refactor: 优化日志系统及自动化备份机制
- 替换 loguru 为原生 logging 库(与pytest兼容性更好)。 - 更新 pytest.ini 统一配置日志格式和基础命令。 - 优化 main.py 增加测试后的日志自动备份与定期清理功能。 - 新增 settings.py 实现配置解耦 - 更新 .gitignore
This commit is contained in:
272
core/run_appium.py
Normal file
272
core/run_appium.py
Normal file
@@ -0,0 +1,272 @@
|
||||
#!/usr/bin/env python
|
||||
# coding=utf-8
|
||||
|
||||
"""
|
||||
@author: CNWei,ChenWei
|
||||
@Software: PyCharm
|
||||
@contact: t6g888@163.com,chenwei@zygj.com
|
||||
@file: test
|
||||
@date: 2026/1/12 10:21
|
||||
@desc:
|
||||
"""
|
||||
import functools
|
||||
import logging
|
||||
import signal
|
||||
import subprocess
|
||||
import time
|
||||
import os
|
||||
import sys
|
||||
import http.client
|
||||
import socket
|
||||
import json
|
||||
from enum import Enum
|
||||
|
||||
from core.settings import BASE_DIR, APPIUM_HOST, APPIUM_PORT
|
||||
|
||||
logger = logging.getLogger(__name__)
|
||||
|
||||
|
||||
# 使用 npm run 确保 APPIUM_HOME=. 变量和本地版本生效
|
||||
# NODE_CMD = f"npm run appium -- -p {APPIUM_PORT}"
|
||||
|
||||
|
||||
class AppiumStatus(Enum):
|
||||
"""Appium 服务状态枚举"""
|
||||
READY = "服务已启动" # 服务和驱动都加载完成 (HTTP 200 + ready: true)
|
||||
INITIALIZING = "驱动正在加载" # 服务已响应但驱动仍在加载 (HTTP 200 + ready: false)
|
||||
CONFLICT = "端口被其他程序占用" # 端口被其他非 Appium 程序占用
|
||||
OFFLINE = "服务未启动" # 服务未启动
|
||||
ERROR = "内部错误"
|
||||
UNKNOWN = "未知状态"
|
||||
|
||||
|
||||
class ServiceRole(Enum):
|
||||
"""服务角色枚举：明确权责"""
|
||||
MANAGED = "由脚本启动 (受控模式)" # 需要负责清理
|
||||
EXTERNAL = "复用外部服务 (共享模式)" # 不负责清理
|
||||
NULL = "无效服务 (空模式)" # 占位或失败状态
|
||||
|
||||
|
||||
def get_appium_command():
|
||||
"""精确定位 Appium 执行文件，避免 npm 包装层"""
|
||||
bin_name = "appium.cmd" if sys.platform == "win32" else "appium"
|
||||
appium_bin = BASE_DIR / "node_modules" / ".bin" / bin_name
|
||||
|
||||
if not appium_bin.exists():
|
||||
# 报错提示：找不到本地 Appium，引导用户安装
|
||||
logger.info(f"\n错误: 在路径 {appium_bin} 未找到 Appium 执行文件。")
|
||||
logger.info("请确保已在项目目录下执行过: npm install appium")
|
||||
sys.exit(1)
|
||||
# 返回执行列表（用于 shell=False）
|
||||
return [str(appium_bin), "-p", str(APPIUM_PORT)]
|
||||
|
||||
|
||||
# 在全局或 start_appium_service 中获取命令
|
||||
APP_CMD_LIST = get_appium_command()
|
||||
|
||||
|
||||
def _cleanup_process_tree(process: subprocess.Popen = None):
|
||||
"""核心清理逻辑：针对方案一的跨平台递归关闭"""
|
||||
if not process or process.poll() is not None:
|
||||
return
|
||||
logger.info(f"正在关闭 Appium 进程树 (PID: {process.pid})...")
