chore(tools): add bypos-collector (条码批量采集器源码备份)

Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>

admin-frontend/src/App.tsx

@@ -4,9 +4,10 @@ import Login from "./components/Login";
 import ProductList from "./components/ProductList";
 import ProductDetail from "./components/ProductDetail";
 import SubmissionsPage from "./components/SubmissionsPage";
-import { Inbox, LogOut, Package } from "lucide-react";
+import ApiKeysPage from "./components/ApiKeysPage";
+import { Inbox, KeyRound, LogOut, Package } from "lucide-react";
 
-type Tab = "products" | "submissions";
+type Tab = "products" | "submissions" | "keys";
 type View = { name: "list" } | { name: "detail"; id: string };
 
 export default function App() {
@@ -99,6 +100,16 @@ export default function App() {
                 </span>
               )}
             </button>
+            <button
+              onClick={() => setTab("keys")}
+              className={`px-3 py-1.5 rounded-md flex items-center gap-1.5 ${
+                tab === "keys"
+                  ? "bg-emerald-50 text-emerald-700"
+                  : "text-gray-600 hover:bg-gray-100"
+              }`}
+            >
+              <KeyRound className="h-4 w-4" /> API 密钥
+            </button>
           </nav>
         </div>
         <div className="flex items-center gap-4 text-sm text-gray-600">
@@ -112,7 +123,9 @@ export default function App() {
         </div>
       </header>
       <main className="flex-1 overflow-auto p-6">
-        {tab === "submissions" ? (
+        {tab === "keys" ? (
+          <ApiKeysPage />
+        ) : tab === "submissions" ? (
           <SubmissionsPage onPending={setPending} />
         ) : view.name === "list" ? (
           <ProductList onOpen={(id) => setView({ name: "detail", id })} />

admin-frontend/src/api.ts

@@ -123,4 +123,18 @@ export const api = {
       method: "POST",
       body: JSON.stringify({ note }),
     }),
+  listApiKeys: () =>
+    request<{ items: import("./types").ApiKey[] }>("/keys"),
+  createApiKey: (body: {
+    name: string;
+    owner_email?: string;
+    tier?: string;
+    rate_limit_per_min?: number;
+  }) =>
+    request<{ key: string; item: import("./types").ApiKey; warning: string }>(
+      "/keys",
+      { method: "POST", body: JSON.stringify(body) },
+    ),
+  revokeApiKey: (id: string) =>
+    request<{ status: string }>(`/keys/${id}`, { method: "DELETE" }),
 };

admin-frontend/src/components/ApiKeysPage.tsx

@@ -0,0 +1,278 @@
+import { useEffect, useState } from "react";
+import { api, ApiError } from "../api";
+import type { ApiKey } from "../types";
+import { Copy, KeyRound, Plus, Trash2 } from "lucide-react";
+
+const TIERS = [
+  { key: "free", label: "免费 (free)", rate: 120 },
+  { key: "partner", label: "合作方 (partner)", rate: 600 },
+  { key: "internal", label: "内部 (internal)", rate: 6000 },
+];
+
+function tierLabel(tier: string): string {
+  return TIERS.find((t) => t.key === tier)?.label ?? tier;
+}
+
+export default function ApiKeysPage() {
+  const [rows, setRows] = useState<ApiKey[]>([]);
+  const [error, setError] = useState("");
+  const [creating, setCreating] = useState(false);
+  const [newKey, setNewKey] = useState<string | null>(null);
+
+  async function load() {
+    setError("");
+    try {
+      const res = await api.listApiKeys();
+      setRows(res.items);
+    } catch (e) {
+      setError(e instanceof ApiError ? e.message : "加载失败");
+    }
+  }
+
+  useEffect(() => {
+    load();
+  }, []);
+
+  async function revoke(id: string, name: string) {
+    if (!confirm(`确认吊销密钥「${name}」？使用该密钥的请求将立即被拒绝。`)) return;
+    try {
+      await api.revokeApiKey(id);
+      await load();
+    } catch (e) {
+      setError(e instanceof ApiError ? e.message : "操作失败");
+    }
+  }
+
+  return (
+    <div className="max-w-4xl">
+      <div className="flex items-center justify-between mb-4">
+        <div>
+          <h2 className="text-lg font-semibold text-gray-800 flex items-center gap-2">
+            <KeyRound className="h-5 w-5 text-emerald-600" /> API 密钥
+          </h2>
+          <p className="text-sm text-gray-500 mt-1">
+            公开 API 免费只读，匿名也可调用（低额度）。签发密钥可获得更高速率上限并归属用量。
+          </p>
+        </div>
+        <button
+          onClick={() => setCreating(true)}
+          className="px-4 py-2 rounded-lg bg-emerald-600 text-white text-sm font-medium hover:bg-emerald-700 flex items-center gap-1.5"
+        >
+          <Plus className="h-4 w-4" /> 签发密钥
+        </button>
+      </div>
+
+      {error && (
+        <div className="mb-3 bg-red-50 text-red-700 text-sm rounded px-4 py-2">{error}</div>
+      )}
+
+      {newKey && (
+        <div className="mb-4 bg-amber-50 border border-amber-200 rounded-lg p-4">
+          <div className="text-sm font-medium text-amber-800 mb-1">
+            密钥已生成，请立即复制保存——它只显示这一次。
+          </div>
+          <div className="flex items-center gap-2">
+            <code className="flex-1 bg-white border rounded px-3 py-2 text-sm break-all">
+              {newKey}
+            </code>
+            <button
+              onClick={() => navigator.clipboard?.writeText(newKey)}
+              className="px-3 py-2 rounded border text-sm text-gray-600 hover:bg-gray-50 flex items-center gap-1"
+            >
+              <Copy className="h-4 w-4" /> 复制
+            </button>
+            <button
+              onClick={() => setNewKey(null)}
+              className="px-3 py-2 rounded text-sm text-gray-500 hover:bg-gray-100"
+            >
+              我已保存
+            </button>
+          </div>
+        </div>
+      )}
+
+      {creating && (
+        <CreateKeyForm
+          onClose={() => setCreating(false)}
+          onCreated={(plaintext) => {
+            setCreating(false);
+            setNewKey(plaintext);
+            load();
+          }}
+        />
+      )}
+
+      <div className="bg-white border rounded-lg overflow-hidden">
+        <table className="w-full text-sm">
+          <thead className="bg-gray-50 text-gray-500 text-left">
+            <tr>
+              <th className="px-4 py-2 font-medium">名称</th>
+              <th className="px-4 py-2 font-medium">前缀</th>
+              <th className="px-4 py-2 font-medium">级别</th>
+              <th className="px-4 py-2 font-medium">速率/分钟</th>
+              <th className="px-4 py-2 font-medium">用量(今日/累计)</th>
+              <th className="px-4 py-2 font-medium">状态</th>
+              <th className="px-4 py-2 font-medium"></th>
+            </tr>
+          </thead>
+          <tbody className="divide-y">
+            {rows.length === 0 ? (
+              <tr>
+                <td colSpan={7} className="px-4 py-8 text-center text-gray-400">
+                  暂无密钥
+                </td>
+              </tr>
+            ) : (
+              rows.map((k) => (
+                <tr key={k.id} className={k.revoked_at ? "opacity-50" : ""}>
+                  <td className="px-4 py-2 text-gray-800">
+                    {k.name}
+                    {k.owner_email && (
+                      <span className="block text-xs text-gray-400">{k.owner_email}</span>
+                    )}
+                  </td>
+                  <td className="px-4 py-2 text-gray-500">
+                    <code>{k.key_prefix}…</code>
+                  </td>
+                  <td className="px-4 py-2 text-gray-600">{tierLabel(k.tier)}</td>
+                  <td className="px-4 py-2 text-gray-600">{k.rate_limit_per_min}</td>
+                  <td className="px-4 py-2 text-gray-600">
+                    {k.usage.today} / {k.usage.total}
+                  </td>
+                  <td className="px-4 py-2">
+                    {k.revoked_at ? (
+                      <span className="text-xs rounded px-2 py-0.5 bg-red-50 text-red-700">
+                        已吊销
+                      </span>
+                    ) : (
+                      <span className="text-xs rounded px-2 py-0.5 bg-emerald-50 text-emerald-700">
+                        启用中
+                      </span>
+                    )}
+                  </td>
+                  <td className="px-4 py-2 text-right">
+                    {!k.revoked_at && (
+                      <button
+                        onClick={() => revoke(k.id, k.name)}
+                        className="text-gray-400 hover:text-red-600"
+                        title="吊销"
+                      >
+                        <Trash2 className="h-4 w-4" />
+                      </button>
+                    )}
+                  </td>
+                </tr>
+              ))
+            )}
+          </tbody>
+        </table>
+      </div>
+    </div>
+  );
+}
+
+function CreateKeyForm({
+  onClose,
+  onCreated,
+}: {
+  onClose: () => void;
+  onCreated: (plaintext: string) => void;
+}) {
+  const [name, setName] = useState("");
+  const [ownerEmail, setOwnerEmail] = useState("");
+  const [tier, setTier] = useState("free");
+  const [rate, setRate] = useState(120);
+  const [busy, setBusy] = useState(false);
+  const [error, setError] = useState("");
+
+  function pickTier(t: string) {
+    setTier(t);
+    const def = TIERS.find((x) => x.key === t);
+    if (def) setRate(def.rate);
+  }
+
+  async function submit() {
+    if (!name.trim()) {
+      setError("名称不能为空");
+      return;
+    }
+    setBusy(true);
+    setError("");
+    try {
+      const res = await api.createApiKey({
+        name: name.trim(),
+        owner_email: ownerEmail.trim() || undefined,
+        tier,
+        rate_limit_per_min: rate,
+      });
+      onCreated(res.key);
+    } catch (e) {
+      setError(e instanceof ApiError ? e.message : "创建失败");
+    } finally {
+      setBusy(false);
+    }
+  }
+
+  return (
+    <div className="mb-4 bg-white border rounded-lg p-5">
+      <h3 className="font-medium text-gray-700 mb-3">签发新密钥</h3>
+      {error && <div className="mb-3 bg-red-50 text-red-700 text-sm rounded px-3 py-2">{error}</div>}
+      <div className="grid grid-cols-2 gap-4">
+        <label className="block">
+          <span className="text-xs text-gray-500">名称 *</span>
+          <input
+            className="w-full border rounded-md px-3 py-2 text-sm mt-1"
+            value={name}
+            onChange={(e) => setName(e.target.value)}
+            placeholder="例如：我的 App / 合作方 X"
+          />
+        </label>
+        <label className="block">
+          <span className="text-xs text-gray-500">负责人邮箱（可选）</span>
+          <input
+            className="w-full border rounded-md px-3 py-2 text-sm mt-1"
+            value={ownerEmail}
+            onChange={(e) => setOwnerEmail(e.target.value)}
+            placeholder="owner@example.com"
+          />
+        </label>
+        <label className="block">
+          <span className="text-xs text-gray-500">级别</span>
+          <select
+            className="w-full border rounded-md px-3 py-2 text-sm mt-1 bg-white"
+            value={tier}
+            onChange={(e) => pickTier(e.target.value)}
+          >
+            {TIERS.map((t) => (
+              <option key={t.key} value={t.key}>
+                {t.label}
+              </option>
+            ))}
+          </select>
+        </label>
+        <label className="block">
+          <span className="text-xs text-gray-500">速率上限（次/分钟）</span>
+          <input
+            type="number"
+            min={1}
+            className="w-full border rounded-md px-3 py-2 text-sm mt-1"
+            value={rate}
+            onChange={(e) => setRate(Math.max(1, parseInt(e.target.value || "1", 10)))}
+          />
+        </label>
+      </div>
+      <div className="mt-4 flex gap-2">
+        <button
+          onClick={submit}
+          disabled={busy}
+          className="px-4 py-2 rounded bg-emerald-600 text-white text-sm hover:bg-emerald-700 disabled:opacity-60"
+        >
+          生成
+        </button>
+        <button onClick={onClose} className="px-4 py-2 rounded border text-sm text-gray-600">
+          取消
+        </button>
+      </div>
+    </div>
+  );
+}

admin-frontend/src/types.ts

@@ -137,6 +137,25 @@ export interface SubmissionDetail {
   existing_product?: ProductDetail;
 }
 
+export interface ApiKeyUsage {
+  total: number;
+  today: number;
+  last_used_at?: number | null;
+}
+
+export interface ApiKey {
+  id: string;
+  name: string;
+  key_prefix: string;
+  owner_email: string | null;
+  tier: string;
+  rate_limit_per_min: number;
+  revoked_at: string | null;
+  created_by: string | null;
+  created_at: string;
+  usage: ApiKeyUsage;
+}
+
 export const FIELD_LABELS: Record<string, string> = {
   name: "名称",
   gtin: "条码",

api/cmd/admin/main.go

@@ -17,6 +17,7 @@ import (
 	"github.com/baicai2026-baicai/goods/api/internal/adminstore"
 	"github.com/baicai2026-baicai/goods/api/internal/adminweb"
 	"github.com/baicai2026-baicai/goods/api/internal/auth"
+	"github.com/baicai2026-baicai/goods/api/internal/ratelimit"
 )
 
 func getenv(key, fallback string) string {
@@ -69,7 +70,9 @@ func main() {
 	}
 
 	authn := auth.New(username, passwordHash, secret, 12*time.Hour)
-	h := adminhandler.New(adminstore.New(pool), authn, basePath, adminweb.Dist())
+	usage := ratelimit.New(getenv("OPENGOODS_REDIS_URL", "redis://localhost:6379/0"))
+	h := adminhandler.New(adminstore.New(pool), authn, basePath, adminweb.Dist()).
+		WithUsage(usage)
 
 	srv := &http.Server{
 		Addr:              addr,

api/cmd/server/main.go

@@ -12,6 +12,7 @@ import (
 	"github.com/baicai2026-baicai/goods/api/internal/config"
 	"github.com/baicai2026-baicai/goods/api/internal/handler"
 	"github.com/baicai2026-baicai/goods/api/internal/publicweb"
+	"github.com/baicai2026-baicai/goods/api/internal/ratelimit"
 	"github.com/baicai2026-baicai/goods/api/internal/store"
 )
 
@@ -31,7 +32,12 @@ func main() {
 		log.Printf("warning: database not reachable at startup: %v", err)
 	}
 
-	h := handler.New(store.New(pool), publicweb.Dist())
+	limiter := ratelimit.New(cfg.RedisURL)
+	if !limiter.Enabled() {
+		log.Print("warning: Redis not configured; public API rate limiting disabled")
+	}
+	h := handler.New(store.New(pool), publicweb.Dist()).
+		WithRateLimit(limiter, cfg.AnonRateLimitPerMin)
 
 	srv := &http.Server{
 		Addr:              cfg.Addr,

api/go.mod

@@ -5,13 +5,17 @@ go 1.23.4
 require (
 	github.com/go-chi/chi/v5 v5.1.0
 	github.com/jackc/pgx/v5 v5.7.2
+	github.com/redis/go-redis/v9 v9.18.0
 	golang.org/x/crypto v0.31.0
 )
 
 require (
+	github.com/cespare/xxhash/v2 v2.3.0 // indirect
+	github.com/dgryski/go-rendezvous v0.0.0-20200823014737-9f7001d12a5f // indirect
 	github.com/jackc/pgpassfile v1.0.0 // indirect
 	github.com/jackc/pgservicefile v0.0.0-20240606120523-5a60cdf6a761 // indirect
 	github.com/jackc/puddle/v2 v2.2.2 // indirect
+	go.uber.org/atomic v1.11.0 // indirect
 	golang.org/x/sync v0.10.0 // indirect
 	golang.org/x/text v0.21.0 // indirect
 )

api/go.sum

@@ -1,6 +1,14 @@
+github.com/bsm/ginkgo/v2 v2.12.0 h1:Ny8MWAHyOepLGlLKYmXG4IEkioBysk6GpaRTLC8zwWs=
+github.com/bsm/ginkgo/v2 v2.12.0/go.mod h1:SwYbGRRDovPVboqFv0tPTcG1sN61LM1Z4ARdbAV9g4c=
+github.com/bsm/gomega v1.27.10 h1:yeMWxP2pV2fG3FgAODIY8EiRE3dy0aeFYt4l7wh6yKA=
+github.com/bsm/gomega v1.27.10/go.mod h1:JyEr/xRbxbtgWNi8tIEVPUYZ5Dzef52k01W3YH0H+O0=
+github.com/cespare/xxhash/v2 v2.3.0 h1:UL815xU9SqsFlibzuggzjXhog7bL6oX9BbNZnL2UFvs=
+github.com/cespare/xxhash/v2 v2.3.0/go.mod h1:VGX0DQ3Q6kWi7AoAeZDth3/j3BFtOZR5XLFGgcrjCOs=
 github.com/davecgh/go-spew v1.1.0/go.mod h1:J7Y8YcW2NihsgmVo/mv3lAwl/skON4iLHjSsI+c5H38=
 github.com/davecgh/go-spew v1.1.1 h1:vj9j/u1bqnvCEfJOwUhtlOARqs3+rkHYY13jYWTU97c=
 github.com/davecgh/go-spew v1.1.1/go.mod h1:J7Y8YcW2NihsgmVo/mv3lAwl/skON4iLHjSsI+c5H38=
+github.com/dgryski/go-rendezvous v0.0.0-20200823014737-9f7001d12a5f h1:lO4WD4F/rVNCu3HqELle0jiPLLBs70cWOduZpkS1E78=
+github.com/dgryski/go-rendezvous v0.0.0-20200823014737-9f7001d12a5f/go.mod h1:cuUVRXasLTGF7a8hSLbxyZXjz+1KgoB3wDUb6vlszIc=
 github.com/go-chi/chi/v5 v5.1.0 h1:acVI1TYaD+hhedDJ3r54HyA6sExp3HfXq7QWEEY/xMw=
 github.com/go-chi/chi/v5 v5.1.0/go.mod h1:DslCQbL2OYiznFReuXYUmQ2hGd1aDpCnlMNITLSKoi8=
 github.com/jackc/pgpassfile v1.0.0 h1:/6Hmqy13Ss2zCq62VdNG8tM1wchn8zjSGOBJ6icpsIM=
@@ -11,13 +19,21 @@ github.com/jackc/pgx/v5 v5.7.2 h1:mLoDLV6sonKlvjIEsV56SkWNCnuNv531l94GaIzO+XI=
 github.com/jackc/pgx/v5 v5.7.2/go.mod h1:ncY89UGWxg82EykZUwSpUKEfccBGGYq1xjrOpsbsfGQ=
 github.com/jackc/puddle/v2 v2.2.2 h1:PR8nw+E/1w0GLuRFSmiioY6UooMp6KJv0/61nB7icHo=
 github.com/jackc/puddle/v2 v2.2.2/go.mod h1:vriiEXHvEE654aYKXXjOvZM39qJ0q+azkZFrfEOc3H4=
+github.com/klauspost/cpuid/v2 v2.0.9 h1:lgaqFMSdTdQYdZ04uHyN2d/eKdOMyi2YLSvlQIBFYa4=
+github.com/klauspost/cpuid/v2 v2.0.9/go.mod h1:FInQzS24/EEf25PyTYn52gqo7WaD8xa0213Md/qVLRg=
 github.com/pmezard/go-difflib v1.0.0 h1:4DBwDE0NGyQoBHbLQYPwSUPoCMWR5BEzIk/f1lZbAQM=
 github.com/pmezard/go-difflib v1.0.0/go.mod h1:iKH77koFhYxTK1pcRnkKkqfTogsbg7gZNVY4sRDYZ/4=
+github.com/redis/go-redis/v9 v9.18.0 h1:pMkxYPkEbMPwRdenAzUNyFNrDgHx9U+DrBabWNfSRQs=
+github.com/redis/go-redis/v9 v9.18.0/go.mod h1:k3ufPphLU5YXwNTUcCRXGxUoF1fqxnhFQmscfkCoDA0=
 github.com/stretchr/objx v0.1.0/go.mod h1:HFkY916IF+rwdDfMAkV7OtwuqBVzrE8GR6GFx+wExME=
 github.com/stretchr/testify v1.3.0/go.mod h1:M5WIy9Dh21IEIfnGCwXGc5bZfKNJtfHm1UVUgZn+9EI=
 github.com/stretchr/testify v1.7.0/go.mod h1:6Fq8oRcR53rry900zMqJjRRixrwX3KX962/h/Wwjteg=
 github.com/stretchr/testify v1.8.1 h1:w7B6lhMri9wdJUVmEZPGGhZzrYTPvgJArz7wNPgYKsk=
 github.com/stretchr/testify v1.8.1/go.mod h1:w2LPCIKwWwSfY2zedu0+kehJoqGctiVI29o6fzry7u4=
+github.com/zeebo/xxh3 v1.0.2 h1:xZmwmqxHZA8AI603jOQ0tMqmBr9lPeFwGg6d+xy9DC0=
+github.com/zeebo/xxh3 v1.0.2/go.mod h1:5NWz9Sef7zIDm2JHfFlcQvNekmcEl9ekUZQQKCYaDcA=
+go.uber.org/atomic v1.11.0 h1:ZvwS0R+56ePWxUNi+Atn9dWONBPp/AUETXlHW0DxSjE=
+go.uber.org/atomic v1.11.0/go.mod h1:LUxbIzbOniOlMKjJjyPfpl4v+PKK2cNJn91OQbhoJI0=
 golang.org/x/crypto v0.31.0 h1:ihbySMvVjLAeSH1IbfcRTkD/iNscyz8rGzjF/E5hV6U=
 golang.org/x/crypto v0.31.0/go.mod h1:kDsLvtWBEx7MV9tJOj9bnXsPbxwJQ6csT/x4KIN4Ssk=
 golang.org/x/sync v0.10.0 h1:3NQrjDixjgGwUOCaF8w2+VYHv0Ve/vGYSbdkTa98gmQ=

