feat: OCR 新增支持波兰(PL,版式 pables_dpd/pables_inpost)

Reviewed-on: #8

.gitea/workflows/ci.yml

@@ -12,7 +12,10 @@ jobs:
       - name: 检出代码
         run: |
           REF="${{ github.event.pull_request.head.ref || github.ref_name }}"
-          git clone --depth 1 --branch "$REF" "${{ github.server_url }}/${{ github.repository }}.git" .
+          git clone --depth 1 "${{ github.server_url }}/${{ github.repository }}.git" .
+          # PR head 可能是完整 ref(AGit 建的 PR 为 refs/pull/N/head),clone --branch 只认分支名,改 fetch 后 checkout
+          git fetch --depth 1 origin "$REF"
+          git checkout --detach FETCH_HEAD
       - name: 安装 uv
         run: curl -LsSf https://astral.sh/uv/install.sh | sh
       - name: 安装依赖

api/ocr/models.py

@@ -37,14 +37,16 @@ class OcrResult(BaseModel):
 
 
 # OCR 服务已有分类器模型的国家(版面见 api.md §4)
-SUPPORTED_COUNTRIES = {"US", "UK", "DE", "MX", "SA"}
+SUPPORTED_COUNTRIES = {"US", "UK", "DE", "MX", "SA", "CA", "ES", "IT", "PL"}
 
 # ISO 国家代码 / 中文国名 → OCR 国家代码
 _COUNTRY_MAP: dict[str, str] = {
     # 代码
     "GB": "UK", "UK": "UK", "US": "US", "DE": "DE", "MX": "MX", "SA": "SA",
+    "CA": "CA", "ES": "ES", "IT": "IT", "PL": "PL",
     # 中文名
     "英国": "UK", "美国": "US", "德国": "DE", "墨西哥": "MX", "沙特": "SA",
+    "加拿大": "CA", "西班牙": "ES", "意大利": "IT", "波兰": "PL",
 }
 
 

docs/sql/2026-09-28-goods-status-allow-pause.sql

@@ -0,0 +1,13 @@
+-- 2026-09-28 Goods.status 约束新增允许 '暂停'
+-- 背景:RIIN 商品级 goodsStatus 复用平台状态枚举,存在 PAUSE(暂停),
+-- 原约束未包含该值,导致映射层只能落 NULL(chk_goods_status 拒绝其余新值)。
+-- 配套:mappers.py 的 RIIN_GOODS_STATUS_MAP 补 COMPLETE/PAUSE。
+-- 执行方式(两步,避免阻塞 ETL 写入;VALIDATE 只锁 SHARE UPDATE EXCLUSIVE):
+ALTER TABLE "Goods" DROP CONSTRAINT chk_goods_status;
+ALTER TABLE "Goods" ADD CONSTRAINT chk_goods_status CHECK (
+    status IS NULL OR status = ANY (ARRAY[
+        '备货中', '已删除', '已取消', '已完成', '待付款', '待确认',
+        '搁置中', '待发货', '已发货', '退款中', '暂停'
+    ])
+) NOT VALID;
+ALTER TABLE "Goods" VALIDATE CONSTRAINT chk_goods_status;

flows/label_ocr.py

@@ -137,12 +137,12 @@ def label_ocr(
     courier_source: str | None = None,
     days: int = 7,
 ) -> dict:
-    """面单 OCR enrichment。默认处理英国+美国,近 7 天下单的订单。
+    """面单 OCR enrichment。默认处理全部已支持国家(见 api.ocr.models.SUPPORTED_COUNTRIES),近 7 天下单的订单。
 
-    手动跑其他国家: label_ocr(country_regions=['德国'])
+    手动跑指定国家: label_ocr(country_regions=['德国'])
     """
     if country_regions is None:
-        country_regions = ['英国', '美国', '墨西哥', '德国', '沙特']
+        country_regions = ['英国', '美国', '墨西哥', '德国', '沙特', '加拿大', '西班牙', '意大利', '波兰']
     if _has_running_peer():
         print("已有 label-ocr 在运行,跳过本次")
         return {"skipped": True}