|
||||
try:
|
||||
if sys.platform == "win32":
|
||||
# Windows 下使用 taskkill 强制关闭进程树 /T
|
||||
subprocess.run(['taskkill', '/F', '/T', '/PID', str(process.pid)],
|
||||
stdout=subprocess.DEVNULL, stderr=subprocess.DEVNULL)
|
||||
else:
|
||||
# Unix/Mac: 获取进程组 ID 并发送终止信号
|
||||
pgid = os.getpgid(process.pid)
|
||||
os.killpg(pgid, signal.SIGTERM)
|
||||
|
||||
process.wait(timeout=5)
|
||||
logger.info("所有相关进程已安全退出。")
|
||||
except Exception as e:
|
||||
logger.error(f"关闭进程时遇到意外: {e}")
|
||||
try:
|
||||
process.kill()
|
||||
except Exception as e:
|
||||
logger.warning(f"强制关闭: {e}")
|
||||
# 这里通常保持安静，因为我们已经尝试过清理了
|
||||
logger.info("服务已完全清理。")
|
||||
|
||||
|
||||
class AppiumService:
|
||||
"""Appium 服务上下文容器"""
|
||||
|
||||
def __init__(self, role: ServiceRole, process: subprocess.Popen = None):
|
||||
self.role = role
|
||||
self.process = process
|
||||
|
||||
def stop(self):
|
||||
"""统一停止接口：根据角色决定是否关闭进程"""
|
||||
match self.role:
|
||||
case ServiceRole.EXTERNAL:
|
||||
logger.info(f"--> [角色: {self.role.value}] 脚本退出，保留原服务运行。")
|
||||
return
|
||||
case ServiceRole.MANAGED:
|
||||
_cleanup_process_tree(self.process)
|
||||
case ServiceRole.NULL:
|
||||
logger.info(f"--> [角色: {self.role.value}] 无需清理。")
|
||||
|
||||
def __repr__(self):
|
||||
return f"<AppiumService 角色='{self.role.value}' 端口={APPIUM_PORT}>"
|
||||
|
||||
|
||||
def get_appium_status() -> AppiumStatus:
|
||||
"""深度探测 Appium 状态"""
|
||||
conn = None
|
||||
try:
|
||||
# 1. 端口开启，尝试获取 Appium 状态接口
|
||||
conn = http.client.HTTPConnection(APPIUM_HOST, APPIUM_PORT, timeout=2)
|
||||
conn.request("GET", "/status")
|
||||
response = conn.getresponse()
|
||||
|
||||
if response.status != 200:
|
||||
return AppiumStatus.CONFLICT
|
||||
|
||||
data = json.loads(response.read().decode())
|
||||
|
||||
# 2. 解析 Appium 3.x 标准响应结构
|
||||
# 即使服务响应了，也要看驱动是否加载完成 (ready 字段)
|
||||
is_ready = data.get("value", {}).get("ready")
|
||||
return AppiumStatus.READY if is_ready else AppiumStatus.INITIALIZING
|
||||
|
||||
except (socket.error, ConnectionRefusedError):
|
||||
# 3. 如果通信拒绝，检查端口是否真的空闲
|
||||
with socket.socket(socket.AF_INET, socket.SOCK_STREAM) as s:
|
||||
try:
|
||||
s.settimeout(1)
|
||||
s.bind((APPIUM_HOST, APPIUM_PORT))
|
||||
return AppiumStatus.OFFLINE # 真正未启动
|
||||
except OSError:
|
||||
return AppiumStatus.CONFLICT # 端口被占用但没响应 HTTP
|
||||
except (http.client.HTTPException, json.JSONDecodeError):
|
||||
return AppiumStatus.UNKNOWN
|
||||
except Exception as e:
|
||||
logger.error(e)
|
||||
return AppiumStatus.ERROR
|
||||
finally:
|
||||
if conn: conn.close()
|
||||
|
||||
|
||||
def start_appium_service() -> AppiumService:
|
||||
"""管理 Appium 服务的生命周期"""
|
||||
process = None # 1. 预先初始化变量，防止作用域错误
|
||||
managed = False
|
||||
# 轮询等待真正就绪
|
||||
max_retries = 40
|
||||
|
||||
for i in range(max_retries):
|
||||
status = get_appium_status()
|
||||
match status: # Python 3.10+ 的模式匹配
|
||||
case AppiumStatus.READY:
|
||||
if managed:
|
||||
# 安全打印 PID
|
||||
pid_str = f"PID: {process.pid}" if process else "EXTERNAL"
|
||||
logger.info(f"Appium 已经完全就绪! ({pid_str})")
|
||||
return AppiumService(ServiceRole.MANAGED, process)
|
||||
else:
|
||||
logger.info(f"--> [复用] 有效的 Appium 服务已在运行 (Port: {APPIUM_PORT})")
|
||||
logger.info("--> [注意] 脚本退出时将保留该服务，不会将其关闭。")
|
||||
return AppiumService(ServiceRole.EXTERNAL, process)
|
||||
case AppiumStatus.CONFLICT:
|
||||
logger.warning(f"\n[!] 错误: 端口 {APPIUM_PORT} 被非 Appium 程序占用。")
|
||||
logger.info("=" * 60)
|
||||
logger.info("请手动执行以下命令释放端口后重试：")
|
||||
if sys.platform == "win32":
|
||||
logger.info(
|
||||
f" CMD: for /f \"tokens=5\" %a in ('netstat -aon ^| findstr :{APPIUM_PORT}') do taskkill /F /PID %a")
|
||||
else:
|
||||
logger.info(f" Terminal: lsof -ti:{APPIUM_PORT} | xargs kill -9")
|
||||
logger.info("=" * 60)
|
||||
sys.exit(1)
|
||||
case AppiumStatus.OFFLINE:
|
||||
|
||||
if not managed:
|
||||
logger.info("Appium 服务未启动...")