api/internal/adminhandler/apikey.go

@@ -0,0 +1,70 @@
+package adminhandler
+
+import (
+	"encoding/json"
+	"net/http"
+	"strings"
+
+	"github.com/go-chi/chi/v5"
+
+	"github.com/baicai2026-baicai/goods/api/internal/adminstore"
+	"github.com/baicai2026-baicai/goods/api/internal/auth"
+	"github.com/baicai2026-baicai/goods/api/internal/ratelimit"
+)
+
+// apiKeyView is an issued key plus its usage counters.
+type apiKeyView struct {
+	adminstore.APIKeyRow
+	Usage ratelimit.UsageStat `json:"usage"`
+}
+
+// ListAPIKeys returns all issued keys with usage stats merged in.
+func (h *Handler) ListAPIKeys(w http.ResponseWriter, r *http.Request) {
+	keys, err := h.store.ListAPIKeys(r.Context())
+	if h.handleErr(w, err) {
+		return
+	}
+	views := make([]apiKeyView, 0, len(keys))
+	for _, k := range keys {
+		v := apiKeyView{APIKeyRow: k}
+		if h.usage != nil {
+			v.Usage = h.usage.Usage(r.Context(), k.ID)
+		}
+		views = append(views, v)
+	}
+	writeJSON(w, http.StatusOK, map[string]any{"items": views})
+}
+
+// CreateAPIKey issues a new key and returns its plaintext exactly once.
+func (h *Handler) CreateAPIKey(w http.ResponseWriter, r *http.Request) {
+	var in adminstore.APIKeyInput
+	if err := json.NewDecoder(r.Body).Decode(&in); err != nil {
+		writeError(w, http.StatusBadRequest, "bad_request", "invalid body")
+		return
+	}
+	if strings.TrimSpace(in.Name) == "" {
+		writeError(w, http.StatusBadRequest, "bad_request", "名称不能为空")
+		return
+	}
+	if in.Tier != "" && in.Tier != "free" && in.Tier != "partner" && in.Tier != "internal" {
+		writeError(w, http.StatusBadRequest, "bad_request", "tier 取值无效")
+		return
+	}
+	plaintext, row, err := h.store.CreateAPIKey(r.Context(), in, auth.UserFrom(r.Context()))
+	if h.handleErr(w, err) {
+		return
+	}
+	writeJSON(w, http.StatusCreated, map[string]any{
+		"key":     plaintext,
+		"item":    row,
+		"warning": "请立即复制保存此密钥，它只显示这一次，无法再次查看。",
+	})
+}
+
+// RevokeAPIKey disables a key. Subsequent requests with it are rejected.
+func (h *Handler) RevokeAPIKey(w http.ResponseWriter, r *http.Request) {
+	if err := h.store.RevokeAPIKey(r.Context(), chi.URLParam(r, "id")); h.handleErr(w, err) {
+		return
+	}
+	writeJSON(w, http.StatusOK, map[string]string{"status": "revoked"})
+}

api/internal/adminhandler/handler.go

@@ -16,6 +16,7 @@ import (
 	"github.com/baicai2026-baicai/goods/api/internal/adminstore"
 	"github.com/baicai2026-baicai/goods/api/internal/auth"
 	"github.com/baicai2026-baicai/goods/api/internal/gtin"
+	"github.com/baicai2026-baicai/goods/api/internal/ratelimit"
 )
 
 // Handler holds the admin dependencies.
@@ -25,6 +26,7 @@ type Handler struct {
 	basePath    string
 	spa         fs.FS
 	submitLimit *rateLimiter
+	usage       *ratelimit.Limiter
 }
 
 // New constructs an admin Handler. basePath is e.g. "/ping" (no trailing slash).
@@ -39,6 +41,13 @@ func New(store *adminstore.Store, authn *auth.Authenticator, basePath string, sp
 	}
 }
 
+// WithUsage attaches a Redis-backed limiter used to read per-key usage counters
+// for the API-key management view. Optional; without it usage shows as zero.
+func (h *Handler) WithUsage(l *ratelimit.Limiter) *Handler {
+	h.usage = l
+	return h
+}
+
 // Router builds the HTTP handler.
 func (h *Handler) Router() http.Handler {
 	r := chi.NewRouter()
@@ -73,6 +82,10 @@ func (h *Handler) Router() http.Handler {
 			r.Get("/api/submissions/{id}", h.GetSubmission)
 			r.Post("/api/submissions/{id}/approve", h.ApproveSubmission)
 			r.Post("/api/submissions/{id}/reject", h.RejectSubmission)
+
+			r.Get("/api/keys", h.ListAPIKeys)
+			r.Post("/api/keys", h.CreateAPIKey)
+			r.Delete("/api/keys/{id}", h.RevokeAPIKey)
 		})
 
 		r.Handle("/*", http.HandlerFunc(h.serveSPA))

api/internal/adminstore/apikey.go

@@ -0,0 +1,130 @@
+package adminstore
+
+import (
+	"context"
+	"errors"
+	"strings"
+	"time"
+
+	"github.com/jackc/pgx/v5/pgconn"
+
+	"github.com/baicai2026-baicai/goods/api/internal/apikey"
+)
+
+// APIKeyRow is an admin-facing view of an issued API key (never the secret).
+type APIKeyRow struct {
+	ID              string  `json:"id"`
+	Name            string  `json:"name"`
+	KeyPrefix       string  `json:"key_prefix"`
+	OwnerEmail      *string `json:"owner_email"`
+	Tier            string  `json:"tier"`
+	RateLimitPerMin int     `json:"rate_limit_per_min"`
+	RevokedAt       *string `json:"revoked_at"`
+	CreatedBy       *string `json:"created_by"`
+	CreatedAt       string  `json:"created_at"`
+}
+
+// APIKeyInput holds the fields accepted when issuing a key.
+type APIKeyInput struct {
+	Name            string `json:"name"`
+	OwnerEmail      string `json:"owner_email"`
+	Tier            string `json:"tier"`
+	RateLimitPerMin int    `json:"rate_limit_per_min"`
+}
+
+// CreateAPIKey issues a new key, returning the one-time plaintext alongside the
+// stored row. Only the SHA-256 hash and a short display prefix are persisted.
+func (s *Store) CreateAPIKey(ctx context.Context, in APIKeyInput, createdBy string) (plaintext string, row APIKeyRow, err error) {
+	tier := in.Tier
+	if tier == "" {
+		tier = "free"
+	}
+	rate := in.RateLimitPerMin
+	if rate <= 0 {
+		rate = 120
+	}
+	var owner *string
+	if e := strings.TrimSpace(in.OwnerEmail); e != "" {
+		owner = &e
+	}
+
+	key, hash, prefix, err := apikey.Generate()
+	if err != nil {
+		return "", row, err
+	}
+
+	var revoked, created *time.Time
+	var createdByOut *string
+	err = s.pool.QueryRow(ctx, `
+INSERT INTO api_key (name, key_prefix, key_hash, owner_email, tier, rate_limit_per_min, created_by)
+VALUES ($1, $2, $3, $4, $5, $6, $7)
+RETURNING id, name, key_prefix, owner_email, tier, rate_limit_per_min, revoked_at, created_by, created_at`,
+		strings.TrimSpace(in.Name), prefix, hash, owner, tier, rate, createdBy,
+	).Scan(&row.ID, &row.Name, &row.KeyPrefix, &row.OwnerEmail, &row.Tier,
+		&row.RateLimitPerMin, &revoked, &createdByOut, &created)
+	if err != nil {
+		return "", row, err
+	}
+	row.CreatedBy = createdByOut
+	if created != nil {
+		row.CreatedAt = created.Format(time.RFC3339)
+	}
+	return key, row, nil
+}
+
+// ListAPIKeys returns all keys (active first, newest first).
+func (s *Store) ListAPIKeys(ctx context.Context) ([]APIKeyRow, error) {
+	rows, err := s.pool.Query(ctx, `
+SELECT id, name, key_prefix, owner_email, tier, rate_limit_per_min, revoked_at, created_by, created_at
+FROM api_key
+ORDER BY (revoked_at IS NULL) DESC, created_at DESC`)
+	if err != nil {
+		return nil, err
+	}
+	defer rows.Close()
+	out := []APIKeyRow{}
+	for rows.Next() {
+		var r APIKeyRow
+		var revoked, created *time.Time
+		if err := rows.Scan(&r.ID, &r.Name, &r.KeyPrefix, &r.OwnerEmail, &r.Tier,
+			&r.RateLimitPerMin, &revoked, &r.CreatedBy, &created); err != nil {
+			return nil, err
+		}
+		if revoked != nil {
+			v := revoked.Format(time.RFC3339)
+			r.RevokedAt = &v
+		}
+		if created != nil {
+			r.CreatedAt = created.Format(time.RFC3339)
+		}
+		out = append(out, r)
+	}
+	return out, rows.Err()
+}
+
+// RevokeAPIKey marks a key revoked. Revoking an already-revoked or missing key
+// returns ErrNotFound.
+func (s *Store) RevokeAPIKey(ctx context.Context, id string) error {
+	tag, err := s.pool.Exec(ctx,
+		"UPDATE api_key SET revoked_at = now() WHERE id = $1 AND revoked_at IS NULL", id)
+	if err != nil {
+		if isInvalidUUID(err) {
+			return ErrNotFound
+		}
+		return err
+	}
+	if tag.RowsAffected() == 0 {
+		return ErrNotFound
+	}
+	return nil
+}
+
+// isInvalidUUID reports whether err is a Postgres invalid-UUID-text error,
+// which happens when a non-UUID id is supplied.
+func isInvalidUUID(err error) bool {
+	var pgErr *pgconn.PgError
+	if errors.As(err, &pgErr) {
+		return pgErr.Code == "22P02"
+	}
+	return false
+}

api/internal/apikey/apikey.go

@@ -0,0 +1,49 @@
+// Package apikey handles generation and hashing of public-API keys.
+//
+// A key looks like "og_live_<random>". Only the SHA-256 hash is ever persisted;
+// the plaintext is returned once at creation time and cannot be recovered.
+package apikey
+
+import (
+	"crypto/rand"
+	"crypto/sha256"
+	"encoding/base64"
+	"encoding/hex"
+	"strings"
+)
+
+// Prefix is the human-readable scheme prefix every key carries.
+const Prefix = "og_live_"
+
+// prefixLen is how many leading characters (including Prefix) are stored in
+// api_key.key_prefix for identifying a key without revealing its secret.
+const prefixLen = 12
+
+// Generate returns a new random key (plaintext), its SHA-256 hash, and a short
+// display prefix. The plaintext must be shown to the caller exactly once.
+func Generate() (key, hash, displayPrefix string, err error) {
+	buf := make([]byte, 24)
+	if _, err = rand.Read(buf); err != nil {
+		return "", "", "", err
+	}
+	// URL-safe, no padding => stable, copy-pasteable token body.
+	body := base64.RawURLEncoding.EncodeToString(buf)
+	key = Prefix + body
+	hash = Hash(key)
+	displayPrefix = key
+	if len(displayPrefix) > prefixLen {
+		displayPrefix = displayPrefix[:prefixLen]
+	}
+	return key, hash, displayPrefix, nil
+}
+
+// Hash returns the hex-encoded SHA-256 of a key, used for storage and lookup.
+func Hash(key string) string {
+	sum := sha256.Sum256([]byte(strings.TrimSpace(key)))
+	return hex.EncodeToString(sum[:])
+}
+
+// Looks like a key issued by this service (cheap pre-check before hashing).
+func IsWellFormed(key string) bool {
+	return strings.HasPrefix(key, Prefix) && len(key) > len(Prefix)+8
+}

api/internal/apikey/apikey_test.go

@@ -0,0 +1,60 @@
+package apikey
+
+import "testing"
+
+func TestGenerate(t *testing.T) {
+	key, hash, prefix, err := Generate()
+	if err != nil {
+		t.Fatalf("Generate: %v", err)
+	}
+	if !IsWellFormed(key) {
+		t.Fatalf("generated key not well-formed: %q", key)
+	}
+	if Hash(key) != hash {
+		t.Fatalf("Hash(key) != returned hash")
+	}
+	if len(prefix) != prefixLen || key[:prefixLen] != prefix {
+		t.Fatalf("prefix %q not a %d-char prefix of key %q", prefix, prefixLen, key)
+	}
+	if len(hash) != 64 {
+		t.Fatalf("hash not hex sha-256: %q", hash)
+	}
+}
+
+func TestGenerateUnique(t *testing.T) {
+	seen := map[string]bool{}
+	for i := 0; i < 100; i++ {
+		k, _, _, err := Generate()
+		if err != nil {
+			t.Fatal(err)
+		}
+		if seen[k] {
+			t.Fatalf("duplicate key generated: %q", k)
+		}
+		seen[k] = true
+	}
+}
+
+func TestHashStableAndTrimmed(t *testing.T) {
+	if Hash("og_live_abc") != Hash("  og_live_abc  ") {
+		t.Fatal("Hash should ignore surrounding whitespace")
+	}
+	if Hash("a") == Hash("b") {
+		t.Fatal("distinct inputs must hash differently")
+	}
+}
+
+func TestIsWellFormed(t *testing.T) {
+	cases := map[string]bool{
+		"og_live_abcdefghijkl": true,  // body longer than 8 chars
+		"og_live_":             false, // empty body
+		"og_live_abc":          false, // body too short
+		"nope_abcdefghijkl":    false, // wrong prefix
+		"":                     false,
+	}
+	for in, want := range cases {
+		if got := IsWellFormed(in); got != want {
+			t.Errorf("IsWellFormed(%q) = %v, want %v", in, got, want)
+		}
+	}
+}

api/internal/config/config.go

@@ -2,26 +2,38 @@ package config
 
 import (
 	"os"
+	"strconv"
 )
 
 // Config holds runtime configuration for the OpenGoods API server.
 // Values are read from environment variables with sensible defaults so the
 // server can boot in a local Docker Compose setup without extra configuration.
 type Config struct {
-	Addr        string
-	DatabaseURL string
-	RedisURL    string
+	Addr                string
+	DatabaseURL         string
+	RedisURL            string
+	AnonRateLimitPerMin int
 }
 
 // Load reads configuration from the environment.
 func Load() Config {
 	return Config{
-		Addr:        getenv("OPENGOODS_ADDR", ":8080"),
-		DatabaseURL: getenv("OPENGOODS_DATABASE_URL", "postgres://opengoods:opengoods@localhost:5432/opengoods?sslmode=disable"),
-		RedisURL:    getenv("OPENGOODS_REDIS_URL", "redis://localhost:6379/0"),
+		Addr:                getenv("OPENGOODS_ADDR", ":8080"),
+		DatabaseURL:         getenv("OPENGOODS_DATABASE_URL", "postgres://opengoods:opengoods@localhost:5432/opengoods?sslmode=disable"),
+		RedisURL:            getenv("OPENGOODS_REDIS_URL", "redis://localhost:6379/0"),
+		AnonRateLimitPerMin: getenvInt("OPENGOODS_ANON_RATE_LIMIT_PER_MIN", 60),
 	}
 }
 
+func getenvInt(key string, fallback int) int {
+	if v, ok := os.LookupEnv(key); ok && v != "" {
+		if n, err := strconv.Atoi(v); err == nil && n > 0 {
+			return n
+		}
+	}
+	return fallback
+}
+
 func getenv(key, fallback string) string {
 	if v, ok := os.LookupEnv(key); ok && v != "" {
 		return v

api/internal/handler/handler.go

@@ -5,6 +5,7 @@
 package handler
 
 import (
+	_ "embed"
 	"encoding/json"
 	"errors"
 	"io/fs"
@@ -15,26 +16,48 @@ import (
 	"github.com/go-chi/chi/v5"
 	"github.com/go-chi/chi/v5/middleware"
 
+	"github.com/baicai2026-baicai/goods/api/internal/ratelimit"
 	"github.com/baicai2026-baicai/goods/api/internal/store"
 )
 
+//go:embed openapi.json
+var openAPISpec []byte
+
 // APIVersion is the current public API version prefix.
 const APIVersion = "v1"
 
 const (
 	defaultPageSize = 20
 	maxPageSize     = 100
+
+	// defaultAnonLimit is the per-minute request budget for unauthenticated
+	// callers (identified by client IP) when none is configured.
+	defaultAnonLimit = 60
 )
 
 // Handler holds dependencies shared by the HTTP routes.
 type Handler struct {
-	store *store.Store
-	spa   fs.FS
+	store     *store.Store
+	spa       fs.FS
+	limiter   *ratelimit.Limiter
+	anonLimit int
 }
 
 // New constructs a Handler backed by the given store. spa may be nil (JSON-only).
+// Rate limiting is disabled until WithRateLimit is called.
 func New(s *store.Store, spa fs.FS) *Handler {
-	return &Handler{store: s, spa: spa}
+	return &Handler{store: s, spa: spa, anonLimit: defaultAnonLimit}
+}
+
+// WithRateLimit attaches a Redis-backed limiter and the anonymous per-minute
+// budget, enabling rate limiting + usage tracking on the public API routes.
+// A non-positive anonPerMin keeps the default.
+func (h *Handler) WithRateLimit(l *ratelimit.Limiter, anonPerMin int) *Handler {
+	h.limiter = l
+	if anonPerMin > 0 {
+		h.anonLimit = anonPerMin
+	}
+	return h
 }
 