flows/platform_order/platform_order_upper_sync.py

@@ -0,0 +1,65 @@
+# -*- coding: utf-8 -*-
+"""平台订单 platform 字段大写同步:把 platform_order_data.platform 统一转为大写。
+
+优化自 platform_order_upper_sync.py,对应 platform-order-upper-sync。
+"""
+
+import time
+
+import psycopg2
+from prefect import flow
+
+# 数据库连接参数
+conn_params = {
+    "host": "192.168.124.92",
+    "port": 5432,
+    "user": "postgres",
+    "password": "Inkreach123456.",
+    "database": "smt",
+}
+
+
+def log_msg(msg, level="info"):
+    ts = time.strftime('%H:%M:%S')
+    tag = {"success": "[OK]", "error": "[ERR]"}.get(level, "")
+    print(f"[{ts}] {tag} {msg}")
+
+
+def update_platform_upper() -> int:
+    conn = None
+    cur = None
+    try:
+        conn = psycopg2.connect(**conn_params)
+        cur = conn.cursor()
+
+        # 仅更新需要变更的行,避免每次都全表写入
+        update_sql = """
+            UPDATE platform_order_data
+            SET platform = UPPER(platform)
+            WHERE platform IS NOT NULL
+              AND platform <> UPPER(platform);
+        """
+        cur.execute(update_sql)
+        affected_rows = cur.rowcount
+        conn.commit()
+        log_msg(f"platform 大写转换完成，共 {affected_rows} 行受影响", "success")
+        return affected_rows
+    except Exception as e:
+        log_msg(f"platform 大写转换失败：{e}", "error")
+        if conn:
+            conn.rollback()
+        raise
+    finally:
+        if cur:
+            cur.close()
+        if conn:
+            conn.close()
+
+
+@flow(name="platform-order-upper-sync", log_prints=True)
+def platform_order_upper_sync() -> int:
+    return update_platform_upper()
+
+
+if __name__ == "__main__":
+    platform_order_upper_sync.serve(name="platform-order-upper-sync", cron="*/2 * * * *")

flows/platform_order/sdspod.py

@@ -359,10 +359,18 @@ def sd_format_item(item):
     create_time = item.get("gmtCreated", "")
     parcel_list = item.get("parcelList", [])
     if parcel_list and isinstance(parcel_list, list):
-        qty = sum(p.get("totalQty", 0) for p in parcel_list)
+        def safe_int(val):
+            try:
+                return int(val)
+            except (ValueError, TypeError):
+                return 0
+
+        qty = sum(safe_int(p.get("totalQty", 0)) for p in parcel_list)
     else:
         qty = 0
 
+
+
     return [
         to_chinese_country(country),
         to_chinese_country(country),

flows/shared/mappers.py

@@ -33,6 +33,10 @@ RIIN_EXPRESS_SOURCE_MAP: dict[int, str] = {
 RIIN_GOODS_STATUS_MAP: dict[str, str] = {
     "NOT_SHIPPED": "待发货", "SHIPPED": "已发货", "CANCEL": "已取消",
     "CANCELLED": "已取消", "REFUNDING": "退款中",
+    # 商品级状态复用平台状态枚举,COMPLETE/PAUSE 缺失曾导致 1/3 商品行 status 为 NULL
+    "COMPLETE": "已完成", "PAUSE": "暂停",
+    # CLOSE 同为商品终态(Goods 约束无"已关闭",取"已取消");UNKNOWN 保留 NULL 并由告警暴露
+    "CLOSE": "已取消",
 }
 
 RIIN_PLATFORM_STATUS_MAP: dict[str, str] = {
@@ -55,6 +59,17 @@ SDS_CHANNEL_TYPE_LABEL_MAP: dict[str, str] = {
 
 # ---------- 日期解析 ----------
 
+# 未知枚举值告警:同值只报一次,避免批量同步刷屏(对齐 TS 版 warnUnknown)
+_warned_values: set[str] = set()
+
+
+def _warn_unknown(map_name: str, value: object) -> None:
+    key = f"{map_name}:{value!r}"
+    if key not in _warned_values:
+        _warned_values.add(key)
+        print(f"[mapper] unknown {map_name} value: {value!r}")
+
+
 def safe_date(value: Any, context: str = "") -> datetime:
     """解析日期字符串(无时区视为 +08:00),返回 tz-aware datetime。"""
     if not value or not isinstance(value, str):
@@ -188,15 +203,20 @@ class GoodsUpsertInput:
 
