diff --git a/package.json b/package.json index 485f7e3..31303c1 100644 --- a/package.json +++ b/package.json @@ -16,6 +16,9 @@ "ops:package": "python tools/refresh_trading_calendar.py && python tools/prepare_upload_zip.py --validation-mode release --profile", "prepare-upload-zip": "python tools/refresh_trading_calendar.py && python tools/prepare_upload_zip.py", "ops:audit": "python tools/harness_coverage_auditor.py", + "build-realized-performance": "python tools/build_realized_performance_v1.py", + "validate-completion-harness": "python tools/validate_completion_harness_instructions_v1.py", + "validate-realized-performance": "python tools/validate_realized_performance_v1.py", "validate-gas-recovery": "python tools/validate_gas_orchestration_recovery_v1.py", "ops:clean": "python tools/clean_temp_artifacts_v1.py", "ops:dev": "node core_satellite_collector.js", @@ -26,14 +29,14 @@ "render-report-json": "python tools/render_operational_report.py --json GatherTradingData.json --output Temp/operational_report.md --report-json-output Temp/operational_report.json" }, "dependencies": { - "cheerio": "latest", + "cheerio": "1.2.0", "googleapis": "^171.4.0", - "iconv-lite": "latest", - "yahoo-finance2": "latest" + "iconv-lite": "0.7.2", + "yahoo-finance2": "3.15.3" }, "optionalDependencies": { - "adm-zip": "latest", - "fast-xml-parser": "latest" + "adm-zip": "0.5.17", + "fast-xml-parser": "5.8.0" }, "devDependencies": { "xlsx": "^0.18.5" diff --git a/spec/41_release_dag.yaml b/spec/41_release_dag.yaml index 1496015..c7bce28 100644 --- a/spec/41_release_dag.yaml +++ b/spec/41_release_dag.yaml @@ -1,5 +1,5 @@ schema_version: release_dag.v3 -step_count: 87 +step_count: 90 goal: Linearize package.json scripts into a validated DAG execution graph. execution_order: # 토폴로지 정렬 기준 병렬 실행 wave (의존성 없는 노드들을 동시에 실행 가능) @@ -21,6 +21,7 @@ execution_order: - validate_calibration - validate_cash_ledger - validate_change_requests + - validate_completion_harness_instructions - validate_factor_lifecycle - validate_factor_lifecycle_completeness - validate_field_dict @@ -41,6 +42,7 @@ execution_order: - build_etf_representative_monitor - build_factor_shadow_eligibility - build_formula_outputs + - build_realized_performance - build_missing_formula_bridge - build_ratchet_trailing_general - build_rebalance_sheet @@ -54,6 +56,7 @@ execution_order: - validate_engine_health_card - validate_module_io_coverage - validate_no_replay_live_mix + - validate_realized_performance - validate_rule_lifecycle - validate_schema_model wave_2: @@ -148,6 +151,18 @@ dag: artifact_policy: "keep" note: "DATA_GATED 항목별 실측 진척도 — non-blocking diagnostic" + build_realized_performance: + id: build_realized_performance + command: ["python", "tools/build_realized_performance_v1.py"] + inputs: ["tools/build_realized_performance_v1.py", "GatherTradingData.json", "Temp/proposal_evaluation_history.json"] + outputs: ["Temp/realized_performance_v1.json"] + depends_on: ["build_data_gated_progress"] + timeout_sec: 30 + cache_key: "build_realized_performance_v1" + strict: false + artifact_policy: "keep" + note: "WBS-4.1 realized performance replay summary — non-blocking diagnostic" + build_factor_shadow_eligibility: id: build_factor_shadow_eligibility command: ["python", "tools/build_factor_shadow_eligibility_v1.py"] @@ -566,6 +581,18 @@ dag: strict: true artifact_policy: "keep" + validate_realized_performance: + id: validate_realized_performance + command: ["python", "tools/validate_realized_performance_v1.py"] + inputs: ["tools/validate_realized_performance_v1.py", "Temp/realized_performance_v1.json"] + outputs: ["Temp/validate_realized_performance_v1.json"] + depends_on: ["build_realized_performance"] + timeout_sec: 30 + cache_key: "validate_realized_performance_v1" + strict: true + artifact_policy: "keep" + note: "REALIZED_PERFORMANCE_V1 output schema and evidence validation" + validate_runtime_source_whitelist: id: validate_runtime_source_whitelist command: ["python", "tools/validate_runtime_source_whitelist_v1.py", "--manifest", "runtime/active_artifact_manifest.yaml", "--scan", "src", "gas_*.gs"] @@ -785,6 +812,34 @@ dag: strict: true artifact_policy: "keep" + validate_completion_harness_instructions: + id: validate_completion_harness_instructions + command: ["python", "tools/validate_completion_harness_instructions_v1.py"] + inputs: + [ + "tools/validate_completion_harness_instructions_v1.py", + "AGENTS.md", + "REPORT_GUIDE.md", + "docs/runbook.md", + "docs/ROADMAP_WBS.md", + "docs/doctrine.md", + "prompts/analysis_prompt.md", + "prompts/review_prompt.md", + "prompts/capture_parse_prompt.md", + "prompts/engine_audit_master_prompt_v2.md", + "prompts/engine_audit_master_prompt_v3.md", + "prompts/engine_audit_prompt.md", + "prompts/low_capability_report_renderer.md", + "prompts/report_renderer_prompt.md", + "prompts/weekly_operational_report_master_prompt_v1.md" + ] + outputs: ["Temp/completion_harness_instructions_v1.json"] + depends_on: [] + timeout_sec: 30 + cache_key: "validate_completion_harness_instructions_v1" + strict: true + artifact_policy: "keep" + validate_engine_health_card: id: validate_engine_health_card command: ["python", "tools/validate_engine_health_card_v1.py", "--json", "Temp/engine_health_card_v1.json"] @@ -1159,7 +1214,7 @@ dag: command: ["python", "tools/prepare_upload_zip.py", "--skip-validate", "--skip-convert", "--validation-mode", "package-only"] inputs: ["tools/prepare_upload_zip.py"] outputs: [] - depends_on: ["audit_entropy", "validate_specs", "validate_active_manifest", "validate_report_sync", "validate_report_numeric_consistency", "validate_field_dict", "validate_provenance", "validate_low_capability", "validate_golden_coverage", "validate_calibration", "validate_schema_model", "validate_gas_adapter", "validate_agents_shrink", "validate_no_replay_live_mix", "validate_runtime_source_whitelist", "validate_cash_ledger", "validate_factor_lifecycle", "validate_factor_lifecycle_completeness", "validate_metric_alias_collision", "validate_architecture_boundaries", "validate_module_io_coverage", "validate_artifact_chain_hash", "validate_artifact_sync", "validate_renderer_no_calc", "validate_packaged_refs", "validate_property_invariants", "validate_anti_late_entry", "validate_rule_lifecycle", "validate_change_requests", "validate_engine_health_card", "validate_llm_regression", "validate_llm_copy_only", "build_final_decision", "build_final_context", "build_provenance_ledger", "build_live_replay_separation", "build_late_chase_attribution", "build_profit_giveback_ratchet", "build_shadow_ledger", "build_operating_cadence_signal", "build_engine_health_card", "build_module_io_coverage", "build_artifact_chain_hash", "build_report", "build_bundle", "build_schema_models", "build_architecture_boundaries", "validate_decision_trace", "validate_factor_conflicts", "validate_no_lookahead", "validate_execution_sim", "validate_render_diff", "build_shadow_promotion", "validate_llm_determinism", "build_time_stop_forecast", "validate_live_activation", "build_rebalance_sheet"] + depends_on: ["audit_entropy", "validate_specs", "validate_active_manifest", "validate_report_sync", "validate_report_numeric_consistency", "validate_field_dict", "validate_provenance", "validate_low_capability", "validate_golden_coverage", "validate_calibration", "validate_schema_model", "validate_gas_adapter", "validate_agents_shrink", "validate_no_replay_live_mix", "validate_realized_performance", "validate_runtime_source_whitelist", "validate_cash_ledger", "validate_factor_lifecycle", "validate_factor_lifecycle_completeness", "validate_metric_alias_collision", "validate_architecture_boundaries", "validate_module_io_coverage", "validate_artifact_chain_hash", "validate_artifact_sync", "validate_renderer_no_calc", "validate_packaged_refs", "validate_property_invariants", "validate_anti_late_entry", "validate_rule_lifecycle", "validate_change_requests", "validate_completion_harness_instructions", "validate_engine_health_card", "validate_llm_regression", "validate_llm_copy_only", "build_final_decision", "build_final_context", "build_provenance_ledger", "build_live_replay_separation", "build_late_chase_attribution", "build_profit_giveback_ratchet", "build_shadow_ledger", "build_operating_cadence_signal", "build_engine_health_card", "build_module_io_coverage", "build_artifact_chain_hash", "build_report", "build_bundle", "build_schema_models", "build_architecture_boundaries", "validate_decision_trace", "validate_factor_conflicts", "validate_no_lookahead", "validate_execution_sim", "validate_render_diff", "build_shadow_promotion", "validate_llm_determinism", "build_time_stop_forecast", "validate_live_activation", "build_rebalance_sheet"] timeout_sec: 60 cache_key: "prepare_zip_v1" strict: true diff --git a/src/quant_engine/prepare_upload_zip.py b/src/quant_engine/prepare_upload_zip.py index ab23aee..8baacee 100644 --- a/src/quant_engine/prepare_upload_zip.py +++ b/src/quant_engine/prepare_upload_zip.py @@ -3,9 +3,12 @@ from __future__ import annotations import argparse import json import zipfile +from functools import lru_cache from datetime import datetime, timezone from pathlib import Path +import yaml + from tools.orchestration_harness_v1 import run_plan from src.quant_engine.pipeline_runtime_anomaly_lib_v1 import finalize_runtime_profile, runtime_profile_from_steps @@ -86,6 +89,8 @@ TEMP_KEEP_FILES = { "canonical_artifact_resolver_v1.json", "final_execution_decision_v2.json", "prediction_accuracy_harness_v2.json", + "realized_performance_v1.json", + "validate_realized_performance_v1.json", "single_truth_ledger_v2.json", "smart_cash_recovery_v7.json", "smart_cash_recovery_v9.json", @@ -107,6 +112,35 @@ UPLOAD_KEEP_DIRS_UPLOAD = { } +@lru_cache(maxsize=1) +def _active_manifest_refs() -> set[str]: + manifest_path = ROOT / "runtime" / "active_artifact_manifest.yaml" + if not manifest_path.exists(): + return set() + try: + manifest = yaml.safe_load(manifest_path.read_text(encoding="utf-8")) or {} + except Exception: + return set() + refs: set[str] = set() + if isinstance(manifest, dict): + canonical = manifest.get("canonical_source") + if isinstance(canonical, str) and canonical.strip(): + refs.add(Path(canonical).as_posix()) + aliases = manifest.get("active_aliases") + if isinstance(aliases, dict): + for val in aliases.values(): + if isinstance(val, str) and val.strip(): + refs.add(Path(val).as_posix()) + rows = manifest.get("manifest_rows") + if isinstance(rows, list): + for row in rows: + if isinstance(row, dict): + active_artifact = row.get("active_artifact") + if