 // Router builds the top-level HTTP handler with middleware and routes mounted.
@@ -47,16 +70,22 @@ func (h *Handler) Router() http.Handler {
 	r.Get("/healthz", h.Healthz)
 
 	r.Route("/api/"+APIVersion, func(r chi.Router) {
-		r.Route("/products", func(r chi.Router) {
-			r.Get("/barcode/{gtin}", h.ProductByBarcode)
-			r.Get("/search", h.SearchProducts)
-			r.Get("/{id}", h.ProductByID)
-			r.Get("/{id}/nutriments", h.ProductNutriments)
-			r.Get("/{id}/msrp", h.ProductMSRP)
+		// Machine-readable spec; not rate limited so tooling can always fetch it.
+		r.Get("/openapi.json", h.OpenAPI)
+
+		r.Group(func(r chi.Router) {
+			r.Use(h.rateLimit)
+			r.Route("/products", func(r chi.Router) {
+				r.Get("/barcode/{gtin}", h.ProductByBarcode)
+				r.Get("/search", h.SearchProducts)
+				r.Get("/{id}", h.ProductByID)
+				r.Get("/{id}/nutriments", h.ProductNutriments)
+				r.Get("/{id}/msrp", h.ProductMSRP)
+			})
+			r.Get("/brands", h.ListBrands)
+			r.Get("/categories", h.ListCategories)
+			r.Get("/sources/{id}", h.SourceByID)
 		})
-		r.Get("/brands", h.ListBrands)
-		r.Get("/categories", h.ListCategories)
-		r.Get("/sources/{id}", h.SourceByID)
 	})
 
 	// Public SPA (homepage + search + contribute). API routes above take
@@ -93,6 +122,12 @@ func (h *Handler) Healthz(w http.ResponseWriter, r *http.Request) {
 	writeJSON(w, http.StatusOK, map[string]string{"status": "ok"})
 }
 
+// OpenAPI serves the embedded OpenAPI 3 specification for the public API.
+func (h *Handler) OpenAPI(w http.ResponseWriter, r *http.Request) {
+	w.Header().Set("Content-Type", "application/json; charset=utf-8")
+	_, _ = w.Write(openAPISpec)
+}
+
 // ProductByBarcode returns a product by its GTIN.
 func (h *Handler) ProductByBarcode(w http.ResponseWriter, r *http.Request) {
 	p, err := h.store.ProductByGTIN(r.Context(), chi.URLParam(r, "gtin"))
@@ -111,13 +146,19 @@ func (h *Handler) ProductByID(w http.ResponseWriter, r *http.Request) {
 	writeJSON(w, http.StatusOK, p)
 }
 
-// SearchProducts runs a fuzzy name search with optional category filter + paging.
+// SearchProducts runs a trigram-fuzzy name search with optional
+// category/brand/country filters, ranked by relevance, plus paging.
 func (h *Handler) SearchProducts(w http.ResponseWriter, r *http.Request) {
-	q := r.URL.Query().Get("q")
-	category := r.URL.Query().Get("category")
+	qv := r.URL.Query()
+	filters := store.SearchFilters{
+		Query:    strings.TrimSpace(qv.Get("q")),
+		Category: strings.TrimSpace(qv.Get("category")),
+		Brand:    strings.TrimSpace(qv.Get("brand")),
+		Country:  strings.TrimSpace(qv.Get("country")),
+	}
 	page, size := pageParams(r)
 
-	items, total, err := h.store.SearchProducts(r.Context(), q, category, size, (page-1)*size)
+	items, total, err := h.store.SearchProducts(r.Context(), filters, size, (page-1)*size)
 	if h.handleErr(w, r, err) {
 		return
 	}

api/internal/handler/handler_db_test.go

@@ -116,6 +116,101 @@ func TestSearchProducts(t *testing.T) {
 	}
 }
 
+func TestSearchFuzzyAndFilters(t *testing.T) {
+	h, _ := newTestHandler(t)
+	ctx := context.Background()
+	dsn := os.Getenv("OPENGOODS_DATABASE_URL")
+	if dsn == "" {
+		dsn = "postgres://opengoods:opengoods@localhost:5432/opengoods?sslmode=disable"
+	}
+	pool, err := pgxpool.New(ctx, dsn)
+	if err != nil {
+		t.Skipf("no database: %v", err)
+	}
+	defer pool.Close()
+
+	_, err = pool.Exec(ctx,
+		"INSERT INTO brand (name, normalized_name) VALUES ('ZZ Test Brand','zz test brand') ON CONFLICT DO NOTHING")
+	if err != nil {
+		t.Fatalf("seed brand: %v", err)
+	}
+	_, err = pool.Exec(ctx, `
+		INSERT INTO product (name, brand_id, country_of_origin, quality_score, status)
+		VALUES ('ZZ Hazelnut Chocolate', (SELECT id FROM brand WHERE name='ZZ Test Brand'), 'Testland', 0.5, 'active')`)
+	if err != nil {
+		t.Fatalf("seed product: %v", err)
+	}
+	t.Cleanup(func() {
+		_, _ = pool.Exec(ctx, "DELETE FROM product WHERE name='ZZ Hazelnut Chocolate'")
+		_, _ = pool.Exec(ctx, "DELETE FROM brand WHERE name='ZZ Test Brand'")
+	})
+
+	decode := func(path string) []store.ProductSummary {
+		rec := doGET(t, h, path)
+		if rec.Code != http.StatusOK {
+			t.Fatalf("%s -> status %d", path, rec.Code)
+		}
+		var body struct {
+			Items []store.ProductSummary `json:"items"`
+		}
+		if err := json.NewDecoder(rec.Body).Decode(&body); err != nil {
+			t.Fatal(err)
+		}
+		return body.Items
+	}
+	has := func(items []store.ProductSummary, name string) *store.ProductSummary {
+		for i := range items {
+			if items[i].Name == name {
+				return &items[i]
+			}
+		}
+		return nil
+	}
+
+	// Typo "choclate" should fuzzy-match via word_similarity and carry a score.
+	got := has(decode("/api/"+APIVersion+"/products/search?q=choclate"), "ZZ Hazelnut Chocolate")
+	if got == nil {
+		t.Fatal("fuzzy query 'choclate' did not match 'ZZ Hazelnut Chocolate'")
+	}
+	if got.Score == nil || *got.Score <= 0 {
+		t.Fatalf("expected positive fuzzy score, got %v", got.Score)
+	}
+
+	// Brand filter.
+	if has(decode("/api/"+APIVersion+"/products/search?brand=ZZ+Test+Brand"), "ZZ Hazelnut Chocolate") == nil {
+		t.Fatal("brand filter did not return the product")
+	}
+	// Country filter (case-insensitive prefix).
+	if has(decode("/api/"+APIVersion+"/products/search?country=test"), "ZZ Hazelnut Chocolate") == nil {
+		t.Fatal("country filter did not return the product")
+	}
+	// Non-matching country excludes it.
+	if has(decode("/api/"+APIVersion+"/products/search?country=france"), "ZZ Hazelnut Chocolate") != nil {
+		t.Fatal("country filter 'france' should not return the product")
+	}
+}
+
+func TestOpenAPISpec(t *testing.T) {
+	h, _ := newTestHandler(t)
+	rec := doGET(t, h, "/api/"+APIVersion+"/openapi.json")
+	if rec.Code != http.StatusOK {
+		t.Fatalf("status = %d", rec.Code)
+	}
+	var spec struct {
+		OpenAPI string         `json:"openapi"`
+		Paths   map[string]any `json:"paths"`
+	}
+	if err := json.NewDecoder(rec.Body).Decode(&spec); err != nil {
+		t.Fatalf("openapi.json is not valid JSON: %v", err)
+	}
+	if spec.OpenAPI == "" || len(spec.Paths) == 0 {
+		t.Fatalf("unexpected spec: %+v", spec)
+	}
+	if _, ok := spec.Paths["/products/search"]; !ok {
+		t.Fatal("spec missing /products/search path")
+	}
+}
+
 func TestListCategories(t *testing.T) {
 	h, _ := newTestHandler(t)
 	rec := doGET(t, h, "/api/"+APIVersion+"/categories")

api/internal/handler/middleware.go

@@ -0,0 +1,90 @@
+package handler
+
+import (
+	"context"
+	"errors"
+	"net"
+	"net/http"
+	"strconv"
+	"strings"
+	"time"
+
+	"github.com/baicai2026-baicai/goods/api/internal/apikey"
+	"github.com/baicai2026-baicai/goods/api/internal/store"
+)
+
+type ctxKey int
+
+const apiKeyIDKey ctxKey = 0
+
+// rateLimit authenticates an optional API key and enforces a per-minute budget
+// on the public API. Anonymous callers are limited by client IP at a lower
+// budget; a valid key raises the budget and attributes usage. An API key that
+// is present but invalid or revoked is rejected with 401. Rate-limit headers
+// are set on every response; over-budget callers get 429 + Retry-After.
+func (h *Handler) rateLimit(next http.Handler) http.Handler {
+	return http.HandlerFunc(func(w http.ResponseWriter, r *http.Request) {
+		id := "ip:" + clientIP(r)
+		limit := h.anonLimit
+		keyID := ""
+
+		if raw := presentedKey(r); raw != "" {
+			if !apikey.IsWellFormed(raw) {
+				writeError(w, r, http.StatusUnauthorized, "invalid_api_key", "API key 格式无效")
+				return
+			}
+			k, err := h.store.APIKeyByHash(r.Context(), apikey.Hash(raw))
+			if errors.Is(err, store.ErrNotFound) {
+				writeError(w, r, http.StatusUnauthorized, "invalid_api_key", "API key 无效或已吊销")
+				return
+			}
+			if err != nil {
+				writeError(w, r, http.StatusInternalServerError, "internal_error", "internal server error")
+				return
+			}
+			keyID = k.ID
+			limit = k.RateLimitPerMin
+			id = "key:" + k.ID
+		}
+
+		res := h.limiter.Allow(r.Context(), id, limit, time.Minute)
+		w.Header().Set("X-RateLimit-Limit", strconv.Itoa(res.Limit))
+		w.Header().Set("X-RateLimit-Remaining", strconv.Itoa(res.Remaining))
+		w.Header().Set("X-RateLimit-Reset", strconv.FormatInt(res.ResetUnix, 10))
+		if !res.Allowed {
+			retry := res.ResetUnix - time.Now().Unix()
+			if retry < 1 {
+				retry = 1
+			}
+			w.Header().Set("Retry-After", strconv.FormatInt(retry, 10))
+			writeError(w, r, http.StatusTooManyRequests, "rate_limited", "请求过于频繁，请稍后再试")
+			return
+		}
+
+		if keyID != "" {
+			h.limiter.RecordUsage(r.Context(), keyID)
+			next.ServeHTTP(w, r.WithContext(context.WithValue(r.Context(), apiKeyIDKey, keyID)))
+			return
+		}
+		next.ServeHTTP(w, r)
+	})
+}
+
+// presentedKey extracts an API key from the X-API-Key header or a Bearer token.
+func presentedKey(r *http.Request) string {
+	if v := strings.TrimSpace(r.Header.Get("X-API-Key")); v != "" {
+		return v
+	}
+	if v := r.Header.Get("Authorization"); strings.HasPrefix(v, "Bearer ") {
+		return strings.TrimSpace(strings.TrimPrefix(v, "Bearer "))
+	}
+	return ""
+}
+
+// clientIP returns the caller IP, preferring chi's RealIP-normalized RemoteAddr.
+func clientIP(r *http.Request) string {
+	if host, _, err := net.SplitHostPort(r.RemoteAddr); err == nil {
+		return host
+	}
+	return r.RemoteAddr
+}

api/internal/handler/middleware_db_test.go

@@ -0,0 +1,120 @@
+package handler
+
+import (
+	"context"
+	"fmt"
+	"net/http"
+	"net/http/httptest"
+	"os"
+	"testing"
+	"time"
+
+	"github.com/jackc/pgx/v5/pgxpool"
+
+	"github.com/baicai2026-baicai/goods/api/internal/apikey"
+	"github.com/baicai2026-baicai/goods/api/internal/ratelimit"
+	"github.com/baicai2026-baicai/goods/api/internal/store"
+)
+
+// newRateLimitedHandler builds a handler backed by the test DB and a live Redis
+// limiter, plus a freshly issued API key with the given per-minute limit. It
+// skips when either backend is unavailable.
+func newRateLimitedHandler(t *testing.T, keyLimit int) (h *Handler, plaintextKey string) {
+	t.Helper()
+	dsn := os.Getenv("OPENGOODS_DATABASE_URL")
+	if dsn == "" {
+		dsn = "postgres://opengoods:opengoods@localhost:5432/opengoods?sslmode=disable"
+	}
+	ctx, cancel := context.WithTimeout(context.Background(), 3*time.Second)
+	defer cancel()
+	pool, err := pgxpool.New(ctx, dsn)
+	if err != nil {
+		t.Skipf("no database: %v", err)
+	}
+	if err := pool.Ping(ctx); err != nil {
+		pool.Close()
+		t.Skipf("database not reachable: %v", err)
+	}
+	var hasTable bool
+	if err := pool.QueryRow(ctx, "SELECT to_regclass('public.api_key') IS NOT NULL").Scan(&hasTable); err != nil || !hasTable {
+		pool.Close()
+		t.Skip("migrations not applied (api_key missing)")
+	}
+
+	redisURL := os.Getenv("OPENGOODS_REDIS_URL")
+	if redisURL == "" {
+		redisURL = "redis://localhost:6379/0"
+	}
+	limiter := ratelimit.New(redisURL)
+	pingCtx, pingCancel := context.WithTimeout(context.Background(), time.Second)
+	defer pingCancel()
+	if err := limiter.Ping(pingCtx); err != nil {
+		pool.Close()
+		t.Skipf("redis not reachable: %v", err)
+	}
+
+	key, hash, prefix, err := apikey.Generate()
+	if err != nil {
+		pool.Close()
+		t.Fatal(err)
+	}
+	name := fmt.Sprintf("test-key-%d", time.Now().UnixNano())
+	if _, err := pool.Exec(context.Background(),
+		`INSERT INTO api_key (name, key_prefix, key_hash, rate_limit_per_min) VALUES ($1,$2,$3,$4)`,
+		name, prefix, hash, keyLimit); err != nil {
+		pool.Close()
+		t.Fatalf("insert api_key: %v", err)
+	}
+	t.Cleanup(func() {
+		_, _ = pool.Exec(context.Background(), "DELETE FROM api_key WHERE key_hash=$1", hash)
+		pool.Close()
+	})
+
+	return New(store.New(pool), nil).WithRateLimit(limiter, 60), key
+}
+
+func TestRateLimitHeadersAndKeyAuth(t *testing.T) {
+	h, key := newRateLimitedHandler(t, 100)
+	req := httptest.NewRequest(http.MethodGet, "/api/"+APIVersion+"/categories", nil)
+	req.Header.Set("X-API-Key", key)
+	rec := httptest.NewRecorder()
+	h.Router().ServeHTTP(rec, req)
+
+	if rec.Code != http.StatusOK {
+		t.Fatalf("status = %d, body = %s", rec.Code, rec.Body.String())
+	}
+	if got := rec.Header().Get("X-RateLimit-Limit"); got != "100" {
+		t.Fatalf("X-RateLimit-Limit = %q, want 100 (key limit)", got)
+	}
+	if rec.Header().Get("X-RateLimit-Remaining") == "" {
+		t.Fatal("missing X-RateLimit-Remaining header")
+	}
+}
+
+func TestInvalidKeyRejected(t *testing.T) {
+	h, _ := newRateLimitedHandler(t, 100)
+	req := httptest.NewRequest(http.MethodGet, "/api/"+APIVersion+"/categories", nil)
+	req.Header.Set("X-API-Key", "og_live_thiskeydoesnotexist123456")
+	rec := httptest.NewRecorder()
+	h.Router().ServeHTTP(rec, req)
+	if rec.Code != http.StatusUnauthorized {
+		t.Fatalf("status = %d, want 401; body = %s", rec.Code, rec.Body.String())
+	}
+}
+
+func TestRateLimitExceeded(t *testing.T) {
+	h, key := newRateLimitedHandler(t, 1)
+	do := func() int {
+		req := httptest.NewRequest(http.MethodGet, "/api/"+APIVersion+"/categories", nil)
+		req.Header.Set("X-API-Key", key)
+		rec := httptest.NewRecorder()
+		h.Router().ServeHTTP(rec, req)
+		return rec.Code
+	}
+	if code := do(); code != http.StatusOK {
+		t.Fatalf("first request status = %d, want 200", code)
+	}
+	if code := do(); code != http.StatusTooManyRequests {
+		t.Fatalf("second request status = %d, want 429", code)
+	}
+}

