auto-sync: feishu-group-msg-sync 2026-04-10_19:53
This commit is contained in:
commit
a3fe9418a2
111
SKILL.md
Normal file
111
SKILL.md
Normal file
@ -0,0 +1,111 @@
|
||||
---
|
||||
name: feishu-group-msg-sync
|
||||
description: 定期同步飞书群聊消息到飞书电子表格。以Bot身份拉取群消息(文本/图片/视频/音频/文件),非文本媒体上传到腾讯COS生成可访问链接,所有记录追加写入飞书电子表格。支持crontab定时执行。触发场景:群消息记录、群聊同步到表格、群聊数据采集、群反馈收集。
|
||||
---
|
||||
|
||||
# 飞书群聊消息同步到电子表格
|
||||
|
||||
定期将飞书群聊消息同步到飞书电子表格,非文本媒体上传到COS。
|
||||
|
||||
## 工作流程
|
||||
|
||||
1. 以Bot身份从飞书群拉取增量消息(基于上次同步时间)
|
||||
2. 文本消息直接记录内容
|
||||
3. 媒体消息(图片/视频/音频/文件)下载后上传到腾讯COS,记录可访问URL
|
||||
4. 所有记录追加写入飞书电子表格
|
||||
|
||||
## 前置条件
|
||||
|
||||
- Bot已加入目标飞书群
|
||||
- Bot有权读取群消息(`im:message` / `im:message.group_at_msg` scope)
|
||||
- Bot有权写入目标电子表格(已被添加为协作者)
|
||||
- 腾讯COS凭证可用(参考 `tencent-cos-upload` skill)
|
||||
- Python依赖:`cos-python-sdk-v5`、`lark-cli`
|
||||
|
||||
## 使用方式
|
||||
|
||||
### 快速部署
|
||||
|
||||
1. 复制脚本模板并修改配置:
|
||||
```bash
|
||||
cp scripts/sync_group_to_sheet.py /path/to/your/scripts/
|
||||
# 编辑脚本顶部的配置区域
|
||||
```
|
||||
|
||||
2. 修改脚本中的配置常量:
|
||||
```python
|
||||
CHAT_ID = "oc_xxx" # 目标群ID
|
||||
SPREADSHEET_TOKEN = "xxx" # 电子表格token
|
||||
SHEET_ID = "xxx" # sheet页ID
|
||||
LARK_CLI_CONFIG = "/root/.openclaw/credentials/<agent>" # Bot凭证目录
|
||||
```
|
||||
|
||||
3. 设置crontab:
|
||||
```bash
|
||||
# 每小时执行
|
||||
0 * * * * /usr/bin/python3 /path/to/sync_group_to_sheet.py >> /var/log/sync.log 2>&1
|
||||
```
|
||||
|
||||
### 关键配置说明
|
||||
|
||||
#### 获取群ID
|
||||
搜索群或从群设置中获取 `chat_id`(格式 `oc_xxx`)。
|
||||
|
||||
#### 获取电子表格信息
|
||||
```bash
|
||||
# 如果是wiki下的表格,先获取obj_token
|
||||
LARKSUITE_CLI_CONFIG_DIR=<credentials> lark-cli wiki spaces get_node \
|
||||
--params '{"token":"<wiki_token>"}' --as bot
|
||||
|
||||
# 获取sheet_id
|
||||
LARKSUITE_CLI_CONFIG_DIR=<credentials> lark-cli sheets +info \
|
||||
--spreadsheet-token <token> --as bot
|
||||
```
|
||||
|
||||
#### 电子表格列结构
|
||||
|
||||
脚本默认写入4列,在表格首行设置表头:
|
||||
|
||||
| 列 | 内容 | 示例 |
|
||||
|----|------|------|
|
||||
| A | 时间 | 2026-04-10 15:30 |
|
||||
| B | 反馈人 | 张三 |
|
||||
| C | 信息类型 | 文本/图片/视频(15s)/语音/文件 |
|
||||
| D | 信息内容(或地址) | 文本内容 或 https://domain/path/file.png |
|
||||
|
||||
## COS目录结构
|
||||
|
||||
媒体文件按类型和日期组织:
|
||||
```
|
||||
{cos_base_path}/
|
||||
├── image/{date}/{msgid}.png
|
||||
├── video/{date}/{msgid}.mp4
|
||||
├── audio/{date}/{msgid}.ogg
|
||||
└── file/{date}/{msgid}.ext
|
||||
```
|
||||
|
||||
## 增量同步机制
|
||||
|
||||
- 同步时间记录在 `LAST_SYNC_FILE`(默认 `/tmp/last_feedback_sync_time`)
|
||||
- 每次同步拉取 `上次同步时间 → 当前时间` 的消息
|
||||
- 仅在写入成功后更新同步时间,确保不丢消息
|
||||
- 首次运行自动从1小时前开始
|
||||
|
||||
## 消息类型处理
|
||||
|
||||
| 消息类型 | 处理方式 |
|
||||
|----------|---------|
|
||||
| text/post | 直接记录文本内容 |
|
||||
| image | 下载 → 上传COS → 记录URL |
|
||||
| media(视频) | 下载 → 上传COS → 记录URL,类型含时长 |
|
||||
| audio | 下载 → 上传COS → 记录URL |
|
||||
