Skip to content

Commit 27520e2

Browse files
Inkman007Inkman007
authored
Merge pull request #669 from WhitesmokeX/issue/435-578-idempotency-recovery
Add idempotency & recovery integration for repeatable pipeline jobs (Issues #435, #578)
2 parents a14481a + 7bb8be4 commit 27520e2

10 files changed

Lines changed: 3250 additions & 0 deletions

‎data/test_criticality.json‎

Lines changed: 166 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,166 @@
1+
{
2+
"version": 1,
3+
"description": "LedgerLens test criticality taxonomy. Maps test file/node-id patterns to CRITICAL, HIGH, or LOW. CRITICAL and HIGH failures block the release gate; LOW is informational only. Patterns use fnmatch glob rules against the pytest node ID.",
4+
"entries": [
5+
{
6+
"pattern": "tests/test_benford.py",
7+
"criticality": "CRITICAL",
8+
"reason": "Core Benford scoring path — chi-square, Z-score, MAD. Any failure here means risk scores are unreliable."
9+
},
10+
{
11+
"pattern": "tests/test_benford_ci.py",
12+
"criticality": "CRITICAL",
13+
"reason": "Benford conformity interval tests — regression here breaks fraud detection thresholds."
14+
},
15+
{
16+
"pattern": "tests/test_feature_engineering.py",
17+
"criticality": "CRITICAL",
18+
"reason": "Feature engineering produces the model input matrix. Silent errors here corrupt every downstream score."
19+
},
20+
{
21+
"pattern": "tests/test_model_inference.py",
22+
"criticality": "CRITICAL",
23+
"reason": "BFT ensemble voting and RiskScorer — the final scoring output consumed by the API and Soroban contract."
24+
},
25+
{
26+
"pattern": "tests/test_model_training.py",
27+
"criticality": "CRITICAL",
28+
"reason": "Model training pipeline — ensures trained artifacts are valid and reproducible."
29+
},
30+
{
31+
"pattern": "tests/test_persistence.py",
32+
"criticality": "CRITICAL",
33+
"reason": "SQLAlchemy persistence layer — failures here mean scored wallets are not stored correctly."
34+
},
35+
{
36+
"pattern": "tests/test_dry_run.py",
37+
"criticality": "CRITICAL",
38+
"reason": "Dry-run safety contract — --dry-run must never write to DB or on-chain."
39+
},
40+
{
41+
"pattern": "tests/test_config.py",
42+
"criticality": "CRITICAL",
43+
"reason": "Config validation — bad config silently corrupts the pipeline at runtime."
44+
},
45+
{
46+
"pattern": "tests/test_inference_shap.py",
47+
"criticality": "HIGH",
48+
"reason": "SHAP explainability — required for auditable scores but not on the direct scoring hot path."
49+
},
50+
{
51+
"pattern": "tests/test_wallet_graph.py",
52+
"criticality": "HIGH",
53+
"reason": "Wallet graph features — funding-graph similarity and ring detection."
54+
},
55+
{
56+
"pattern": "tests/test_wallet_graph_advanced.py",
57+
"criticality": "HIGH",
58+
"reason": "Advanced ring detection edge cases."
59+
},
60+
{
61+
"pattern": "tests/test_drift_monitor.py",
62+
"criticality": "HIGH",
63+
"reason": "Feature drift detection — PSI-based retraining triggers."
64+
},
65+
{
66+
"pattern": "tests/test_retrain_trigger.py",
67+
"criticality": "HIGH",
68+
"reason": "Automated retraining pipeline — promotion gate and exit codes."
69+
},
70+
{
71+
"pattern": "tests/test_cross_venue_features.py",
72+
"criticality": "HIGH",
73+
"reason": "Cross-venue coordination features — SDEX + AMM wash-trade signal."
74+
},
75+
{
76+
"pattern": "tests/test_cross_asset_features.py",
77+
"criticality": "HIGH",
78+
"reason": "Cross-asset coordination features."
79+
},
80+
{
81+
"pattern": "tests/test_alert_dispatcher.py",
82+
"criticality": "HIGH",
83+
"reason": "Alert delivery — dedup and dispatch correctness."
84+
},
85+
{
86+
"pattern": "tests/test_contract_client.py",
87+
"criticality": "HIGH",
88+
"reason": "Soroban contract client — on-chain score submission."
89+
},
90+
{
91+
"pattern": "tests/test_adversarial.py",
92+
"criticality": "HIGH",
93+
"reason": "Adversarial robustness — FGSM/PGD evasion attack coverage."
94+
},
95+
{
96+
"pattern": "tests/test_dp_shap.py",
97+
"criticality": "HIGH",
98+
"reason": "Differential-privacy SHAP — privacy budget enforcement."
99+
},
100+
{
101+
"pattern": "tests/test_ensemble_calibrator.py",
102+
"criticality": "HIGH",
103+
"reason": "Ensemble calibration — NSGA-II Pareto front correctness."
104+
},
105+
{
106+
"pattern": "tests/test_conformal.py",
107+
"criticality": "HIGH",
108+
"reason": "Conformal prediction intervals — uncertainty quantification."
109+
},
110+
{
111+
"pattern": "tests/test_orderbook.py",
112+
"criticality": "HIGH",
113+
"reason": "Order-book event ingestion — cancellation rate feature."
114+
},
115+
{
116+
"pattern": "tests/test_streaming_pipeline.py",
117+
"criticality": "HIGH",
118+
"reason": "Streaming pipeline backend routing."
119+
},
120+
{
121+
"pattern": "tests/test_idempotency.py",
122+
"criticality": "CRITICAL",
123+
"reason": "Idempotency guarantees — double-writes corrupt the risk-score store."
124+
},
125+
{
126+
"pattern": "tests/test_recovery.py",
127+
"criticality": "CRITICAL",
128+
"reason": "Partial-run recovery — failures here mean crashed pipelines corrupt data on restart."
129+
},
130+
{
131+
"pattern": "tests/test_release_gate.py",
132+
"criticality": "HIGH",
133+
"reason": "Release gate itself — meta-test ensuring the gate logic is correct."
134+
},
135+
{
136+
"pattern": "tests/test_quality_metrics.py",
137+
"criticality": "HIGH",
138+
"reason": "Engineering quality scorecard — ensures metric collection is accurate."
139+
},
140+
{
141+
"pattern": "tests/test_backtest.py",
142+
"criticality": "LOW",
143+
"reason": "Historical backtesting — slow and data-dependent; informational only."
144+
},
145+
{
146+
"pattern": "tests/test_federated.py",
147+
"criticality": "LOW",
148+
"reason": "Federated learning — experimental; not on the production scoring path."
149+
},
150+
{
151+
"pattern": "tests/test_async_federated.py",
152+
"criticality": "LOW",
153+
"reason": "Async federated learning — experimental."
154+
},
155+
{
156+
"pattern": "tests/test_gnn_encoder.py",
157+
"criticality": "LOW",
158+
"reason": "GNN encoder — embedding generation for future use."
159+
},
160+
{
161+
"pattern": "tests/integration/*",
162+
"criticality": "LOW",
163+
"reason": "Live integration tests — require external services; not run in standard CI."
164+
}
165+
]
166+
}

‎pipeline/__init__.py‎

Lines changed: 1 addition & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1 @@
1+
"""Pipeline utilities package: idempotency guarantees and partial-run recovery."""

0 commit comments

Comments
 (0)