chore: establish project baseline

2026-09-21 15:51:47 -03:00
commit 9dcc263217
459 changed files with 447890 additions and 0 deletions
@@ -0,0 +1,46 @@
+"""Promote CFOP/FCP/ANP/NCM_UTRIB/RTC_BIOFUEL_INDEX/RTC_CCREDPRES from the
+official spreadsheets the operator placed in sources/raw/bootstrap_v1/ after
+the Portal NF-e/SPED downloads were session-blocked. No network fetch.
+"""
+import json
+from datetime import datetime, timezone
+from pathlib import Path
+
+from central.app.datasets import bootstrap_v1_extra as parsers
+from central.app.datasets.service import DatasetService
+from central.app.db import SessionLocal
+
+ROOT = Path("sources/raw/bootstrap_v1")
+
+JOBS = [
+    ("CFOP", ROOT / "IT_2023_002_v2_10_Tabela_CFOP_Descrição.xlsx", parsers.parse_cfop, "import_cfop"),
+    ("FCP", ROOT / "Tabela_Alíquotas_FCP_por_UF_2025-02_07.xlsx", parsers.parse_fcp, "import_fcp"),
+    ("ANP", ROOT / "Tabela_Codigo_Produto_ANP-20241118.xlsx", parsers.parse_anp, "import_anp"),
+    ("NCM_UTRIB", ROOT / "IT2024.001v2.40 - Tabela NCM 2022 com Utrib_Comércio Exterior_vigência 01.10.26.xlsx", parsers.parse_ncm_utrib, "import_ncm_utrib"),
+    ("RTC_BIOFUEL_INDEX", ROOT / "IT 2025.004 v.1.20 - Tabela de Índice.xlsx", parsers.parse_rtc_biofuel_index, "import_rtc_biofuel_index"),
+    ("RTC_CCREDPRES", ROOT / "cCredPres_2026-06-22.xlsx", parsers.parse_rtc_cred_pres, "import_rtc_cred_pres"),
+    ("RTC_CCLASSTRIB", ROOT / "cClassTrib 2026-06-22.xlsx", parsers.parse_rtc_cclasstrib, "import_rtc_cclasstrib"),
+    ("RTC_CBS_RATES", ROOT / "IT 2026.002 v.1.00 - Aliquotas da CBS.xlsx", parsers.parse_rtc_cbs_rates, "import_rtc_cbs_rates"),
+]
+
+
+def main():
+    results = {}
+    with SessionLocal() as session:
+        service = DatasetService(session)
+        for domain, path, parse_fn, method_name in JOBS:
+            try:
+                if not path.exists():
+                    results[domain] = {"status": "MISSING_BOOTSTRAP_FILE", "path": str(path)}
+                    continue
+                payload = parse_fn(path.read_bytes())
+                dataset = getattr(service, method_name)(payload, f"file://{path.name}", datetime.now(timezone.utc))
+                results[domain] = {"status": dataset.status, "dataset_id": dataset.id, "sha256": dataset.sha256, "record_count": dataset.record_count}
+            except Exception as error:  # noqa: BLE001 — isolate one domain's failure from the rest
+                results[domain] = {"status": "FAILED", "reason": f"{type(error).__name__}: {error}"}
+
+    print(json.dumps(results, indent=2, ensure_ascii=False))
+
+
+if __name__ == "__main__":
+    main()