| file | 下载 → 上传COS → 记录URL |
|
||||
| sticker | 记录为"表情包" |
|
||||
| system | 跳过(入群退群等系统消息) |
|
||||
| 已删除 | 跳过 |
|
||||
|
||||
## 自定义扩展
|
||||
|
||||
如需修改表格列结构、增加字段(如message_id、是否为回复等),编辑脚本中的 `process_message()` 函数返回的行数据。
|
||||
|
||||
如需同步多个群到同一表格,复制脚本并修改 `CHAT_ID`,或改造为从配置文件读取多群列表。
|
||||
76
references/lark-cli-cheatsheet.md
Normal file
76
references/lark-cli-cheatsheet.md
Normal file
@ -0,0 +1,76 @@
|
||||
# lark-cli 消息和表格操作速查
|
||||
|
||||
## 拉取群消息
|
||||
|
||||
```bash
|
||||
lark-cli im +chat-messages-list \
|
||||
--chat-id "oc_xxx" \
|
||||
--start "2026-04-10T00:00:00+08:00" \
|
||||
--end "2026-04-10T23:59:59+08:00" \
|
||||
--sort asc \
|
||||
--page-size 50 \
|
||||
--as bot
|
||||
```
|
||||
|
||||
注意:
|
||||
- `--start`/`--end` 使用 ISO 8601 格式(含时区)
|
||||
- `--sort` 值为 `asc` 或 `desc`(不是 `create_time_asc`)
|
||||
- `--page-size` 范围 1-50
|
||||
|
||||
## 下载消息资源
|
||||
|
||||
```bash
|
||||
# 图片(--type image)
|
||||
lark-cli im +messages-resources-download \
|
||||
--message-id "om_xxx" \
|
||||
--file-key "img_xxx" \
|
||||
--type image \
|
||||
--output "filename.png" \
|
||||
--as bot
|
||||
|
||||
# 文件/视频/音频(--type file)
|
||||
lark-cli im +messages-resources-download \
|
||||
--message-id "om_xxx" \
|
||||
--file-key "file_xxx" \
|
||||
--type file \
|
||||
--output "filename.mp4" \
|
||||
--as bot
|
||||
```
|
||||
|
||||
注意:
|
||||
- `--output` 不支持绝对路径,必须使用相对路径
|
||||
- 需要先 `cd` 到目标目录再执行
|
||||
|
||||
## 电子表格操作
|
||||
|
||||
### 查看表格信息
|
||||
```bash
|
||||
lark-cli sheets +info --spreadsheet-token "xxx" --as bot
|
||||
```
|
||||
|
||||
### 读取数据
|
||||
```bash
|
||||
lark-cli sheets +read \
|
||||
--spreadsheet-token "xxx" \
|
||||
--range "{sheet_id}!A1:D10" \
|
||||
--as bot
|
||||
```
|
||||
|
||||
### 追加行
|
||||
```bash
|
||||
lark-cli sheets +append \
|
||||
--spreadsheet-token "xxx" \
|
||||
--sheet-id "{sheet_id}" \
|
||||
--range "{sheet_id}!A:D" \
|
||||
--values '[["val1","val2","val3","val4"]]' \
|
||||
--as bot
|
||||
```
|
||||
|
||||
### 覆盖写入
|
||||
```bash
|
||||
lark-cli sheets +write \
|
||||
--spreadsheet-token "xxx" \
|
||||
--range "{sheet_id}!A2:D10" \
|
||||
--values '[["a","b","c","d"]]' \
|
||||
--as bot
|
||||
```
|
||||
283
scripts/sync_group_to_sheet.py
Normal file
283
scripts/sync_group_to_sheet.py
Normal file
@ -0,0 +1,283 @@
|
||||
#!/usr/bin/env python3
|
||||
"""
|
||||
飞书群聊消息同步到电子表格
|
||||
- Bot身份拉取群消息
|
||||
- 非文本媒体上传到腾讯COS
|
||||
- 记录写入飞书电子表格
|
||||
|
||||
使用前修改下方配置常量。
|
||||
"""
|
||||
import os
|
||||
import sys
|
||||
import json
|
||||
import subprocess
|
||||
import logging
|
||||
import re
|
||||
from datetime import datetime, timezone, timedelta
|
||||
from pathlib import Path
|
||||
|
||||
# ============ 配置(使用前必须修改)============
|
||||
|
||||
# 飞书群
|
||||
CHAT_ID = "oc_xxx" # 目标群ID
|
||||
|
||||
# 飞书电子表格
|
||||
SPREADSHEET_TOKEN = "xxx" # 电子表格token
|
||||
SHEET_ID = "xxx" # sheet页ID