 # ---------- RIIN 映射 ----------
 
-def map_riin_system_status(order_status: int) -> str | None:
-    return RIIN_SYSTEM_STATUS_MAP.get(order_status)
+def map_riin_system_status(order_status: int | None) -> str | None:
+    if order_status is None:
+        return None
+    v = RIIN_SYSTEM_STATUS_MAP.get(order_status)
+    if v is None:
+        _warn_unknown("RIIN systemStatus", order_status)
+    return v
 
 
 def map_riin_order(raw: Any) -> OrderUpsertInput:
     platform_oid = raw.platform_oid
     ordered_at = _parse(raw.push_time, f"RIIN order {platform_oid} pushTime")
     confirmed_at = _parse(raw.distributor_push_time, f"RIIN order {platform_oid} distributorPushTime")
-    system_status = map_riin_system_status(raw.order_status) if raw.order_status is not None else None
+    system_status = map_riin_system_status(raw.order_status)
     is_terminal = system_status in ("已发货", "已关闭", "已取消")
     completed_at: datetime | None = None
     if is_terminal:
@@ -236,6 +256,15 @@ def map_riin_order(raw: Any) -> OrderUpsertInput:
     )
 
 
+def map_riin_goods_status(goods_status: str | None) -> str | None:
+    if not goods_status:
+        return None
+    v = RIIN_GOODS_STATUS_MAP.get(goods_status)
+    if v is None:
+        _warn_unknown("RIIN goodsStatus", goods_status)
+    return v
+
+
 def map_riin_goods(raw: Any, parent_order_no: str, parent_created_at: datetime) -> GoodsUpsertInput:
     product_code = raw.product_code
     item_name = product_code.split("-")[0] if product_code else None
@@ -250,7 +279,7 @@ def map_riin_goods(raw: Any, parent_order_no: str, parent_created_at: datetime)
         itemSize=raw.size_code,
         itemStyleNo=style_no,
         quantity=raw.num if raw.num is not None else 1,
-        status=RIIN_GOODS_STATUS_MAP.get(raw.goods_status) if raw.goods_status else None,
+        status=map_riin_goods_status(raw.goods_status),
     )
 
 
@@ -262,13 +291,27 @@ def map_sds_channel_type(channel_type: Any) -> str | None:
     return SDS_CHANNEL_TYPE_LABEL_MAP.get(channel_type)
 
 
-def map_sds_system_status(status: int) -> str | None:
-    return SDS_SYSTEM_STATUS_MAP.get(status)
+def map_sds_system_status(status: int | None) -> str | None:
+    if status is None:
+        return None
+    v = SDS_SYSTEM_STATUS_MAP.get(status)
+    if v is None:
+        _warn_unknown("SDS systemStatus", status)
+    return v
+
+
+def map_sds_goods_status(status: int | None) -> str | None:
+    if status is None:
+        return None
+    v = SDS_GOODS_STATUS_MAP.get(status)
+    if v is None:
+        _warn_unknown("SDS goodsStatus", status)
+    return v
 
 
 def map_sds_order(raw: Any, channel_type_map: dict[int, str | None], pre_courier: str | None = None) -> OrderUpsertInput:
     ordered_at = _parse(raw.gmt_pay, f"SDS order {raw.no} gmtPay")
-    system_status = map_sds_system_status(raw.status) if raw.status is not None else None
+    system_status = map_sds_system_status(raw.status)
     is_terminal = system_status in ("已完成", "已取消")
     completed_at: datetime | None = None
     if is_terminal and raw.finish_time and raw.finish_time > 0:
@@ -369,7 +412,7 @@ def map_sds_goods(raw: Any, parent_order_no: str, parent_created_at: datetime) -
         itemSize=raw.product_size,
         itemStyleNo=_parse_sds_style_no(product_name),
         quantity=raw.num if raw.num is not None else 1,
-        status=SDS_GOODS_STATUS_MAP.get(raw.status) if raw.status is not None else None,
+        status=map_sds_goods_status(raw.status),
     )
 
