refactor: countryRegion 统一中文归一化; label-ocr 改用 countryRegion; destinationCountry 还原 ISO2

flows/label_ocr.py

@@ -115,16 +115,16 @@ def process_all(pending: list) -> dict[str, int]:
 
 @flow(name="label-ocr", log_prints=True)
 def label_ocr(
-    destination_countries: list[str] | None = None,
+    country_regions: list[str] | None = None,
     courier_source: str | None = None,
     days: int = 7,
 ) -> dict:
-    """面单 OCR enrichment。默认处理英国(GB)+美国(US),近 7 天下单的订单。
+    """面单 OCR enrichment。默认处理英国+美国,近 7 天下单的订单。
 
-    手动跑其他国家: label_ocr(destination_countries=['德国'])
+    手动跑其他国家: label_ocr(country_regions=['德国'])
     """
-    if destination_countries is None:
+    if country_regions is None:
-        destination_countries = ['英国', '美国']
+        country_regions = ['英国', '美国']
     if _has_running_peer():
         print("已有 label-ocr 在运行,跳过本次")
         return {"skipped": True}
@@ -133,11 +133,11 @@ def label_ocr(
     print(f"OCR 状态(开始): {before}")
 
     pending = db.get_ocr_pending(
-        destination_countries=destination_countries,
+        country_regions=country_regions,
         courier_source=courier_source,
         days=days,
     )
-    scope = f" [{'/'.join(destination_countries)}{f' {courier_source}' if courier_source else ''}]"
+    scope = f" [{'/'.join(country_regions)}{f' {courier_source}' if courier_source else ''}]"
     print(f"待 OCR: {len(pending)} 单{scope} (RIIN+SDS, 近{days}天)")
     if not pending:
         emit_sync_summary(f"面单 OCR{scope}", build_stats_rows("ALL", {}, [

flows/shared/db.py

@@ -102,25 +102,25 @@ class SyncDB:
 
     def get_ocr_pending(
         self, limit: int = 0,
-        destination_countries: list[str] | None = None,
+        country_regions: list[str] | None = None,
         courier_source: str | None = None,
         days: int = 0,
     ) -> list[tuple[str, datetime, str, str | None]]:
-        """查待 OCR 订单(所有系统):(orderNo, createdAt, labelPdfUrl, destinationCountry)。
+        """查待 OCR 订单(所有系统):(orderNo, createdAt, labelPdfUrl, countryRegion)。
 
         条件:有 labelPdfUrl、ocrLayout 未处理(Null)。
-        limit=0 表示不限。可选按 destinationCountries / courierSource 筛选。
+        limit=0 表示不限。可选按 countryRegions / courierSource 筛选。
         days>0 时只取近 N 天下单的订单。
         """
         sql = (
-            'SELECT "orderNo","createdAt","labelPdfUrl","destinationCountry" '
+            'SELECT "orderNo","createdAt","labelPdfUrl","countryRegion" '
             'FROM "Orders" '
             'WHERE "labelPdfUrl" IS NOT NULL AND "ocrLayout" IS NULL'
         )
         params: list = []
-        if destination_countries:
+        if country_regions:
-            sql += ' AND "destinationCountry"=ANY(%s)'
+            sql += ' AND "countryRegion"=ANY(%s)'
-            params.append(destination_countries)
+            params.append(country_regions)
         if courier_source:
             sql += ' AND "courierSource"=%s'
             params.append(courier_source)

flows/shared/mappers.py

@@ -79,7 +79,7 @@ def _from_epoch_ms(value: Any) -> datetime | None:
 # ---------- 国家/地区归一 ----------
 
 _US_ALIASES = {"US", "UNITED STATES", "USA", "ESTADOS UNIDOS", "美国加州发货仓"}
-_MX_ALIASES = {"MX", "MEXICO", "MÉXICO", "MEX", "ME", "MÉXICO"}
+_MX_ALIASES = {"MX", "MEXICO", "MÉXICO", "MEX", "ME"}
 _CA_ALIASES = {"加拿大工厂", "CA"}
 _PL_ALIASES = {"欧州波兰发货仓", "欧洲波兰法兰克福发货仓"}
 _WAREHOUSE_MAP = {
@@ -102,6 +102,7 @@ _ISO2_CN: dict[str, str] = {
     "NO": "挪威", "IS": "冰岛", "TR": "土耳其", "AE": "阿联酋", "IL": "以色列",
     "NZ": "新西兰", "SG": "新加坡", "MY": "马来西亚", "TH": "泰国", "PH": "菲律宾",
     "ID": "印度尼西亚", "VN": "越南", "IN": "印度", "RU": "俄罗斯", "UA": "乌克兰",
+    "RS": "塞尔维亚",
 }
 
