🛠️ Add capture reprocess batch helper script

This commit is contained in:
d-robotics 2026-10-08 03:30:38 -07:00
parent 8f36cd1705
commit 7a98233bf0

166
scripts/capture_reprocess.py Executable file
View File

@ -0,0 +1,166 @@
#!/usr/bin/env python3
"""批量查看并重新处理失败的 Capture 记录。
用法(先看失败原因,确认后再触发):
# 1) 列出某工作流所有 failed / cancelled 记录及失败原因(不触发任何操作)
export DROBOTICFLOW_PASSWORD='...'
python3 scripts/capture_reprocess.py --base-url https://<线上地址> \
--phone <管理员手机号> --workflow workflow-xxxx
# 2) 确认失败原因与数据已修复后,真正触发批量重新处理
python3 scripts/capture_reprocess.py ... --go --reason "源数据已修复"
说明:
- 重新处理接口对每条记录单独调用 POST
/api/workflows/{id}/capture-automation/records/{recordId}/reprocess,
服务端要求记录状态为 failed 或 cancelled、工作流为持续 Capture 类型。
- 默认每条之间间隔 2 秒,避免瞬间压入过多 Run;可用 --sleep 调整。
- 账号需要组织管理员权限(与页面上「重新处理」按钮一致)。
- 列表接口一次最多返回 200 条记录;更多记录需分批处理。
- 密码只从环境变量 DROBOTICFLOW_PASSWORD 读取,不进命令行历史。
"""
import argparse
import json
import os
import sys
import time
import urllib.error
import urllib.request
import http.cookiejar
from typing import Optional
def build_client(base_url: str) -> tuple:
jar = http.cookiejar.CookieJar()
opener = urllib.request.build_opener(urllib.request.HTTPCookieProcessor(jar))
return opener, jar
def request_json(opener, method: str, url: str, payload: Optional[dict], csrf: Optional[str]) -> dict:
data = None
headers = {"Accept": "application/json"}
if payload is not None:
data = json.dumps(payload).encode("utf-8")
headers["Content-Type"] = "application/json"
if csrf:
headers["X-Droboticflow-CSRF-Token"] = csrf
req = urllib.request.Request(url, data=data, headers=headers, method=method)
try:
with opener.open(req, timeout=120) as response:
body = response.read().decode("utf-8")
return json.loads(body) if body.strip() else {}
except urllib.error.HTTPError as error:
body = error.read().decode("utf-8", errors="replace")
try:
message = json.loads(body).get("error", body)
except json.JSONDecodeError:
message = body
raise RuntimeError(f"HTTP {error.code}: {message}") from error
def csrf_from_jar(jar) -> Optional[str]:
for cookie in jar:
if cookie.name == "droboticflow_csrf":
return cookie.value
return None
def login(opener, jar, base_url: str, phone: str, password: str) -> str:
request_json(opener, "POST", f"{base_url}/api/auth/login", {"phone": phone, "password": password}, None)
csrf = csrf_from_jar(jar)
if not csrf:
raise RuntimeError("登录成功但未获得 CSRF cookie,无法执行写操作")
return csrf
def truncate(text: str, width: int) -> str:
text = " ".join((text or "").split())
return text if len(text) <= width else text[: width - 1] + "…"
def main() -> int:
parser = argparse.ArgumentParser(description="批量查看 / 重新处理失败的 Capture 记录")
parser.add_argument("--base-url", required=True, help="平台地址,如 https://example.com(不带末尾斜杠)")
parser.add_argument("--workflow", required=True, help="工作流 ID(workflow-…)")
parser.add_argument("--phone", help="管理员手机号")
parser.add_argument("--password", help="密码(建议改用环境变量 DROBOTICFLOW_PASSWORD)")
parser.add_argument("--status", default="failed,cancelled", help="要处理的记录状态,逗号分隔(默认 failed,cancelled)")
parser.add_argument("--limit", type=int, default=0, help="最多处理多少条(0 = 全部匹配)")
parser.add_argument("--go", action="store_true", help="真正触发重新处理;缺省只列出记录与失败原因")
parser.add_argument("--reason", default="数据已修复,批量重新处理", help="重新处理原因(记录在操作历史里)")
parser.add_argument("--sleep", type=float, default=2.0, help="每条请求之间的间隔秒数(默认 2)")
args = parser.parse_args()
password = args.password or os.environ.get("DROBOTICFLOW_PASSWORD")
if not password:
print("缺少密码:设置环境变量 DROBOTICFLOW_PASSWORD,或用 --password", file=sys.stderr)
return 2
if not args.phone:
print("缺少手机号:用 --phone", file=sys.stderr)
return 2
base_url = args.base_url.rstrip("/")
wanted = {item.strip() for item in args.status.split(",") if item.strip()}
opener, jar = build_client(base_url)
try:
csrf = login(opener, jar, base_url, args.phone, password)
except RuntimeError as error:
print(f"登录失败:{error}", file=sys.stderr)
return 1
print(f"已登录,开始查询工作流 {args.workflow} …\n")
try:
automation = request_json(opener, "GET", f"{base_url}/api/workflows/{args.workflow}/capture-automation", None, None)
except RuntimeError as error:
print(f"读取记录失败:{error}", file=sys.stderr)
return 1
counts = automation.get("counts") or {}
print("状态计数:", ",".join(f"{key}={value}" for key, value in sorted(counts.items()) if value))
records = [r for r in (automation.get("records") or []) if r.get("status") in wanted]
if args.limit > 0:
records = records[: args.limit]
print(f"匹配 {args.status} 的记录共 {len(records)} 条(接口单次最多返回 200 条)\n")
if not records:
print("没有匹配的记录。列表按更新时间倒序,失败记录可能被新记录挤出前 200 条。")
return 0
for record in records:
run_id = record.get("runId") or "—"
error = record.get("error") or record.get("failedNodeID") or "(无错误信息)"
print(f"{record['id']}")
print(f" Capture: {record.get('captureId', '—')} 状态: {record.get('status')} Run: {run_id}")
print(f" 失败信息: {truncate(error, 120)}")
if not args.go:
print(f"\n以上为只读预览。确认数据已修复后加 --go 触发重新处理(原因可用 --reason 指定)。")
return 0
print(f"\n开始批量重新处理(原因:{args.reason}),间隔 {args.sleep}s …")
ok, failed = 0, 0
for index, record in enumerate(records, 1):
record_id = record["id"]
url = f"{base_url}/api/workflows/{args.workflow}/capture-automation/records/{record_id}/reprocess"
try:
result = request_json(opener, "POST", url, {"reason": args.reason}, csrf)
except RuntimeError as error:
failed += 1
print(f"[{index}/{len(records)}] {record_id} ✗ {error}")
else:
after = next((r for r in (result.get("records") or []) if r.get("id") == record_id), {})
ok += 1
print(f"[{index}/{len(records)}] {record_id} ✓ 现在状态: {after.get('status', 'requeueing/queued')} Run: {after.get('runId', '—')}")
if index < len(records):
time.sleep(args.sleep)
print(f"\n完成:成功 {ok} 条,失败 {failed} 条。")
print("注意:排队成功不代表处理成功;失败原因相同的数据请先修复,否则会再次失败。")
return 1 if failed else 0
if __name__ == "__main__":
sys.exit(main())