isinstance(active_artifact, str) and active_artifact.strip(): + refs.add(Path(active_artifact).as_posix()) + return refs + + def _load_json(path: Path) -> dict: if not path.exists(): return {} @@ -151,6 +185,8 @@ def should_include(path: Path, mode: str, include_xlsx: bool, include_backups: b return False if path.name == DEFAULT_OUTPUT.name: return False + if mode == "upload" and rel.as_posix() in _active_manifest_refs(): + return True if parts[0] == "Temp": if path.name in TEMP_EXCLUDED_FILES: return False diff --git a/tools/build_realized_performance_v1.py b/tools/build_realized_performance_v1.py index bc70869..d7db7df 100644 --- a/tools/build_realized_performance_v1.py +++ b/tools/build_realized_performance_v1.py @@ -300,7 +300,7 @@ def main() -> int: out_path.write_text(json.dumps(result, ensure_ascii=False, indent=2) + "\n", encoding="utf-8") s = result["summary"] print( - f"[{FORMULA_ID}] T20_replay_CAGR(est)={s.get('best_estimated_cagr_pct')}% " + f"[{FORMULA_ID}] T20_replay_CAGR(est)={s.get('best_estimated_cagr')}% " f"Sharpe(est)={s.get('best_estimated_sharpe')} " f"MDD_worst={s.get('worst_case_mdd_pct')}% " f"T5_win_rate={s.get('t5_win_rate_pct')}% -> {out_path}" diff --git a/tools/run_release_dag_v3.py b/tools/run_release_dag_v3.py index 657b78a..e586051 100644 --- a/tools/run_release_dag_v3.py +++ b/tools/run_release_dag_v3.py @@ -82,7 +82,7 @@ def main() -> int: def add_to_closure(nid): if nid not in closure: closure.add(nid) - for dep in nodes[nid].get("depends_on") or []: + for dep in sorted(nodes[nid].get("depends_on") or []): if dep in nodes: add_to_closure(dep) @@ -99,14 +99,14 @@ def main() -> int: raise ValueError(f"Cycle detected involving {nid}") if nid not in visited: temp.add(nid) - for dep in nodes[nid].get("depends_on") or []: + for dep in sorted(nodes[nid].get("depends_on") or []): if dep in closure: visit(dep) temp.remove(nid) visited.add(nid) order.append(nid) - for nid in closure: + for nid in sorted(closure): if nid not in visited: try: visit(nid) @@ -116,6 +116,8 @@ def main() -> int: steps_run = [] success = True + warning_failed = False + warning_failures: list[dict[str, object]] = [] LINEAGE_LOG.parent.mkdir(parents=True, exist_ok=True) @@ -180,6 +182,11 @@ def main() -> int: print(proc.stderr) if node.get("warn_only", False): print(f"Node {nid} is warn_only - continuing") + warning_failed = True + warning_failures.append({ + "node_id": nid, + "returncode": proc.returncode, + }) else: success = False if node.get("strict", True) or args.strict: @@ -191,17 +198,23 @@ def main() -> int: "formula_id": "RELEASE_DAG_RUN_V4", "mode": args.mode, "steps": steps_run, - "gate": "PASS" if success else "FAIL" + "warning_failures": warning_failures, + "gate": "PASS" if success and not warning_failed else "PASS_WITH_WARNINGS" if success else "FAIL" }, ensure_ascii=False, indent=2), encoding="utf-8") print(json.dumps({ "formula_id": "RELEASE_DAG_RUN_V4", "mode": args.mode, "step_count": len(steps_run), - "gate": "PASS" if success else "FAIL" + "warning_failure_count": len(warning_failures), + "gate": "PASS" if success and not warning_failed else "PASS_WITH_WARNINGS" if success else "FAIL" }, ensure_ascii=True, indent=2)) - return 0 if success else 1 + if not success: + return 1 + if args.strict and warning_failed: + return 1 + return 0 if __name__ == "__main__": diff --git a/tools/validate_completion_harness_instructions_v1.py b/tools/validate_completion_harness_instructions_v1.py new file mode 100644 index 0000000..ce67b45 --- /dev/null +++ b/tools/validate_completion_harness_instructions_v1.py @@ -0,0 +1,148 @@ +#!