modified: app/ali_oss.py

new file:   app/amazon/del_brand.py
	new file:   app/amazon/main.py
	modified:   app/blueprints/__pycache__/brand.cpython-39.pyc
	modified:   app/blueprints/__pycache__/communication.cpython-39.pyc
	modified:   app/blueprints/__pycache__/main.cpython-39.pyc
	modified:   app/blueprints/brand.py
	new file:   "app/blueprints/brand_\345\244\207\344\273\275.py"
	modified:   app/blueprints/communication.py
	modified:   app/blueprints/main.py
	modified:   app/config.py
	modified:   app/main.py
This commit is contained in:
铭坤
2026-04-01 11:53:12 +08:00
parent b6dd1d7931
commit 785d7563f9
12 changed files with 1779 additions and 81 deletions

View File

@@ -14,6 +14,7 @@ def upload_file(file_content: bytes, key: str):
cfg.credentials_provider = credentials_provider
cfg.region = region
cfg.endpoint = endpoint
cfg.retry_max_attempts = 3
client = oss.Client(cfg)
result = client.put_object(

519
app/amazon/del_brand.py Normal file
View File

@@ -0,0 +1,519 @@
import traceback
import winreg
import subprocess
import time
import uuid
import requests
import json
import os
from typing import Literal
from DrissionPage import Chromium
from DrissionPage.common import By
def kill_process(version: Literal["v5", "v6"]):
"""杀紫鸟客户端进程(独立函数版本)"""
driver = ZiniaoDriver({})
driver.kill_process(version)
class ZiniaoDriver:
"""紫鸟浏览器自动化驱动类"""
def __init__(self, user_info: dict, socket_port: int = 19890):
"""
初始化紫鸟浏览器驱动
Args:
user_info: 用户信息字典,包含 company, username, password
socket_port: 客户端通信端口,默认 19890
"""
self.user_info = user_info
self.socket_port = socket_port
self.client_path = None
self.browser = None
self.tab = None
self.store_id = None
def get_zinaio_exe(self, protocol_name: str = "superbrowser"):
"""
获取紫鸟安装目录
Args:
protocol_name: 协议名称,默认 "superbrowser"
Returns:
exe_path: 紫鸟浏览器可执行文件路径
"""
try:
key_path = rf"SOFTWARE\Classes\{protocol_name}\shell\open\command"
key = winreg.OpenKey(winreg.HKEY_CURRENT_USER, key_path)
command, _ = winreg.QueryValueEx(key, "")
winreg.CloseKey(key)
if isinstance(command, str):
exe_path = command.split()[0]
else:
exe_path = command[0]
return exe_path
except FileNotFoundError:
try:
key = winreg.OpenKey(winreg.HKEY_LOCAL_MACHINE, key_path)
command, _ = winreg.QueryValueEx(key, "")
winreg.CloseKey(key)
if isinstance(command, str):
exe_path = command.split()[0]
else:
exe_path = command[0]
return exe_path
except FileNotFoundError:
return None
def update_core(self):
"""
下载所有内核打开店铺前调用需客户端版本5.285.7以上
因为http有超时时间所以这个action适合循环调用直到返回成功
"""
data = {
"action": "updateCore",
"requestId": str(uuid.uuid4()),
}
data.update(self.user_info)
while True:
url = f'http://127.0.0.1:{self.socket_port}'
response = requests.post(url, json.dumps(data).encode('utf-8'), timeout=120)
result = response.json()
print(result)
if result is None:
print("等待客户端启动...")
time.sleep(2)
continue
if result.get("statusCode") is None or result.get("statusCode") == -10003:
print("当前版本不支持此接口,请升级客户端")
return
elif result.get("statusCode") == 0:
print("更新内核完成")
return
else:
print(f"等待更新内核: {json.dumps(result)}")
time.sleep(2)
def kill_process(self, version: Literal["v5", "v6"]):
"""
杀紫鸟客户端进程
Args:
version: 客户端版本
"""
if version == "v5":
process_name = 'SuperBrowser.exe'
else:
process_name = 'ziniao.exe'
os.system('taskkill /f /t /im ' + process_name)
time.sleep(3)
def get_browser_list(self) -> list:
"""
获取浏览器列表
Returns:
list: 浏览器列表
"""
request_id = str(uuid.uuid4())
data = {
"action": "getBrowserList",
"requestId": request_id
}
data.update(self.user_info)
url = f'http://127.0.0.1:{self.socket_port}'
response = requests.post(url, json.dumps(data).encode('utf-8'), timeout=120)
r = response.json()
if str(r.get("statusCode")) == "0":
print(r)
return r.get("browserList")
elif str(r.get("statusCode")) == "-10003":
print(f"login Err {json.dumps(r, ensure_ascii=False)}")
exit()
else:
print(f"Fail {json.dumps(r, ensure_ascii=False)} ")
exit()
def open_store(self, store_info, isWebDriverReadOnlyMode=0, isprivacy=0,
isHeadless=0, cookieTypeSave=0, jsInfo=""):
"""
打开店铺
Args:
store_info: 店铺信息browserId 或 browserOauth
isWebDriverReadOnlyMode: 是否只读模式,默认 0
isprivacy: 隐私模式,默认 0
isHeadless: 无头模式,默认 0
cookieTypeSave: cookie保存类型默认 0
jsInfo: 注入的JS信息默认 ""
Returns:
dict: 返回结果
"""
request_id = str(uuid.uuid4())
data = {
"action": "startBrowser",
"isWaitPluginUpdate": 0,
"isHeadless": isHeadless,
"requestId": request_id,
"isWebDriverReadOnlyMode": isWebDriverReadOnlyMode,
"cookieTypeLoad": 0,
"cookieTypeSave": cookieTypeSave,
"runMode": "1",
"isLoadUserPlugin": False,
"pluginIdType": 1,
"privacyMode": isprivacy
}
data.update(self.user_info)
if store_info.isdigit():
data["browserId"] = store_info
else:
data["browserOauth"] = store_info
if len(str(jsInfo)) > 2:
data["injectJsInfo"] = json.dumps(jsInfo)
url = f'http://127.0.0.1:{self.socket_port}'
response = requests.post(url, json.dumps(data).encode('utf-8'), timeout=120)
r = response.json()
if str(r.get("statusCode")) == "0":
return r
elif str(r.get("statusCode")) == "-10003":
print(f"login Err {json.dumps(r, ensure_ascii=False)}")
exit()
else:
print(f"Fail {json.dumps(r, ensure_ascii=False)} ")
exit()
def get_browser(self, port) -> Chromium:
"""
获取浏览器实例
Args:
port: 调试端口
Returns:
Chromium: DrissionPage浏览器实例
"""
browser = Chromium(port)
return browser
def start_client(self):
"""启动紫鸟客户端"""
self.client_path = self.get_zinaio_exe("superbrowser").strip('"')
print(self.client_path)
cmd = [self.client_path, '--run_type=web_driver', '--ipc_type=http',
'--port=' + str(self.socket_port)]
print(" ".join(cmd))
subprocess.Popen(cmd)
time.sleep(5)
def open_shop(self, shop_name: str):
"""
打开指定店铺
Args:
shop_name: 店铺名称
Returns:
Chromium or str: 成功返回浏览器实例,失败返回错误信息
"""
# 启动客户端
self.start_client()
# 更新内核
self.update_core()
# 获取店铺列表
shop_ls = self.get_browser_list()
self.store_id = None
# print(shop_ls)
for shop in shop_ls:
if shop.get("browserName") == shop_name:
self.store_id = shop.get('browserOauth')
break
if not self.store_id:
print("店铺不存在")
return "店铺不存在"
# 打开店铺
ret_json = self.open_store(self.store_id)
print(ret_json)
self.store_id = ret_json.get("browserOauth")
if self.store_id is None:
self.store_id = ret_json.get("browserId")
# 获取drissionpage浏览器会话
self.browser = self.get_browser(ret_json.get('debuggingPort'))
ip_check_url = ret_json.get("ipDetectionPage")
if not ip_check_url:
print("ip检测页地址为空请升级紫鸟浏览器到最新版")
print(f"=====关闭店铺:{shop_name}=====")
self.close_store(self.store_id)
exit()
ip_usable = self.open_ip_check(self.browser, ip_check_url)