api/internal/handler/openapi.json

@@ -0,0 +1,186 @@
+{
+  "openapi": "3.0.3",
+  "info": {
+    "title": "OpenGoods / 天工商品档案公共仓 API",
+    "version": "1.0.0",
+    "description": "Public, read-only product-facts REST API. Anonymous access is allowed at a lower per-minute rate; an optional API key grants a higher rate limit and attributes usage. No purchase or commerce endpoints by design.",
+    "license": { "name": "Data under each source's license (e.g. ODbL)" }
+  },
+  "servers": [{ "url": "https://goods.tangshasha.com/api/v1" }],
+  "tags": [
+    { "name": "products" },
+    { "name": "catalog" },
+    { "name": "meta" }
+  ],
+  "security": [{ "ApiKeyHeader": [] }, { "BearerKey": [] }, {}],
+  "paths": {
+    "/products/search": {
+      "get": {
+        "tags": ["products"],
+        "summary": "Search products",
+        "description": "Trigram-fuzzy name search (typo-tolerant) with optional category/brand/country filters, ranked by name similarity blended with data quality_score.",
+        "parameters": [
+          { "name": "q", "in": "query", "schema": { "type": "string" }, "description": "Keyword (name or barcode); fuzzy-matched. Empty returns all, ordered by quality_score." },
+          { "name": "category", "in": "query", "schema": { "type": "string" }, "description": "Category code (matches the subtree), e.g. food.beverages." },
+          { "name": "brand", "in": "query", "schema": { "type": "string" }, "description": "Brand name (fuzzy)." },
+          { "name": "country", "in": "query", "schema": { "type": "string" }, "description": "Country of origin (case-insensitive prefix)." },
+          { "name": "page", "in": "query", "schema": { "type": "integer", "default": 1, "minimum": 1 } },
+          { "name": "size", "in": "query", "schema": { "type": "integer", "default": 20, "maximum": 100 } }
+        ],
+        "responses": {
+          "200": {
+            "description": "Paged search results.",
+            "headers": {
+              "X-RateLimit-Limit": { "schema": { "type": "integer" }, "description": "Max requests in the current window." },
+              "X-RateLimit-Remaining": { "schema": { "type": "integer" }, "description": "Remaining requests in the window." },
+              "X-RateLimit-Reset": { "schema": { "type": "integer" }, "description": "Unix timestamp when the window resets." }
+            },
+            "content": {
+              "application/json": {
+                "schema": {
+                  "type": "object",
+                  "properties": {
+                    "items": { "type": "array", "items": { "$ref": "#/components/schemas/ProductSummary" } },
+                    "page": { "type": "integer" },
+                    "size": { "type": "integer" },
+                    "total": { "type": "integer" }
+                  }
+                }
+              }
+            }
+          },
+          "401": { "$ref": "#/components/responses/InvalidApiKey" },
+          "429": { "$ref": "#/components/responses/RateLimited" }
+        }
+      }
+    },
+    "/products/barcode/{gtin}": {
+      "get": {
+        "tags": ["products"],
+        "summary": "Get product by barcode (GTIN)",
+        "parameters": [{ "name": "gtin", "in": "path", "required": true, "schema": { "type": "string" } }],
+        "responses": {
+          "200": { "description": "Product", "content": { "application/json": { "schema": { "$ref": "#/components/schemas/Product" } } } },
+          "404": { "$ref": "#/components/responses/NotFound" },
+          "429": { "$ref": "#/components/responses/RateLimited" }
+        }
+      }
+    },
+    "/products/{id}": {
+      "get": {
+        "tags": ["products"],
+        "summary": "Get product detail by UUID",
+        "parameters": [{ "name": "id", "in": "path", "required": true, "schema": { "type": "string", "format": "uuid" } }],
+        "responses": {
+          "200": { "description": "Product", "content": { "application/json": { "schema": { "$ref": "#/components/schemas/Product" } } } },
+          "404": { "$ref": "#/components/responses/NotFound" }
+        }
+      }
+    },
+    "/products/{id}/nutriments": {
+      "get": {
+        "tags": ["products"],
+        "summary": "Get product nutriments",
+        "parameters": [{ "name": "id", "in": "path", "required": true, "schema": { "type": "string", "format": "uuid" } }],
+        "responses": { "200": { "description": "Nutriments" }, "404": { "$ref": "#/components/responses/NotFound" } }
+      }
+    },
+    "/products/{id}/msrp": {
+      "get": {
+        "tags": ["products"],
+        "summary": "Get manufacturer suggested retail price snapshots (reference only)",
+        "parameters": [{ "name": "id", "in": "path", "required": true, "schema": { "type": "string", "format": "uuid" } }],
+        "responses": { "200": { "description": "MSRP snapshots" } }
+      }
+    },
+    "/brands": {
+      "get": {
+        "tags": ["catalog"],
+        "summary": "List brands",
+        "parameters": [
+          { "name": "page", "in": "query", "schema": { "type": "integer", "default": 1 } },
+          { "name": "size", "in": "query", "schema": { "type": "integer", "default": 20, "maximum": 100 } }
+        ],
+        "responses": { "200": { "description": "Paged brands" } }
+      }
+    },
+    "/categories": {
+      "get": { "tags": ["catalog"], "summary": "List the category tree", "responses": { "200": { "description": "Category tree" } } }
+    },
+    "/sources/{id}": {
+      "get": {
+        "tags": ["meta"],
+        "summary": "Get a data source",
+        "parameters": [{ "name": "id", "in": "path", "required": true, "schema": { "type": "string", "format": "uuid" } }],
+        "responses": { "200": { "description": "Source" }, "404": { "$ref": "#/components/responses/NotFound" } }
+      }
+    }
+  },
+  "components": {
+    "securitySchemes": {
+      "ApiKeyHeader": { "type": "apiKey", "in": "header", "name": "X-API-Key", "description": "API key, e.g. og_live_xxx. Optional." },
+      "BearerKey": { "type": "http", "scheme": "bearer", "description": "Authorization: Bearer og_live_xxx. Optional." }
+    },
+    "responses": {
+      "NotFound": {
+        "description": "Resource not found.",
+        "content": { "application/json": { "schema": { "$ref": "#/components/schemas/Error" } } }
+      },
+      "RateLimited": {
+        "description": "Rate limit exceeded.",
+        "headers": { "Retry-After": { "schema": { "type": "integer" }, "description": "Seconds to wait before retrying." } },
+        "content": { "application/json": { "schema": { "$ref": "#/components/schemas/Error" } } }
+      },
+      "InvalidApiKey": {
+        "description": "API key invalid or revoked.",
+        "content": { "application/json": { "schema": { "$ref": "#/components/schemas/Error" } } }
+      }
+    },
+    "schemas": {
+      "Error": {
+        "type": "object",
+        "properties": {
+          "error": {
+            "type": "object",
+            "properties": {
+              "code": { "type": "string" },
+              "message": { "type": "string" },
+              "request_id": { "type": "string" }
+            }
+          }
+        }
+      },
+      "ProductSummary": {
+        "type": "object",
+        "properties": {
+          "id": { "type": "string", "format": "uuid" },
+          "gtin": { "type": "string", "nullable": true },
+          "name": { "type": "string" },
+          "brand": { "type": "string", "nullable": true },
+          "category_path": { "type": "string", "nullable": true },
+          "country_of_origin": { "type": "string", "nullable": true },
+          "quality_score": { "type": "number", "format": "float" },
+          "score": { "type": "number", "format": "float", "nullable": true, "description": "Relevance (name word-similarity) when q is provided; null otherwise." }
+        }
+      },
+      "Product": {
+        "type": "object",
+        "properties": {
+          "id": { "type": "string", "format": "uuid" },
+          "gtin": { "type": "string", "nullable": true },
+          "name": { "type": "string" },
+          "brand": { "type": "string", "nullable": true },
+          "category_path": { "type": "string", "nullable": true },
+          "net_content_value": { "type": "number", "nullable": true },
+          "net_content_unit": { "type": "string", "nullable": true },
+          "country_of_origin": { "type": "string", "nullable": true },
+          "quality_score": { "type": "number", "format": "float" },
+          "nutriments": { "type": "object", "additionalProperties": true, "nullable": true },
+          "nutrition_basis": { "type": "string", "nullable": true },
+          "nutri_score": { "type": "string", "nullable": true },
+          "ingredients_text": { "type": "string", "nullable": true }
+        }
+      }
+    }
+  }
+}

api/internal/ratelimit/ratelimit.go

@@ -0,0 +1,132 @@
+// Package ratelimit provides a Redis-backed fixed-window rate limiter and
+// lightweight per-key usage counters for the public API.
+//
+// All state lives in Redis so it is shared across API replicas and visible to
+// the admin console, and so the public server keeps its read-only contract
+// against PostgreSQL. Every operation fails open: if Redis is unavailable the
+// limiter allows the request rather than taking the API down.
+package ratelimit
+
+import (
+	"context"
+	"fmt"
+	"log"
+	"time"
+
+	"github.com/redis/go-redis/v9"
+)
+
+// Limiter throttles callers and records usage. A nil-backed Limiter (when Redis
+// could not be configured) disables limiting and usage tracking.
+type Limiter struct {
+	rdb *redis.Client
+}
+
+// Result describes the outcome of an Allow check and the headers to surface.
+type Result struct {
+	Allowed   bool
+	Limit     int
+	Remaining int
+	ResetUnix int64
+}
+
+// UsageStat is the aggregated usage for a single API key.
+type UsageStat struct {
+	Total      int64  `json:"total"`
+	Today      int64  `json:"today"`
+	LastUsedAt *int64 `json:"last_used_at,omitempty"`
+}
+
+// New builds a Limiter from a redis:// URL. On a parse error it logs and returns
+// a fail-open limiter (Redis disabled) so the server still boots.
+func New(redisURL string) *Limiter {
+	opt, err := redis.ParseURL(redisURL)
+	if err != nil {
+		log.Printf("ratelimit: invalid redis url %q: %v (rate limiting disabled)", redisURL, err)
+		return &Limiter{}
+	}
+	return &Limiter{rdb: redis.NewClient(opt)}
+}
+
+// Enabled reports whether a Redis backend is configured.
+func (l *Limiter) Enabled() bool { return l != nil && l.rdb != nil }
+
+// Ping verifies the Redis backend is reachable. Returns an error if disabled or
+// unreachable.
+func (l *Limiter) Ping(ctx context.Context) error {
+	if !l.Enabled() {
+		return redis.ErrClosed
+	}
+	return l.rdb.Ping(ctx).Err()
+}
+
+// Allow records a hit for id within a fixed window and reports whether the
+// caller is under limit. Fails open (Allowed=true) on any Redis error.
+func (l *Limiter) Allow(ctx context.Context, id string, limit int, window time.Duration) Result {
+	reset := func() int64 {
+		win := int64(window / time.Second)
+		if win < 1 {
+			win = 1
+		}
+		return (time.Now().Unix()/win + 1) * win
+	}
+	if !l.Enabled() {
+		return Result{Allowed: true, Limit: limit, Remaining: limit, ResetUnix: reset()}
+	}
+	win := int64(window / time.Second)
+	if win < 1 {
+		win = 1
+	}
+	bucket := time.Now().Unix() / win
+	key := fmt.Sprintf("rl:%s:%d", id, bucket)
+
+	n, err := l.rdb.Incr(ctx, key).Result()
+	if err != nil {
+		return Result{Allowed: true, Limit: limit, Remaining: limit, ResetUnix: (bucket + 1) * win}
+	}
+	if n == 1 {
+		l.rdb.Expire(ctx, key, time.Duration(win)*time.Second)
+	}
+	remaining := limit - int(n)
+	if remaining < 0 {
+		remaining = 0
+	}
+	return Result{
+		Allowed:   int(n) <= limit,
+		Limit:     limit,
+		Remaining: remaining,
+		ResetUnix: (bucket + 1) * win,
+	}
+}
+
+// RecordUsage increments total/daily counters and stamps last-used for a key.
+// Best-effort: errors are ignored.
+func (l *Limiter) RecordUsage(ctx context.Context, keyID string) {
+	if !l.Enabled() || keyID == "" {
+		return
+	}
+	now := time.Now()
+	day := now.Format("20060102")
+	pipe := l.rdb.Pipeline()
+	pipe.Incr(ctx, "usage:total:"+keyID)
+	dayKey := "usage:day:" + keyID + ":" + day
+	pipe.Incr(ctx, dayKey)
+	pipe.Expire(ctx, dayKey, 90*24*time.Hour)
+	pipe.Set(ctx, "usage:last:"+keyID, now.Unix(), 0)
+	_, _ = pipe.Exec(ctx)
+}
+
+// Usage reads aggregated usage for a key. Returns a zero-value stat on error.
+func (l *Limiter) Usage(ctx context.Context, keyID string) UsageStat {
+	var st UsageStat
+	if !l.Enabled() || keyID == "" {
+		return st
+	}
+	day := time.Now().Format("20060102")
+	st.Total, _ = l.rdb.Get(ctx, "usage:total:"+keyID).Int64()
+	st.Today, _ = l.rdb.Get(ctx, "usage:day:"+keyID+":"+day).Int64()
+	if v, err := l.rdb.Get(ctx, "usage:last:"+keyID).Int64(); err == nil {
+		st.LastUsedAt = &v
+	}
+	return st
+}

api/internal/ratelimit/ratelimit_test.go

@@ -0,0 +1,94 @@
+package ratelimit
+
+import (
+	"context"
+	"fmt"
+	"os"
+	"testing"
+	"time"
+)
+
+// TestDisabledFailsOpen verifies that a Limiter without a Redis backend allows
+// all requests and reports usage as zero rather than erroring.
+func TestDisabledFailsOpen(t *testing.T) {
+	l := New("not-a-valid-url") // parse error => disabled
+	if l.Enabled() {
+		t.Fatal("expected limiter to be disabled for invalid url")
+	}
+	res := l.Allow(context.Background(), "x", 1, time.Minute)
+	if !res.Allowed || res.Remaining != 1 {
+		t.Fatalf("disabled limiter must fail open: %+v", res)
+	}
+	// Must not panic and must return zero usage.
+	l.RecordUsage(context.Background(), "k1")
+	if u := l.Usage(context.Background(), "k1"); u.Total != 0 {
+		t.Fatalf("disabled usage should be zero, got %+v", u)
+	}
+}
+
+// TestNilReceiverSafe ensures a nil *Limiter is safe to use (handler default).
+func TestNilReceiverSafe(t *testing.T) {
+	var l *Limiter
+	if l.Enabled() {
+		t.Fatal("nil limiter must report disabled")
+	}
+	res := l.Allow(context.Background(), "x", 5, time.Minute)
+	if !res.Allowed {
+		t.Fatal("nil limiter must fail open")
+	}
+	l.RecordUsage(context.Background(), "k")
+	_ = l.Usage(context.Background(), "k")
+}
+
+func testLimiter(t *testing.T) *Limiter {
+	t.Helper()
+	url := os.Getenv("OPENGOODS_REDIS_URL")
+	if url == "" {
+		url = "redis://localhost:6379/0"
+	}
+	l := New(url)
+	if !l.Enabled() {
+		t.Skip("redis not configured")
+	}
+	ctx, cancel := context.WithTimeout(context.Background(), time.Second)
+	defer cancel()
+	if err := l.rdb.Ping(ctx).Err(); err != nil {
+		t.Skipf("redis not reachable: %v", err)
+	}
+	return l
+}
+
+func TestAllowFixedWindow(t *testing.T) {
+	l := testLimiter(t)
+	ctx := context.Background()
+	id := fmt.Sprintf("test:%d", time.Now().UnixNano())
+
+	for i := 1; i <= 2; i++ {
+		if res := l.Allow(ctx, id, 2, time.Minute); !res.Allowed {
+			t.Fatalf("request %d should be allowed: %+v", i, res)
+		}
+	}
+	res := l.Allow(ctx, id, 2, time.Minute)
+	if res.Allowed {
+		t.Fatalf("3rd request over limit 2 should be denied: %+v", res)
+	}
+	if res.Remaining != 0 {
+		t.Fatalf("remaining should be 0 when over limit, got %d", res.Remaining)
+	}
+}
+
+func TestRecordAndReadUsage(t *testing.T) {
+	l := testLimiter(t)
+	ctx := context.Background()
+	key := fmt.Sprintf("usagekey:%d", time.Now().UnixNano())
+
+	l.RecordUsage(ctx, key)
+	l.RecordUsage(ctx, key)
+	u := l.Usage(ctx, key)
+	if u.Total != 2 || u.Today != 2 {
+		t.Fatalf("expected total=2 today=2, got %+v", u)
+	}
+	if u.LastUsedAt == nil {
+		t.Fatal("expected last-used timestamp to be set")
+	}
+}

api/internal/store/store.go

@@ -82,11 +82,22 @@ func (s *Store) ProductBarcodes(ctx context.Context, productID string) ([]Barcod
 
 // ProductSummary is a lightweight row used in search/listing responses.
 type ProductSummary struct {
-	ID           string  `json:"id"`
-	GTIN         *string `json:"gtin"`
-	Name         string  `json:"name"`
-	Brand        *string `json:"brand"`
-	CategoryPath *string `json:"category_path"`
+	ID           string   `json:"id"`
+	GTIN         *string  `json:"gtin"`
+	Name         string   `json:"name"`
+	Brand        *string  `json:"brand"`
+	CategoryPath *string  `json:"category_path"`
+	Country      *string  `json:"country_of_origin"`
+	QualityScore float64  `json:"quality_score"`
+	Score        *float64 `json:"score,omitempty"`
+}
+
+// SearchFilters bundles the optional filters accepted by SearchProducts.
+type SearchFilters struct {
+	Query    string // fuzzy name / barcode query
+	Category string // ltree path; matches the subtree
+	Brand    string // fuzzy brand name
+	Country  string // country_of_origin prefix (case-insensitive)
 }
 
 const productSelect = `
@@ -148,34 +159,67 @@ func (s *Store) ProductByID(ctx context.Context, id string) (*Product, error) {
 	return p, nil
 }
 
-// SearchProducts performs a fuzzy name search with optional category subtree filter.
-func (s *Store) SearchProducts(ctx context.Context, q, category string, limit, offset int) ([]ProductSummary, int, error) {
+// fuzzyThreshold is the minimum word_similarity for a name to be considered a
+// fuzzy match. ~0.42 tolerates common typos (e.g. "choclate"→"Chocolate")
+// without returning unrelated products.
+const fuzzyThreshold = "0.42"
+
+// SearchProducts runs a trigram-fuzzy name search with optional category /
+// brand / country filters. When a query is present, matching is inclusive
+// (substring OR trigram-similar OR barcode), and results are ranked by name
+// similarity blended with quality_score so the best, most-complete records
+// surface first. Without a query, results are ordered by quality_score.
+func (s *Store) SearchProducts(ctx context.Context, f SearchFilters, limit, offset int) ([]ProductSummary, int, error) {
 	args := []any{}
 	where := "WHERE p.status = 'active'"
-	if q != "" {
-		args = append(args, q)
-		where += ` AND (p.name ILIKE '%' || $1 || '%'
+
+	qIdx := 0
+	if f.Query != "" {
+		args = append(args, f.Query)
+		qIdx = len(args)
+		q := "$" + strconv.Itoa(qIdx)
+		where += ` AND (p.name ILIKE '%' || ` + q + ` || '%'
+			OR word_similarity(` + q + `, p.name) >= ` + fuzzyThreshold + `
 			OR EXISTS (SELECT 1 FROM product_barcode pb
-			           WHERE pb.product_id = p.id AND pb.gtin ILIKE '%' || $1 || '%'))`
+			           WHERE pb.product_id = p.id AND pb.gtin ILIKE '%' || ` + q + ` || '%'))`
 	}
-	if category != "" {
-		args = append(args, category)
+	if f.Category != "" {
+		args = append(args, f.Category)
 		where += " AND c.path <@ $" + strconv.Itoa(len(args)) + "::ltree"
 	}
+	if f.Brand != "" {
+		args = append(args, f.Brand)
+		where += " AND b.name ILIKE '%' || $" + strconv.Itoa(len(args)) + " || '%'"
+	}
+	if f.Country != "" {
+		args = append(args, f.Country)
+		where += " AND p.country_of_origin ILIKE $" + strconv.Itoa(len(args)) + " || '%'"
+	}
+
+	from := `FROM product p
+LEFT JOIN brand b ON b.id = p.brand_id
+LEFT JOIN category c ON c.id = p.category_id `
 
-	countSQL := "SELECT count(*) FROM product p LEFT JOIN category c ON c.id = p.category_id " + where
 	var total int
-	if err := s.pool.QueryRow(ctx, countSQL, args...).Scan(&total); err != nil {
+	if err := s.pool.QueryRow(ctx, "SELECT count(*) "+from+where, args...).Scan(&total); err != nil {
 		return nil, 0, err
 	}
 
+	// Ranking: when querying, similarity drives order, multiplied by a
+	// quality factor floored at 0.5 so low-quality records aren't zeroed out.
+	scoreExpr := "NULL::real"
+	orderBy := "p.quality_score DESC, p.name"
+	if f.Query != "" {
+		q := "$" + strconv.Itoa(qIdx)
+		scoreExpr = "word_similarity(" + q + ", p.name)"
+		orderBy = scoreExpr + " * (0.5 + p.quality_score) DESC, p.quality_score DESC, p.name"
+	}
+
 	args = append(args, limit, offset)
-	listSQL := `
-SELECT p.id, p.gtin, p.name, b.name, c.path::text
-FROM product p
-LEFT JOIN brand b ON b.id = p.brand_id
-LEFT JOIN category c ON c.id = p.category_id ` + where +
-		" ORDER BY p.name LIMIT $" + strconv.Itoa(len(args)-1) + " OFFSET $" + strconv.Itoa(len(args))
+	listSQL := "SELECT p.id, p.gtin, p.name, b.name, c.path::text, p.country_of_origin, p.quality_score, " +
+		scoreExpr + " AS score " + from + where +
+		" ORDER BY " + orderBy +
+		" LIMIT $" + strconv.Itoa(len(args)-1) + " OFFSET $" + strconv.Itoa(len(args))
 
 	rows, err := s.pool.Query(ctx, listSQL, args...)
 	if err != nil {
@@ -186,7 +230,8 @@ LEFT JOIN category c ON c.id = p.category_id ` + where +
 	out := []ProductSummary{}
 	for rows.Next() {
 		var ps ProductSummary
-		if err := rows.Scan(&ps.ID, &ps.GTIN, &ps.Name, &ps.Brand, &ps.CategoryPath); err != nil {
+		if err := rows.Scan(&ps.ID, &ps.GTIN, &ps.Name, &ps.Brand, &ps.CategoryPath,
+			&ps.Country, &ps.QualityScore, &ps.Score); err != nil {
 			return nil, 0, err
 		}
 		out = append(out, ps)
@@ -305,6 +350,30 @@ func (s *Store) ListCategories(ctx context.Context) ([]Category, error) {
 	return out, rows.Err()
 }
 
