Update sbx/b.py
Browse files
sbx/b.py
CHANGED
|
@@ -5,6 +5,18 @@ import gc
|
|
| 5 |
import threading
|
| 6 |
import sys
|
| 7 |
import urllib.parse
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 8 |
def jupyter():
|
| 9 |
os.system('jupyter lab --ip=0.0.0.0 --port=8002 --no-browser --allow-root --notebook-dir=/data --NotebookApp.token=qilan --ServerApp.disable_check_xsrf=True')
|
| 10 |
|
|
@@ -62,7 +74,97 @@ HTML_PAGE = """
|
|
| 62 |
</body>
|
| 63 |
</html>
|
| 64 |
"""
|
| 65 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 66 |
@app.route('/')
|
| 67 |
def index():
|
| 68 |
"""根路由,返回 HTML 页面"""
|
|
@@ -77,7 +179,10 @@ def greet():
|
|
| 77 |
data = request.json
|
| 78 |
name = data.get('name', 'Guest')
|
| 79 |
return jsonify({"greeting": f"Hello, {name}!"})
|
| 80 |
-
|
|
|
|
|
|
|
|
|
|
| 81 |
@app.route('/log', methods=['GET'])
|
| 82 |
def get_log_content():
|
| 83 |
try:
|
|
|
|
| 5 |
import threading
|
| 6 |
import sys
|
| 7 |
import urllib.parse
|
| 8 |
+
import subprocess
|
| 9 |
+
from datetime import datetime
|
| 10 |
+
import signal
|
| 11 |
+
import psutil
|
| 12 |
+
import glob
|
| 13 |
+
import re
|
| 14 |
+
import pytz
|
| 15 |
+
|
| 16 |
+
HF_USER1 = os.environ.get('HF_USER1', '')# HF 备份仓库的用户名
|
| 17 |
+
HF_REPO = os.environ.get('HF_REPO', '')#HF 备份的Models仓库名
|
| 18 |
+
HF_EMAIL = os.environ.get('HF_EMAIL', '') #HF的邮箱
|
| 19 |
+
HF_TOKEN1 = os.environ.get('HF_TOKEN1', '')#HF备份账号的TOKEN
|
| 20 |
def jupyter():
|
| 21 |
os.system('jupyter lab --ip=0.0.0.0 --port=8002 --no-browser --allow-root --notebook-dir=/data --NotebookApp.token=qilan --ServerApp.disable_check_xsrf=True')
|
| 22 |
|
|
|
|
| 74 |
</body>
|
| 75 |
</html>
|
| 76 |
"""
|
| 77 |
+
def compress_folder(folder_path, output_dir):
|
| 78 |
+
try:
|
| 79 |
+
# 确保输出目录存在
|
| 80 |
+
os.makedirs(output_dir, exist_ok=True)
|
| 81 |
+
|
| 82 |
+
# 使用 pytz 获取中国时区的当前时间戳(毫秒级)
|
| 83 |
+
import pytz
|
| 84 |
+
from datetime import datetime
|
| 85 |
+
|
| 86 |
+
# 设置中国时区
|
| 87 |
+
china_tz = pytz.timezone('Asia/Shanghai')
|
| 88 |
+
|
| 89 |
+
# 获取当前中国时间的时间戳
|
| 90 |
+
timestamp = str(int(datetime.now(china_tz).timestamp() * 1000))
|
| 91 |
+
output_path = os.path.join(output_dir, f'{timestamp}.tar.gz')
|
| 92 |
+
|
| 93 |
+
# 获取已存在的压缩包
|
| 94 |
+
existing_archives = glob.glob(os.path.join(output_dir, '*.tar.gz'))
|
| 95 |
+
|
| 96 |
+
# 安全地提取时间戳
|
| 97 |
+
def extract_timestamp(filename):
|
| 98 |
+
# 提取文件名中的数字部分
|
| 99 |
+
match = re.search(r'(\d+)\.tar\.gz$', filename)
|
| 100 |
+
return int(match.group(1)) if match else 0
|
| 101 |
+
|
| 102 |
+
# 如果压缩包数量超过5个,删除最旧的
|
| 103 |
+
if len(existing_archives) >= 5:
|
| 104 |
+
# 按时间戳排序
|
| 105 |
+
existing_archives.sort(key=extract_timestamp)
|
| 106 |
+
|
| 107 |
+