if ip_usable:
print("ip检测通过打开店铺平台主页")
self.open_launcher_page(ret_json.get("launcherPage"), self.browser)
else:
print("IP检测不通过")
return self.browser
def close_store(self, browser_oauth=None):
"""
关闭店铺
Args:
browser_oauth: 店铺OAuth标识如果不提供则使用当前打开的店铺
Returns:
dict: 返回结果
"""
if browser_oauth is None:
browser_oauth = self.store_id
request_id = str(uuid.uuid4())
data = {
"action": "stopBrowser",
"requestId": request_id,
"duplicate": 0,
"browserOauth": browser_oauth
}
data.update(self.user_info)
url = f'http://127.0.0.1:{self.socket_port}'
response = requests.post(url, json.dumps(data).encode('utf-8'), timeout=120)
r = response.json()
if str(r.get("statusCode")) == "0":
return r
elif str(r.get("statusCode")) == "-10003":
print(f"login Err {json.dumps(r, ensure_ascii=False)}")
exit()
else:
print(f"Fail {json.dumps(r, ensure_ascii=False)} ")
exit()
def open_launcher_page(self, launcher_page: str, browser: Chromium = None):
"""
打开启动页面
Args:
launcher_page: 要打开的页面URL
browser: 浏览器实例,如果不提供则使用当前浏览器实例
"""
if browser is None:
browser = self.browser
tab = browser.new_tab(url=launcher_page)
self.tab = tab
return tab
def open_ip_check(self, browser: Chromium, ip_check_url: str):
"""
打开ip检测页检测ip是否正常
:param browser: drissionpage浏览器会话
:param ip_check_url ip检测页地址
:return 检测结果
"""
try:
tab = browser.latest_tab
tab.get(ip_check_url)
success_button = tab.ele((By.XPATH, '//button[contains(@class, "styles_btn--success")]'),
timeout=60) # 等待查找元素60秒
if success_button:
print("ip检测成功")
return True
else:
print("ip检测超时")
return False
except Exception as e:
print("ip检测异常:" + traceback.format_exc())
return False
class AmazoneDriver(ZiniaoDriver):
"""亚马逊专用驱动类继承自ZiniaoDriver"""
def SwitchingCountries(self, country_name: str):
"""
切换国家
操作:
1、//div[@class="dropdown-account-switcher-header-label"]/span[last()] 获取此元素文本,判断当前国家,如果与目标国家相同则不操作,否则执行下一步
2、点击 //div[@class="dropdown-account-switcher-header-label"] 打开下拉框
3、点击 //div[@class="dropdown-account-switcher-list-item"] 第一个展开国家列表
4、点击 //div[@class="dropdown-account-switcher-list-item dropdown-account-switcher-list-item-indented" and @title="国家名"] 切换到目标国家
5、等待页面加载完成判断国家是否切换成功成功则返回True否则返回False
Args:
country_name: 目标国家名称
Returns:
bool: 切换成功返回True失败返回False
"""
try:
if self.browser is None:
print("浏览器实例不存在,请先打开店铺")
return False
# 获取当前标签页
tab = self.tab
# 步骤1获取当前国家名称
print(f"正在检查当前国家...")
current_country_ele = tab.ele('xpath://div[@class="dropdown-account-switcher-header-label"]/span[last()]',
timeout=10)
if current_country_ele:
current_country = current_country_ele.text.strip()
print(f"当前国家:{current_country}")
# 判断是否与目标国家相同
if current_country == country_name:
print(f"当前已经是目标国家 {country_name},无需切换")
return True
else:
print("无法获取当前国家信息")
return False
# 步骤2点击打开下拉框
print(f"正在打开国家切换下拉框...")
dropdown_header = tab.ele('xpath://div[@class="dropdown-account-switcher-header-label"]', timeout=10)
if not dropdown_header:
print("找不到国家切换下拉框")
return False
dropdown_header.click()
time.sleep(1) # 等待下拉框展开
# 步骤3点击第一个展开国家列表
print(f"正在展开国家列表...")
first_item = tab.ele('xpath://div[@class="dropdown-account-switcher-list-item"]', timeout=10)
if not first_item:
print("找不到国家列表项")
return False
first_item.click()
time.sleep(1) # 等待国家列表展开
# 步骤4点击目标国家
print(f"正在切换到国家:{country_name}")
target_country_xpath = f'//div[@class="dropdown-account-switcher-list-item dropdown-account-switcher-list-item-indented" and @title="{country_name}"]'
target_country = tab.ele(f'xpath:{target_country_xpath}', timeout=10)
if not target_country:
print(f"找不到目标国家:{country_name}")
return False
target_country.click()
# 步骤5等待页面加载完成并验证切换结果
print(f"等待页面加载...")
# time.sleep(3) # 等待页面加载
self.tab.wait.doc_loaded()
# 再次检查当前国家
new_country_ele = tab.ele('xpath://div[@class="dropdown-account-switcher-header-label"]/span[last()]',
timeout=10)
if new_country_ele:
new_country = new_country_ele.text.strip()
if new_country == country_name:
print(f"国家切换成功:{new_country}")
return True
else:
print(f"国家切换失败,当前国家:{new_country},目标国家:{country_name}")
return False
else:
print("无法验证切换结果")
return False
except Exception as e:
print(f"切换国家时发生异常:{traceback.format_exc()}")
return False
def SwitchPage(self):
"""
切换至 管理所有库存页面
1、等待 //navigation-favorites-bar[@class="hydrated"] 出现
"""
navigation = self.tab.ele('xpath://navigation-favorites-bar[@class="hydrated"]')
navigation.wait.displayed(raise_err=False)
page_btn = navigation.sr('xpath://internal-fav-bar-links[@data-internal="navigation"]').sr(
'xpath://a[@data-page-id="ezdpc-gui-inventory-mons"]')
page_btn.wait.displayed(raise_err=False)
page_btn.click(timeout=5)
self.tab.wait.doc_loaded()
# 等待搜索框出现
search_region = self.tab.ele('xpath://div[@id="searchBoxContainer"]//kat-input-group')
search_region.wait.displayed(raise_err=False)
def search(self, asin):
search_region = self.tab.ele('xpath://div[@id="searchBoxContainer"]//kat-input-group')
search_region.wait.displayed(raise_err=False)
time.sleep(0.6)
search_input = self.tab.ele("xpath://kat-input[contains(@class,'SearchBox-module__searchInput')]").sr(
'xpath://span[@class="container"]//input[@part="input"]')
search_input.input(asin)
for _ in range(3):
search_btn = self.tab.ele("xpath://kat-icon[@name='search']")
search_btn.click()
load_ele = self.tab.ele("xpath://div[contains(@class,'Loader-module__loader')]")
load_ele.wait.hidden(timeout=20)
time.sleep(1)
sku_ls = self.tab.eles("//div[@data-sku]")
return sku_ls
def del_action(self, sku_ele):
dropdown = sku_ele.ele('xpath:.//kat-dropdown-button[@variant="secondary"]')
dropdown.click()
time.sleep(1)
del_btn = dropdown.ele("xpath:.//button[@role='menuitem' and text()='删除商品信息']")
del_btn.wait.displayed(raise_err=False)
del_btn.click()
confirm_btn = self.tab.ele('xpath://kat-modal[@data-testid="action-modal"]//kat-button[@variant="primary"]')
confirm_btn.wait.displayed(raise_err=False)
confirm_btn.wait.enabled()
confirm_btn.click()
suc_alert = self.tab.ele("xpath://kat-alert[@variant='success']")
suc = suc_alert.wait.displayed(raise_err=False, timeout=5)
return suc
if __name__ == '__main__':
# 使用示例
user_info = {
"company": "rongchuang123",
"username": "自动化_Robot",
"password": "#20zsg25"
}
shop_name = "郭亚芳"
kill_process("v6")
# 创建驱动实例
driver = AmazoneDriver(user_info)
# 打开店铺并获取浏览器实例
browser = driver.open_shop(shop_name)
country = "西班牙"
asin = "Voanos"
sw_suc = driver.SwitchingCountries(country)
driver.SwitchPage()
sku_ls = driver.search(asin=asin)
print("查询结果有:", len(sku_ls))
for sku in sku_ls:
suc = driver.del_action(sku)
print(sku, "删除结果", suc)