|
||||
logger.info(f"正在准备启动本地 Appium 服务 (Port: {APPIUM_PORT})...")
|
||||
|
||||
# 注入环境变量，确保 Appium 寻找项目本地的驱动
|
||||
env_vars = os.environ.copy()
|
||||
env_vars["APPIUM_HOME"] = str(BASE_DIR)
|
||||
|
||||
try:
|
||||
process = subprocess.Popen(
|
||||
APP_CMD_LIST,
|
||||
stdout=subprocess.DEVNULL,
|
||||
stderr=subprocess.DEVNULL,
|
||||
env=env_vars,
|
||||
cwd=BASE_DIR,
|
||||
# Windows 和 Linux/Mac 的处理方式不同
|
||||
creationflags=subprocess.CREATE_NEW_PROCESS_GROUP if sys.platform == "win32" else 0,
|
||||
preexec_fn=os.setsid if sys.platform != "win32" else None
|
||||
)
|
||||
managed = True
|
||||
except Exception as e:
|
||||
logger.error(f"启动过程发生异常: {e}")
|
||||
sys.exit(1)
|
||||
else:
|
||||
if process and process.poll() is not None:
|
||||
logger.warning("Appium 进程启动后异常退出。")
|
||||
sys.exit(1)
|
||||
case AppiumStatus.INITIALIZING:
|
||||
if managed and process and process.poll() is not None:
|
||||
logger.warning("Appium 驱动加载期间进程崩溃。")
|
||||
_cleanup_process_tree(process)
|
||||
sys.exit(1)
|
||||
if i % 4 == 0: # 每 2 秒提醒一次，避免刷屏
|
||||
logger.info("Appium 正在加载驱动/插件，请稍候...")
|
||||
case AppiumStatus.ERROR:
|
||||
logger.error("探测接口发生内部错误（可能是解析失败或严重网络异常），脚本终止。")
|
||||
if managed and process:
|
||||
_cleanup_process_tree(process)
|
||||
sys.exit(1)
|
||||
case _:
|
||||
logger.error("Appium 启动异常")
|
||||
sys.exit(1)
|
||||
|
||||
time.sleep(0.5)
|
||||
|
||||
logger.warning("启动超时：Appium 在规定时间内未完成初始化。")
|
||||
_cleanup_process_tree(process)
|
||||
sys.exit(1)
|
||||
|
||||
|
||||
def stop_appium_service(server: AppiumService):
|
||||
# """安全关闭服务"""
|
||||
server.stop()
|
||||
|
||||
|
||||
# --- 装饰器实现 ---
|
||||
def with_appium(func):
|
||||
@functools.wraps(func)
|
||||
def wrapper(*args, **kwargs):
|
||||
service = start_appium_service()
|
||||
try:
|
||||
# return func(service, *args, **kwargs)
|
||||
return func(*args, **kwargs)
|
||||
finally:
|
||||
service.stop()
|
||||
|
||||
return wrapper
|
||||
|
||||
|
||||
if __name__ == "__main__":
|
||||
# 使用示例：作为一个上下文管理器或简单的生命周期示例
|
||||
appium_proc = None
|
||||
try:
|
||||
appium_proc = start_appium_service()
|
||||
print(f"\n[项目路径] {BASE_DIR}")
|
||||
print("\n[提示] 现在可以手动或通过其他脚本运行测试用例。")
|
||||
print("[提示] 按下 Ctrl+C 可停止由本脚本启动的服务。")
|
||||
|
||||
# 保持运行，直到手动停止（在实际测试框架中，这里会被替换为测试执行逻辑）
|
||||
while True:
|
||||
time.sleep(1)
|
||||
|
||||
except KeyboardInterrupt:
|
||||
pass
|
||||
finally:
|
||||
stop_appium_service(appium_proc)
|
||||
Reference in New Issue
Block a user