None, "Agent should be retrievable"
assert persisted_agent.status == AgentStatus.INTERN.value, "Status should persist"
assert persisted_agent.confidence_score == 0.6, "Confidence score should persist"
# ============================================================================
# Task 5: Cross-Cutting Integration Tests
# ============================================================================
@pytest.mark.integration
class TestCrossCuttingConcerns:
"""
Tests covering shared concerns across all critical paths:
- Governance enforcement
- Data integrity
- Audit trail completeness
- Error recovery
- Concurrency
"""
def test_governance_bypass_prevention(self, db_session: Session):
"""
Test governance enforcement at all maturity levels.
Scenario:
- Test governance check at all maturity levels
- Verify STUDENT cannot perform restricted actions
- Verify AUTONOMOUS can perform all actions
"""
user = UserFactory(role=UserRole.MEMBER.value, _session=db_session)
db_session.add(user)
db_session.commit()
governance_service = AgentGovernanceService(db_session)
# Create STUDENT agent
student = AgentFactory(
status=AgentStatus.STUDENT.value,
confidence_score=0.3,
_session=db_session
)
db_session.add(student)
# Create INTERN agent
intern = AgentFactory(
status=AgentStatus.INTERN.value,
confidence_score=0.6,
_session=db_session
)
db_session.add(intern)
# Create AUTONOMOUS agent
autonomous = AgentFactory(
status=AgentStatus.AUTONOMOUS.value,
confidence_score=0.95,
_session=db_session
)
db_session.add(autonomous)
db_session.commit()
# Test execute action
# STUDENT: Blocked
student_result = governance_service.can_perform_action(
agent_id=student.id,
action_type="execute"
)
assert student_result["allowed"] is False, "STUDENT should be blocked from execute"
# AUTONOMOUS: Allowed
auto_result = governance_service.can_perform_action(
agent_id=autonomous.id,
action_type="execute"
)
assert auto_result["allowed"] is True, "AUTONOMOUS should be allowed for execute"
def test_data_integrity_across_paths(self, db_session: Session):
"""
Test data consistency across critical paths.
Scenario:
- Create data in agent execution flow
- Access from episode flow (same agent)
- Verify data consistency
- Verify foreign keys maintained
"""
user = UserFactory(_session=db_session)
db_session.add(user)
db_session.commit()
agent = AgentFactory(_session=db_session)
db_session.add(agent)
db_session.commit()
db_session.refresh(agent)
# Create execution (agent execution flow)
execution = AgentExecutionFactory(
agent_id=agent.id,
status="completed",
_session=db_session
)
db_session.add(execution)
# Create episode (episode creation flow) for same agent
episode = EpisodeFactory(
agent_id=agent.id,
user_id=user.id,
_session=db_session
)
db_session.add(episode)
# Create canvas (canvas presentation flow) for same agent
canvas_audit = CanvasAuditFactory(
canvas_id=str(uuid.uuid4()),
agent_id=agent.id,
user_id=user.id,
canvas_type="line_chart",
action="create",
_session=db_session
)
db_session.add(canvas_audit)
db_session.commit()
# Verify data consistency - all records link to same agent
assert execution.agent_id == agent.id, "Execution should link to agent"
assert episode.agent_id == agent.id, "Episode should link to agent"
assert canvas_audit.agent_id == agent.id, "Canvas audit should link to agent"
def test_audit_trail_completeness(self, db_session: Session):
"""
Test audit trail completeness across all paths.
