bypos-collector: 移除硬编码 sdogid、添加重试、补全 CSV 导出、增加测试

改进内容:
- sdogid 不再硬编码，优先从 $BYPOS_SDOGID 环境变量读取，其次 -sdogid 参数，
  也可在 Web 控制台输入框填写；三者都未设时启动警告、采集报错
- 网络错误自动重试（最多 3 次，指数退避 500ms/1s/2s）
- Web 控制台采集参数区增加 sdogid 输入框
- CSV 导出补全 retmsg 和 source 两列
- 新增 Go 单元测试（ean13Check、md5hex、sanitizeBarcodes、resolveSdogID 等）
- README 更新配置说明

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

tools/bypos-collector/README.md

@@ -58,9 +58,10 @@ GET http://zc.bypos.net/byGoodsService/byMessage.asmx/GetGoodsinfo
 
 ## 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。
+- `sdogid`：中心库账号 id（本项目所属云店账号的授权 id）。
+  优先从环境变量 `BYPOS_SDOGID` 读取，其次是命令行 `-sdogid` 参数，
+  也可以在 Web 控制台的输入框中填写。**三者都未设时启动会警告、开始采集时会报错。**
+- 网络错误自动重试（最多 3 次，指数退避 500ms/1s/2s）。
 
 ## 导入 goods/天工库
 

tools/bypos-collector/collect.go

@@ -20,12 +20,21 @@ import (
 )
 
 // ---- 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"
 
+// maxRetries is the number of retry attempts for transient network errors.
+const maxRetries = 3
+
+// resolveSdogID reads the account id from $BYPOS_SDOGID, falling back to the
+// explicit argument (CLI flag or Web UI input). Returns empty if neither set.
+func resolveSdogID(explicit string) string {
+	if v := os.Getenv("BYPOS_SDOGID"); v != "" {
+		return v
+	}
+	return explicit
+}
+
 var stringTagRe = regexp.MustCompile(`(?s)<string[^>]*>(.*)</string>`)
 
 // Product is the normalized record we persist (one JSON object per line).
@@ -87,8 +96,8 @@ func ean13Check(body string) (string, bool) {
 	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) {
+// lookupOnce performs a single HTTP request to the upstream central library.
+func (c *Collector) lookupOnce(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)
@@ -147,6 +156,30 @@ func (c *Collector) lookup(ctx context.Context, barcode string) (*Product, error
 	return p, nil
 }
 
+// lookup queries the upstream with up to maxRetries retries on transient errors.
+func (c *Collector) lookup(ctx context.Context, barcode string) (*Product, error) {
+	var lastErr error
+	for attempt := 0; attempt <= maxRetries; attempt++ {
+		if ctx.Err() != nil {
+			return nil, ctx.Err()
+		}
+		p, err := c.lookupOnce(ctx, barcode)
+		if err == nil {
+			return p, nil
+		}
+		lastErr = err
+		if attempt < maxRetries {
+			backoff := time.Duration(1<<uint(attempt)) * 500 * time.Millisecond
+			select {
+			case <-time.After(backoff):
+			case <-ctx.Done():
+				return nil, ctx.Err()
+			}
+		}
+	}
+	return nil, lastErr
+}
+
 // ---- job / collector state ----
 
 type Stats struct {
@@ -187,9 +220,7 @@ type Collector struct {
 }
 
 func NewCollector(sdogID string) *Collector {
-	if sdogID == "" {
-		sdogID = defaultSdogID
-	}
+	sdogID = resolveSdogID(sdogID)
 	c := &Collector{
 		client: &http.Client{Timeout: 25 * time.Second},
 		sdogID: sdogID,
@@ -308,6 +339,9 @@ func (c *Collector) Start(req JobReq) error {
 	if req.SdogID != "" {
 		c.sdogID = req.SdogID
 	}
+	if c.sdogID == "" {
+		return fmt.Errorf("未指定 sdogid，请通过控制台输入框、环境变量 $BYPOS_SDOGID 或 -sdogid 参数配置")
+	}
 