421
app/amazon/main.py Normal file
View File

@@ -0,0 +1,421 @@
import time
import traceback
import requests
from datetime import datetime
from typing import Dict, Any, List
from config import JSON_TASK_QUEUE, runing_task, runing_shop, DELETE_BRAND_API_BASE, ZN_COMPANY, ZN_USERNAME, ZN_PASSWORD
from amazon.del_brand import AmazoneDriver, kill_process
class TaskMonitor:
"""任务监控器:负责监控队列并执行品牌删除任务"""
def __init__(self):
"""初始化任务监控器"""
self.running = True
self.user_info = {
"company": ZN_COMPANY,
"username": ZN_USERNAME,
"password": ZN_PASSWORD
}
self.chunk_index = 1 # 当前处理的分块索引
def log(self, message: str, level: str = "INFO"):
"""日志输出
Args:
message: 日志消息
level: 日志级别
"""
timestamp = datetime.now().strftime("%Y-%m-%d %H:%M:%S")
print(f"[{timestamp}] [{level}] {message}")
def start(self):
"""启动任务监控(阻塞式循环)"""
self.log("任务监控器启动,开始监听队列...")
while self.running:
try:
# 阻塞式获取任务避免CPU空转
task_data = JSON_TASK_QUEUE.get(block=True, timeout=1)
# 检查任务类型
task_type = task_data.get("type", "")
if task_type != "delete-brand-run":
self.log(f"未知任务类型: {task_type},跳过", "WARNING")
continue
# 处理任务
self.log(f"接收到删除品牌任务,开始处理...")
self.process_task(task_data)
except Exception as e:
if "Empty" not in str(e): # 忽略队列为空的超时异常
self.log(f"任务处理异常: {traceback.format_exc()}", "ERROR")
time.sleep(0.1) # 短暂等待避免异常循环
def process_task(self, task_data: Dict[str, Any]):
"""处理单个任务
Args:
task_data: 任务数据
"""
try:
data = task_data.get("data", {})
task_id = data.get("taskId")
items = data.get("items", [])
if not task_id:
self.log("任务ID为空跳过", "WARNING")
return
# 初始化任务状态
self.update_task_status(
task_id,
status="running",
start_time=datetime.now().strftime("%Y-%m-%d %H:%M:%S"),
total_shops=len(items),
processed_shops=0,
total_asins=sum(shop.get("totalRows", 0) for shop in items),
processed_asins=0,
success_count=0,
failed_count=0,
stop_requested=False # 暂停请求标志
)
self.log(f"开始处理任务 {task_id},共 {len(items)} 个店铺")
# 遍历处理每个店铺
for idx, shop_data in enumerate(items, 1):
shop_name = shop_data.get("shopName", "未知店铺")
self.log(f"[{idx}/{len(items)}] 开始处理店铺: {shop_name}")
try:
self.process_shop(shop_data, task_id)
# 更新已处理店铺数
if task_id in runing_task:
runing_task[task_id]["processed_shops"] += 1
except Exception as e:
self.log(f"处理店铺 {shop_name} 失败: {str(e)}", "ERROR")
self.log(traceback.format_exc(), "ERROR")
# 检查任务最终状态
if task_id in runing_task and runing_task[task_id].get("stop_requested", False):
self.update_task_status(task_id, status="stopped")
self.log(f"任务 {task_id} 已被暂停!")
else:
self.update_task_status(task_id, status="completed")
self.log(f"任务 {task_id} 处理完成!")
except Exception as e:
self.log(f"任务处理失败: {traceback.format_exc()}", "ERROR")
if task_id:
self.update_task_status(task_id, status="failed", error=str(e))
def process_shop(self, shop_data: Dict[str, Any], task_id: int):
"""处理单个店铺(包含重试逻辑)
Args:
shop_data: 店铺数据
task_id: 任务ID
"""
shop_name = shop_data.get("shopName", "未知店铺")
result_id = shop_data.get("resultId")
countries = shop_data.get("countries", [])
if not countries:
self.log(f"店铺 {shop_name} 没有国家数据,跳过", "WARNING")
return
# 更新当前处理的店铺
self.update_task_status(task_id, current_shop=shop_name)
# 将店铺添加到正在执行中的店铺列表
start_time = datetime.now().strftime("%Y-%m-%d %H:%M:%S")
runing_shop[shop_name] = start_time
self.log(f"店铺 {shop_name} 已添加到执行列表,开始时间: {start_time}")
# 店铺打开重试最多3次
driver = None
browser = None
max_retries = 3
for retry in range(max_retries):
try:
self.log(f"尝试打开店铺 {shop_name} (第 {retry + 1}/{max_retries} 次)")
# 如果不是第一次尝试,先杀进程
if retry > 0:
self.log("重试前先杀掉浏览器进程...")
kill_process("v6")
time.sleep(2)
# 创建驱动并打开店铺
driver = AmazoneDriver(self.user_info)
browser = driver.open_shop(shop_name)
if browser and browser != "店铺不存在":
self.log(f"成功打开店铺 {shop_name}")
break
else:
self.log(f"打开店铺失败: {browser}", "WARNING")
driver = None
except Exception as e:
self.log(f"打开店铺异常: {str(e)}", "ERROR")
driver = None
# 如果还有重试机会,等待后继续
if retry < max_retries - 1:
time.sleep(3)
# 检查是否成功打开
if not driver or not browser or browser == "店铺不存在":
self.log(f"店铺 {shop_name} 打开失败,已重试 {max_retries} 次,跳过该店铺", "ERROR")
return
try:
# 处理每个国家
for country_data in countries:
# 检查是否收到暂停请求
if task_id in runing_task and runing_task[task_id].get("stop_requested", False):
self.log(f"检测到任务 {task_id} 的暂停请求,停止处理国家", "WARNING")
break # 跳出循环进入finally关闭店铺
try:
self.process_country(driver, country_data, task_id, result_id, shop_data)
except Exception as e:
country_name = country_data.get("country", "未知")
self.log(f"处理国家 {country_name} 失败: {str(e)}", "ERROR")
self.log(traceback.format_exc(), "ERROR")
finally:
# 关闭店铺
try:
if driver:
self.log(f"关闭店铺 {shop_name}")
driver.close_store()
time.sleep(2)
except Exception as e:
self.log(f"关闭店铺失败: {str(e)}", "WARNING")
# 从正在执行中的店铺列表中移除
if shop_name in runing_shop:
del runing_shop[shop_name]
self.log(f"店铺 {shop_name} 已从执行列表中移除")
def process_country(self, driver: AmazoneDriver, country_data: Dict[str, Any],
task_id: int, result_id: int, shop_data: Dict[str, Any]):
"""处理单个国家的所有ASIN
Args:
driver: 亚马逊驱动实例
country_data: 国家数据
task_id: 任务ID
result_id: 结果ID
shop_data: 店铺数据
"""
country = country_data.get("country", "未知")
items = country_data.get("items", [])
self.log(f"开始处理国家: {country},共 {len(items)} 个ASIN")
self.update_task_status(task_id, current_country=country)
# 切换国家
try:
switch_success = driver.SwitchingCountries(country)
if not switch_success:
self.log(f"切换到国家 {country} 失败,跳过该国家", "ERROR")
return
except Exception as e:
self.log(f"切换国家 {country} 异常: {str(e)}", "ERROR")
return
# 切换到库存管理页面
try:
driver.SwitchPage()
self.log(f"已切换到库存管理页面")
except Exception as e:
self.log(f"切换页面失败: {str(e)}", "ERROR")
return
# 处理每个ASIN
file_key = shop_data.get("fileKey", "")
source_filename = shop_data.get("sourceFilename", "")
total_rows = shop_data.get("totalRows", 0)
for idx, asin_item in enumerate(items, 1):
# 检查是否收到暂停请求
if task_id in runing_task and runing_task[task_id].get("stop_requested", False):
self.log(f"检测到任务 {task_id} 的暂停请求,停止处理", "WARNING")
return # 返回到上层,会触发店铺关闭
try:
self.log(f"[{idx}/{len(items)}] 处理ASIN: {asin_item.get('asin', '')}")
self.process_asin(driver, asin_item, country, task_id, result_id,
file_key, source_filename, total_rows, idx)
except Exception as e:
asin = asin_item.get("asin", "未知")
self.log(f"处理ASIN {asin} 失败: {str(e)}", "ERROR")
# 继续处理下一个ASIN
def process_asin(self, driver: AmazoneDriver, asin_item: Dict[str, Any],
country: str, task_id: int, result_id: int, file_key: str,
source_filename: str, total_rows: int):
"""处理单个ASIN并回传结果
Args:
driver: 亚马逊驱动实例
asin_item: ASIN数据项
country: 国家名称
task_id: 任务ID
result_id: 结果ID
file_key: 文件KEY
source_filename: 源文件名
total_rows: 总行数
chunk_index: 当前处理索引
"""
asin = asin_item.get("asin", "")
row_index = asin_item.get("rowIndex", 0)
# 更新当前处理的ASIN
self.update_task_status(task_id, current_asin=asin)
status = "失败"
try:
# 搜索ASIN
sku_ls = driver.search(asin=asin)
self.log(f"搜索到 {len(sku_ls)} 个SKU")
if len(sku_ls) == 0:
status = "查询不到"
self.log(f"ASIN {asin} 未找到商品", "WARNING")
else:
# 删除所有找到的SKU
success_count = 0
for sku in sku_ls:
try:
suc = driver.del_action(sku)
if suc:
success_count += 1
self.log(f"SKU 删除成功")
else:
self.log(f"SKU 删除失败", "WARNING")
except Exception as e:
self.log(f"删除SKU异常: {str(e)}", "ERROR")
if success_count > 0:
status = "成功"
self.log(f"ASIN {asin} 删除成功 ({success_count}/{len(sku_ls)})")
# 更新成功计数
if task_id in runing_task:
runing_task[task_id]["success_count"] += 1
else:
status = "失败"
self.log(f"ASIN {asin} 所有SKU删除失败", "ERROR")
# 更新失败计数
if task_id in runing_task:
runing_task[task_id]["failed_count"] += 1
except Exception as e:
status = "删除异常"
self.log(f"处理ASIN {asin} 异常: {str(e)}", "ERROR")
# 更新失败计数
if task_id in runing_task:
runing_task[task_id]["failed_count"] += 1
# 更新已处理ASIN计数
if task_id in runing_task:
runing_task[task_id]["processed_asins"] += 1
# 回传结果到API
try:
payload = {
"submissionId": "",
"files": [{
"fileKey": file_key,
"sourceFilename": source_filename,
"chunkIndex": self.chunk_index,
"chunkTotal": total_rows,
"processedRows": self.chunk_index,
"totalRows": total_rows,
"currentCountry": country,
"currentAsin": asin,
"countries": [{
"country": country,
"items": [{
"asin": asin,
"status": status
}]
}]
}]
}
self.post_result(task_id, payload)
self.log(f"ASIN {asin} 结果已回传,状态: {status}")
except Exception as e:
self.log(f"回传结果失败: {str(e)}", "ERROR")
def post_result(self, task_id: int, payload: Dict[str, Any]):
"""回传结果到API
Args:
task_id: 任务ID
payload: 结果数据
"""
url = f"{DELETE_BRAND_API_BASE}/newApi/api/delete-brand/tasks/{task_id}/result"
try:
response = requests.post(
url,
json=payload,
headers={"Content-Type": "application/json"},
timeout=30,
verify=False # 忽略SSL证书验证
)
if response.status_code == 200:
self.log(f"结果回传成功: {url}")
else:
self.log(f"结果回传失败,状态码: {response.status_code}", "WARNING")
except Exception as e:
self.log(f"调用API异常: {str(e)}", "ERROR")
def update_task_status(self, task_id: int, **kwargs):
"""更新任务状态
Args:
task_id: 任务ID
**kwargs: 要更新的字段
"""
if task_id not in runing_task:
runing_task[task_id] = {}
runing_task[task_id].update(kwargs)
def stop(self):
"""停止监控"""
self.running = False
self.log("任务监控器已停止")
def main():
"""主函数:启动任务监控"""
# 创建并启动任务监控器
monitor = TaskMonitor()
try:
monitor.start()
except KeyboardInterrupt:
monitor.log("接收到中断信号,正在停止...")
monitor.stop()
except Exception as e:
monitor.log(f"监控器异常退出: {traceback.format_exc()}", "ERROR")
if __name__ == "__main__":
main()