/usr/bin/env python3 +from __future__ import annotations + +import json +from pathlib import Path + + +ROOT = Path(__file__).resolve().parents[1] + + +def _read(rel: str) -> str: + path = ROOT / rel + if not path.exists(): + return "" + return path.read_text(encoding="utf-8", errors="replace").lower() + + +def _require_any(text: str, options: list[str]) -> bool: + return any(option.lower() in text for option in options) + + +def _require_all(text: str, groups: list[list[str]]) -> bool: + return all(_require_any(text, group) for group in groups) + + +def main() -> int: + files: dict[str, list[list[str]]] = { + "AGENTS.md": [ + ["0b. 기본 하네스 완료 조건"], + ["yaml"], + ["코드"], + ["데이터 실체"], + ["검증 증빙"], + ], + "docs/runbook.md": [ + ["complete only when", "완료"], + ["yaml"], + ["code"], + ["data artifact", "data artifacts", "data/artifact"], + ["validation evidence", "검증 증빙"], + ], + "docs/ROADMAP_WBS.md": [ + ["완료 조건"], + ["yaml"], + ["코드"], + ["데이터 실체"], + ["검증 증빙"], + ], + "REPORT_GUIDE.md": [ + ["completion harness"], + ["yaml"], + ["code"], + ["data artifact", "data artifacts", "data/artifact"], + ["validation evidence", "검증 증빙"], + ], + "docs/doctrine.md": [ + ["completion harness"], + ["yaml"], + ["code"], + ["data artifact", "data artifacts", "data/artifact"], + ["validation evidence", "검증 증빙"], + ], + "prompts/analysis_prompt.md": [ + ["default completion harness"], + ["yaml"], + ["코드"], + ["데이터 실체"], + ["검증 증빙"], + ], + "prompts/review_prompt.md": [ + ["default completion harness"], + ["yaml"], + ["code"], + ["data artifact", "data/artifact"], + ["validation evidence", "검증 증빙"], + ], + "prompts/capture_parse_prompt.md": [ + ["기본 완료 조건"], + ["yaml"], + ["코드"], + ["데이터 실체"], + ["검증 증빙"], + ], + "prompts/engine_audit_master_prompt_v2.md": [ + ["default completion harness"], + ["yaml"], + ["code"], + ["data artifact", "data/artifact"], + ["validation evidence", "검증 증빙"], + ], + "prompts/engine_audit_master_prompt_v3.md": [ + ["default completion harness"], + ["yaml"], + ["code"], + ["data artifact", "data/artifact"], + ["validation evidence", "검증 증빙"], + ], + "prompts/engine_audit_prompt.md": [ + ["yaml"], + ["code"], + ["data artifact", "data/artifact"], + ["validation evidence", "검증 증빙"], + ], + "prompts/low_capability_report_renderer.md": [ + ["default completion harness"], + ["yaml"], + ["code"], + ["data artifact", "data/artifact"], + ["validation evidence", "검증 증빙"], + ], + "prompts/report_renderer_prompt.md": [ + ["yaml"], + ["code"], + ["data artifact", "data/artifact"], + ["validation evidence", "검증 증빙"], + ], + "prompts/weekly_operational_report_master_prompt_v1.md": [ + ["default completion harness"], + ["yaml"], + ["code"], + ["data artifact", "data/artifact"], + ["validation evidence", "검증 증빙"], + ], + } + + missing: list[dict[str, object]] = [] + for rel, groups in files.items(): + text = _read(rel) + if not text: + missing.append({"file": rel, "reason": "missing_file"}) + continue + if not _require_all(text, groups): + missing.append({"file": rel, "reason": "missing_required_text", "required_groups": groups}) + + result = { + "formula_id": "COMPLETION_HARNESS_INSTRUCTIONS_V1", + "gate": "PASS" if not missing else "FAIL", + "checked_files": len(files), + "missing": missing, + } + out = ROOT / "Temp" / "completion_harness_instructions_v1.json" + out.write_text(json.dumps(result, ensure_ascii=False, indent=2), encoding="utf-8") + print(json.dumps(result, ensure_ascii=False, indent=2)) + return 0 if