修改 README

.claude/settings.local.json

@@ -6,7 +6,9 @@
       "Bash(python -c \"import ast, sys; ast.parse\\(open\\('sap-c4c-AttachmentFolder.py'\\).read\\(\\)\\); print\\('语法检查通过'\\)\")",
       "Bash(git commit:*)",
       "Bash(head:*)",
-      "Bash(python3:*)"
+      "Bash(python3:*)",
+      "Bash(curl -sI \"https://cdn.playwright.dev/chrome-for-testing-public/145.0.7632.6/linux64/chrome-headless-shell-linux64.zip\")",
+      "Bash(pip3 show:*)"
     ]
   }
 }

README.md

@@ -32,6 +32,7 @@ python -m playwright install-deps chromium
 | `datasource/` | 存放 SAP Analytics 导出的 CSV 文件 |
 | `downloads/` | 本地临时下载目录（上传 DSM 后自动清理） |
 | `error_log.txt` | 错误日志，每次批量运行时重置 |
+| `failed_tickets.csv` | 失败的 Ticket ID 列表，可用于下次重试 |
 
 ---
 
@@ -83,11 +84,37 @@ python sap-c4c-AttachmentFolder.py ... --json
 python batch_download.py
 ```
 
-- 默认读取 CSV 第二列的 Ticket ID，取前 10 个
+- 读取 datasource 目录下 CSV/xlsx 的第二列，获取所有 Ticket ID
 - 最多 5 个任务并行执行
 - 每个 Ticket 使用独立子目录 `downloads/{ticket_id}`，避免并行冲突
 - 上传 DSM 完成后自动清理本地文件和子目录
-- 错误记录到 `error_log.txt`
+- 错误记录到 `error_log.txt`，失败的 Ticket ID 保存到 `failed_tickets.csv`
+
+### 守护进程
+
+```shell
+  #启动后台下载
+  nohup python3 batch_download.py > run.log 2>&1 &
+  echo $! > batch.pid
+
+  #查看实时进度
+  tail -f run.log
+
+  #查看进程是否还在运行
+  cat batch.pid | xargs ps -p
+
+  #结束进程
+  cat batch.pid | xargs kill
+```
+
+### 重试失败的 Ticket
+
+将上一次运行生成的 `failed_tickets.csv` 移动到 `datasource/` 目录，然后重新运行：
+
+```bash
+mv failed_tickets.csv datasource/
+python batch_download.py
+```
 
 ### 修改并行数或 Ticket 数量
 

batch_download.py

@@ -24,29 +24,45 @@ WORKERS     = 5
 
 SCRIPT      = os.path.join(os.path.dirname(__file__), "sap-c4c-AttachmentFolder.py")
 ERROR_LOG   = os.path.join(os.path.dirname(__file__), "error_log.txt")
+FAILED_CSV  = os.path.join(os.path.dirname(__file__), "failed_tickets.csv")
 DATASOURCE  = os.path.join(os.path.dirname(__file__), "datasource")
 
 print_lock = threading.Lock()
+failed_lock = threading.Lock()
+failed_ids = set()
 # ─────────────────────────────────────────────────────────────────────────────
 
 
-def get_ticket_ids(limit=10):
+def get_ticket_ids():
     ids, seen = [], set()
+
+    def _add(val):
+        val = str(val).strip()
+        if val and val not in seen:
+            seen.add(val)
+            ids.append(val)
+
     for csv_file in glob.glob(os.path.join(DATASOURCE, "*.csv")):
         with open(csv_file, encoding="utf-8-sig") as f:
             reader = csv.reader(f)
             next(reader, None)
             for row in reader:
-                if len(row) < 2:
+                if len(row) >= 2:
+                    _add(row[1])
+
+    for xlsx_file in glob.glob(os.path.join(DATASOURCE, "*.xlsx")):
+        import openpyxl
+        wb = openpyxl.load_workbook(xlsx_file, read_only=True, data_only=True)
+        ws = wb.active
+        first = True
+        for row in ws.iter_rows(min_col=2, max_col=2, values_only=True):
+            if first:
+                first = False
                 continue
-                val = row[1].strip()
+            if row[0] is not None:
-                if val and val not in seen:
+                _add(row[0])
-                    seen.add(val)
+        wb.close()
-                    ids.append(val)
+
-                if len(ids) >= limit:
-                    break
-        if len(ids) >= limit:
-            break
     return ids
 
 
@@ -54,6 +70,11 @@ def log_error(ticket_id, message):
     ts = datetime.datetime.now().strftime("%Y-%m-%d %H:%M:%S")
     with open(ERROR_LOG, "a", encoding="utf-8") as f:
         f.write(f"[{ts}] Ticket {ticket_id}: {message}\n")
+    with failed_lock:
+        if ticket_id not in failed_ids:
+            failed_ids.add(ticket_id)
+            with open(FAILED_CSV, "a", encoding="utf-8", newline="") as f:
+                csv.writer(f).writerow([ticket_id])
 
 
 def run_ticket(ticket_id, index, total):
@@ -117,16 +138,19 @@ def run_ticket(ticket_id, index, total):
 
 
 def main():
+    global failed_ids
     print("读取 Ticket ID ...")
-    ids = get_ticket_ids(10)
+    ids = get_ticket_ids()
     if not ids:
         print("未找到任何 Ticket ID，请检查 datasource 目录")
         sys.exit(1)
 
     print(f"共 {len(ids)} 个 Ticket: {', '.join(ids)}")
 
-    # 清空/创建 error_log
+    # 清空/创建 error_log 和 failed_tickets.csv
     open(ERROR_LOG, "w").close()
+    open(FAILED_CSV, "w", encoding="utf-8", newline="").close()
+    failed_ids.clear()
 
     with ThreadPoolExecutor(max_workers=WORKERS) as executor:
         futures = {executor.submit(run_ticket, tid, i, len(ids)): tid
@@ -135,10 +159,12 @@ def main():
             future.result()  # 触发异常传播（已在 run_ticket 内处理）
 
     print("\n全部完成。")
+    if failed_ids:
+        print(f"失败 {len(failed_ids)} 个 Ticket，已保存到 {FAILED_CSV}")
         if os.path.getsize(ERROR_LOG) > 0:
-        print(f"有错误，详见 {ERROR_LOG}")
+            print(f"错误详情见 {ERROR_LOG}")
     else:
-        print("无错误。")
+        print("全部成功，无失败。")
 
 
 if __name__ == "__main__":

requirements.txt

@@ -1,5 +1,6 @@
 requests
 urllib3
+openpyxl
 # 可选：Salesforce 外链附件爬取
 scrapling[all]
 playwright