View File

@@ -187,8 +187,8 @@ def _run_brand_single(taskid,brand_ls, fileUrl,strategy,totalLines,chunkTotal,ch
# "invalidBrands": invalidBrands,"queryFailedBrands": queryFailedBrands
# }]})
json=data)
print(data)
print(taskid,brand_ls, fileUrl)
# print(data)
# print(taskid,brand_ls, fileUrl)
print("提交结果",data,"\n-->",resp.text)
return True

View File

@@ -0,0 +1,726 @@
"""
品牌爬虫蓝图:展开文件夹、运行任务、任务列表/详情、下载结果
"""
import os
import json
import threading
import traceback
import zipfile
import io
import time
import requests
from queue import Queue, Empty
from concurrent.futures import ThreadPoolExecutor, as_completed
from urllib.parse import urlparse
from flask import Blueprint, request, jsonify, session, send_file, redirect, Response
from app_common import get_db, login_required, BASE_DIR
from config import bucket_path,JAVA_API_BASE
from brand_spider.main import single_file_handle, TaskCancelledError
brand_bp = Blueprint('brand', __name__)
BRAND_OUTPUT_DIR = os.path.join(BASE_DIR, 'brand_output')
OSS_PREFIX = "brand_results"
# 任务完成时推送给前端的 SSE 队列task_id -> [Queue, ...]
_task_event_queues = {}
_task_event_lock = threading.Lock()
# 任务行级进度(当前处理文件的行数/总行数),仅存内存,不落库:
# task_id -> {
# 'file_index': int, # 当前处理第几个文件
# 'file_total': int, # 总文件数
# 'file_path': str, # 当前文件路径
# 'current_line': int, # 当前行号(或当前品牌序号)
# 'total_lines': int, # 总行数(或总品牌数)
# }
_task_line_progress = {}
_task_line_progress_lock = threading.Lock()
_cancelled_brand_tasks = set()
_cancelled_brand_tasks_lock = threading.Lock()
def _mark_task_cancelled(task_id):
with _cancelled_brand_tasks_lock:
_cancelled_brand_tasks.add(task_id)
def _is_task_cancelled(task_id):
with _cancelled_brand_tasks_lock:
return task_id in _cancelled_brand_tasks
def _get_uid_from_request_headers():
"""
从请求头读取 uid。
前端会在 headers 里携带 uid用于确定本次请求的归属用户。
"""
uid = request.headers.get('uid')
if uid is None:
return None
uid = str(uid).strip()
if not uid:
return None
try:
return int(uid)
except Exception:
return None
def _resolve_user_id():
"""
优先使用请求头 uid没有请求头 uid 时回退到 session['user_id']。
若二者同时存在但不一致,则视为无效请求并返回 None。
"""
req_uid = _get_uid_from_request_headers()
if req_uid is not None:
sess_uid = session.get('user_id')
if sess_uid is not None:
try:
if int(sess_uid) != int(req_uid):
return None
except Exception:
if str(sess_uid) != str(req_uid):
return None
return req_uid
return session.get('user_id')
def _get_user_id_or_error():
user_id = _resolve_user_id()
if not user_id:
return None, (jsonify({'success': False, 'error': '未登录或缺少uid'}), 401)
try:
return int(user_id), None
except Exception:
return None, (jsonify({'success': False, 'error': 'uid无效'}), 400)
def _push_task_event(task_id, event):
"""向订阅了该任务的所有 SSE 连接推送事件,并移除该任务的队列列表。
同时清理内存中的行级进度,不通过数据库中转行级进度。"""
with _task_event_lock:
queues = _task_event_queues.pop(task_id, [])
for q in queues:
try:
q.put_nowait(event)
except Exception:
pass
# 任务进入终态时,顺便清理行级进度缓存
with _task_line_progress_lock:
_task_line_progress.pop(task_id, None)
status = (event or {}).get('status') if isinstance(event, dict) else None
if status in ('success', 'failed', 'cancelled'):
with _cancelled_brand_tasks_lock:
_cancelled_brand_tasks.discard(task_id)
def _expand_folder_xlsx(folder_path):
"""返回文件夹下所有 .xlsx 文件的绝对路径列表"""
if not folder_path or not os.path.isdir(folder_path):
return []
paths = []
for name in os.listdir(folder_path):
if name.endswith('.xlsx') or name.endswith('.XLSX'):
paths.append(os.path.normpath(os.path.join(folder_path, name)))
return paths
def _upload_local_xlsx_to_oss(local_path, user_id):
"""将本地 xlsx 文件上传到 OSS返回 (url, None) 或 (None, error_message)。"""
if not local_path or not os.path.isfile(local_path):
return None, "文件不存在"
try:
from ali_oss import upload_file as oss_upload_file
except ImportError:
return None, "OSS 模块未配置"
base = os.path.basename(local_path)
safe_base = "".join(c if c.isalnum() or c in '-_.' else '_' for c in base)
if not safe_base.endswith('.xlsx'):
safe_base = safe_base + '.xlsx'
key = f"{bucket_path}brand_input/{user_id}/{int(time.time() * 1000)}/{safe_base}"
try:
with open(local_path, "rb") as f:
content = f.read()
url = oss_upload_file(content, key)
return url, None
except Exception as e:
return None, str(e)
def _run_brand_single(taskid,brand_ls, fileUrl,strategy,totalLines,chunkTotal,chunkIndex):
""""""
invalidBrands = [] # 不符合品牌的数据
queryFailedBrands = [] # 查询失败的数据
keptRows = []
for brand in brand_ls:
if _is_task_cancelled(taskid):
raise TaskCancelledError("任务已取消")
faild_data,query_faild_data = single_file_handle(brand,strategy)
invalidBrands.extend(faild_data)
queryFailedBrands.extend(query_faild_data)
if len(faild_data) == 0 and len(query_faild_data) == 0:
keptRows.append(brand)
data = { "strategy": strategy ,
"files": [
{
"fileUrl": fileUrl,
"originalFilename": "",
"relativePath": "",
"mainSheetName": "",
"chunkIndex": chunkIndex,
"chunkTotal": chunkTotal,
"totalLines": totalLines,
"keptRows": keptRows,
"invalidBrands": invalidBrands,
"queryFailedBrands": queryFailedBrands
}
]
}
resp = requests.post(f"{JAVA_API_BASE}/api/brand/tasks/{taskid}/result",headers={"accept":"application/json"},
# json={ "strategy": strategy ,
# "files": [
# { "fileUrl": fileUrl,"originalFilename": "","relativePath": "",
# "mainSheetName": "","columns": [],"keptRows": [],
# "invalidBrands": invalidBrands,"queryFailedBrands": queryFailedBrands
# }]})
json=data)
# print(data)
# print(taskid,brand_ls, fileUrl)
print("提交结果",data,"\n-->",resp.text)
return True
def _background_brand_task(task_id,data):
"""
后台执行爬虫任务
参数示例:
data : {
"taskId": 348,
"strategy": "Terms",
"files": [
{
"fileIndex": 1,
"fileUrl": "ab8cce4878754bfd89b4cd3ed20e1395",
"originalFilename": "品牌样例.xlsx",
"relativePath": "店铺A/品牌样例.xlsx",
"sheetName": "Sheet1",
"columns": [
"品牌",
"ASIH",
"状态",
"时间"
],
"rows": [
{
"品牌": "YQAUTEC",
"ASIH": "B0F28NZ752",
"状态": "",
"时间": "",
"__rowIndex": 2
}
],
"uniqueBrands": [
"YQAUTEC"
]
}
]
}
"""
try:
file_ls = data.get("files")
strategy = data.get("strategy")
if not isinstance(file_ls, list):
file_ls = []
tasks = []
for file_data in file_ls:
rows = file_data.get("rows") or []
brand_ls = [i.get("品牌") for i in rows if i.get("品牌")]
fileUrl = file_data.get("fileUrl")
if not brand_ls:
continue
for i in range(0, len(brand_ls), 5):
chunk = brand_ls[i:i + 5]
tasks.append((chunk, fileUrl))
if tasks:
max_workers = min(8, len(tasks))
with ThreadPoolExecutor(max_workers=max_workers) as executor:
if _is_task_cancelled(task_id):
_push_task_event(task_id, {'status': 'cancelled'})
return
futures = [
executor.submit(_run_brand_single, task_id, chunk, file_url, strategy,len(brand_ls),len(tasks),chunkIndex+1)
for chunkIndex,(chunk, file_url) in enumerate(tasks)
]
for future in as_completed(futures):
if _is_task_cancelled(task_id):
