qrpa 1.0.89__tar.gz → 1.0.91__tar.gz

{qrpa-1.0.89 → qrpa-1.0.91}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: qrpa
-Version: 1.0.89
+Version: 1.0.91
 Summary: qsir's rpa library
 Author: QSir
 Author-email: QSir <1171725650@qq.com>

{qrpa-1.0.89 → qrpa-1.0.91}/pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "setuptools.build_meta"
 
 [project]
 name = "qrpa"
-version = "1.0.89"
+version = "1.0.91"
 description = "qsir's rpa library"
 authors = [{ name = "QSir", email = "1171725650@qq.com" }]
 readme = "README.md"

{qrpa-1.0.89 → qrpa-1.0.91}/qrpa/db_migrator.py

@@ -580,7 +580,7 @@ def create_default_migrator(silent: bool = False) -> DatabaseMigrator:
     )
     
     remote_config = RemoteConfig(
-        ssh_host="git@e3",
+        ssh_host="git@ecslz",
         temp_dir="/tmp/db_migration",
         database=remote_db
     )

{qrpa-1.0.89 → qrpa-1.0.91}/qrpa/fun_excel.py

@@ -14,6 +14,9 @@ import threading
 from playwright.sync_api import sync_playwright
 import psutil
 
+import os, sys
+from pathlib import Path
+
 from .fun_base import log, sanitize_filename, create_file_path, copy_file, add_https, send_exception
 
 excel_color_index = {
@@ -195,13 +198,13 @@ def merge_by_column_v2(sheet, column_name, other_columns):
     # 更安全的数据获取方式，确保获取完整的数据范围
     last_row = get_last_row(sheet, col_letter)
     data = sheet.range(f'{col_letter}1:{col_letter}{last_row}').value
-    
+
     # 确保data是列表格式
     if not isinstance(data, list):
         data = [data]
-    
+
     log(f'数据范围: {col_letter}1:{col_letter}{last_row}, 数据长度: {len(data)}')
-    
+
     start_row = 2  # 从第2行开始，跳过表头
     merge_row_ranges = []  # 用来存储需要合并的行范围 (start_row, end_row)
 
@@ -211,13 +214,13 @@ def merge_by_column_v2(sheet, column_name, other_columns):
         col_name = find_column_by_data(sheet, 1, col)
         if col_name:
             all_columns.append(col_name)
-    
+
     log(f'需要合并的列: {all_columns}')
 
     # 遍历数据行，从第3行开始比较（因为第1行是表头，第2行是第一个数据行）
     for row in range(3, len(data) + 1):
-        log(f'查找 {row}/{len(data)}, 当前值: {data[row-1] if row-1 < len(data) else "超出范围"}, 前一个值: {data[row-2] if row-2 < len(data) else "超出范围"}')
-        
+        log(f'查找 {row}/{len(data)}, 当前值: {data[row - 1] if row - 1 < len(data) else "超出范围"}, 前一个值: {data[row - 2] if row - 2 < len(data) else "超出范围"}')
+
         # 检查值是否发生变化
         if row <= len(data) and data[row - 1] != data[row - 2]:
             # 值发生变化，处理前一组
@@ -233,7 +236,7 @@ def merge_by_column_v2(sheet, column_name, other_columns):
         merge_row_ranges.append((start_row, end_row))
 
     log(f'行合并范围: {merge_row_ranges}')
-    
+
     # 对每个行范围，在所有指定列中执行合并
     for start_row, end_row in merge_row_ranges:
         if start_row < end_row:  # 只有当开始行小于结束行时才合并（多行）
@@ -756,9 +759,33 @@ def download_images_concurrently(image_urls, platform='shein', img_save_dir=None
         results = list(executor.map(lambda url: download_img_v2(url, platform, img_save_path=img_save_dir), image_urls))
     return results
 
+def get_chromium_executable():
+    """
+    返回 Chromium 可执行文件路径，兼容 PyInstaller 打包后的 exe
+    """
+    # PyInstaller 临时目录
+    if getattr(sys, 'frozen', False):
+        base_path = Path(sys._MEIPASS)
+    else:
+        base_path = Path(__file__).parent
+
+    # 尝试查找打包后的浏览器文件
+    possible_path = base_path / "playwright" / "driver" / "package" / "chromium_headless_shell-1169" / "chrome-win" / "headless_shell.exe"
+    if possible_path.exists():
+        return str(possible_path)
+
+    # fallback: 系统 Playwright 安装目录
+    local_appdata = Path(os.environ.get("LOCALAPPDATA", ""))
+    fallback_path = local_appdata / "ms-playwright" / "chromium" / "chrome-win" / "chrome.exe"
+    if fallback_path.exists():
+        return str(fallback_path)
+
+    raise FileNotFoundError("Chromium 可执行文件未找到，请先执行 'playwright install' 下载浏览器。")
+
 def download_img_by_chrome(image_url, save_name):
+    chromium_path = get_chromium_executable()
     with sync_playwright() as p:
-        browser = p.chromium.launch(headless=True)  # 运行时可以看到浏览器
+        browser = p.chromium.launch(headless=True, executable_path=chromium_path)  # 运行时可以看到浏览器
         context = browser.new_context()
         page = context.new_page()
         # 直接通过Playwright下载图片