+// APIKey is the minimal metadata the public API needs to authorize a caller.
+type APIKey struct {
+	ID              string
+	Name            string
+	RateLimitPerMin int
+}
+
+// APIKeyByHash returns the active (non-revoked) key matching a SHA-256 hash,
+// or ErrNotFound if no such active key exists.
+func (s *Store) APIKeyByHash(ctx context.Context, hash string) (*APIKey, error) {
+	var k APIKey
+	err := s.pool.QueryRow(ctx,
+		`SELECT id, name, rate_limit_per_min
+		 FROM api_key WHERE key_hash = $1 AND revoked_at IS NULL`, hash,
+	).Scan(&k.ID, &k.Name, &k.RateLimitPerMin)
+	if errors.Is(err, pgx.ErrNoRows) {
+		return nil, ErrNotFound
+	}
+	if err != nil {
+		return nil, err
+	}
+	return &k, nil
+}
+
 // Source describes a data source with its license and trust weight.
 type Source struct {
 	ID          string  `json:"id"`

docs/api.md

@@ -0,0 +1,117 @@
+# OpenGoods 公共 API 开发者文档
+
+天工商品档案公共仓（OpenGoods）提供**公开、只读**的商品事实 REST API：按条码/名称查询商品的客观资料（品牌、品类、净含量、产地、配料、营养成分、Nutri-Score、厂商建议零售价快照等）。返回均为 JSON（UTF-8）。**本服务不含任何购买/交易接口。**
+
+- 基础地址：`https://goods.tangshasha.com/api/v1`
+- 机器可读规范（OpenAPI 3）：`GET /api/v1/openapi.json`
+- 交互式文档：站点「API 调用说明」页
+
+## 鉴权
+
+API 默认**匿名可用**，无需任何凭证即可调用。匿名请求按来源 IP 计入一个较低的默认每分钟额度。
+
+如需更高额度并让用量归属到你，可向运营方申请一枚 **API Key**（形如 `og_live_xxxxxxxx`），请求时二选一携带：
+
+```bash
+curl -H "X-API-Key: og_live_xxxxxxxx" \
+  "https://goods.tangshasha.com/api/v1/products/search?q=牛奶"
+
+# 或
+curl -H "Authorization: Bearer og_live_xxxxxxxx" \
+  "https://goods.tangshasha.com/api/v1/products/search?q=牛奶"
+```
+
+> 仅在创建时返回一次明文 Key，请妥善保存。服务端只存储其 SHA-256 哈希。
+
+## 限流
+
+采用**固定窗口**限流（每分钟）。每个响应都会回写以下响应头：
+
+| 响应头 | 含义 |
+| --- | --- |
+| `X-RateLimit-Limit` | 当前窗口允许的最大请求数 |
+| `X-RateLimit-Remaining` | 当前窗口剩余可用次数 |
+| `X-RateLimit-Reset` | 窗口重置的 Unix 时间戳（秒） |
+| `Retry-After` | 仅在超额（429）时返回，建议等待的秒数 |
+
+- 超过额度：`429 Too Many Requests`，错误码 `rate_limited`。
+- Key 无效或已吊销：`401 Unauthorized`，错误码 `invalid_api_key`。
+
+## 错误格式
+
+非 2xx 响应体统一为：
+
+```json
+{ "error": { "code": "not_found", "message": "…", "request_id": "…" } }
+```
+
+## 分页
+
+列表类接口支持 `page`（默认 `1`）与 `size`（默认 `20`，最大 `100`），响应含 `page`/`size`/`total`。
+
+## 端点
+
+### `GET /products/search` — 搜索商品
+
+按名称做三元组（trigram）模糊搜索，**可容忍错别字**；支持品类/品牌/产地过滤；结果按相关度（名称相似度 × 数据质量分）排序。
+
+| 参数 | 必填 | 说明 |
+| --- | --- | --- |
+| `q` | 否 | 关键词（名称/条码），模糊匹配；留空则按质量分返回全部 |
+| `category` | 否 | 品类编码（含子树），如 `food.beverages` |
+| `brand` | 否 | 品牌名（模糊匹配），如 `Ferrero` |
+| `country` | 否 | 产地前缀（不区分大小写），如 `China` |
+| `page` | 否 | 页码，默认 1 |
+| `size` | 否 | 每页条数，默认 20，最大 100 |
+
+```bash
+curl "https://goods.tangshasha.com/api/v1/products/search?q=nutela&country=Italy"
+```
+
+```json
+{
+  "items": [
+    {
+      "id": "…",
+      "gtin": "3017624010701",
+      "name": "Nutella",
+      "brand": "Ferrero",
+      "category_path": "food.snacks.chocolate",
+      "country_of_origin": "Italy",
+      "quality_score": 0.81,
+      "score": 0.71
+    }
+  ],
+  "page": 1,
+  "size": 20,
+  "total": 1
+}
+```
+
+`score` 为名称相关度（提供 `q` 时返回，0–1），未提供 `q` 时为 `null`。
+
+### `GET /products/barcode/{gtin}` — 按条码查询
+
+```bash
+curl "https://goods.tangshasha.com/api/v1/products/barcode/5449000000996"
+```
+
+### `GET /products/{id}` — 商品详情
+
+按商品 UUID 获取完整档案（含配料、营养、添加剂、图片、MSRP 等）。
+
+### `GET /products/{id}/nutriments` — 商品营养成分
+
+### `GET /products/{id}/msrp` — 厂商建议零售价快照
+
+官方建议零售价历史快照，仅供参考，不含任何购买入口。
+
+### `GET /brands` — 品牌列表（分页）
+
+### `GET /categories` — 品类树
+
+### `GET /sources/{id}` — 数据来源
+
+## 免责声明
+
+数据可能存在误差或滞后，按「现状」提供，不构成医疗/购买建议。商品资料版权归各原始来源所有，请遵循其许可（如 OpenFoodFacts 的 ODbL），引用时请注明天工商品档案公共仓及原始来源。

ingestion/opengoods/adapters/openfoodfacts.py

@@ -27,6 +27,9 @@ _DEFAULT_MIN_INTERVAL = 4.0
 _API_URL = "https://world.openfoodfacts.org/api/v2/product/{barcode}.json"
 _SEARCH_URL = "https://world.openfoodfacts.org/api/v2/search"
 
+# HTTP statuses worth retrying: rate limiting and transient server errors.
+_RETRY_STATUS = frozenset({429, 500, 502, 503, 504})
+
 # Fields requested from the search API so a returned product can be transformed
 # without an extra per-barcode round trip.
 _SEARCH_FIELDS = (
@@ -46,9 +49,13 @@ class OpenFoodFactsAdapter:
         self,
         client: httpx.Client | None = None,
         min_interval: float = _DEFAULT_MIN_INTERVAL,
+        max_retries: int = 4,
+        backoff_base: float = 2.0,
     ) -> None:
         self._client = client or httpx.Client(headers={"User-Agent": USER_AGENT}, timeout=30.0)
         self._min_interval = min_interval
+        self._max_retries = max_retries
+        self._backoff_base = backoff_base
         self._last_call = 0.0
 
     def _throttle(self) -> None:
@@ -58,11 +65,49 @@ class OpenFoodFactsAdapter:
             time.sleep(wait)
         self._last_call = time.monotonic()
 
+    def _get(self, url: str, params: dict | None = None) -> httpx.Response:
+        """GET with throttling and retry/backoff on transient errors.
+
+        Retries on connection/timeout errors and on retryable HTTP statuses
+        (429 and 5xx, which OFF returns intermittently when overloaded), using
+        exponential backoff that honours a ``Retry-After`` header when present.
+        """
+        last_exc: Exception | None = None
+        for attempt in range(self._max_retries + 1):
+            self._throttle()
+            try:
+                resp = self._client.get(url, params=params)
+            except httpx.TransportError as exc:
+                last_exc = exc
+            else:
+                if resp.status_code < 400 or resp.status_code not in _RETRY_STATUS:
+                    resp.raise_for_status()
+                    return resp
+                last_exc = httpx.HTTPStatusError(
+                    f"retryable status {resp.status_code}", request=resp.request, response=resp
+                )
+            if attempt < self._max_retries:
+                retry_after = self._retry_after(last_exc)
+                time.sleep(retry_after if retry_after is not None else self._backoff_base**attempt)
+        assert last_exc is not None
+        raise last_exc
+
+    @staticmethod
+    def _retry_after(exc: Exception | None) -> float | None:
+        resp = getattr(exc, "response", None)
+        if resp is None:
+            return None
+        value = resp.headers.get("Retry-After")
+        if not value:
+            return None
+        try:
+            return float(value)
+        except ValueError:
+            return None
+
     def fetch_barcode(self, barcode: str) -> dict | None:
         """Fetch a single product by barcode; return the raw `product` dict."""
-        self._throttle()
-        resp = self._client.get(_API_URL.format(barcode=barcode))
-        resp.raise_for_status()
+        resp = self._get(_API_URL.format(barcode=barcode))
         payload = resp.json()
         if payload.get("status") != 1:
             return None
@@ -91,8 +136,7 @@ class OpenFoodFactsAdapter:
         ``last_modified_t`` they processed as the next watermark.
         """
         for page in range(1, max_pages + 1):
-            self._throttle()
-            resp = self._client.get(
+            resp = self._get(
                 _SEARCH_URL,
                 params={
                     "fields": _SEARCH_FIELDS,
@@ -101,7 +145,6 @@ class OpenFoodFactsAdapter:
                     "page_size": page_size,
                 },
             )
-            resp.raise_for_status()
             products = resp.json().get("products") or []
             if not products:
                 return
@@ -114,6 +157,60 @@ class OpenFoodFactsAdapter:
             if reached_old or len(products) < page_size:
                 return
 
+    def fetch_by_country(
+        self,
+        country: str,
+        *,
+        page_size: int = 100,
+        max_pages: int = 10,
+        sort_by: str = "unique_scans_n",
+    ) -> Iterator[dict]:
+        """Yield products sold in ``country`` (an OFF ``countries_tags_en`` slug).
+
+        Used to seed a market-specific catalogue (e.g. ``china``). Results are
+        sorted by ``sort_by`` (default ``unique_scans_n`` so the most-scanned,
+        best-known products come first) and de-duplicated across pages, since
+        OFF's popularity ordering is not stable between page requests.
+        """
+        seen: set[str] = set()
+        for page in range(1, max_pages + 1):
+            resp = self._get(
+                _SEARCH_URL,
+                params={
+                    "fields": _SEARCH_FIELDS,
+                    "countries_tags_en": country,
+                    "sort_by": sort_by,
+                    "page": page,
+                    "page_size": page_size,
+                },
+            )
+            products = resp.json().get("products") or []
+            if not products:
+                return
+            new_on_page = 0
+            for prod in products:
+                code = str(prod.get("code") or "")
+                if code and code in seen:
+                    continue
+                if code:
+                    seen.add(code)
+                new_on_page += 1
+                yield prod
+            if len(products) < page_size or new_on_page == 0:
+                return
+
+
+def is_cn_gs1(code: str | None) -> bool:
+    """Return True for a GS1 China company prefix (barcodes starting 690-699).
+
+    These identify products registered with GS1 China, i.e. genuinely domestic
+    items, as opposed to imported goods merely tagged as sold in China.
+    """
+    if not code:
+        return False
+    code = code.strip()
+    return len(code) >= 3 and code[:2] == "69" and code[2].isdigit()
+
 
 def read_dump(path: str | Path) -> Iterator[dict]:
     """Yield raw product records from an OFF JSONL dump file.

ingestion/opengoods/etl/load.py

@@ -7,6 +7,7 @@ source with field-level provenance in `product_source`.
 from __future__ import annotations
 
 import json
+import logging
 import os
 from typing import Any
 
@@ -18,6 +19,8 @@ from opengoods.etl.quality import update_quality
 
 OFF_HOMEPAGE = "https://world.openfoodfacts.org"
 
+logger = logging.getLogger(__name__)
+
 
 def default_dsn() -> str:
     return os.environ.get(
@@ -196,6 +199,25 @@ def load_record(conn: psycopg.Connection, rec: dict[str, Any], source_id: str, r
     return product_id
 
 
+def load_record_safe(
+    conn: psycopg.Connection, rec: dict[str, Any], source_id: str, raw: dict
+) -> bool:
+    """Load one record inside a savepoint.
+
+    On success the record's writes stay in the surrounding transaction. On any
+    error, only this record's writes are rolled back (to the savepoint) and the
+    batch continues, so a single malformed source record cannot abort a large
+    import. Returns True if loaded, False if skipped due to an error.
+    """
+    try:
+        with conn.transaction():
+            load_record(conn, rec, source_id, raw)
+        return True
+    except Exception as exc:  # noqa: BLE001 - per-record isolation is intentional
+        logger.warning("skipping record gtin=%s: %s", rec.get("gtin"), exc)
+        return False
+
+
 def _jsonable(raw: dict) -> dict:
     """Drop values that are not JSON-serializable from a raw record."""
     try:

ingestion/opengoods/etl/transform.py

@@ -78,6 +78,14 @@ def map_category(raw: dict) -> str | None:
     return None
 
 
+def _clamp(value: str | None, max_len: int) -> str | None:
+    """Trim a string to fit a bounded DB column; external data length varies."""
+    if value is None:
+        return None
+    value = value.strip()
+    return value[:max_len] or None
+
+
 def _clean_tags(tags: list[str] | None, prefix: str = "") -> list[str]:
     out: list[str] = []
     for t in tags or []:
@@ -143,16 +151,16 @@ def transform(raw: dict) -> dict | None:
         "brand": brand,
         "category_path": map_category(raw),
         "net_content_value": net_value,
-        "net_content_unit": net_unit,
+        "net_content_unit": _clamp(net_unit, 16),
         "net_content_canonical": net_canonical,
-        "country_of_origin": (raw.get("countries") or "").split(",")[0].strip() or None,
+        "country_of_origin": _clamp((raw.get("countries") or "").split(",")[0].strip() or None, 64),
         "food": {
             "ingredients_text": raw.get("ingredients_text") or None,
             "allergens": _clean_tags(raw.get("allergens_tags")),
             "additives": _clean_tags(raw.get("additives_tags")),
             "nutriments": transform_nutriments(raw.get("nutriments") or {}),
             "nutrition_basis": "per_100g",
-            "serving_size": raw.get("serving_size") or None,
+            "serving_size": _clamp(raw.get("serving_size") or None, 32),
             "nutri_score": (raw.get("nutriscore_grade") or "").upper()[:1] or None,
         },
         "image_url": raw.get("image_front_url") or raw.get("image_url") or None,

ingestion/opengoods/jobs/seed_off.py

@@ -7,6 +7,11 @@ Usage:
     # from a downloaded OFF JSONL dump (optionally .gz), limited to N records
     python -m opengoods.jobs.seed_off --dump products.jsonl.gz --limit 1000
 
+    # market-focused: the most-scanned products sold in China, restricted to
+    # genuine GS1-China (69x) barcodes
+    python -m opengoods.jobs.seed_off --country china --domestic-only \
+        --max-pages 20 --limit 1000
+
 The OFF read API is rate-limited client-side; for large imports use a dump.
 """
 
@@ -18,25 +23,38 @@ from collections.abc import Iterator
 
 import psycopg
 
-from opengoods.adapters.openfoodfacts import OpenFoodFactsAdapter, read_dump
-from opengoods.etl.load import default_dsn, ensure_source, load_record
+from opengoods.adapters.openfoodfacts import (
+    OpenFoodFactsAdapter,
+    is_cn_gs1,
+    read_dump,
+)
+from opengoods.etl.load import default_dsn, ensure_source, load_record_safe
 from opengoods.etl.transform import transform
 
 
 def _raw_records(args: argparse.Namespace) -> Iterator[dict]:
     if args.dump:
-        records = read_dump(args.dump)
+        records: Iterator[dict] = read_dump(args.dump)
+    elif args.country:
+        adapter = OpenFoodFactsAdapter(min_interval=args.min_interval)
+        records = adapter.fetch_by_country(
+            args.country, page_size=args.page_size, max_pages=args.max_pages
+        )
     else:
         adapter = OpenFoodFactsAdapter(min_interval=args.min_interval)
         records = adapter.fetch(args.barcodes)
-    for i, rec in enumerate(records):
-        if args.limit and i >= args.limit:
+    yielded = 0
+    for rec in records:
+        if args.domestic_only and not is_cn_gs1(str(rec.get("code") or "")):
+            continue
+        if args.limit and yielded >= args.limit:
             break
+        yielded += 1
         yield rec
 
 
 def run(args: argparse.Namespace) -> int:
-    loaded = skipped = 0
+    loaded = skipped = errored = 0
     with psycopg.connect(args.dsn, autocommit=False) as conn:
         source_id = ensure_source(conn)
         for raw in _raw_records(args):
@@ -44,10 +62,12 @@ def run(args: argparse.Namespace) -> int:
             if rec is None:
                 skipped += 1
                 continue
-            load_record(conn, rec, source_id, raw)
-            loaded += 1
+            if load_record_safe(conn, rec, source_id, raw):
+                loaded += 1
+            else:
+                errored += 1
         conn.commit()
-    print(f"loaded={loaded} skipped={skipped}")
+    print(f"loaded={loaded} skipped={skipped} errored={errored}")
     return 0
 
 
@@ -56,7 +76,18 @@ def main(argv: list[str] | None = None) -> int:
     src = parser.add_mutually_exclusive_group(required=True)
     src.add_argument("--barcodes", nargs="+", help="barcodes to fetch via the OFF API")
     src.add_argument("--dump", help="path to an OFF JSONL dump (.jsonl or .jsonl.gz)")
+    src.add_argument(
+        "--country",
+        help="OFF countries_tags_en slug to seed from, e.g. 'china' (most-scanned first)",
+    )
+    parser.add_argument(
+        "--domestic-only",
+        action="store_true",
+        help="keep only genuine GS1-China (69x) barcodes; drop imported goods",
+    )
     parser.add_argument("--limit", type=int, default=0, help="max records to load (0 = all)")
+    parser.add_argument("--page-size", type=int, default=100, help="search page size")
+    parser.add_argument("--max-pages", type=int, default=10, help="max search pages (country mode)")
     parser.add_argument("--min-interval", type=float, default=4.0, help="API throttle seconds")
     parser.add_argument("--dsn", default=default_dsn(), help="PostgreSQL DSN")
     return run(parser.parse_args(argv))

ingestion/opengoods/jobs/update_off.py

@@ -17,14 +17,14 @@ import sys
 import psycopg
 
 from opengoods.adapters.openfoodfacts import SOURCE_NAME, OpenFoodFactsAdapter
-from opengoods.etl.load import default_dsn, ensure_source, load_record
+from opengoods.etl.load import default_dsn, ensure_source, load_record_safe
 from opengoods.etl.state import get_watermark, set_watermark
 from opengoods.etl.transform import transform
 
 
 def run(args: argparse.Namespace) -> int:
     adapter = OpenFoodFactsAdapter(min_interval=args.min_interval)
-    loaded = skipped = 0
+    loaded = skipped = errored = 0
     high_watermark = 0
     with psycopg.connect(args.dsn, autocommit=False) as conn:
         source_id = ensure_source(conn)
@@ -38,16 +38,21 @@ def run(args: argparse.Namespace) -> int:
             if rec is None:
                 skipped += 1
                 continue
-            load_record(conn, rec, source_id, raw)
-            loaded += 1
+            if load_record_safe(conn, rec, source_id, raw):
+                loaded += 1
+            else:
+                errored += 1
         set_watermark(
             conn,
             SOURCE_NAME,
             high_watermark,
-            stats={"loaded": loaded, "skipped": skipped, "since": since},
+            stats={"loaded": loaded, "skipped": skipped, "errored": errored, "since": since},
         )
         conn.commit()
-    print(f"since={since} loaded={loaded} skipped={skipped} watermark={high_watermark}")
+    print(
+        f"since={since} loaded={loaded} skipped={skipped} "
+        f"errored={errored} watermark={high_watermark}"
+    )
     return 0
 