for f in futures:
f.cancel()
_push_task_event(task_id, {'status': 'cancelled'})
return
future.result()
_push_task_event(task_id, {'status': 'success'})
except Exception as e:
print("执行出错",traceback.format_exc())
_push_task_event(task_id, {'status': 'failed', 'error_message': str(e)})
print("执行完成")
def _parse_json(val, default=None):
if val is None:
return default if default is not None else []
if isinstance(val, (list, dict)):
return val
try:
return json.loads(val)
except Exception:
return default if default is not None else []
def _is_url(s):
if not isinstance(s, str) or not s.strip():
return False
return s.strip().startswith('http://') or s.strip().startswith('https://')
def _normalize_result_paths(raw):
"""
将 result_paths 规范化为 (url_list, zip_url)。
支持旧格式 list 或新格式 dict {"urls": [...], "zip_url": "..."}。
"""
if not raw:
return [], None
if isinstance(raw, dict):
urls = raw.get('urls') or []
if not isinstance(urls, list):
urls = []
zip_url = raw.get('zip_url')
if zip_url and not isinstance(zip_url, str):
zip_url = None
return urls, zip_url
if isinstance(raw, list):
return raw, None
return [], None
@brand_bp.route('/api/brand/expand-folder', methods=['POST'])
@login_required
def api_brand_expand_folder():
"""展开文件夹,返回其下所有 .xlsx 文件路径列表"""
try:
_, err = _get_user_id_or_error()
if err:
return err
data = request.get_json() or {}
folder = (data.get('folder') or '').strip()
if not folder:
return jsonify({'success': False, 'error': '请提供 folder 路径'}), 400
paths = _expand_folder_xlsx(folder)
return jsonify({'success': True, 'paths': paths})
except Exception as e:
return jsonify({'success': False, 'error': str(e)}), 500
def _expand_folder_xlsx_recursive_items(folder_path):
"""返回文件夹下所有 .xlsx 文件的绝对路径和相对路径列表(递归)"""
if not folder_path or not os.path.isdir(folder_path):
return []
items = []
root = os.path.normpath(folder_path)
for current_root, _, filenames in os.walk(root):
for name in filenames:
if not (name.endswith('.xlsx') or name.endswith('.XLSX')):
continue
absolute_path = os.path.normpath(os.path.join(current_root, name))
relative_path = os.path.relpath(absolute_path, root).replace('\\', '/')
items.append({
'absolutePath': absolute_path,
'relativePath': relative_path,
})
items.sort(key=lambda item: item['relativePath'])
return items
@brand_bp.route('/api/brand/expand-folder-recursive', methods=['POST'])
@login_required
def api_brand_expand_folder_recursive():
"""递归展开文件夹,返回 .xlsx 文件绝对路径及相对路径列表"""
try:
_, err = _get_user_id_or_error()
if err:
return err
data = request.get_json() or {}
folder = (data.get('folder') or '').strip()
if not folder:
return jsonify({'success': False, 'error': '请提供 folder 路径'}), 400
items = _expand_folder_xlsx_recursive_items(folder)
return jsonify({'success': True, 'items': items})
except Exception as e:
return jsonify({'success': False, 'error': str(e)}), 500
@brand_bp.route('/api/brand/run', methods=['POST'])
# @login_required
def api_brand_run():
"""立即运行:创建任务并后台执行,立即返回 task_id前端可轮询进度与取消"""
try:
data = request.get_json() or {}
paths = data.get('paths') or []
# task_type: 1=立即执行2=添加任务;此接口默认 1
try:
task_type = int(data.get('task_type') or 1)
except Exception:
task_type = 1
if task_type not in (1, 2):
task_type = 1
strategy = (data.get('strategy') or 'Terms').strip()
if strategy not in ('Terms', 'Simple'):
strategy = 'Terms'
if not isinstance(paths, list):
paths = []
paths = [p.strip() for p in paths if p and isinstance(p, str) and os.path.isfile(p.strip())]
if not paths:
return jsonify({'success': False, 'error': '没有有效的 xlsx 文件路径'}), 400
# 先上传到 OSS获取链接再入库
user_id, err = _get_user_id_or_error()
if err:
return err
urls = []
for p in paths:
url, err = _upload_local_xlsx_to_oss(p, user_id)
if err:
return jsonify({'success': False, 'error': f'上传文件失败: {os.path.basename(p)} - {err}'}), 500
base_name = os.path.basename(p)
urls.append({"fileUrl": url,"originalFilename": base_name,"relativePath": p })
# 请求提交
resp = requests.post(f"{JAVA_API_BASE}/api/brand/tasks?userId={user_id}",headers={
"content-type":"application/json",
},data=json.dumps({ "files": urls, "strategy": strategy,"taskType": 1,"archiveName": ""}))
print({ "files": urls, "strategy": strategy,"taskType": 1,"archiveName": ""})
print(f"{JAVA_API_BASE}/api/brand/tasks?userId={user_id}")
resp_data = resp.json()
# print(f"{JAVA_API_BASE}/api/brand/tasks?userId={user_id} 返回:",resp_data,"状态:",resp.status_code)
if resp_data.get("success"):
task_id = resp_data.get("data").get("taskId")
data = resp_data.get("data")
threading.Thread(target=_background_brand_task, args=(task_id,data), daemon=True).start()
return jsonify({'success': True, 'task_id': task_id})
else:
print(resp_data)
return jsonify({'success': False, 'error': "请求接口失败"}), 500
except Exception as e:
traceback.print_exc()
return jsonify({'success': False, 'error': str(e)}), 500
@brand_bp.route('/api/brand/tasks', methods=['GET', 'POST'])
# @login_required
def api_brand_tasks():
"""GET: 获取当前用户的任务列表POST: 添加任务(后台执行)"""
if request.method == 'GET':
try:
user_id, err = _get_user_id_or_error()
if err:
return err
resp = requests.get(f"{JAVA_API_BASE}/api/brand/tasks", params={
"userId": user_id
})
resp_data = resp.json()
print("获取列表",resp.text)
if resp_data.get("success"):
items = resp_data.get("data").get("items")
return jsonify({'success': True, 'items': items})
return jsonify({'success': True, 'items': []})
except Exception as e:
traceback.print_exc()
print("获取列表失败",e)
return jsonify({'success': False, 'error': str(e)}), 500
# POST
try:
data = request.get_json() or {}
paths = data.get('paths') or []
# task_type: 1=立即执行2=添加任务;此接口默认 2
try:
task_type = int(data.get('task_type') or 2)
except Exception:
task_type = 2
if task_type not in (1, 2):
task_type = 2
strategy = (data.get('strategy') or 'Terms').strip()
if strategy not in ('Terms', 'Simple'):
strategy = 'Terms'
if not isinstance(paths, list):
paths = []
paths = [p.strip() for p in paths if p and isinstance(p, str)]
if not paths:
return jsonify({'success': False, 'error': '请提供至少一个文件路径或链接'}), 400
user_id, err = _get_user_id_or_error()
if err:
return err
urls = []
for p in paths:
url, err = _upload_local_xlsx_to_oss(p, user_id)
if err:
return jsonify({'success': False, 'error': f'上传文件失败: {os.path.basename(p)} - {err}'}), 500
base_name = os.path.basename(p)
urls.append({"fileUrl": url, "originalFilename": base_name, "relativePath": p})
params = {
"userId": user_id
}
# resp = requests.post(f"{JAVA_API_BASE}/api/brand/tasks?userId={user_id}",headers={
# "content-type":"application/json",
# },data=json.dumps({ "files": urls, "strategy": strategy,"taskType": 1,"archiveName": ""}))
req_data = {"files": urls, "strategy": strategy, "taskType": 2, "archiveName": ""}