{qrpa-1.0.89 → qrpa-1.0.91}/qrpa/fun_web.py

@@ -146,6 +146,54 @@ def full_screen_shot(web_page: Page, config):
     web_page.screenshot(path=full_screenshot_image_path, full_page=True)
     return full_screenshot_image_path
 
+def fetch_get(page: Page, url: str, headers: Optional[dict] = None, config: Optional[dict] = None) -> dict:
+    """
+    发送 HTTP GET 请求，支持自定义 headers 和配置。
+    
+    :param page: Playwright 的 Page 对象
+    :param url: 请求地址
+    :param headers: 自定义 headers 字典
+    :param config: 请求配置字典，可包含 credentials, mode, referrer, referrerPolicy 等
+    :return: 服务器返回的 JSON 响应（dict）
+    """
+    if headers is not None and not isinstance(headers, dict):
+        raise ValueError("headers 参数必须是 dict 或 None")
+    if config is not None and not isinstance(config, dict):
+        raise ValueError("config 参数必须是 dict 或 None")
+
+    try:
+        page.wait_for_load_state('load')
+        response = page.evaluate("""
+            async ({ url, extraHeaders, config }) => {
+                try {
+                    const defaultHeaders = {
+                        'user-agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/130.0.0.0 Safari/537.36',
+                    };
+
+                    const defaultConfig = {
+                        method: 'GET',
+                        credentials: 'include',
+                        mode: 'cors'
+                    };
+
+                    const headers = Object.assign({}, defaultHeaders, extraHeaders || {});
+                    const options = Object.assign({}, defaultConfig, config || {}, { headers: headers });
+
+                    const response = await fetch(url, options);
+                    if (!response.ok) {
+                        throw new Error(`HTTP ${response.status} - ${response.statusText}`);
+                    }
+                    return await response.json();
+                } catch (error) {
+                    return { "error": "fetch_failed", "message": error.message };
+                }
+            }
+        """, {"url": url, "extraHeaders": headers, "config": config})
+
+        return response
+    except Exception as e:
+        raise send_exception()
+
 def safe_goto(page, url, **kwargs):
     caller = inspect.stack()[1]
     log(f"[DEBUG] goto called from {caller.filename}:{caller.lineno} url={url}")

{qrpa-1.0.89 → qrpa-1.0.91}/qrpa/mysql_module/shein_product_model.py

@@ -64,6 +64,9 @@ class SheinProductSkc(Base):
     created_at = Column(DateTime, default=datetime.now, comment='创建时间')
     updated_at = Column(DateTime, default=datetime.now, onupdate=datetime.now, comment='更新时间')
 
+    # 用户备注字段（供后续web界面使用）
+    user_notes = Column(Text, nullable=True, comment='用户备注')
+
     # 定义索引
     __table_args__ = (
         Index('ix_skc_id', 'skc_id'),
@@ -108,6 +111,7 @@ class SheinProductSku(Base):
     price = Column(DECIMAL(10, 2), nullable=True, comment='价格')
     erp_cost_price = Column(DECIMAL(10, 2), nullable=True, comment='ERP成本价')
     erp_supplier_name = Column(String(100), nullable=True, comment='ERP默认供货商')
+    erp_stock = Column(Integer, nullable=True, comment='ERP库存')
 
     # 时间戳
     created_at = Column(DateTime, default=datetime.now, comment='创建时间')
@@ -218,10 +222,10 @@ class SheinProductManager:
         session = self.Session()
         try:
             for data in data_list:
-                
+
                 # 处理SKC数据
                 skc_record = self._upsert_skc_data(session, data)
-                
+
                 # 处理SKU数据
                 for sku_data in data.get('skuList', []):
                     sku_data['local_skc_id'] = skc_record.id
@@ -319,6 +323,7 @@ class SheinProductManager:
             existing_sku.price = sku_data.get('price')
             existing_sku.erp_cost_price = sku_data.get('erp_cost_price')
             existing_sku.erp_supplier_name = sku_data.get('erp_supplier_name')
+            existing_sku.erp_stock = sku_data.get('erp_stock')
             existing_sku.updated_at = datetime.now()
         else:
             # 插入新记录
@@ -473,4 +478,7 @@ def example_usage2():
 
 if __name__ == "__main__":
     pass
-    # example_usage()
+    database_url = "mysql+pymysql://root:123wyk@localhost:3306/lz"
+    manager = SheinProductManager(database_url)
+    manager.create_tables()
+    # example_usage()