 

ingestion/tests/test_load_integration.py

@@ -11,7 +11,7 @@ from pathlib import Path
 
 import pytest
 
-from opengoods.etl.load import default_dsn, ensure_source, load_record
+from opengoods.etl.load import default_dsn, ensure_source, load_record, load_record_safe
 from opengoods.etl.transform import transform
 
 psycopg = pytest.importorskip("psycopg")
@@ -59,3 +59,25 @@ def test_load_record_roundtrip(conn):
     assert prov[0] >= 1
 
     conn.rollback()  # keep the test DB clean
+
+
+def test_load_record_safe_isolates_bad_record(conn):
+    source_id = ensure_source(conn)
+
+    # Unique gtin so the good record is a fresh INSERT, not an upsert/update.
+    good = transform(FIXTURE)
+    good["gtin"] = "4006381333931"
+    assert load_record_safe(conn, good, source_id, FIXTURE) is True
+    after_good = conn.execute("SELECT count(*) FROM product").fetchone()[0]
+
+    # A record whose name violates NOT NULL must not abort the batch.
+    bad = dict(good)
+    bad["gtin"] = "5000112637922"
+    bad["name"] = None
+    assert load_record_safe(conn, bad, source_id, {}) is False
+
+    # The good record survived the bad one's rollback-to-savepoint.
+    after_bad = conn.execute("SELECT count(*) FROM product").fetchone()[0]
+    assert after_bad == after_good
+
+    conn.rollback()  # keep the test DB clean

ingestion/tests/test_off_country.py

@@ -0,0 +1,53 @@
+import httpx
+
+from opengoods.adapters.openfoodfacts import OpenFoodFactsAdapter, is_cn_gs1
+
+
+def _adapter(handler, **kwargs):
+    client = httpx.Client(transport=httpx.MockTransport(handler))
+    return OpenFoodFactsAdapter(client=client, min_interval=0, backoff_base=0, **kwargs)
+
+
+def test_is_cn_gs1():
+    assert is_cn_gs1("6901234567892")  # GS1 China prefix
+    assert is_cn_gs1("690")
+    assert not is_cn_gs1("3017624010701")  # France
+    assert not is_cn_gs1("5449000000996")  # Belgium
+    assert not is_cn_gs1("")
+    assert not is_cn_gs1(None)
+    assert not is_cn_gs1("69")  # too short to carry a prefix digit
+
+
+def test_fetch_by_country_passes_filter_and_paginates():
+    seen_params = []
+
+    def handler(request: httpx.Request) -> httpx.Response:
+        seen_params.append(dict(request.url.params))
+        page = int(request.url.params.get("page", "1"))
+        if page == 1:
+            return httpx.Response(
+                200,
+                json={"products": [{"code": "6901"}, {"code": "6902"}]},
+            )
+        return httpx.Response(200, json={"products": []})
+
+    adapter = _adapter(handler)
+    out = list(adapter.fetch_by_country("china", page_size=2, max_pages=5))
+    assert [p["code"] for p in out] == ["6901", "6902"]
+    assert seen_params[0]["countries_tags_en"] == "china"
+    assert seen_params[0]["sort_by"] == "unique_scans_n"
+
+
+def test_fetch_by_country_dedupes_across_pages():
+    def handler(request: httpx.Request) -> httpx.Response:
+        page = int(request.url.params.get("page", "1"))
+        if page == 1:
+            return httpx.Response(200, json={"products": [{"code": "6901"}, {"code": "6902"}]})
+        if page == 2:
+            # OFF popularity ordering is unstable; a repeat appears on page 2
+            return httpx.Response(200, json={"products": [{"code": "6902"}, {"code": "6903"}]})
+        return httpx.Response(200, json={"products": []})
+
+    adapter = _adapter(handler)
+    out = [p["code"] for p in adapter.fetch_by_country("china", page_size=2, max_pages=5)]
+    assert out == ["6901", "6902", "6903"]

ingestion/tests/test_off_retry.py

@@ -0,0 +1,59 @@
+import httpx
+import pytest
+
+from opengoods.adapters.openfoodfacts import OpenFoodFactsAdapter
+
+
+def _adapter(handler, **kwargs):
+    client = httpx.Client(transport=httpx.MockTransport(handler))
+    return OpenFoodFactsAdapter(client=client, min_interval=0, backoff_base=0, **kwargs)
+
+
+def test_retries_transient_5xx_then_succeeds():
+    calls = {"n": 0}
+
+    def handler(request: httpx.Request) -> httpx.Response:
+        calls["n"] += 1
+        if calls["n"] < 3:
+            return httpx.Response(503)
+        return httpx.Response(200, json={"status": 1, "product": {"code": "x"}})
+
+    adapter = _adapter(handler, max_retries=4)
+    assert adapter.fetch_barcode("x") == {"code": "x"}
+    assert calls["n"] == 3  # two 503s retried, third succeeds
+
+
+def test_retries_exhausted_raises():
+    def handler(request: httpx.Request) -> httpx.Response:
+        return httpx.Response(503)
+
+    adapter = _adapter(handler, max_retries=2)
+    with pytest.raises(httpx.HTTPStatusError):
+        adapter.fetch_barcode("x")
+
+
+def test_non_retryable_4xx_not_retried():
+    calls = {"n": 0}
+
+    def handler(request: httpx.Request) -> httpx.Response:
+        calls["n"] += 1
+        return httpx.Response(404)
+
+    adapter = _adapter(handler, max_retries=4)
+    with pytest.raises(httpx.HTTPStatusError):
+        adapter.fetch_barcode("x")
+    assert calls["n"] == 1  # 404 is not retried
+
+
+def test_retries_connection_error_then_succeeds():
+    calls = {"n": 0}
+
+    def handler(request: httpx.Request) -> httpx.Response:
+        calls["n"] += 1
+        if calls["n"] == 1:
+            raise httpx.ConnectError("boom")
+        return httpx.Response(200, json={"status": 1, "product": {"code": "y"}})
+
+    adapter = _adapter(handler, max_retries=4)
+    assert adapter.fetch_barcode("y") == {"code": "y"}
+    assert calls["n"] == 2

ingestion/tests/test_transform.py

@@ -65,3 +65,14 @@ def test_transform_full_record():
 
 def test_transform_drops_unnamed():
     assert transform({"code": "0000000000000"}) is None
+
+
+def test_transform_clamps_long_serving_size():
+    raw = {
+        "code": "3017624010701",
+        "product_name": "X",
+        "serving_size": "1 portion (30 g) / 1 portion (30 g) / extra long descriptive text here",
+    }
+    rec = transform(raw)
+    assert rec is not None
+    assert len(rec["food"]["serving_size"]) == 32

migrations/0008_api_key.down.sql

@@ -0,0 +1 @@
+DROP TABLE IF EXISTS api_key;

migrations/0008_api_key.up.sql

@@ -0,0 +1,24 @@
+-- API keys for the public read-only API. Keys grant higher rate limits and let
+-- usage be attributed to a caller; the API itself stays free and read-only.
+-- Only the SHA-256 hash of a key is stored; the plaintext is shown once at
+-- creation time. Keys are issued/revoked from the admin console. The public
+-- server only ever SELECTs from this table (request counting lives in Redis),
+-- preserving its read-only contract against PostgreSQL.
+CREATE TABLE IF NOT EXISTS api_key (
+    id                 UUID PRIMARY KEY DEFAULT gen_random_uuid(),
+    name               TEXT NOT NULL,
+    key_prefix         VARCHAR(20) NOT NULL,         -- shown for identification, e.g. og_live_AbC1
+    key_hash           TEXT NOT NULL UNIQUE,         -- hex SHA-256 of the full key
+    owner_email        TEXT,
+    tier               VARCHAR(16) NOT NULL DEFAULT 'free',
+    rate_limit_per_min INT NOT NULL DEFAULT 120,
+    revoked_at         TIMESTAMPTZ,
+    created_by         TEXT,
+    created_at         TIMESTAMPTZ NOT NULL DEFAULT now(),
+    CONSTRAINT api_key_tier_chk CHECK (tier IN ('free', 'partner', 'internal')),
+    CONSTRAINT api_key_rate_chk CHECK (rate_limit_per_min > 0)
+);
+
+-- Fast lookup of active keys by their hash on every authenticated request.
+CREATE INDEX IF NOT EXISTS idx_api_key_active_hash
+    ON api_key (key_hash) WHERE revoked_at IS NULL;

migrations/0009_search.down.sql

@@ -0,0 +1,2 @@
+DROP INDEX IF EXISTS idx_product_country;
+DROP INDEX IF EXISTS idx_brand_name_trgm;

migrations/0009_search.up.sql

@@ -0,0 +1,9 @@
+-- Search upgrade: trigram-based fuzzy matching + brand/country filters.
+-- pg_trgm and the product.name GIN index already exist (see 0001). Add a
+-- matching trigram index on brand.name so brand filtering / fuzzy brand
+-- lookups can use an index instead of a sequential scan.
+CREATE INDEX IF NOT EXISTS idx_brand_name_trgm ON brand USING gin (name gin_trgm_ops);
+
+-- country_of_origin is filtered by exact/prefix match; a plain btree index
+-- keeps that cheap as the catalog grows.
+CREATE INDEX IF NOT EXISTS idx_product_country ON product (country_of_origin);

public-frontend/src/api.ts

@@ -25,11 +25,22 @@ export interface SearchResult {
   total: number;
 }
 
+export interface SearchFilters {
+  brand?: string;
+  country?: string;
+}
+
 export const api = {
-  search: (q: string, page = 1, size = 20) =>
-    req<SearchResult>(
-      `/api/v1/products/search?q=${encodeURIComponent(q)}&page=${page}&size=${size}`,
-    ),
+  search: (q: string, page = 1, size = 20, filters: SearchFilters = {}) => {
+    const params = new URLSearchParams({
+      q,
+      page: String(page),
+      size: String(size),
+    });
+    if (filters.brand) params.set("brand", filters.brand);
+    if (filters.country) params.set("country", filters.country);
+    return req<SearchResult>(`/api/v1/products/search?${params.toString()}`);
+  },
   product: (id: string) => req<Product>(`/api/v1/products/${id}`),
   categories: () => req<{ items: Category[] }>(`/api/v1/categories`),
   submit: (input: SubmissionInput) =>

public-frontend/src/components/ApiDocs.tsx

@@ -117,7 +117,7 @@ export default function ApiDocs() {
             基础地址：<code className="font-mono bg-gray-100 rounded px-1.5 py-0.5">{BASE}</code>
           </div>
           <ul className="mt-2 list-disc pl-5 text-gray-600 space-y-1">
-            <li>无需 API Key / Token，直接 GET 即可。</li>
+            <li>无需 API Key / Token 即可直接 GET；带 Key 可获得更高频率上限（见下文「鉴权与限流」）。</li>
             <li>
               分页参数 <code className="font-mono">page</code>（默认 1）、
               <code className="font-mono">size</code>（默认 20，最大 100）。
@@ -131,6 +131,53 @@ export default function ApiDocs() {
         </div>
       </div>
 
+      <div className="bg-white border rounded-lg p-5">
+        <h2 className="text-lg font-semibold text-gray-800">鉴权与限流</h2>
+        <p className="mt-2 text-gray-600 text-sm leading-relaxed">
+          API 默认<strong>匿名可用</strong>：无需任何凭证即可调用，按来源 IP 计入一个较低的默认频率额度。
+          如需更高额度并让用量归属到你，可在运营方申请一枚 API Key，请求时通过请求头携带：
+        </p>
+        <div className="mt-3">
+          <Code>{`# 二选一
+curl -H "X-API-Key: og_live_xxxxxxxx" ${BASE}/products/search?q=牛奶
+curl -H "Authorization: Bearer og_live_xxxxxxxx" ${BASE}/products/search?q=牛奶`}</Code>
+        </div>
+        <p className="mt-3 text-gray-600 text-sm leading-relaxed">
+          采用<strong>固定窗口</strong>限流（每分钟）。每个响应都会回写以下响应头，便于客户端自适应：
+        </p>
+        <table className="mt-3 w-full text-sm">
+          <thead className="text-gray-400 text-left">
+            <tr>
+              <th className="font-medium pr-4 pb-1">响应头</th>
+              <th className="font-medium pb-1">含义</th>
+            </tr>
+          </thead>
+          <tbody className="align-top">
+            <tr>
+              <td className="pr-4 py-0.5 font-mono text-gray-700">X-RateLimit-Limit</td>
+              <td className="py-0.5 text-gray-600">当前窗口允许的最大请求数</td>
+            </tr>
+            <tr>
+              <td className="pr-4 py-0.5 font-mono text-gray-700">X-RateLimit-Remaining</td>
+              <td className="py-0.5 text-gray-600">当前窗口剩余可用次数</td>
+            </tr>
+            <tr>
+              <td className="pr-4 py-0.5 font-mono text-gray-700">X-RateLimit-Reset</td>
+              <td className="py-0.5 text-gray-600">窗口重置的 Unix 时间戳（秒）</td>
+            </tr>
+            <tr>
+              <td className="pr-4 py-0.5 font-mono text-gray-700">Retry-After</td>
+              <td className="py-0.5 text-gray-600">超额时返回，建议等待的秒数</td>
+            </tr>
+          </tbody>
+        </table>
+        <p className="mt-3 text-gray-600 text-sm leading-relaxed">
+          超过额度返回 <code className="font-mono">429 Too Many Requests</code>，错误码
+          <code className="font-mono">rate_limited</code>；无效或已吊销的 Key 返回
+          <code className="font-mono">401</code>，错误码 <code className="font-mono">invalid_api_key</code>。
+        </p>
+      </div>
+
       <Endpoint
         method="GET"
         path="/healthz"
@@ -164,19 +211,24 @@ export default function ApiDocs() {
         method="GET"
         path="/api/v1/products/search"
         title="搜索商品"
-        desc="按名称模糊搜索，可按品类过滤，支持分页。"
+        desc="按名称做三元组（trigram）模糊搜索，可容忍错别字；支持品类/品牌/产地过滤；结果按相关度（名称相似度 × 数据质量分）排序。"
         params={[
-          { name: "q", desc: "关键词（名称/条码），留空返回全部" },
-          { name: "category", desc: "品类编码过滤，如 food.beverages" },
+          { name: "q", desc: "关键词（名称/条码），支持模糊匹配；留空则按质量分返回全部" },
+          { name: "category", desc: "品类编码（含子树），如 food.beverages" },
+          { name: "brand", desc: "品牌名（模糊匹配），如 Ferrero" },
+          { name: "country", desc: "产地前缀（不区分大小写），如 China" },
           { name: "page", desc: "页码，默认 1" },
           { name: "size", desc: "每页条数，默认 20，最大 100" },
         ]}
-        example={`curl "${BASE}/products/search?q=nutella&page=1&size=20"`}
+        example={`curl "${BASE}/products/search?q=nutela&country=Italy&page=1&size=20"`}
         response={`{
   "items": [
     { "id": "…", "gtin": "3017624010701",
       "name": "Nutella", "brand": "Ferrero",
-      "category_path": "food.snacks.chocolate" }
+      "category_path": "food.snacks.chocolate",
+      "country_of_origin": "Italy",
+      "quality_score": 0.81,
+      "score": 0.71 }
   ],
   "page": 1, "size": 20, "total": 1
 }`}

public-frontend/src/components/Home.tsx

@@ -13,6 +13,8 @@ export default function Home({
   onApi: () => void;
 }) {
   const [q, setQ] = useState("");
+  const [brand, setBrand] = useState("");
+  const [country, setCountry] = useState("");
   const [items, setItems] = useState<ProductSummary[]>([]);
   const [total, setTotal] = useState(0);
   const [searched, setSearched] = useState(false);
@@ -24,7 +26,10 @@ export default function Home({
     setLoading(true);
     setError("");
     try {
-      const res = await api.search(q.trim(), 1, 30);
+      const res = await api.search(q.trim(), 1, 30, {
+        brand: brand.trim() || undefined,
+        country: country.trim() || undefined,
+      });
       setItems(res.items);
       setTotal(res.total);
       setSearched(true);
@@ -61,6 +66,20 @@ export default function Home({
             {loading ? "检索中…" : "检索"}
           </button>
         </form>
+        <div className="mt-3 max-w-2xl mx-auto flex flex-wrap items-center justify-center gap-2 text-sm">
+          <input
+            value={brand}
+            onChange={(e) => setBrand(e.target.value)}
+            placeholder="按品牌筛选（如 Ferrero）"
+            className="flex-1 min-w-[14rem] bg-white border rounded-lg px-3 py-2 outline-none focus:ring-2 focus:ring-emerald-300"
+          />
+          <input
+            value={country}
+            onChange={(e) => setCountry(e.target.value)}
+            placeholder="按产地筛选（如 China）"
+            className="flex-1 min-w-[14rem] bg-white border rounded-lg px-3 py-2 outline-none focus:ring-2 focus:ring-emerald-300"
+          />
+        </div>
         <button
           onClick={onApi}
           className="mt-4 inline-flex items-center gap-1.5 text-sm text-emerald-700 hover:underline"
@@ -105,7 +124,12 @@ export default function Home({
                         {p.gtin ? ` · ${p.gtin}` : ""}
                       </div>
                     </div>
-                    <span className="text-xs text-gray-400">{p.category_path || ""}</span>
+                    <span className="text-xs text-gray-400 text-right">
+                      <span className="block">{p.category_path || ""}</span>
+                      {p.country_of_origin ? (
+                        <span className="block text-gray-400">产地：{p.country_of_origin}</span>
+                      ) : null}
+                    </span>
                   </button>
                 </li>
               ))}

public-frontend/src/types.ts

@@ -4,6 +4,9 @@ export interface ProductSummary {
   name: string;
   brand: string | null;
   category_path: string | null;
+  country_of_origin?: string | null;
+  quality_score?: number;
+  score?: number | null;
 }
 
 export interface Barcode {

tools/bypos-collector/.gitignore

@@ -0,0 +1,9 @@
+# build artifacts
+bypos-collector.exe
+bypos-collector
+bypos-collector-linux
+*.exe
+# collected data / outputs
+*.jsonl
+*.csv
+*.log