resp = requests.post(f"{JAVA_API_BASE}/api/brand/tasks?userId={user_id}", headers={"content-type": "application/json"},
data=json.dumps(req_data))
print(req_data)
# print(f"{JAVA_API_BASE}/api/brand/tasks?userId={user_id},返回", resp.text)
resp_data = resp.json()
if resp_data.get("success"):
task_id = resp_data.get("data").get("taskId")
data = resp_data.get("data")
return jsonify({'success': True, 'task_id': task_id})
return jsonify({'success': False, 'task_id': "请求异常"})
except Exception as e:
return jsonify({'success': False, 'error': str(e)}), 500
@brand_bp.route('/api/brand/tasks/<int:task_id>')
# @login_required
def api_brand_task_detail(task_id):
"""获取单个任务详情(用于轮询状态)"""
try:
resp = requests.get(f"{JAVA_API_BASE}/api/brand/tasks/{task_id}")
resp_data = resp.json()
if resp_data.get("success"):
task = resp_data["data"]["task"]
return jsonify({
'success': True,
'task': task
})
except Exception as e:
return jsonify({'success': False, 'error': str(e)}), 500
@brand_bp.route('/api/brand/tasks/<int:task_id>/line-progress')
# @login_required
def api_brand_task_line_progress(task_id):
"""
获取任务的“当前文件行级进度”(当前行数/总行数)。
该信息仅存放在内存字典 _task_line_progress 中,不写入数据库。
"""
try:
resp = requests.get(f"{JAVA_API_BASE}/api/brand/tasks/{task_id}")
resp_data = resp.json()
if resp_data.get("success"):
if resp_data["data"]["line_progress"]["has_progress"]:
info = resp_data["data"]["line_progress"]["info"]
resp = {
'file_index': int(info.get('file_index') or 0),
'file_total': int(info.get('file_total') or 0),
'file_name': info.get("file_name",""),
'current_line': int(info.get('current_line') or 0),
'total_lines': int(info.get('total_lines') or 0),
}
return jsonify({'success': True, 'has_progress': True, 'info': resp})
raise RuntimeError("获取进度失败")
except Exception as e:
return jsonify({'success': False, 'error': str(e)}), 500
@brand_bp.route('/api/brand/tasks/<int:task_id>/events')
# @login_required
def api_brand_task_events(task_id):
"""SSE任务完成时后端主动推送事件前端监听后隐藏进度条与取消按钮"""
user_id, err = _get_user_id_or_error()
if err:
return err
def _task_status_event():
conn = get_db()
try:
with conn.cursor() as cur:
cur.execute(
"SELECT status, error_message FROM brand_crawl_tasks WHERE id = %s AND user_id = %s",
(task_id, user_id)
)
row = cur.fetchone()
finally:
conn.close()
if not row:
return None
st = (row.get('status') or '').lower()
if st in ('success', 'failed', 'cancelled'):
return {'status': st, 'error_message': (row.get('error_message') or '').strip() or None}
return None
# 若任务已处于终态,直接返回一条事件后结束
ev = _task_status_event()
if ev is not None:
def _one_shot():
yield "data: " + json.dumps(ev, ensure_ascii=False) + "\n\n"
return Response(
_one_shot(),
mimetype='text/event-stream',
headers={'Cache-Control': 'no-cache', 'X-Accel-Buffering': 'no'}
)
# 否则注册队列,等待后台任务完成时推送
q = Queue()
with _task_event_lock:
_task_event_queues.setdefault(task_id, []).append(q)
def _stream():
try:
while True:
try:
event = q.get(timeout=20)
yield "data: " + json.dumps(event, ensure_ascii=False) + "\n\n"
return
except Empty:
yield ": keepalive\n\n"
finally:
with _task_event_lock:
lst = _task_event_queues.get(task_id, [])
if q in lst:
lst.remove(q)
if not lst:
_task_event_queues.pop(task_id, None)
return Response(
_stream(),
mimetype='text/event-stream',
headers={'Cache-Control': 'no-cache', 'X-Accel-Buffering': 'no'}
)
@brand_bp.route('/api/brand/tasks/<int:task_id>/cancel', methods=['POST'])
# @login_required
def api_brand_task_cancel(task_id):
"""取消正在执行或等待中的任务pending/running 均可取消)"""
try:
user_id, err = _get_user_id_or_error()
if err:
return err
resp = requests.post(
f"{JAVA_API_BASE}/api/brand/tasks/{task_id}/cancel",
headers={"accept": "application/json"},
timeout=20
)
try:
resp_data = resp.json()
except Exception:
return jsonify({'success': False, 'error': '取消接口返回非 JSON'}), 500
if not resp_data.get('success'):
return jsonify({'success': False, 'error': '第三方取消任务失败'}), 400
_mark_task_cancelled(task_id)
_push_task_event(task_id, {'status': 'cancelled'})
return jsonify({'success': True})
except Exception as e:
return jsonify({'success': False, 'error': str(e)}), 500
@brand_bp.route('/api/brand/tasks/<int:task_id>', methods=['DELETE'])
@login_required
def api_brand_task_delete(task_id):
"""删除任务(仅限非 running 状态)"""
try:
user_id, err = _get_user_id_or_error()
if err:
return err
resp = requests.delete(
f"{JAVA_API_BASE}/api/brand/tasks/{task_id}",
headers={"accept": "application/json"},
timeout=20
)
try:
resp_data = resp.json()
except Exception:
return jsonify({'success': False, 'error': '取消接口返回非 JSON'}), 500
if not resp_data.get('success'):
return jsonify({'success': False, 'error': '第三方取消任务失败'}), 400
return jsonify({'success': True})
except Exception as e:
return jsonify({'success': False, 'error': str(e)}), 500
@brand_bp.route('/api/brand/download/<int:task_id>')
@login_required
def api_brand_download(task_id):
"""下载任务结果:优先使用 OSS zip_url 重定向;否则本地/单链接/多链接按原逻辑处理"""
try:
user_id, err = _get_user_id_or_error()
if err:
return err
conn = get_db()
with conn.cursor() as cur:
cur.execute(
"SELECT result_paths FROM brand_crawl_tasks WHERE id = %s AND user_id = %s",
(task_id, user_id)
)
row = cur.fetchone()
conn.close()
if not row or not row.get('result_paths'):
return jsonify({'success': False, 'error': '无结果可下载'}), 404
raw = row['result_paths']
if isinstance(raw, str):
raw = _parse_json(raw)
url_list, zip_url = _normalize_result_paths(raw)
# 新格式:有 zip_url 直接重定向到 OSS
if zip_url and _is_url(zip_url):
return redirect(zip_url, code=302)
# 兼容旧格式result_paths 可能为 list本地路径或 url
if not url_list and isinstance(raw, list):
url_list = raw
local_paths = [p for p in url_list if isinstance(p, str) and not _is_url(p) and os.path.isfile(p)]
url_paths = [p.strip() for p in url_list if isinstance(p, str) and _is_url(p)]
if len(url_paths) == 1 and not local_paths:
return redirect(url_paths[0], code=302)
if len(local_paths) == 1 and not url_paths:
return send_file(
local_paths[0],
as_attachment=True,
download_name=os.path.basename(local_paths[0])
)
if local_paths or url_paths:
buf = io.BytesIO()
with zipfile.ZipFile(buf, 'w', zipfile.ZIP_DEFLATED) as zf:
for p in local_paths:
zf.write(p, os.path.basename(p))
for url in url_paths:
try:
import requests as req
r = req.get(url, timeout=30, stream=True)
r.raise_for_status()
name = os.path.basename(urlparse(url).path) or ('file_%s' % (url_paths.index(url)))
if not name or name == 'file_%s' % url_paths.index(url):
name = 'download_%s' % url_paths.index(url)
zf.writestr(name, r.content)
except Exception:
pass
buf.seek(0)
return send_file(
buf,
mimetype='application/zip',
as_attachment=True,
download_name='brand_task_%s.zip' % task_id
)
return jsonify({'success': False, 'error': '结果文件不存在或链接不可用'}), 404
except Exception as e:
return jsonify({'success': False, 'error': str(e)}), 500