|
||||
|
||||
# Bot凭证
|
||||
LARK_CLI_CONFIG = "/root/.openclaw/credentials/xiaokui" # Bot凭证目录
|
||||
|
||||
# 同步状态
|
||||
LAST_SYNC_FILE = "/tmp/last_feedback_sync_time" # 同步时间记录文件
|
||||
WORK_DIR = "/tmp/feedback_sync_workdir" # 临时工作目录
|
||||
|
||||
# COS(从 tencent-cos-upload skill 引用)
|
||||
COS_BASE_PATH = "vala_llm/user_feedback" # COS上的基础路径
|
||||
|
||||
# 时区
|
||||
TZ = timezone(timedelta(hours=8))
|
||||
|
||||
# ============ 配置结束 ============
|
||||
|
||||
logging.basicConfig(level=logging.INFO, format='[%(asctime)s] %(message)s', datefmt='%Y-%m-%d %H:%M:%S')
|
||||
log = logging.getLogger(__name__)
|
||||
|
||||
# 引入COS上传器
|
||||
sys.path.insert(0, '/root/.openclaw/skills/tencent-cos-upload/scripts')
|
||||
from cos_upload import CosUploader
|
||||
|
||||
|
||||
def lark_cli(*args, cwd=None):
|
||||
"""调用lark-cli并返回解析后的JSON"""
|
||||
env = os.environ.copy()
|
||||
env['LARKSUITE_CLI_CONFIG_DIR'] = LARK_CLI_CONFIG
|
||||
cmd = ['lark-cli'] + list(args)
|
||||
result = subprocess.run(cmd, capture_output=True, text=True, env=env, cwd=cwd or WORK_DIR)
|
||||
if result.returncode != 0 and not result.stdout:
|
||||
log.error(f"lark-cli error: {result.stderr}")
|
||||
return None
|
||||
try:
|
||||
return json.loads(result.stdout)
|
||||
except json.JSONDecodeError:
|
||||
log.error(f"lark-cli JSON parse error: {result.stdout[:200]}")
|
||||
return None
|
||||
|
||||
|
||||
def fetch_messages(start_iso: str, end_iso: str) -> list:
|
||||
"""拉取群消息"""
|
||||
result = lark_cli(
|
||||
'im', '+chat-messages-list',
|
||||
'--chat-id', CHAT_ID,
|
||||
'--start', start_iso,
|
||||
'--end', end_iso,
|
||||
'--sort', 'asc',
|
||||
'--page-size', '50',
|
||||
'--as', 'bot'
|
||||
)
|
||||
if not result or not result.get('ok'):
|
||||
err = result.get('error', {}).get('message', 'unknown') if result else 'no response'
|
||||
log.error(f"拉取消息失败: {err}")
|
||||
return []
|
||||
return result.get('data', {}).get('messages', [])
|
||||
|
||||
|
||||
def download_resource(msg_id: str, file_key: str, res_type: str, filename: str) -> str:
|
||||
"""下载飞书消息中的资源,返回本地路径"""
|
||||
result = lark_cli(
|
||||
'im', '+messages-resources-download',
|
||||
'--message-id', msg_id,
|
||||
'--file-key', file_key,
|
||||
'--type', res_type,
|
||||
'--output', filename,
|
||||
'--as', 'bot',
|
||||
cwd=WORK_DIR
|
||||
)
|
||||
local_path = os.path.join(WORK_DIR, filename)
|
||||
if result and result.get('ok') and os.path.exists(local_path) and os.path.getsize(local_path) > 0:
|
||||
return local_path
|
||||
return None
|
||||
|
||||
|
||||
def append_to_sheet(rows: list):
|
||||
"""追加行到电子表格"""
|
||||
if not rows:
|
||||
return
|
||||
values_json = json.dumps(rows, ensure_ascii=False)
|
||||
result = lark_cli(
|
||||
'sheets', '+append',
|
||||
'--spreadsheet-token', SPREADSHEET_TOKEN,
|
||||
'--sheet-id', SHEET_ID,
|
||||
'--range', f'{SHEET_ID}!A:D',