tools/bypos-collector/README.md

@@ -0,0 +1,67 @@
+# bypos-collector
+
+按条码批量采集商品档案的小工具(单文件 Windows/Linux 程序,自带本地 Web 控制台)。
+数据来源是云店「新增商品」输入条码时所查的同一个中心商品库 `zc.bypos.net`。
+采集结果落地为 JSONL,供后续导入本项目(天工/goods)。
+
+> 这是一个**独立模块**(有自己的 `go.mod`),与 `api/` 主服务互不影响,
+> CI 不会编译它。放在 `tools/` 下仅作代码留存与后期迭代。
+
+## 目录
+
+| 文件 | 说明 |
+| --- | --- |
+| `main.go` | 入口:本地 HTTP 服务 + 启动浏览器 + API 路由(start/stop/stats/download/export.csv) |
+| `collect.go` | 核心:签名、EAN-13 校验位、范围/清单枚举、并发+限速、JSONL 落库、断点续采 |
+| `web/index.html` | 内嵌(`go:embed`)的控制台界面 |
+| `build.sh` | 交叉编译出 `bypos-collector.exe`(windows/amd64)与 linux 测试二进制 |
+| `使用说明.md` | 面向使用者的操作说明 |
+
+## 构建
+
+```bash
+./build.sh
+# 产物:bypos-collector.exe(发给 Windows 用户)/ bypos-collector-linux(本地测试)
+```
+
+二进制与采集产物(`*.jsonl`/`*.csv`)已在 `.gitignore` 中排除,不入库。
+
+## 接口与签名(逆向所得,后期迭代参考)
+
+云店新增商品页输入条码时,前端经服务端代理 `/prod-api/ZmSvr/httpUtil/getGet`
+转发到中心库:
+
+```
+GET http://zc.bypos.net/byGoodsService/byMessage.asmx/GetGoodsinfo
+    ?sdogid=<账号id>&regnum=1&barcode=<条码>
+    &sparm1=<md5(sdogid)>            # 常量,随账号固定
+    &sparm2=<md5(barcode + tsMs)>    # tsMs = 当前秒*1000(末尾恒为 000)
+    &sparm3=<tsMs 前 10 位 = 秒级时间戳>
+    &sparm4=&barcodetype=yunpos
+```
+
+返回 `<string>{...json...}</string>`,内层 JSON 字段:
+
+| 上游字段 | 含义 | 归一化字段 |
+| --- | --- | --- |
+| item_name | 品名 | name |
+| item_size | 规格 | spec |
+| unit_no | 单位 | unit |
+| item_area | 产地/地区 | area |
+| birth_com | 生产企业(常空) | manufacturer |
+| birth_doc | 生产许可(常空) | license |
+| inprice | 建议进价 | in_price |
+| sellprice | 建议零售价 | sell_price |
+| retcode | 1=命中,0=失败 | status(hit/miss/invalid) |
+
+`retmsg` 含「非国标条码 / 参数异常」=> invalid;含「条码不存在」=> miss。
+
+## 配置
+
+- `sdogid`:中心库账号 id(本项目所属云店账号的授权 id)。默认值见
+  `collect.go` 的 `defaultSdogID`,也可用 `-sdogid` 参数或控制台覆盖。
+  **这是账号级凭证**——若本仓库对外公开,建议改为从环境变量/外部配置读取。
+
+## 注意
+
+批量自动查询比页面逐条更"重",上游可能对账号限频。请低速、分前缀/品类分批采集。

tools/bypos-collector/build.sh

@@ -0,0 +1,11 @@
+#!/usr/bin/env bash
+# Build the bypos-collector for Windows (and a Linux binary for local testing).
+set -euo pipefail
+cd "$(dirname "$0")"
+gofmt -w ./*.go
+go vet ./...
+echo "building windows/amd64 .exe ..."
+GOOS=windows GOARCH=amd64 go build -ldflags "-s -w" -o bypos-collector.exe .
+echo "building linux/amd64 (for testing) ..."
+go build -o bypos-collector-linux .
+ls -la bypos-collector.exe bypos-collector-linux

tools/bypos-collector/collect.go

@@ -0,0 +1,481 @@
+package main
+
+import (
+	"context"
+	"crypto/md5"
+	"encoding/hex"
+	"encoding/json"
+	"fmt"
+	"io"
+	"net/http"
+	"net/url"
+	"os"
+	"path/filepath"
+	"regexp"
+	"strconv"
+	"strings"
+	"sync"
+	"sync/atomic"
+	"time"
+)
+
+// ---- upstream config ----
+// sdogid is the 云店 account license id observed in the live request. It is
+// configurable so the tool is not tied to a single account.
+const defaultSdogID = "137966"
+
+const endpoint = "http://zc.bypos.net/byGoodsService/byMessage.asmx/GetGoodsinfo"
+
+var stringTagRe = regexp.MustCompile(`(?s)<string[^>]*>(.*)</string>`)
+
+// Product is the normalized record we persist (one JSON object per line).
+type Product struct {
+	Barcode      string `json:"barcode"`
+	Name         string `json:"name"`         // item_name 品名
+	Spec         string `json:"spec"`         // item_size 规格
+	Unit         string `json:"unit"`         // unit_no 单位
+	Area         string `json:"area"`         // item_area 产地/地区
+	Manufacturer string `json:"manufacturer"` // birth_com 生产企业
+	License      string `json:"license"`      // birth_doc 生产许可
+	InPrice      string `json:"in_price"`     // 建议进价
+	SellPrice    string `json:"sell_price"`   // 建议零售价
+	Status       string `json:"status"`       // hit / miss / invalid / error
+	RetMsg       string `json:"retmsg"`       // 原始返回信息
+	FetchedAt    string `json:"fetched_at"`   // RFC3339
+	Source       string `json:"source"`       // zc.bypos.net
+}
+
+// upstream raw fields
+type rawResp struct {
+	RetCode   string `json:"retcode"`
+	RetMsg    string `json:"retmsg"`
+	Barcode   string `json:"barcode"`
+	ItemName  string `json:"item_name"`
+	UnitNo    string `json:"unit_no"`
+	ItemSize  string `json:"item_size"`
+	ItemArea  string `json:"item_area"`
+	BirthCom  string `json:"birth_com"`
+	BirthDoc  string `json:"birth_doc"`
+	InPrice   string `json:"inprice"`
+	SellPrice string `json:"sellprice"`
+}
+
+func md5hex(s string) string {
+	h := md5.Sum([]byte(s))
+	return hex.EncodeToString(h[:])
+}
+
+// ean13Check computes the EAN-13 check digit for a 12-digit body.
+func ean13Check(body string) (string, bool) {
+	if len(body) != 12 {
+		return "", false
+	}
+	sum := 0
+	for i := 0; i < 12; i++ {
+		c := body[i]
+		if c < '0' || c > '9' {
+			return "", false
+		}
+		d := int(c - '0')
+		if i%2 == 0 {
+			sum += d
+		} else {
+			sum += d * 3
+		}
+	}
+	chk := (10 - (sum % 10)) % 10
+	return body + strconv.Itoa(chk), true
+}
+
+// lookup queries the upstream central library for one barcode.
+func (c *Collector) lookup(ctx context.Context, barcode string) (*Product, error) {
+	tsMs := strconv.FormatInt(time.Now().Unix()*1000, 10) // always ends in 000
+	sparm1 := md5hex(c.sdogID)
+	sparm2 := md5hex(barcode + tsMs)
+	sparm3 := tsMs[:10]
+	q := url.Values{}
+	q.Set("sdogid", c.sdogID)
+	q.Set("regnum", "1")
+	q.Set("barcode", barcode)
+	q.Set("sparm1", sparm1)
+	q.Set("sparm2", sparm2)
+	q.Set("sparm3", sparm3)
+	q.Set("sparm4", "")
+	q.Set("barcodetype", "yunpos")
+	reqURL := endpoint + "?" + q.Encode()
+
+	req, _ := http.NewRequestWithContext(ctx, "GET", reqURL, nil)
+	req.Header.Set("User-Agent", "Mozilla/5.0")
+	resp, err := c.client.Do(req)
+	if err != nil {
+		return nil, err
+	}
+	defer resp.Body.Close()
+	b, err := io.ReadAll(resp.Body)
+	if err != nil {
+		return nil, err
+	}
+	inner := b
+	if m := stringTagRe.FindSubmatch(b); m != nil {
+		inner = m[1]
+	}
+	var r rawResp
+	if err := json.Unmarshal(inner, &r); err != nil {
+		return nil, fmt.Errorf("parse: %v (body=%.120s)", err, string(b))
+	}
+	p := &Product{
+		Barcode:   barcode,
+		RetMsg:    r.RetMsg,
+		FetchedAt: time.Now().Format(time.RFC3339),
+		Source:    "zc.bypos.net",
+	}
+	if r.RetCode == "1" {
+		p.Status = "hit"
+		p.Name = strings.TrimSpace(r.ItemName)
+		p.Spec = strings.TrimSpace(r.ItemSize)
+		p.Unit = strings.TrimSpace(r.UnitNo)
+		p.Area = strings.TrimSpace(r.ItemArea)
+		p.Manufacturer = strings.TrimSpace(r.BirthCom)
+		p.License = strings.TrimSpace(r.BirthDoc)
+		p.InPrice = strings.TrimSpace(r.InPrice)
+		p.SellPrice = strings.TrimSpace(r.SellPrice)
+	} else if strings.Contains(r.RetMsg, "非国标") || strings.Contains(r.RetMsg, "参数异常") {
+		p.Status = "invalid"
+	} else {
+		p.Status = "miss"
+	}
+	return p, nil
+}
+
+// ---- job / collector state ----
+
+type Stats struct {
+	Running   bool   `json:"running"`
+	Total     int64  `json:"total"`
+	Done      int64  `json:"done"`
+	Hits      int64  `json:"hits"`
+	Miss      int64  `json:"miss"`
+	Invalid   int64  `json:"invalid"`
+	Errors    int64  `json:"errors"`
+	Skipped   int64  `json:"skipped"`
+	Current   string `json:"current"`
+	OutFile   string `json:"out_file"`
+	StartedAt string `json:"started_at"`
+	Message   string `json:"message"`
+}
+
+type Collector struct {
+	mu      sync.Mutex
+	client  *http.Client
+	sdogID  string
+	outPath string
+	outFile *os.File
+
+	cancel context.CancelFunc
+	wg     sync.WaitGroup
+
+	// atomic counters
+	total, done, hits, miss, invalid, errors, skipped int64
+	running                                           int32
+
+	current   atomic.Value // string
+	startedAt string
+	message   string
+
+	seen   map[string]struct{} // barcodes already in output (dedupe / resume)
+	recent []Product           // ring of last results for UI
+}
+
+func NewCollector(sdogID string) *Collector {
+	if sdogID == "" {
+		sdogID = defaultSdogID
+	}
+	c := &Collector{
+		client: &http.Client{Timeout: 25 * time.Second},
+		sdogID: sdogID,
+		seen:   map[string]struct{}{},
+	}
+	c.current.Store("")
+	return c
+}
+
+func (c *Collector) isRunning() bool { return atomic.LoadInt32(&c.running) == 1 }
+
+func (c *Collector) snapshot() Stats {
+	cur, _ := c.current.Load().(string)
+	c.mu.Lock()
+	msg := c.message
+	out := c.outPath
+	started := c.startedAt
+	c.mu.Unlock()
+	return Stats{
+		Running:   c.isRunning(),
+		Total:     atomic.LoadInt64(&c.total),
+		Done:      atomic.LoadInt64(&c.done),
+		Hits:      atomic.LoadInt64(&c.hits),
+		Miss:      atomic.LoadInt64(&c.miss),
+		Invalid:   atomic.LoadInt64(&c.invalid),
+		Errors:    atomic.LoadInt64(&c.errors),
+		Skipped:   atomic.LoadInt64(&c.skipped),
+		Current:   cur,
+		OutFile:   out,
+		StartedAt: started,
+		Message:   msg,
+	}
+}
+
+func (c *Collector) recentResults() []Product {
+	c.mu.Lock()
+	defer c.mu.Unlock()
+	out := make([]Product, len(c.recent))
+	copy(out, c.recent)
+	return out
+}
+
+func (c *Collector) pushRecent(p Product) {
+	c.mu.Lock()
+	c.recent = append(c.recent, p)
+	if len(c.recent) > 60 {
+		c.recent = c.recent[len(c.recent)-60:]
+	}
+	c.mu.Unlock()
+}
+
+// loadSeen reads an existing output file to build the dedupe set (for resume).
+func (c *Collector) loadSeen(path string) error {
+	c.seen = map[string]struct{}{}
+	f, err := os.Open(path)
+	if err != nil {
+		if os.IsNotExist(err) {
+			return nil
+		}
+		return err
+	}
+	defer f.Close()
+	dec := json.NewDecoder(f)
+	for {
+		var p Product
+		if err := dec.Decode(&p); err != nil {
+			break
+		}
+		if p.Barcode != "" {
+			c.seen[p.Barcode] = struct{}{}
+		}
+	}
+	return nil
+}
+
+type JobReq struct {
+	Mode        string `json:"mode"`        // "range" | "list"
+	StartBody   string `json:"start_body"`  // 12-digit body (range mode)
+	EndBody     string `json:"end_body"`    // 12-digit body (range mode)
+	List        string `json:"list"`        // newline/space separated barcodes (list mode)
+	Concurrency int    `json:"concurrency"` // parallel requests
+	DelayMs     int    `json:"delay_ms"`    // min interval between request starts
+	LogMiss     bool   `json:"log_miss"`    // also write miss/invalid lines
+	OutFile     string `json:"out_file"`
+	SdogID      string `json:"sdog_id"`
+}
+
+func sanitizeBarcodes(s string) []string {
+	fields := regexp.MustCompile(`[^0-9]+`).Split(s, -1)
+	var out []string
+	for _, f := range fields {
+		if f != "" {
+			out = append(out, f)
+		}
+	}
+	return out
+}
+
+// Start launches a collection job. Returns error if validation fails or busy.
+func (c *Collector) Start(req JobReq) error {
+	if c.isRunning() {
+		return fmt.Errorf("已有任务在运行")
+	}
+	if req.Concurrency <= 0 {
+		req.Concurrency = 3
+	}
+	if req.Concurrency > 20 {
+		req.Concurrency = 20
+	}
+	if req.DelayMs < 0 {
+		req.DelayMs = 0
+	}
+	if req.OutFile == "" {
+		req.OutFile = "products.jsonl"
+	}
+	if req.SdogID != "" {
+		c.sdogID = req.SdogID
+	}
+
+	// Build the list of barcodes to query.
+	var barcodes []string
+	switch req.Mode {
+	case "list":
+		barcodes = sanitizeBarcodes(req.List)
+		if len(barcodes) == 0 {
+			return fmt.Errorf("条码清单为空")
+		}
+	case "range":
+		start, err := strconv.ParseInt(req.StartBody, 10, 64)
+		if err != nil || len(req.StartBody) != 12 {
+			return fmt.Errorf("起始码必须是 12 位数字(不含校验位)")
+		}
+		end, err := strconv.ParseInt(req.EndBody, 10, 64)
+		if err != nil || len(req.EndBody) != 12 {
+			return fmt.Errorf("结束码必须是 12 位数字(不含校验位)")
+		}
+		if end < start {
+			return fmt.Errorf("结束码不能小于起始码")
+		}
+		if end-start+1 > 5_000_000 {
+			return fmt.Errorf("单次范围过大(>500万),请缩小区间分批采集")
+		}
+		for v := start; v <= end; v++ {
+			body := fmt.Sprintf("%012d", v)
+			full, ok := ean13Check(body)
+			if ok {
+				barcodes = append(barcodes, full)
+			}
+		}
+	default:
+		return fmt.Errorf("未知模式: %s", req.Mode)
+	}
+
+	abs, _ := filepath.Abs(req.OutFile)
+	if err := c.loadSeen(abs); err != nil {
+		return fmt.Errorf("读取已有文件失败: %v", err)
+	}
+	f, err := os.OpenFile(abs, os.O_CREATE|os.O_WRONLY|os.O_APPEND, 0644)
+	if err != nil {
+		return fmt.Errorf("打开输出文件失败: %v", err)
+	}
+	c.outFile = f
+	c.outPath = abs
+
+	// reset counters
+	atomic.StoreInt64(&c.total, int64(len(barcodes)))
+	atomic.StoreInt64(&c.done, 0)
+	atomic.StoreInt64(&c.hits, 0)
+	atomic.StoreInt64(&c.miss, 0)
+	atomic.StoreInt64(&c.invalid, 0)
+	atomic.StoreInt64(&c.errors, 0)
+	atomic.StoreInt64(&c.skipped, 0)
+	c.mu.Lock()
+	c.recent = nil
+	c.startedAt = time.Now().Format(time.RFC3339)
+	c.message = ""
+	c.mu.Unlock()
+	atomic.StoreInt32(&c.running, 1)
+
+	ctx, cancel := context.WithCancel(context.Background())
+	c.cancel = cancel
+
+	go c.run(ctx, barcodes, req)
+	return nil
+}
+
+func (c *Collector) Stop() {
+	if c.cancel != nil {
+		c.cancel()
+	}
+}
+
+func (c *Collector) run(ctx context.Context, barcodes []string, req JobReq) {
+	defer func() {
+		atomic.StoreInt32(&c.running, 0)
+		if c.outFile != nil {
+			c.outFile.Sync()
+			c.outFile.Close()
+			c.outFile = nil
+		}
+		c.current.Store("")
+	}()
+
+	jobs := make(chan string, req.Concurrency*2)
+	var writeMu sync.Mutex
+
+	// global rate limiter: one token every DelayMs
+	var ticker *time.Ticker
+	if req.DelayMs > 0 {
+		ticker = time.NewTicker(time.Duration(req.DelayMs) * time.Millisecond)
+		defer ticker.Stop()
+	}
+
+	worker := func() {
+		defer c.wg.Done()
+		for bc := range jobs {
+			if ctx.Err() != nil {
+				return
+			}
+			if ticker != nil {
+				select {
+				case <-ticker.C:
+				case <-ctx.Done():
+					return
+				}
+			}
+			c.current.Store(bc)
+			p, err := c.lookup(ctx, bc)
+			if err != nil {
+				if ctx.Err() != nil {
+					return
+				}
+				atomic.AddInt64(&c.errors, 1)
+				atomic.AddInt64(&c.done, 1)
+				ep := Product{Barcode: bc, Status: "error", RetMsg: err.Error(), FetchedAt: time.Now().Format(time.RFC3339), Source: "zc.bypos.net"}
+				c.pushRecent(ep)
+				continue
+			}
+			switch p.Status {
+			case "hit":
+				atomic.AddInt64(&c.hits, 1)
+			case "miss":
+				atomic.AddInt64(&c.miss, 1)
+			case "invalid":
+				atomic.AddInt64(&c.invalid, 1)
+			}
+			atomic.AddInt64(&c.done, 1)
+			c.pushRecent(*p)
+			if p.Status == "hit" || req.LogMiss {
+				line, _ := json.Marshal(p)
+				writeMu.Lock()
+				c.outFile.Write(line)
+				c.outFile.Write([]byte("\n"))
+				c.seen[bc] = struct{}{}
+				writeMu.Unlock()
+			}
+		}
+	}
+
+	for i := 0; i < req.Concurrency; i++ {
+		c.wg.Add(1)
+		go worker()
+	}
+
+	for _, bc := range barcodes {
+		if ctx.Err() != nil {
+			break
+		}
+		if _, ok := c.seen[bc]; ok {
+			atomic.AddInt64(&c.skipped, 1)
+			atomic.AddInt64(&c.done, 1)
+			continue
+		}
+		select {
+		case jobs <- bc:
+		case <-ctx.Done():
+		}
+	}
+	close(jobs)
+	c.wg.Wait()
+
+	c.mu.Lock()
+	if ctx.Err() != nil {
+		c.message = "已停止"
+	} else {
+		c.message = "采集完成"
+	}
+	c.mu.Unlock()
+}