 _EN_CN: dict[str, str] = {
@@ -110,7 +111,12 @@ _EN_CN: dict[str, str] = {
     "SPAIN": "西班牙", "BRAZIL": "巴西", "ITALY": "意大利", "CANADA": "加拿大",
     "FRANCE": "法国", "POLAND": "波兰", "NETHERLANDS": "荷兰", "AUSTRALIA": "澳大利亚",
     "CROATIA": "克罗地亚", "CZECH REPUBLIC": "捷克", "ESTADOS UNIDOS": "美国",
-    "SLOVAKIA (SLOVAK REPUBLIC)": "斯洛伐克", "AUSTRALIA": "澳大利亚",
+    "SLOVAKIA (SLOVAK REPUBLIC)": "斯洛伐克", "ROMANIA": "罗马尼亚",
+    "HUNGARY": "匈牙利", "BELGIUM": "比利时", "SWEDEN": "瑞典", "DENMARK": "丹麦",
+    "FINLAND": "芬兰", "IRELAND": "爱尔兰", "PORTUGAL": "葡萄牙", "AUSTRIA": "奥地利",
+    "GREECE": "希腊", "LUXEMBOURG": "卢森堡", "BULGARIA": "保加利亚",
+    "ESTONIA": "爱沙尼亚", "LATVIA": "拉脱维亚", "LITHUANIA": "立陶宛",
+    "SLOVENIA": "斯洛文尼亚",
 }
 
 _CN_NORMALIZE: dict[str, str] = {
@@ -118,8 +124,10 @@ _CN_NORMALIZE: dict[str, str] = {
 }
 
 
-def normalize_destination_country(raw: str | None) -> str | None:
+def normalize_country_region(raw: str | None, factory_name: str | None = None) -> str | None:
-    """将 destinationCountry 归一为中文国名。"""
+    """将国家/地区原始值统一归一为中文国名。"""
+    if factory_name == "广州印美达":
+        return "中国"
     if not raw:
         return None
     trimmed = raw.strip()
@@ -134,42 +142,17 @@ def normalize_destination_country(raw: str | None) -> str | None:
     if trimmed in _CN_NORMALIZE:
         return _CN_NORMALIZE[trimmed]
     upper = trimmed.upper()
-    if upper in _ISO2_CN:
-        return _ISO2_CN[upper]
-    if upper in _EN_CN:
-        return _EN_CN[upper]
-    return trimmed
-_EXCLUDED_COUNTRIES = {
-    "Croatia", "Czech Republic", "丹麦", "保加利亚", "克罗地亚", "卢森堡", "奥地利", "希腊",
-    "拉脱维亚", "捷克", "斯洛伐克", "斯洛文尼亚", "比利时", "法国", "爱尔兰", "爱沙尼亚",
-    "瑞典", "瑞士", "立陶宛", "罗马尼亚", "芬兰", "荷兰", "Denmark", "Austria", "Greece",
-    "Latvia", "Slovakia", "Slovenia", "Belgium", "France", "Ireland", "Estonia", "Sweden",
-    "Switzerland", "Lithuania", "Romania", "Finland", "Netherlands", "Luxembourg", "Bulgaria",
-}
-
-
-def normalize_country_region(raw: str | None, factory_name: str | None = None) -> str | None:
-    if factory_name == "广州印美达":
-        return "中国"
-    if not raw:
-        return None
-    trimmed = raw.strip()
-    if trimmed in _EXCLUDED_COUNTRIES:
-        return None
-    upper = trimmed.upper()
     if upper in _US_ALIASES:
         return "美国"
     if upper in _MX_ALIASES:
         return "墨西哥"
     if upper in _CA_ALIASES:
         return "加拿大"
-    if trimmed in _PL_ALIASES:
+    if upper in _ISO2_CN:
-        return "波兰"
+        return _ISO2_CN[upper]
-    if trimmed in _WAREHOUSE_MAP:
+    if upper in _EN_CN:
-        return _WAREHOUSE_MAP[trimmed]
+        return _EN_CN[upper]
-    if raw in _DIRECT_MAP:
+    return trimmed
-        return _DIRECT_MAP[raw]
-    return raw
 
 
 # ---------- 统一写入模型 ----------
@@ -239,7 +222,7 @@ def map_riin_order(raw: Any) -> OrderUpsertInput:
         completedAt=completed_at,
         shipDeadline=ship_deadline,
         consigneeName=raw.consignee_name,
-        destinationCountry=normalize_destination_country(raw.receiver_country),
+        destinationCountry=raw.receiver_country,
         stateProvince=raw.receiver_province,
         city=raw.receiver_city,
         addressLine1=raw.address,
@@ -339,7 +322,7 @@ def map_sds_order(raw: Any, channel_type_map: dict[int, str | None], pre_courier
         completedAt=completed_at,
         shipDeadline=ship_deadline,
         consigneeName=addr.consignee if addr else None,
-        destinationCountry=normalize_destination_country(addr.country) if addr else None,
+        destinationCountry=addr.country if addr else None,
         stateProvince=addr.province if addr else None,
         city=addr.city if addr else None,
         addressLine1=addr.address_detail1 if addr else None,