{qrpa-1.0.89 → qrpa-1.0.91}/qrpa/shein_excel.py

@@ -2341,7 +2341,7 @@ class SheinExcel:
         self.dealFormula(sheet)  # 有空再封装优化
         colorize_by_field(sheet, 'SPU')
         autofit_column(sheet, ['商品信息', '店铺名称', 'SKC点击率/SKC转化率', '自主参与活动'])
-        column_to_left(sheet, ['店铺名称', 'SKC点击率/SKC转化率', '自主参与活动'])
+        column_to_left(sheet, ['店铺名称', 'SKC点击率/SKC转化率', '自主参与活动','近7天SKU销量/SKC销量/SKC曝光'])
         specify_column_width(sheet, ['商品标题'], 150 / 6)
         add_borders(sheet)
         InsertImageV2(sheet, ['SKC图片', 'SKU图片'], 'shein', 120, None, None, True)

{qrpa-1.0.89 → qrpa-1.0.91}/qrpa/shein_lib.py

@@ -1,6 +1,6 @@
 from .fun_file import read_dict_from_file, write_dict_to_file, read_dict_from_file_ex, write_dict_to_file_ex
 from .fun_base import log, send_exception, md5_string, get_safe_value, NetWorkIdleTimeout
-from .fun_web import fetch, full_screen_shot, safe_goto
+from .fun_web import fetch, fetch_get, full_screen_shot, safe_goto
 from .time_utils import TimeUtils
 from .wxwork import WxWorkBot
 
@@ -36,7 +36,7 @@ class SheinLib:
             try:
                 current_url = web_page.url
                 log(f"尝试获取页面信息 - URL: {current_url}", self.store_username, self.store_name)
-                
+
                 # 检查是否在认证页面，如果是则直接跳转到目标页面
                 if '/auth/SSLS' in current_url:
                     log("检测到SSLS认证页面，直接跳转到首页", self.store_username, self.store_name)
@@ -44,17 +44,17 @@ class SheinLib:
                     web_page.wait_for_timeout(3000)
                     current_url = web_page.url
                     log(f"跳转后URL: {current_url}", self.store_username, self.store_name)
-                
+
                 # 等待导航完成
                 web_page.wait_for_load_state("domcontentloaded", timeout=6000)
-                
+
                 final_url = web_page.url
                 final_title = web_page.title()
                 log(f"页面稳定 - URL: {final_url}, 标题: {final_title}", self.store_username, self.store_name)
                 break
-                
+
             except Exception as e:
-                log(f"第{attempt+1}次等待页面稳定失败: {e}", self.store_username, self.store_name)
+                log(f"第{attempt + 1}次等待页面稳定失败: {e}", self.store_username, self.store_name)
                 if "crashed" in str(e) or "Target" in str(e):
                     log("页面稳定检查时崩溃，直接继续", self.store_username, self.store_name)
                     break
@@ -66,7 +66,7 @@ class SheinLib:
                     log("页面稳定等待最终失败，继续执行", self.store_username, self.store_name)
                     break
                 web_page.wait_for_timeout(2000)
-        
+
         web_page.wait_for_timeout(2000)
 