View File

@@ -4,7 +4,7 @@ from queue import Empty, Full
from flask import Blueprint, jsonify, request
from config import JSON_TASK_QUEUE
from config import JSON_TASK_QUEUE,runing_task
communication_bp = Blueprint('communication', __name__)
@@ -65,3 +65,18 @@ def del_brand():
_latest_del_brand_ts = time.time()
print("【删除品牌】返回任务数据", item)
return jsonify({'success': True, 'data': item})
@communication_bp.route('/api/amazon/get_all_detail', methods=['GET', 'POST'])
def get_all_detail():
"""获取正在执行的删除品牌任务详情"""
request_id = _resolve_request_id()
if not request_id: #如果没有提供 id 参数,直接返回所有正在执行的任务详情
return jsonify({'success': True, 'data': runing_task})
task_info = runing_task.get(request_id)
if not task_info:
return jsonify({'success': False, 'message': '任务不存在', 'data': None})
return jsonify({'success': True, 'data': task_info})

View File

@@ -110,6 +110,14 @@ def proxy(path):
if key.lower() in ['host', 'content-length', 'connection']:
continue
headers[key] = value
try:
print("=============================")
print("target_url",target_url)
print("params",params)
print("data",request.get_data())
print("=============================")
except Exception as e:
print("打印失败",e)
try:
# 使用流式请求

View File

@@ -20,6 +20,14 @@ proxy_mode = int(os.getenv("proxy_mode",1))
client_name=os.getenv("client_name") + ".exe"
JAVA_API_BASE = os.getenv("java_api_base", "http://127.0.0.1:18080")
# 紫鸟浏览器配置
ZN_COMPANY = os.getenv("zn_company", "")
ZN_USERNAME = os.getenv("zn_username", "")
ZN_PASSWORD = os.getenv("zn_password", "")
# 删除品牌API配置
DELETE_BRAND_API_BASE = os.getenv("DELETE_BRAND_API_BASE", JAVA_API_BASE)
cache_path = "./user_data"
@@ -40,7 +48,7 @@ os.environ['SECRET_KEY'] = "ddffc7c1d02121d9554d7b080b2511b6"
debug = True
version = "1.0.22"
version = "1.0.13"
APP_UPDATE_URL = f"{_base_url}/api/version/latest" # 检测更新接口地址,返回 { "file_url": "...", "version": "x.x.x" }
os.environ['APP_VERSION'] = version
os.environ['APP_UPDATE_URL'] = APP_UPDATE_URL
@@ -49,6 +57,34 @@ os.environ['APP_UPDATE_URL'] = APP_UPDATE_URL
# 队列
JSON_TASK_QUEUE = Queue()
#删除品牌,正在执行中的任务
runing_task = {
}
# 正在执行中的店铺
runing_shop = {}
#
# runing_task = {
# "497": {
# "status": "running", # running/completed/failed
# "start_time": "2026-04-01 10:00:00",
# "stop_requested": False,
# "current_shop": "魏振峰",
# "current_country": "德国",
# "current_asin": "B0D451RQRG",
# "total_shops": 1,
# "processed_shops": 0,
# "total_asins": 14,
# "processed_asins": 3,
# "success_count": 2,
# "failed_count": 1
# }
# }