Scenario:
- Execute agent request
- Create episode
- Create canvas
- Verify all actions logged in audit trail
- Verify timestamps sequential
"""
user = UserFactory(_session=db_session)
db_session.add(user)
db_session.commit()
agent = AgentFactory(_session=db_session)
db_session.add(agent)
db_session.commit()
db_session.refresh(agent)
timestamps = []
# Execute agent request
execution = AgentExecutionFactory(
agent_id=agent.id,
status="completed",
_session=db_session
)
db_session.add(execution)
db_session.commit()
timestamps.append(("execution", execution.started_at))
# Create episode
episode = EpisodeFactory(
agent_id=agent.id,
user_id=user.id,
_session=db_session
)
db_session.add(episode)
db_session.commit()
timestamps.append(("episode", episode.created_at))
# Create canvas
canvas_audit = CanvasAuditFactory(
canvas_id=str(uuid.uuid4()),
agent_id=agent.id,
user_id=user.id,
canvas_type="line_chart",
action="create",
_session=db_session
)
db_session.add(canvas_audit)
db_session.commit()
timestamps.append(("canvas", canvas_audit.created_at))
# Verify all actions logged
assert len(timestamps) == 3, "Should have 3 audit entries"
def test_error_recovery_across_paths(self, db_session: Session):
"""
Test error recovery and isolation between paths.
Scenario:
- Trigger error in agent execution
- Verify cleanup occurs
- Continue to episode creation
- Verify no corruption from previous error
"""
user = UserFactory(_session=db_session)
db_session.add(user)
db_session.commit()
agent = AgentFactory(_session=db_session)
db_session.add(agent)
db_session.commit()
db_session.refresh(agent)
# Trigger error in agent execution (failed execution)
failed_execution = AgentExecutionFactory(
agent_id=agent.id,
status="failed",
error_message="Simulated error",
_session=db_session
)
db_session.add(failed_execution)
db_session.commit()
# Verify error logged
assert failed_execution.status == "failed", "Failed execution should be logged"
assert failed_execution.error_message is not None, "Error message should be present"
# Continue to episode creation (should work independently)
episode = EpisodeFactory(
agent_id=agent.id,
user_id=user.id,
_session=db_session
)
db_session.add(episode)
db_session.commit()
# Verify episode created successfully despite previous error
assert episode.id is not None, "Episode should be created"
assert episode.agent_id == agent.id, "Episode should link to agent"
def test_concurrency_across_paths(self, db_session: Session):
"""
Test concurrent operations across paths.
Scenario:
- Execute agent request while creating episode
- Verify no race conditions
- Verify both operations complete
"""
user = UserFactory(_session=db_session)
db_session.add(user)
db_session.commit()
agent = AgentFactory(_session=db_session)
db_session.add(agent)
db_session.commit()
db_session.refresh(agent)
# Simulate concurrent operations by creating records
# (SQLite has limited concurrency, but this tests the transaction logic)
# Operation 1: Execute agent
execution = AgentExecutionFactory(
agent_id=agent.id,
status="completed",
_session=db_session
)
db_session.add(execution)
# Operation 2: Create episode (simulating concurrent access)
episode = EpisodeFactory(
agent_id=agent.id,
user_id=user.id,
_session=db_session
)
db_session.add(episode)
# Operation 3: Create canvas
canvas_audit = CanvasAuditFactory(
canvas_id=str(uuid.uuid4()),
agent_id=agent.id,
user_id=user.id,
canvas_type="line_chart",
action="create",
_session=db_session
)
db_session.add(canvas_audit)
# Commit all operations
db_session.commit()
# Verify both operations complete without race conditions
assert execution.id is not None, "Execution should complete"
assert episode.id is not None, "Episode should complete"
assert canvas_audit.id is not None, "Canvas should complete"
# Verify no duplicate records
executions = db_session.query(AgentExecution).filter(
AgentExecution.agent_id == agent.id
).all()
assert len(executions) == 1, "Should have exactly 1 execution"
episodes = db_session.query(Episode).filter(
Episode.agent_id == agent.id
).all()
assert len(episodes) == 1, "Should have exactly 1 episode"
canvases = db_session.query(CanvasAudit).filter(
CanvasAudit.agent_id == agent.id
).all()
assert len(canvases) == 1, "Should have exactly 1 canvas"
|