not missing else 1 + + +if __name__ == "__main__": + raise SystemExit(main()) diff --git a/tools/validate_realized_performance_v1.py b/tools/validate_realized_performance_v1.py new file mode 100644 index 0000000..0afb055 --- /dev/null +++ b/tools/validate_realized_performance_v1.py @@ -0,0 +1,96 @@ +"""validate_realized_performance_v1.py — REALIZED_PERFORMANCE_VALIDATE_V1 + +Temp/realized_performance_v1.json의 존재와 기본 스키마를 검증한다. +데이터가 충분하지 않은 구간은 allow-listed insufficient_data 상태로만 허용한다. +""" +from __future__ import annotations + +import argparse +import json +from pathlib import Path +from typing import Any + +ROOT = Path(__file__).resolve().parents[1] +DEFAULT_INPUT = ROOT / "Temp" / "realized_performance_v1.json" +DEFAULT_OUT = ROOT / "Temp" / "validate_realized_performance_v1.json" +FORMULA_ID = "REALIZED_PERFORMANCE_VALIDATE_V1" + + +def _load(path: Path) -> Any: + if not path.exists(): + return {} + try: + return json.loads(path.read_text(encoding="utf-8")) + except Exception: + return {} + + +def _is_nonempty_dict(v: Any) -> bool: + return isinstance(v, dict) and len(v) > 0 + + +def main() -> int: + ap = argparse.ArgumentParser() + ap.add_argument("--input", default=str(DEFAULT_INPUT)) + ap.add_argument("--out", default=str(DEFAULT_OUT)) + args = ap.parse_args() + + input_path = Path(args.input) + input_path = input_path if input_path.is_absolute() else ROOT / input_path + out_path = Path(args.out) + out_path = out_path if out_path.is_absolute() else ROOT / out_path + + payload = _load(input_path) + missing: list[dict[str, object]] = [] + + if not _is_nonempty_dict(payload): + missing.append({"field": "file", "reason": "missing_or_invalid_json"}) + else: + if payload.get("formula_id") != "REALIZED_PERFORMANCE_V1": + missing.append({ + "field": "formula_id", + "reason": "unexpected_value", + "expected": "REALIZED_PERFORMANCE_V1", + }) + + for field in ["performance_metrics", "current_portfolio_mdd", "insufficient_data_items", "summary"]: + if field not in payload: + missing.append({"field": field, "reason": "missing_key"}) + + perf = payload.get("performance_metrics") + if isinstance(perf, dict): + for key in ["t1_operational", "t5_operational", "t20_replay_estimated"]: + block = perf.get(key) + if not _is_nonempty_dict(block): + missing.append({"field": f"performance_metrics.{key}", "reason": "missing_block"}) + continue + if "label" not in block or "n" not in block: + missing.append({"field": f"performance_metrics.{key}", "reason": "missing_label_or_n"}) + + summary = payload.get("summary") + if isinstance(summary, dict): + if "disclaimer" not in summary: + missing.append({"field": "summary.disclaimer", "reason": "missing_key"}) + if "best_estimated_source" not in summary: + missing.append({"field": "summary.best_estimated_source", "reason": "missing_key"}) + + mdd = payload.get("current_portfolio_mdd") + if not _is_nonempty_dict(mdd): + missing.append({"field": "current_portfolio_mdd", "reason": "missing_block"}) + else: + if "worst_case_scenario" not in mdd: + missing.append({"field": "current_portfolio_mdd.worst_case_scenario", "reason": "missing_key"}) + + result = { + "formula_id": FORMULA_ID, + "gate": "PASS" if not missing else "FAIL", + "checked_file": str(Path(args.input).as_posix()), + "missing": missing, + } + out_path.write_text(json.dumps(result, ensure_ascii=False, indent=2) + "\n", encoding="utf-8") + print(json.dumps(result, ensure_ascii=False, indent=2)) + return 0 if not missing else 1 + + +if __name__ == "__main__": + raise SystemExit(main())