|
||||
'--values', values_json,
|
||||
'--as', 'bot'
|
||||
)
|
||||
if result and result.get('ok'):
|
||||
log.info(f"电子表格写入成功: {len(rows)} 行")
|
||||
else:
|
||||
err = result.get('error', {}).get('message', 'unknown') if result else 'no response'
|
||||
log.error(f"电子表格写入失败: {err}")
|
||||
|
||||
|
||||
def extract_file_key(msg_type: str, content: str) -> tuple:
|
||||
"""从消息内容中提取file_key和资源类型"""
|
||||
if msg_type == 'image':
|
||||
match = re.search(r'img_[a-zA-Z0-9_-]+', content)
|
||||
return (match.group(0), 'image') if match else (None, None)
|
||||
elif msg_type in ('media', 'audio'):
|
||||
match = re.search(r'file_[a-zA-Z0-9_-]+', content)
|
||||
return (match.group(0), 'file') if match else (None, None)
|
||||
elif msg_type == 'file':
|
||||
try:
|
||||
c = json.loads(content)
|
||||
return (c.get('file_key'), 'file')
|
||||
except Exception:
|
||||
match = re.search(r'file_[a-zA-Z0-9_-]+', content)
|
||||
return (match.group(0), 'file') if match else (None, None)
|
||||
return None, None
|
||||
|
||||
|
||||
def get_media_info(msg_type: str, content: str) -> tuple:
|
||||
"""返回 (cos子目录, 文件扩展名, content_type)"""
|
||||
if msg_type == 'image':
|
||||
return 'image', '.png', 'image/png'
|
||||
elif msg_type == 'media':
|
||||
name_match = re.search(r'name="([^"]*)"', content)
|
||||
ext = os.path.splitext(name_match.group(1))[1] if name_match else '.mp4'
|
||||
return 'video', ext or '.mp4', 'video/mp4'
|
||||
elif msg_type == 'audio':
|
||||
return 'audio', '.ogg', 'audio/ogg'
|
||||
elif msg_type == 'file':
|
||||
try:
|
||||
c = json.loads(content)
|
||||
ext = os.path.splitext(c.get('file_name', ''))[1]
|
||||
except Exception:
|
||||
ext = ''
|
||||
return 'file', ext or '.bin', 'application/octet-stream'
|
||||
return 'other', '', 'application/octet-stream'
|
||||
|
||||
|
||||
def process_message(msg: dict, cos_uploader, date_str: str) -> list:
|
||||
"""处理单条消息,返回表格行 [时间, 反馈人, 类型, 内容/URL] 或 None"""
|
||||
msg_id = msg.get('message_id', '')
|
||||
sender_name = msg.get('sender', {}).get('name', '未知')
|
||||
create_time = msg.get('create_time', '')
|
||||
msg_type = msg.get('msg_type', '')
|
||||
content = msg.get('content', '')
|
||||
deleted = msg.get('deleted', False)
|
||||
|
||||
if deleted or msg_type == 'system':
|
||||
return None
|
||||
|
||||
# 文本消息
|
||||
if msg_type in ('text', 'post'):
|
||||
text = re.sub(r'<[^>]*>', '', content).strip()
|
||||
return [create_time, sender_name, '文本', text]
|
||||
|
||||
# 表情包
|
||||
if msg_type == 'sticker':
|
||||
return [create_time, sender_name, '表情', '(表情包)']
|
||||
|
||||
# 媒体消息
|
||||
cos_subdir, ext, content_type = get_media_info(msg_type, content)
|
||||
file_key, res_type = extract_file_key(msg_type, content)
|
||||
|
||||
type_labels = {'image': '图片', 'media': '视频', 'audio': '语音', 'file': '文件'}
|
||||
type_label = type_labels.get(msg_type, msg_type)