 	// Build the list of barcodes to query.
 	var barcodes []string

tools/bypos-collector/collect_test.go

@@ -0,0 +1,113 @@
+package main
+
+import (
+	"os"
+	"testing"
+)
+
+func TestEan13Check_valid(t *testing.T) {
+	tests := []struct {
+		body string
+		want string
+	}{
+		{"692045990501", "6920459905012"},
+		{"690100000000", "6901000000004"},
+		{"690100000099", "6901000000998"},
+	}
+	for _, tt := range tests {
+		got, ok := ean13Check(tt.body)
+		if !ok {
+			t.Errorf("ean13Check(%q) returned not ok", tt.body)
+			continue
+		}
+		if got != tt.want {
+			t.Errorf("ean13Check(%q) = %q, want %q", tt.body, got, tt.want)
+		}
+	}
+}
+
+func TestEan13Check_invalid(t *testing.T) {
+	cases := []string{"", "12345", "12345678901", "1234567890123", "69010000a000"}
+	for _, body := range cases {
+		_, ok := ean13Check(body)
+		if ok {
+			t.Errorf("ean13Check(%q) should return not ok", body)
+		}
+	}
+}
+
+func TestMd5hex(t *testing.T) {
+	got := md5hex("137966")
+	if len(got) != 32 {
+		t.Errorf("md5hex should return 32-char hex, got len=%d", len(got))
+	}
+	if got != md5hex("137966") {
+		t.Error("md5hex should be deterministic")
+	}
+	if got == md5hex("other") {
+		t.Error("md5hex should differ for different inputs")
+	}
+}
+
+func TestSanitizeBarcodes(t *testing.T) {
+	got := sanitizeBarcodes("6920459905012\n6901028941068  123")
+	want := []string{"6920459905012", "6901028941068", "123"}
+	if len(got) != len(want) {
+		t.Fatalf("len = %d, want %d", len(got), len(want))
+	}
+	for i := range want {
+		if got[i] != want[i] {
+			t.Errorf("got[%d] = %q, want %q", i, got[i], want[i])
+		}
+	}
+}
+
+func TestSanitizeBarcodes_empty(t *testing.T) {
+	got := sanitizeBarcodes("  \n\t ")
+	if len(got) != 0 {
+		t.Errorf("expected empty, got %v", got)
+	}
+}
+
+func TestResolveSdogID_envOverride(t *testing.T) {
+	os.Setenv("BYPOS_SDOGID", "999999")
+	defer os.Unsetenv("BYPOS_SDOGID")
+	got := resolveSdogID("111111")
+	if got != "999999" {
+		t.Errorf("expected env var override, got %q", got)
+	}
+}
+
+func TestResolveSdogID_fallback(t *testing.T) {
+	os.Unsetenv("BYPOS_SDOGID")
+	got := resolveSdogID("111111")
+	if got != "111111" {
+		t.Errorf("expected fallback to explicit, got %q", got)
+	}
+}
+
+func TestResolveSdogID_empty(t *testing.T) {
+	os.Unsetenv("BYPOS_SDOGID")
+	got := resolveSdogID("")
+	if got != "" {
+		t.Errorf("expected empty, got %q", got)
+	}
+}
+
+func TestNewCollector_sdogFromEnv(t *testing.T) {
+	os.Setenv("BYPOS_SDOGID", "888888")
+	defer os.Unsetenv("BYPOS_SDOGID")
+	c := NewCollector("")
+	if c.sdogID != "888888" {
+		t.Errorf("expected sdogID from env, got %q", c.sdogID)
+	}
+}
+
+func TestStartRejectsEmptySdogID(t *testing.T) {
+	os.Unsetenv("BYPOS_SDOGID")
+	c := NewCollector("")
+	err := c.Start(JobReq{Mode: "list", List: "6920459905012"})
+	if err == nil {
+		t.Fatal("expected error for empty sdogid")
+	}
+}

tools/bypos-collector/main.go

@@ -20,15 +20,16 @@ import (
 //go:embed web/*
 var webFS embed.FS
 
-var collector = NewCollector("")
+var collector *Collector
 
 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)")
+	sdog := flag.String("sdogid", "", "中心库账号 id（优先读 $BYPOS_SDOGID 环境变量）")
 	flag.Parse()
-	if *sdog != "" {
-		collector.sdogID = *sdog
+	collector = NewCollector(*sdog)
+	if collector.sdogID == "" {
+		log.Println("警告: 未配置 sdogid，请通过 $BYPOS_SDOGID 环境变量、-sdogid 参数或控制台输入框指定")
 	}
 
 	sub, _ := fs.Sub(webFS, "web")
@@ -143,14 +144,14 @@ func handleExportCSV(w http.ResponseWriter, r *http.Request) {
 	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"})
+	cw.Write([]string{"barcode", "name", "spec", "unit", "area", "manufacturer", "license", "in_price", "sell_price", "status", "retmsg", "fetched_at", "source"})
 	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.Write([]string{p.Barcode, p.Name, p.Spec, p.Unit, p.Area, p.Manufacturer, p.License, p.InPrice, p.SellPrice, p.Status, p.RetMsg, p.FetchedAt, p.Source})
 	}
 	cw.Flush()
 }

tools/bypos-collector/web/index.html

@@ -79,6 +79,10 @@
   <div class="card">
     <h3>② 采集参数</h3>
     <div class="row">
+      <div>
+        <label>中心库账号 sdogid</label>
+        <input type="text" id="sdogid" placeholder="留空则使用环境变量或启动参数"/>
+      </div>
       <div>
         <label>并发数</label>
         <input type="number" id="concurrency" value="3" min="1" max="20"/>
@@ -165,6 +169,7 @@ async function start(){
     start_body: document.getElementById('start').value.trim(),
     end_body: document.getElementById('end').value.trim(),
     list: document.getElementById('list').value,
+    sdog_id: document.getElementById('sdogid').value.trim(),
     concurrency: parseInt(document.getElementById('concurrency').value)||3,
     delay_ms: parseInt(document.getElementById('delay').value)||0,
     log_miss: document.getElementById('logmiss').checked,