섹터 유니버스 분리와 월간 갱신 정합화

src/quant_engine/convert_xlsx_to_json.py

@@ -174,6 +174,28 @@ def normalize_legacy_source_markers(sheet: str, records: list[dict[str, Any]]) -
         source = record.get("Source")
         if isinstance(source, str) and "sector_targets.json" in source:
             record["Source"] = source.replace("sector_targets.json", "sector_universe")
+        source_url = str(record.get("Source_URL") or "").strip()
+        transport_mode = str(record.get("Transport_Mode") or "").strip()
+        if record.get("Source") in (None, "", "DEFAULT_TEMPLATE"):
+            if "finance.naver.com/item/main.naver?code=" in source_url:
+                record["Source"] = "NAVER_ETF_PAGE"
+                if not transport_mode:
+                    record["Transport_Mode"] = "HTML_SERVER_RENDERED"
+            elif source_url:
+                record["Source"] = "SHEET_INPUT"
+                if not transport_mode:
+                    record["Transport_Mode"] = "MANUAL_OR_TEMPLATE"
+            else:
+                record["Source"] = "SHEET_INPUT"
+                if not transport_mode:
+                    record["Transport_Mode"] = "MANUAL_OR_TEMPLATE"
+        elif record.get("Source") == "NAVER_ETF_PAGE_FAIL_LAYOUT_CHANGED" and not transport_mode:
+            record["Transport_Mode"] = "LAYOUT_CHANGED"
+        elif record.get("Source") == "REPRESENTATIVE_STOCK_PROXY" and not transport_mode:
+            record["Transport_Mode"] = "HTML_SERVER_RENDERED"
+        sector = str(record.get("Sector") or "").strip()
+        if sector:
+            record["Sector_Check"] = sector
     return records
 
 
@@ -1428,6 +1450,80 @@ def convert_xlsx_to_json(xlsx_path: Path, output_path: Path) -> None:
             result["data"][sheet] = normalize_legacy_source_markers(sheet, dataframe_records(df))
         result["metadata"]["sheets_included"].append(sheet)
 
+    sector_source_map: dict[str, str] = {}
+    sector_universe_rows = result["data"].get("sector_universe")
+    if isinstance(sector_universe_rows, list):
+        for row in sector_universe_rows:
+            if not isinstance(row, dict):
+                continue
+            sector = str(row.get("Sector") or "").strip()
+            if not sector:
+                continue
+            source = str(row.get("Source") or "").strip() or "SHEET_INPUT"
+            sector_source_map.setdefault(sector, source)
+
+    sector_flow_rows = result["data"].get("sector_flow")
+    if isinstance(sector_flow_rows, list):
+        split_finance_map = {
+            "금융/은행": [
+                ("은행", "091170", "KODEX 은행"),
+                ("증권", "0111J0", "HANARO 증권고배당TOP3플러스"),
+                ("지주회사", "307520", "TIGER 지주회사"),
+            ]
+        }
+        normalized_rows: list[dict[str, Any]] = []
+        for row in sector_flow_rows:
+            if not isinstance(row, dict):
+                continue
+            sector = str(row.get("Sector") or "").strip()
+            if not sector:
+                continue
+            source = str(row.get("Universe_Source") or "").strip() or sector_source_map.get(sector, "SHEET_INPUT")
+            row["Universe_Source"] = source
+            if sector in split_finance_map:
+                for split_sector, split_ticker, split_name in split_finance_map[sector]:
+                    cloned = dict(row)
+                    cloned["Sector"] = split_sector
+                    cloned["Proxy_Ticker"] = split_ticker
+                    cloned["Proxy_Name"] = split_name
+                    cloned["Proxy_Type"] = "ETF"
+                    cloned["ETF_Code"] = split_ticker
+                    cloned["Reason"] = "PRE_SPLIT_FINANCE_FLOW_CARRYOVER"
+                    cloned["Universe_Source"] = "NAVER_ETF_PAGE"
+                    normalized_rows.append(cloned)
+            else:
+                normalized_rows.append(row)
+        result["data"]["sector_flow"] = normalized_rows
+
+    sector_flow_history_rows = result["data"].get("sector_flow_history")
+    if isinstance(sector_flow_history_rows, list):
+        split_finance_map = {
+            "금융/은행": [
+                ("은행", "091170", "KODEX 은행"),
+                ("증권", "0111J0", "HANARO 증권고배당TOP3플러스"),
+                ("지주회사", "307520", "TIGER 지주회사"),
+            ]
+        }
+        normalized_history: list[dict[str, Any]] = []
+        for row in sector_flow_history_rows:
+            if not isinstance(row, dict):
+                continue
+            sector = str(row.get("Sector") or "").strip()
+            if not sector:
+                continue
+            if sector in split_finance_map:
+                for split_sector, split_ticker, split_name in split_finance_map[sector]:
+                    cloned = dict(row)
+                    cloned["Sector"] = split_sector
+                    cloned["Proxy_Ticker"] = split_ticker
+                    cloned["Proxy_Name"] = split_name
+                    cloned["Proxy_Type"] = "ETF"
+                    cloned["Reason"] = "PRE_SPLIT_FINANCE_FLOW_CARRYOVER"
+                    normalized_history.append(cloned)
+            else:
+                normalized_history.append(row)
+        result["data"]["sector_flow_history"] = normalized_history
+
     # harness_context 시트가 없으면 메타에 경고 기록
     if "_harness_context" not in result["data"]:
         result["metadata"]["harness_context_missing"] = (