 

prefect.yaml

@@ -97,7 +97,14 @@ deployments:
 - name: platform-order-sdspod
   entrypoint: flows/platform_order/sdspod.py:sdspod_platform_sync
   schedules:
-  - cron: '*/30 * * * *'
+  - cron: 0 * * * *
+    timezone: Asia/Shanghai
+  work_pool:
+    name: inkreach-pool
+- name: platform-order-upper-sync
+  entrypoint: flows/platform_order/platform_order_upper_sync.py:platform_order_upper_sync
+  schedules:
+  - cron: '*/2 * * * *'
     timezone: Asia/Shanghai
   work_pool:
     name: inkreach-pool

scripts/backfill_riin_goods_status.py

@@ -0,0 +1,72 @@
+"""一次性回填:RIIN Goods.status 为 NULL 的历史行(终态"已发货"订单)。
+
+背景:RIIN 商品级 goods_status 复用平台状态枚举,COMPLETE/PAUSE 此前无映射,
+约 1/3 商品行 status 被写为 NULL(mappers.py 已补映射)。本脚本按订单终态
+将近似值回填;"已关闭"与非终态订单不在此列,走 riin-by-orders 重拉取真值。
+幂等:只更新 status IS NULL 的行。分批提交,避免长事务。
+
+运行:uv run python scripts/backfill_riin_goods_status.py [--dry-run] [--batch N]
+"""
+
+from __future__ import annotations
+
+import argparse
+import os
+
+import psycopg2
+from dotenv import load_dotenv
+
+load_dotenv(os.path.join(os.path.dirname(__file__), "..", ".env"))
+
+DEFAULT_BATCH = 50_000
+
+SQL_COUNT = """
+SELECT count(*) FROM "Goods" g
+JOIN "Orders" o ON o."orderNo" = g."orderNo" AND o."createdAt" = g."createdAt"
+WHERE o."systemName" = 'RIIN' AND o."systemStatus" = '已发货' AND g."status" IS NULL
+"""
+
+# ctid 分批:PostgreSQL 的 UPDATE 不支持 LIMIT,用子查询圈定每批行
+SQL_BACKFILL_BATCH = """
+UPDATE "Goods" g SET "status" = '已发货'
+WHERE g.ctid IN (
+    SELECT g2.ctid FROM "Goods" g2
+    JOIN "Orders" o ON o."orderNo" = g2."orderNo" AND o."createdAt" = g2."createdAt"
+    WHERE o."systemName" = 'RIIN' AND o."systemStatus" = '已发货' AND g2."status" IS NULL
+    LIMIT %s
+)
+"""
+
+
+def main() -> None:
+    parser = argparse.ArgumentParser(description="回填 RIIN Goods.status NULL(已发货订单)")
+    parser.add_argument("--dry-run", action="store_true", help="只统计待回填行数,不写入")
+    parser.add_argument("--batch", type=int, default=DEFAULT_BATCH, help="每事务回填行数")
+    args = parser.parse_args()
+
+    conn = psycopg2.connect(os.environ["DB_DSN"])
+    try:
+        with conn.cursor() as cur:
+            cur.execute(SQL_COUNT)
+            (pending,) = cur.fetchone()
+        print(f"[backfill] 待回填(status IS NULL 且订单已发货): {pending} 行")
+        if args.dry_run or pending == 0:
+            return
+
+        total = 0
+        while True:
+            with conn.cursor() as cur:
+                cur.execute(SQL_BACKFILL_BATCH, (args.batch,))
+                updated = cur.rowcount
+            conn.commit()
+            total += updated
+            print(f"[backfill] 本批 {updated} 行,累计 {total}")
+            if updated < args.batch:
+                break
+        print(f"[backfill] 完成,共回填 {total} 行")
+    finally:
+        conn.close()
+
+
+if __name__ == "__main__":
+    main()