tools/bypos-collector/go.mod

@@ -0,0 +1,3 @@
+module byposcollector
+
+go 1.23.4

tools/bypos-collector/main.go

@@ -0,0 +1,156 @@
+package main
+
+import (
+	"embed"
+	"encoding/csv"
+	"encoding/json"
+	"flag"
+	"fmt"
+	"io"
+	"io/fs"
+	"log"
+	"net"
+	"net/http"
+	"os"
+	"os/exec"
+	"runtime"
+	"time"
+)
+
+//go:embed web/*
+var webFS embed.FS
+
+var collector = NewCollector("")
+
+func main() {
+	addr := flag.String("addr", "127.0.0.1:8765", "本地监听地址")
+	noOpen := flag.Bool("no-open", false, "不自动打开浏览器")
+	sdog := flag.String("sdogid", "", "中心库账号 id(默认使用内置值)")
+	flag.Parse()
+	if *sdog != "" {
+		collector.sdogID = *sdog
+	}
+
+	sub, _ := fs.Sub(webFS, "web")
+	mux := http.NewServeMux()
+	mux.Handle("/", http.FileServer(http.FS(sub)))
+	mux.HandleFunc("/api/start", handleStart)
+	mux.HandleFunc("/api/stop", handleStop)
+	mux.HandleFunc("/api/stats", handleStats)
+	mux.HandleFunc("/api/download", handleDownload)
+	mux.HandleFunc("/api/export.csv", handleExportCSV)
+
+	ln, err := net.Listen("tcp", *addr)
+	if err != nil {
+		log.Fatalf("无法监听 %s: %v", *addr, err)
+	}
+	realAddr := ln.Addr().String()
+	urlStr := "http://" + realAddr + "/"
+	fmt.Println("==============================================")
+	fmt.Println(" 中心库商品采集器  bypos-collector")
+	fmt.Println(" 控制台: " + urlStr)
+	fmt.Println(" 关闭本窗口即停止程序")
+	fmt.Println("==============================================")
+	if !*noOpen {
+		go openBrowser(urlStr)
+	}
+	log.Fatal(http.Serve(ln, mux))
+}
+
+func openBrowser(url string) {
+	time.Sleep(600 * time.Millisecond)
+	var cmd string
+	var args []string
+	switch runtime.GOOS {
+	case "windows":
+		cmd = "rundll32"
+		args = []string{"url.dll,FileProtocolHandler", url}
+	case "darwin":
+		cmd = "open"
+		args = []string{url}
+	default:
+		cmd = "xdg-open"
+		args = []string{url}
+	}
+	_ = exec.Command(cmd, args...).Start()
+}
+
+func writeJSON(w http.ResponseWriter, code int, v interface{}) {
+	w.Header().Set("Content-Type", "application/json; charset=utf-8")
+	w.WriteHeader(code)
+	json.NewEncoder(w).Encode(v)
+}
+
+func handleStart(w http.ResponseWriter, r *http.Request) {
+	if r.Method != "POST" {
+		http.Error(w, "method", 405)
+		return
+	}
+	var req JobReq
+	if err := json.NewDecoder(r.Body).Decode(&req); err != nil {
+		writeJSON(w, 400, map[string]string{"error": "请求格式错误"})
+		return
+	}
+	if err := collector.Start(req); err != nil {
+		writeJSON(w, 400, map[string]string{"error": err.Error()})
+		return
+	}
+	writeJSON(w, 200, map[string]string{"ok": "started"})
+}
+
+func handleStop(w http.ResponseWriter, r *http.Request) {
+	collector.Stop()
+	writeJSON(w, 200, map[string]string{"ok": "stopping"})
+}
+
+func handleStats(w http.ResponseWriter, r *http.Request) {
+	writeJSON(w, 200, map[string]interface{}{
+		"stats":  collector.snapshot(),
+		"recent": collector.recentResults(),
+	})
+}
+
+func handleDownload(w http.ResponseWriter, r *http.Request) {
+	s := collector.snapshot()
+	if s.OutFile == "" {
+		http.Error(w, "no output yet", 404)
+		return
+	}
+	f, err := os.Open(s.OutFile)
+	if err != nil {
+		http.Error(w, err.Error(), 404)
+		return
+	}
+	defer f.Close()
+	w.Header().Set("Content-Type", "application/x-ndjson; charset=utf-8")
+	w.Header().Set("Content-Disposition", "attachment; filename=products.jsonl")
+	io.Copy(w, f)
+}
+
+func handleExportCSV(w http.ResponseWriter, r *http.Request) {
+	s := collector.snapshot()
+	if s.OutFile == "" {
+		http.Error(w, "no output yet", 404)
+		return
+	}
+	f, err := os.Open(s.OutFile)
+	if err != nil {
+		http.Error(w, err.Error(), 404)
+		return
+	}
+	defer f.Close()
+	w.Header().Set("Content-Type", "text/csv; charset=utf-8")
+	w.Header().Set("Content-Disposition", "attachment; filename=products.csv")
+	w.Write([]byte{0xEF, 0xBB, 0xBF}) // UTF-8 BOM so Excel reads Chinese correctly
+	cw := csv.NewWriter(w)
+	cw.Write([]string{"barcode", "name", "spec", "unit", "area", "manufacturer", "license", "in_price", "sell_price", "status", "fetched_at"})
+	dec := json.NewDecoder(f)
+	for {
+		var p Product
+		if err := dec.Decode(&p); err != nil {
+			break
+		}
+		cw.Write([]string{p.Barcode, p.Name, p.Spec, p.Unit, p.Area, p.Manufacturer, p.License, p.InPrice, p.SellPrice, p.Status, p.FetchedAt})
+	}
+	cw.Flush()
+}

tools/bypos-collector/web/index.html

@@ -0,0 +1,206 @@
+<!DOCTYPE html>
+<html lang="zh-CN">
+<head>
+<meta charset="utf-8"/>
+<meta name="viewport" content="width=device-width, initial-scale=1"/>
+<title>中心库商品采集器</title>
+<style>
+  * { box-sizing: border-box; }
+  body { font-family: -apple-system, "Microsoft YaHei", Arial, sans-serif; margin: 0; background:#f4f6f9; color:#222; }
+  header { background:#1f6feb; color:#fff; padding:14px 22px; font-size:18px; font-weight:600; }
+  .wrap { max-width:1080px; margin:18px auto; padding:0 16px; }
+  .card { background:#fff; border:1px solid #e3e8ef; border-radius:10px; padding:18px 20px; margin-bottom:16px; }
+  .card h3 { margin:0 0 12px; font-size:15px; color:#1f6feb; }
+  label { display:block; font-size:13px; color:#555; margin:8px 0 4px; }
+  input[type=text], input[type=number], textarea, select {
+    width:100%; padding:8px 10px; border:1px solid #cdd5e0; border-radius:6px; font-size:14px;
+  }
+  textarea { height:90px; font-family:monospace; }
+  .row { display:flex; gap:14px; flex-wrap:wrap; }
+  .row > div { flex:1; min-width:160px; }
+  .tabs { display:flex; gap:8px; margin-bottom:12px; }
+  .tab { padding:7px 16px; border:1px solid #cdd5e0; border-radius:20px; cursor:pointer; font-size:13px; background:#fff; }
+  .tab.active { background:#1f6feb; color:#fff; border-color:#1f6feb; }
+  button.primary { background:#1f6feb; color:#fff; border:none; padding:10px 22px; border-radius:6px; font-size:14px; cursor:pointer; }
+  button.danger { background:#d1242f; color:#fff; border:none; padding:10px 22px; border-radius:6px; font-size:14px; cursor:pointer; }
+  button.ghost { background:#fff; color:#1f6feb; border:1px solid #1f6feb; padding:8px 16px; border-radius:6px; cursor:pointer; font-size:13px; }
+  button:disabled { opacity:.5; cursor:not-allowed; }
+  .stats { display:flex; gap:10px; flex-wrap:wrap; }
+  .stat { flex:1; min-width:90px; background:#f7f9fc; border:1px solid #e3e8ef; border-radius:8px; padding:10px; text-align:center; }
+  .stat .n { font-size:22px; font-weight:700; }
+  .stat .l { font-size:12px; color:#777; margin-top:2px; }
+  .bar { height:10px; background:#e3e8ef; border-radius:6px; overflow:hidden; margin:10px 0; }
+  .bar > div { height:100%; background:#2da44e; width:0%; transition:width .4s; }
+  table { width:100%; border-collapse:collapse; font-size:13px; }
+  th, td { text-align:left; padding:6px 8px; border-bottom:1px solid #eef1f5; white-space:nowrap; overflow:hidden; text-overflow:ellipsis; max-width:180px; }
+  th { color:#888; font-weight:600; }
+  .hit { color:#2da44e; } .miss { color:#999; } .invalid { color:#d1242f; } .error { color:#bf8700; }
+  .hint { font-size:12px; color:#888; margin-top:6px; line-height:1.5; }
+  .est { font-size:13px; color:#1f6feb; margin-top:6px; }
+</style>
+</head>
+<body>
+<header>中心库商品采集器 · bypos-collector</header>
+<div class="wrap">
+
+  <div class="card">
+    <h3>① 规划采集范围</h3>
+    <div class="tabs">
+      <div class="tab active" data-mode="range" onclick="setMode('range')">按条码范围</div>
+      <div class="tab" data-mode="list" onclick="setMode('list')">按条码清单</div>
+    </div>
+
+    <div id="pane-range">
+      <div class="hint">EAN-13 国标条码共 13 位,最后一位是校验位由程序自动计算。下面填<b>前 12 位</b>(本体),程序逐个枚举并补校验位查询。常见前缀:69 开头为中国大陆。</div>
+      <label>快捷填充前缀(可选)</label>
+      <div class="row">
+        <div><input type="text" id="prefix" placeholder="如 690100,点下方按钮自动算区间"/></div>
+        <div style="flex:0"><button class="ghost" onclick="fillFromPrefix()">用前缀填充区间</button></div>
+      </div>
+      <div class="row">
+        <div>
+          <label>起始本体(12 位)</label>
+          <input type="text" id="start" value="690100000000" maxlength="12"/>
+        </div>
+        <div>
+          <label>结束本体(12 位)</label>
+          <input type="text" id="end" value="690100000999" maxlength="12"/>
+        </div>
+      </div>
+      <div class="est" id="est"></div>
+    </div>
+
+    <div id="pane-list" style="display:none">
+      <label>粘贴条码清单(每行一个,或用空格/逗号分隔)</label>
+      <textarea id="list" placeholder="6901028941068&#10;6920202888883"></textarea>
+    </div>
+  </div>
+
+  <div class="card">
+    <h3>② 采集参数</h3>
+    <div class="row">
+      <div>
+        <label>并发数</label>
+        <input type="number" id="concurrency" value="3" min="1" max="20"/>
+      </div>
+      <div>
+        <label>每次请求间隔(毫秒)</label>
+        <input type="number" id="delay" value="300" min="0"/>
+      </div>
+      <div>
+        <label>输出文件名</label>
+        <input type="text" id="outfile" value="products.jsonl"/>
+      </div>
+    </div>
+    <label style="margin-top:12px"><input type="checkbox" id="logmiss"/> 同时记录未命中/无效条码(默认只存命中)</label>
+    <div class="hint">速度越快越容易触发上游频控。建议并发 3、间隔 300ms 起步,稳定后再调。已采过的条码会自动跳过(断点续采)。</div>
+  </div>
+
+  <div class="card">
+    <h3>③ 运行</h3>
+    <div style="margin-bottom:12px">
+      <button class="primary" id="btnStart" onclick="start()">开始采集</button>
+      <button class="danger" id="btnStop" onclick="stop()" disabled>停止</button>
+      <button class="ghost" onclick="window.open('/api/download')">下载 JSONL</button>
+      <button class="ghost" onclick="window.open('/api/export.csv')">导出 CSV(Excel)</button>
+    </div>
+    <div class="bar"><div id="prog"></div></div>
+    <div class="stats">
+      <div class="stat"><div class="n" id="s-done">0</div><div class="l">已处理</div></div>
+      <div class="stat"><div class="n" id="s-total">0</div><div class="l">总计</div></div>
+      <div class="stat"><div class="n hit" id="s-hits">0</div><div class="l">命中</div></div>
+      <div class="stat"><div class="n miss" id="s-miss">0</div><div class="l">未命中</div></div>
+      <div class="stat"><div class="n invalid" id="s-invalid">0</div><div class="l">无效</div></div>
+      <div class="stat"><div class="n error" id="s-errors">0</div><div class="l">错误</div></div>
+      <div class="stat"><div class="n" id="s-skipped">0</div><div class="l">跳过</div></div>
+    </div>
+    <div class="hint" id="msg"></div>
+  </div>
+
+  <div class="card">
+    <h3>④ 实时结果(最近 60 条)</h3>
+    <div style="max-height:340px; overflow:auto">
+      <table>
+        <thead><tr><th>条码</th><th>品名</th><th>规格</th><th>单位</th><th>产地</th><th>进价</th><th>零售价</th><th>状态</th></tr></thead>
+        <tbody id="rows"></tbody>
+      </table>
+    </div>
+  </div>
+</div>
+
+<script>
+let mode = 'range';
+function setMode(m){
+  mode = m;
+  document.querySelectorAll('.tab').forEach(t=>t.classList.toggle('active', t.dataset.mode===m));
+  document.getElementById('pane-range').style.display = m==='range'?'block':'none';
+  document.getElementById('pane-list').style.display = m==='list'?'block':'none';
+}
+function fillFromPrefix(){
+  let p = document.getElementById('prefix').value.replace(/[^0-9]/g,'');
+  if(!p){ alert('请先填前缀'); return; }
+  if(p.length>=12){ alert('前缀太长,应少于 12 位'); return; }
+  let pad = 12 - p.length;
+  document.getElementById('start').value = p + '0'.repeat(pad);
+  document.getElementById('end').value   = p + '9'.repeat(pad);
+  updateEst();
+}
+function updateEst(){
+  let s = document.getElementById('start').value.replace(/[^0-9]/g,'');
+  let e = document.getElementById('end').value.replace(/[^0-9]/g,'');
+  if(s.length===12 && e.length===12){
+    let n = (BigInt(e) - BigInt(s)) + 1n;
+    document.getElementById('est').textContent = '本次将查询约 ' + n.toString() + ' 个条码';
+  } else {
+    document.getElementById('est').textContent = '';
+  }
+}
+document.getElementById('start').addEventListener('input', updateEst);
+document.getElementById('end').addEventListener('input', updateEst);
+updateEst();
+
+async function start(){
+  let body = {
+    mode: mode,
+    start_body: document.getElementById('start').value.trim(),
+    end_body: document.getElementById('end').value.trim(),
+    list: document.getElementById('list').value,
+    concurrency: parseInt(document.getElementById('concurrency').value)||3,
+    delay_ms: parseInt(document.getElementById('delay').value)||0,
+    log_miss: document.getElementById('logmiss').checked,
+    out_file: document.getElementById('outfile').value.trim()
+  };
+  let r = await fetch('/api/start', {method:'POST', headers:{'Content-Type':'application/json'}, body:JSON.stringify(body)});
+  let j = await r.json();
+  if(j.error){ alert('启动失败: ' + j.error); return; }
+}
+async function stop(){ await fetch('/api/stop', {method:'POST'}); }
+
+function esc(s){ return (s||'').replace(/[&<>]/g, c=>({'&':'&amp;','<':'&lt;','>':'&gt;'}[c])); }
+
+async function poll(){
+  try{
+    let r = await fetch('/api/stats'); let j = await r.json();
+    let s = j.stats;
+    document.getElementById('s-done').textContent = s.done;
+    document.getElementById('s-total').textContent = s.total;
+    document.getElementById('s-hits').textContent = s.hits;
+    document.getElementById('s-miss').textContent = s.miss;
+    document.getElementById('s-invalid').textContent = s.invalid;
+    document.getElementById('s-errors').textContent = s.errors;
+    document.getElementById('s-skipped').textContent = s.skipped;
+    let pct = s.total>0 ? Math.floor(s.done*100/s.total) : 0;
+    document.getElementById('prog').style.width = pct + '%';
+    document.getElementById('msg').textContent = (s.running? ('采集中… 当前 '+s.current) : (s.message||'空闲'));
+    document.getElementById('btnStart').disabled = s.running;
+    document.getElementById('btnStop').disabled = !s.running;
+    let rows = (j.recent||[]).slice().reverse().map(p=>
+      '<tr><td>'+esc(p.barcode)+'</td><td>'+esc(p.name)+'</td><td>'+esc(p.spec)+'</td><td>'+esc(p.unit)+'</td><td>'+esc(p.area)+'</td><td>'+esc(p.in_price)+'</td><td>'+esc(p.sell_price)+'</td><td class="'+p.status+'">'+esc(p.status)+'</td></tr>'
+    ).join('');
+    document.getElementById('rows').innerHTML = rows;
+  }catch(e){}
+}
+setInterval(poll, 1000); poll();
+</script>
+</body>
+</html>

tools/bypos-collector/使用说明.md

@@ -0,0 +1,68 @@
+# 中心库商品采集器 bypos-collector 使用说明
+
+一个单文件 Windows 小程序,通过云店「新增商品」用到的同一个中心商品库
+(`zc.bypos.net`)按条码批量采集商品档案(品名/规格/单位/产地/厂商/建议进价/建议零售价),
+存到本地,供后续导入天工(goods)系统。
+
+## 一、运行
+
+1. 把 `bypos-collector.exe` 放到任意空文件夹(采集结果会生成在同一文件夹)。
+2. 双击运行。会弹出一个黑色命令行窗口(不要关它),并自动打开浏览器控制台
+   `http://127.0.0.1:8765/`。
+   - 若没自动打开,手动在浏览器输入上面这个地址。
+3. 用完直接关掉那个命令行窗口即可退出。
+
+## 二、采集
+
+控制台分四步:
+
+**① 规划采集范围** —— 两种方式二选一:
+- **按条码范围**:EAN-13 国标条码共 13 位,最后一位是校验位,程序自动算。
+  你只填**前 12 位**的起止区间。可在「快捷填充前缀」里填如 `690100`,
+  点按钮自动生成区间(`690100000000` ~ `690100999999`)。
+- **按条码清单**:直接粘贴一批条码(每行一个,或空格/逗号分隔)。
+
+**② 采集参数**:
+- 并发数(默认 3)、请求间隔(默认 300ms):**越慢越安全**,上游可能对账号限频。
+- 输出文件名(默认 `products.jsonl`)。
+- 「同时记录未命中/无效条码」:默认只存命中的;勾上会把未命中也记下来。
+
+**③ 运行**:点「开始采集」。进度、命中/未命中/错误实时显示。
+已经采过的条码会自动跳过(可随时停了再开,断点续采)。
+
+**④ 实时结果**:最近 60 条滚动显示。
+
+## 三、导出
+
+- 「下载 JSONL」:原始数据(每行一个 JSON),用于导入天工系统。
+- 「导出 CSV」:Excel 可直接打开查看。
+
+## 四、字段说明(JSONL 每行)
+
+| 字段 | 含义 |
+| --- | --- |
+| barcode | 条码(GTIN/EAN-13) |
+| name | 品名 |
+| spec | 规格 |
+| unit | 单位 |
+| area | 产地/地区 |
+| manufacturer | 生产企业(常为空) |
+| license | 生产许可(常为空) |
+| in_price | 建议进价 |
+| sell_price | 建议零售价 |
+| status | hit=命中 / miss=不存在 / invalid=非国标条码 / error=请求出错 |
+| fetched_at | 采集时间 |
+
+## 五、注意
+
+- 这是用云店账号授权去查上游中心库,**批量自动**比页面里一条条查更"重",
+  上游厂商可能对账号做频控/限额。请低速、分批("一点点采"),发现大量报错就降速。
+- 全量 69 段是个天文数字,不要无脑全跑;建议按你关心的品牌/品类前缀分批。
+
+## 六、命令行参数(可选)
+
+```
+bypos-collector.exe -addr 127.0.0.1:8765   # 改监听端口
+bypos-collector.exe -no-open               # 不自动开浏览器
+bypos-collector.exe -sdogid 137966         # 指定中心库账号 id
+```