|
||||
|
||||
if msg_type == 'media':
|
||||
dur = re.search(r'duration="([^"]*)"', content)
|
||||
if dur:
|
||||
type_label += f'({dur.group(1)})'
|
||||
|
||||
if not file_key:
|
||||
return [create_time, sender_name, type_label, f'(无法提取资源key)']
|
||||
|
||||
# 文件名:纯ASCII
|
||||
short_id = msg_id[-12:]
|
||||
if msg_type == 'media':
|
||||
name_match = re.search(r'name="([^"]*)"', content)
|
||||
orig_ext = os.path.splitext(name_match.group(1))[1] if name_match else ext
|
||||
filename = f'{short_id}{orig_ext or ext}'
|
||||
elif msg_type == 'file':
|
||||
try:
|
||||
c = json.loads(content)
|
||||
orig_ext = os.path.splitext(c.get('file_name', ''))[1]
|
||||
except Exception:
|
||||
orig_ext = ext
|
||||
filename = f'{short_id}{orig_ext or ext}'
|
||||
else:
|
||||
filename = f'{short_id}{ext}'
|
||||
|
||||
cos_key = f'{COS_BASE_PATH}/{cos_subdir}/{date_str}/{filename}'
|
||||
|
||||
# 下载
|
||||
local_path = download_resource(msg_id, file_key, res_type, filename)
|
||||
if not local_path:
|
||||
log.warning(f"资源下载失败: {msg_id} ({msg_type})")
|
||||
return [create_time, sender_name, type_label, '(下载失败)']
|
||||
|
||||
# 上传COS
|
||||
try:
|
||||
url = cos_uploader.upload(local_path, cos_key, content_type)
|
||||
log.info(f"COS上传成功: {cos_key}")
|
||||
except Exception as e:
|
||||
log.error(f"COS上传失败: {cos_key} - {e}")
|
||||
url = f'(上传失败)'
|
||||
finally:
|
||||
try:
|
||||
os.remove(local_path)
|
||||
except Exception:
|
||||
pass
|
||||
|
||||
return [create_time, sender_name, type_label, url]
|
||||
|
||||
|
||||
def main():
|
||||
os.makedirs(WORK_DIR, exist_ok=True)
|
||||
|
||||
# 读取上次同步时间
|
||||
if os.path.exists(LAST_SYNC_FILE):
|
||||
with open(LAST_SYNC_FILE) as f:
|
||||
last_sync_iso = f.read().strip()
|
||||
else:
|
||||
last_sync_iso = (datetime.now(TZ) - timedelta(hours=1)).isoformat()
|
||||
|
||||
current_iso = datetime.now(TZ).isoformat()
|
||||
log.info(f"开始同步: {last_sync_iso} -> {current_iso}")
|
||||
|
||||
# 拉取消息
|
||||
messages = fetch_messages(last_sync_iso, current_iso)
|
||||
if not messages:
|
||||
log.info("没有新消息,同步结束")
|
||||
with open(LAST_SYNC_FILE, 'w') as f:
|
||||
f.write(current_iso)
|
||||
return
|
||||
|
||||
log.info(f"发现 {len(messages)} 条新消息")
|
||||
|
||||
# 初始化COS
|
||||
cos_uploader = CosUploader()
|
||||
date_str = datetime.now(TZ).strftime('%Y-%m-%d')
|
||||
|
||||
# 处理消息
|
||||
rows = []
|
||||
for msg in messages:
|
||||
row = process_message(msg, cos_uploader, date_str)
|
||||
if row:
|
||||
rows.append(row)
|
||||
|
||||
# 写入表格
|
||||
if rows:
|
||||
append_to_sheet(rows)
|
||||
log.info(f"同步完成: {len(rows)} 条记录")
|
||||
else:
|
||||
log.info("无有效消息需要写入")
|
||||
|
||||
# 更新同步时间
|
||||
with open(LAST_SYNC_FILE, 'w') as f:
|
||||
f.write(current_iso)
|
||||
|
||||
|
||||
if __name__ == '__main__':
|
||||
main()
|
||||
Loading…
Reference in New Issue
Block a user