View File

@@ -1,26 +1,35 @@
"""
基于 pywebview 实现的跨平台 UI需先登录方可使用
"""
from config import cache_path,debug,version,JAVA_API_BASE,JSON_TASK_QUEUE
from config import cache_path,debug,version,JAVA_API_BASE,JSON_TASK_QUEUE,runing_task,runing_shop
import datetime
import json
import sys
import shutil
import os
os.makedirs(cache_path,exist_ok=True)
if not debug:
today = datetime.datetime.now().strftime("%Y_%m_%d")
sys.stdout = open(os.path.join(cache_path,f'{today}.log'), 'a',buffering=1)
# sys.stderr = sys.stdout
sys.stderr = sys.stdout
import webview
import threading
import time
import requests
import subprocess
with open("version.txt","w",encoding="utf-8") as file:
file.write(json.dumps({"version":version}))
print(f"""
========================================================
版本: {version}
========================================================
""")
# 获取当前脚本所在目录
BASE_DIR = os.path.dirname(os.path.abspath(__file__))
@@ -170,7 +179,6 @@ class WindowAPI:
def save_template_xlsx(self):
"""弹窗选择保存位置,将品牌文档格式模板 xlsx 保存到用户选择的位置。"""
import shutil
template_name = '品牌文档格式_模板.xlsx'
template_path = os.path.join(BASE_DIR, 'static', template_name)
if not os.path.isfile(template_path):
@@ -189,7 +197,6 @@ class WindowAPI:
except Exception as e:
return {'success': False, 'error': str(e)}
def save_template_zip(self):
"""弹窗选择保存位置,将品牌文档格式模板 xlsx 保存到用户选择的位置。"""
import shutil
@@ -234,6 +241,21 @@ class WindowAPI:
except Exception as e:
return {'success': False, 'error': str(e)}
def open_external_url(self, url):
if not url or not str(url).strip():
return {'success': False, 'error': '链接为空'}
target = str(url).strip()
try:
if sys.platform.startswith('win'):
os.startfile(target)
elif sys.platform == 'darwin':
subprocess.Popen(['open', target])
else:
subprocess.Popen(['xdg-open', target])
return {'success': True}
except Exception as e:
return {'success': False, 'error': str(e)}
def enqueue_json(self, data):
"""保存任务到队列"""
try:
@@ -245,83 +267,33 @@ class WindowAPI:
payload = json.loads(data)
if not isinstance(payload, (dict, list)):
return {'success': False, 'error': '仅支持 JSON 对象或数组'}
if payload.get("type") == "delete-brand-run" and payload.get("data").get("taskId") in runing_task:
return {'success': False, 'error': '已存在正在执行的删除品牌任务'}
# 判断当前店铺是否正在运行中
if payload.get("data").get("items"):
shop_name = payload["data"]["items"][0]["shopName"]
if shop_name in runing_shop:
return {'success': False, 'error': '当前店铺正在执行中,请等待店铺完成之后重试'}
JSON_TASK_QUEUE.put(payload)
return {'success': True, 'queue_size': JSON_TASK_QUEUE.qsize()}
except Exception as e:
return {'success': False, 'error': str(e)}
def _select_folder_win32():
"""Windows: 用 ctypes 调用系统文件夹选择对话框,打包成 exe 后也能正常弹窗(不依赖 tkinter"""
try:
import ctypes
from ctypes import wintypes
BIF_RETURNONLYFSDIRS = 0x00000001
BIF_NEWDIALOGSTYLE = 0x00000040
MAX_PATH = 260
shell32 = ctypes.windll.shell32 # type: ignore[attr-defined]
# BROWSEINFOW
class BROWSEINFOW(ctypes.Structure):
_fields_ = [
("hwndOwner", wintypes.HWND),
("pidlRoot", ctypes.c_void_p),
("pszDisplayName", ctypes.c_wchar_p),
("lpszTitle", ctypes.c_wchar_p),
("ulFlags", wintypes.UINT),
("lpfn", ctypes.c_void_p),
("lParam", wintypes.LPARAM),
("iImage", ctypes.c_int),
]
display_name_buf = ctypes.create_unicode_buffer(MAX_PATH)
bi = BROWSEINFOW()
bi.hwndOwner = 0
bi.pidlRoot = 0
bi.pszDisplayName = ctypes.cast(display_name_buf, ctypes.c_wchar_p)
bi.lpszTitle = "选择保存图片的文件夹"
bi.ulFlags = BIF_RETURNONLYFSDIRS | BIF_NEWDIALOGSTYLE
bi.lpfn = 0
bi.lParam = 0
bi.iImage = 0
pidl = shell32.SHBrowseForFolderW(ctypes.byref(bi))
if not pidl:
return ''
path_buf = ctypes.create_unicode_buffer(MAX_PATH)
if not shell32.SHGetPathFromIDListW(pidl, path_buf):
return ''
ole32 = ctypes.windll.ole32 # type: ignore[attr-defined]
ole32.CoTaskMemFree(pidl)
return path_buf.value or ''
except Exception as e:
print("Windows 文件夹选择失败:", e)
return ''
def _select_folder_tk():
import tkinter as tk
from tkinter import filedialog
root = tk.Tk()
root.withdraw()
root.attributes('-topmost', True)
# 使用 after 确保 askdirectory 在主循环启动后执行
def ask():
nonlocal path
path = filedialog.askdirectory(title='选择保存图片的文件夹')
root.quit() # 关闭主循环
path = ''
root.after(100, ask)
root.mainloop() # 启动主循环,直到 root.quit() 被调用
try:
root.destroy()
except:
pass
return path
def select_folder():
"""弹窗选择文件夹返回路径空字符串表示取消。Windows 打包 exe 时优先用系统 API 避免 tkinter 不弹窗。"""
return _select_folder_tk()
def get_detail_del(self,task_id):
"""获取正在执行的删除品牌任务详情"""
task_info = runing_task.get(task_id)
if not task_info:
return {'success': False, 'error': '任务不存在'}
return {'success': True, 'task_info': task_info}
def stop_task(self,task_id):
"""停止正在执行的删除品牌任务"""
task_info = runing_task.get(task_id)
if not task_info:
return {'success': False, 'error': '任务不存在'}
# 这里可以设置一个标志位,实际的删除品牌任务需要定期检查这个标志位来决定是否停止
task_info['stop_requested'] = True
return {'success': True, 'message': '已请求停止任务'}
def start_flask():
@@ -350,7 +322,7 @@ def main():
)
api = WindowAPI(window)
window.expose(api.close, api.minimize, api.maximize, api.toggle_maximize, generate_images, api.save_image, api.save_image_to_folder, api.select_folder, api.select_brand_xlsx_files, api.select_brand_folder,
api.save_file_from_url, api.save_template_xlsx,api.save_template_zip,api.upload_file_to_java,api.enqueue_json)
api.save_file_from_url, api.save_template_xlsx,api.save_template_zip,api.upload_file_to_java,api.open_external_url,api.enqueue_json,api.get_detail_del)
webview.start(
debug=True,
storage_path=cache_path,