tests/test_mappers.py

@@ -0,0 +1,126 @@
+"""mappers 状态映射单测。
+
+运行:uv run python -m unittest discover -s tests -v
+"""
+
+from __future__ import annotations
+
+import io
+import unittest
+from contextlib import redirect_stdout
+from datetime import datetime, timedelta, timezone
+from types import SimpleNamespace
+
+from flows.shared import mappers
+from flows.shared.mappers import (
+    map_riin_goods,
+    map_riin_goods_status,
+    map_riin_system_status,
+    map_sds_goods_status,
+    map_sds_system_status,
+)
+
+TZ_CN = timezone(timedelta(hours=8))
+
+
+class WarnUnknownTest(unittest.TestCase):
+    def setUp(self):
+        mappers._warned_values.clear()
+
+    def _capture(self):
+        return io.StringIO()
+
+    def test_未知值告警且同值只报一次(self):
+        buf = self._capture()
+        with redirect_stdout(buf):
+            self.assertIsNone(map_riin_goods_status("SOMETHING_NEW"))
+            self.assertIsNone(map_riin_goods_status("SOMETHING_NEW"))  # 第二次不再报
+        lines = [ln for ln in buf.getvalue().splitlines() if "SOMETHING_NEW" in ln]
+        self.assertEqual(len(lines), 1)
+        self.assertIn("[mapper] unknown RIIN goodsStatus", lines[0])
+
+
+class RiinGoodsStatusTest(unittest.TestCase):
+    def setUp(self):
+        mappers._warned_values.clear()
+
+    def test_全部已知枚举可映射(self):
+        cases = {
+            "NOT_SHIPPED": "待发货",
+            "SHIPPED": "已发货",
+            "CANCEL": "已取消",
+            "CANCELLED": "已取消",
+            "REFUNDING": "退款中",
+            "COMPLETE": "已完成",  # 本次补齐(曾导致 NULL)
+            "PAUSE": "暂停",       # 本次补齐(曾导致 NULL)
+            "CLOSE": "已取消",     # 回填重拉时告警发现的新枚举
+        }
+        for raw, want in cases.items():
+            with self.subTest(raw=raw):
+                self.assertEqual(map_riin_goods_status(raw), want)
+
+    def test_空值返回None(self):
+        self.assertIsNone(map_riin_goods_status(None))
+        self.assertIsNone(map_riin_goods_status(""))
+
+    def test_map_riin_goods_接线_PAUSE与COMPLETE不再落NULL(self):
+        # 复现线上订单 GSU1S555400MFLL:4 行商品 goods_status=PAUSE/COMPLETE
+        created = datetime(2026, 9, 14, 9, 39, 24, tzinfo=TZ_CN)
+        base = dict(
+            platform_oid="GSU1S555400MFLL", product_code="DG001-BK",
+            color_code="BK", size_code="L", style_code=None, num=1,
+        )
+        g1 = map_riin_goods(SimpleNamespace(id="P1", goods_status="PAUSE", **base),
+                            "GSU1S555400MFLL", created)
+        g2 = map_riin_goods(SimpleNamespace(id="P2", goods_status="COMPLETE", **base),
+                            "GSU1S555400MFLL", created)
+        self.assertEqual(g1.status, "暂停")
+        self.assertEqual(g2.status, "已完成")
+        self.assertEqual(g1.productionNo, "P1")
+        self.assertEqual(g1.quantity, 1)
+        self.assertEqual(g1.itemName, "DG001")
+
+
+class RiinSystemStatusTest(unittest.TestCase):
+    def setUp(self):
+        mappers._warned_values.clear()
+
+    def test_已知与空值(self):
+        self.assertEqual(map_riin_system_status(12), "已发货")
+        self.assertEqual(map_riin_system_status(-4), "待推送")
+        self.assertIsNone(map_riin_system_status(None))
+
+    def test_未知值返回None并告警(self):
+        buf = io.StringIO()
+        with redirect_stdout(buf):
+            self.assertIsNone(map_riin_system_status(999))
+        self.assertIn("unknown RIIN systemStatus", buf.getvalue())
+
+
+class SdsStatusTest(unittest.TestCase):
+    def setUp(self):
+        mappers._warned_values.clear()
+
+    def test_商品状态含已删除(self):
+        self.assertEqual(map_sds_goods_status(99), "已删除")
+        self.assertEqual(map_sds_goods_status(98), "已取消")
+        self.assertEqual(map_sds_goods_status(4), "已完成")
+        self.assertIsNone(map_sds_goods_status(None))
+
+    def test_未知商品状态返回None并告警(self):
+        buf = io.StringIO()
+        with redirect_stdout(buf):
+            self.assertIsNone(map_sds_goods_status(6))
+        self.assertIn("unknown SDS goodsStatus", buf.getvalue())
+
+    def test_订单状态与未知告警(self):
+        self.assertEqual(map_sds_system_status(3), "备货中")
+        self.assertIsNone(map_sds_system_status(None))
+        buf = io.StringIO()
+        with redirect_stdout(buf):
+            self.assertIsNone(map_sds_system_status(42))
+        self.assertIn("unknown SDS systemStatus", buf.getvalue())
+
+
+if __name__ == "__main__":
+    unittest.main()