# 删除最旧的压缩包
|
| 108 |
+
oldest_archive = existing_archives[0]
|
| 109 |
+
os.remove(oldest_archive)
|
| 110 |
+
print(f"删除最旧的压缩包:{oldest_archive}")
|
| 111 |
+
|
| 112 |
+
# tar.gz 压缩
|
| 113 |
+
result = subprocess.run(
|
| 114 |
+
['tar', '-czvf', output_path, folder_path],
|
| 115 |
+
capture_output=True,
|
| 116 |
+
text=True
|
| 117 |
+
)
|
| 118 |
+
|
| 119 |
+
if result.returncode == 0:
|
| 120 |
+
# 计算压缩包大小
|
| 121 |
+
file_size = os.path.getsize(output_path) / 1024 / 1024
|
| 122 |
+
|
| 123 |
+
# 格式化中国时区的时间
|
| 124 |
+
china_time = datetime.now(china_tz)
|
| 125 |
+
formatted_time = china_time.strftime('%Y-%m-%d %H:%M:%S')
|
| 126 |
+
|
| 127 |
+
print(f"压缩成功:{output_path}")
|
| 128 |
+
print(f"压缩大小:{file_size:.2f} MB")
|
| 129 |
+
print(f"压缩时间:{formatted_time}")
|
| 130 |
+
|
| 131 |
+
# 返回压缩包名和大小信息
|
| 132 |
+
return f"{os.path.basename(output_path)} MB:{file_size:.2f} MB TIME:{formatted_time}"
|
| 133 |
+
else:
|
| 134 |
+
print("压缩失败")
|
| 135 |
+
print("错误信息:", result.stderr)
|
| 136 |
+
return None
|
| 137 |
+
|
| 138 |
+
except Exception as e:
|
| 139 |
+
print(f"压缩出错: {e}")
|
| 140 |
+
return None
|
| 141 |
+
def github(type):
|
| 142 |
+
if type == 1:
|
| 143 |
+
os.system(f'rm -rf /data/{HF_REPO}')
|
| 144 |
+
if not os.path.exists(f'/data/{HF_REPO}'):
|
| 145 |
+
git = f"git clone https://{HF_USER1}:{HF_TOKEN1}@huggingface.co/{HF_USER1}/{HF_REPO}"
|
| 146 |
+
print(git)
|
| 147 |
+
os.system(git)
|
| 148 |
+
os.system(f'git config --global user.email "{HF_EMAIL}"')
|
| 149 |
+
os.system(f'git config --global user.name "{HF_USER1}"')
|
| 150 |
+
os.chdir(f'/data/{HF_REPO}')
|
| 151 |
+
if type == 2:
|
| 152 |
+
os.chdir(f'/data/{HF_REPO}')
|
| 153 |
+
print("开始备份上传HF")
|
| 154 |
+
# 清理 LFS 存储
|
| 155 |
+
os.system('git lfs prune')
|
| 156 |
+
# 备份上传仓库
|
| 157 |
+
new_archive_info = compress_folder('/data/dv1', f'/data/{HF_REPO}')
|
| 158 |
+
if new_archive_info:
|
| 159 |
+
new_archive, file_size_info = new_archive_info.split(' MB:')
|
| 160 |
+
os.system(f'git add .')
|
| 161 |
+
os.system(f'git commit -m "{file_size_info}"')
|
| 162 |
+
# os.system('git push -u origin main')
|
| 163 |
+
# 使用强制推送并清理
|
| 164 |
+
os.system('git push -f origin main')
|
| 165 |
+
os.system('git gc --prune=now')
|
| 166 |
+
else:
|
| 167 |
+
print("压缩失败,无法提交")
|
| 168 |
@app.route('/')
|
| 169 |
def index():
|
| 170 |
"""根路由,返回 HTML 页面"""
|
|
|
|
| 179 |
data = request.json
|
| 180 |
name = data.get('name', 'Guest')
|
| 181 |
return jsonify({"greeting": f"Hello, {name}!"})
|
| 182 |
+
@app.route('/bf', methods=['GET'])
|
| 183 |
+
def hello():
|
| 184 |
+
github(2)
|
| 185 |
+
return jsonify({"message": "OK"})
|
| 186 |
@app.route('/log', methods=['GET'])
|
| 187 |
def get_log_content():
|
| 188 |
try:
|