         # 定义最大重试次数
@@ -83,19 +83,19 @@ class SheinLib:
                 retries += 1
 
                 while not web_page.locator('//div[contains(text(),"商家后台")]').nth(1).is_visible():
-                    
+
                     try:
                         current_url = web_page.url
                         current_title = web_page.title()
                         log(f"循环检查 - URL: {current_url}, 标题: {current_title}", self.store_username, self.store_name)
-                        
+
                         # 如果在认证页面且出现问题，直接跳转
                         if '/auth/SSLS' in current_url:
                             log("在主循环中检测到SSLS认证页面，跳转到首页", self.store_username, self.store_name)
                             web_page.goto('https://sso.geiwohuo.com/#/home', wait_until='domcontentloaded', timeout=15000)
                             web_page.wait_for_timeout(3000)
                             continue
-                            
+
                     except Exception as status_error:
                         log(f"获取页面状态失败: {status_error}", self.store_username, self.store_name)
                         if "crashed" in str(status_error):
@@ -200,13 +200,13 @@ class SheinLib:
             except Exception as e:
                 log(f"错误发生: {e}, 重试中...({self.store_username}, {self.store_name})")
                 log(traceback.format_exc())
-                
+
                 # 收集崩溃时的详细信息
                 try:
                     crash_url = web_page.url
                     crash_title = web_page.title()
                     log(f"崩溃时页面信息 - URL: {crash_url}, 标题: {crash_title}", self.store_username, self.store_name)
-                    
+
                     # 尝试截图保存崩溃现场
                     try:
                         screenshot_path = f"crash_screenshot_{self.store_username}_{int(time.time())}.png"
@@ -214,19 +214,19 @@ class SheinLib:
                         log(f"已保存崩溃截图: {screenshot_path}", self.store_username, self.store_name)
                     except:
                         log("无法截取崩溃时的页面截图", self.store_username, self.store_name)
-                        
+
                 except:
                     log("无法获取崩溃时的页面信息", self.store_username, self.store_name)
-                
+
                 # 检查特定类型的错误
                 if any(keyword in str(e).lower() for keyword in ['memory', 'out of memory', 'oom']):
                     log("检测到内存相关崩溃", self.store_username, self.store_name)
-                
+
                 if "destroyed" in str(e) or "navigation" in str(e):
                     log("检测到导航中断，等待页面稳定后重试", self.store_username, self.store_name)
                     web_page.wait_for_timeout(5000)
                     continue
-                
+
                 if 'crashed' in str(e) or 'Target' in str(e):
                     log("检测到页面或目标崩溃，直接退出当前循环", self.store_username, self.store_name)
                     raise e
@@ -241,6 +241,44 @@ class SheinLib:
         # web_page.wait_for_load_state("networkidle")
         web_page.wait_for_timeout(3000)
 
+    # 获取用户信息
+    def get_user(self, uuid=None):
+        log(f'获取用户信息:{self.store_username} {self.store_name}')
+
+        # 生成 uuid 参数，如果没有提供则使用时间戳
+        if uuid is None:
+            import time
+            uuid = str(int(time.time() * 1000))
+
+        url = f"https://sso.geiwohuo.com/sso-prefix/auth/getUser?uuid={uuid}"
+
+        # 设置请求头，根据 Chrome 请求
+        headers = {
+            "gmpsso-language": "CN",
+            "origin-url"     : "https://sso.geiwohuo.com/#/home/",
+            "x-sso-scene"    : "gmpsso"
+        }
+
+        # 特定于此请求的配置
+        fetch_config = {
+            "credentials"   : "omit",
+            "referrer"      : "https://sso.geiwohuo.com/",
+            "referrerPolicy": "strict-origin-when-cross-origin"
+        }
+
+        response_text = fetch_get(self.web_page, url, headers, fetch_config)
+        error_code = response_text.get('code')
+        if str(error_code) != '0':
+            raise send_exception(json.dumps(response_text, ensure_ascii=False))
+        info = response_text.get('info', {})
+        log(info)
+        cache_file = f'{self.config.auto_dir}/shein_user.json'
+        info['store_username'] = self.store_username
+        info['store_name'] = self.store_name
+        write_dict_to_file_ex(cache_file, {self.store_username: info}, [self.store_username])
+
+        return info
+
     # 获取质检报告pdf地址
     def get_qc_report_url(self, deliverCode, purchaseCode):
         log(f'获取质检报告:{deliverCode} {purchaseCode}')
@@ -849,6 +887,8 @@ class SheinLib:
                 cost_price = self.bridge.get_sku_cost(sku_item['supplierSku'], self.config.erp_source)
                 sku_item['erp_cost_price'] = cost_price if isinstance(cost_price, (int, float)) else None
                 sku_item['erp_supplier_name'] = self.bridge.get_sku_supplier(sku_item['supplierSku'], self.config.erp_source)
+                stock = self.bridge.get_sku_stock(sku_item['supplierSku'], self.config.erp_source)
+                sku_item['erp_stock'] = stock if isinstance(stock, (int, float)) else None
 
         cache_file = f'{self.config.auto_dir}/shein/product/skc_list_{self.store_username}.json'
         write_dict_to_file_ex(cache_file, {self.store_username: skc_list}, [self.store_username])
@@ -2276,6 +2316,7 @@ class SheinLib:
             skc = str(spu_info['skc'])
             # if not shein_db.exists_sales_1_days_ago(skc):
             #     log(f'未查到昨天销量: {skc}')
+            self.get_skc_week_actual_sales(skc)
             self.get_skc_sales(skc, date_60_days_ago, date_1_days_ago)
             skcCode = spu_info['supplierCode']
             product_name = DictSpuInfo[spu]['product_name_en']
@@ -2396,7 +2437,7 @@ class SheinLib:
                 # SKC趋势数据
                 sku_item.append(skc_trend.get('saleCnt', 0))  # SKC近7天销量
                 sku_item.append(skc_trend.get('epsUvIdx', 0))  # SKC近7天曝光人数
-                sku_item.append(skc_trend.get('goodsUvIdx', 0))  # SKC近7天商详访客
+                sku_item.append(skc_trend.get('goodsUv', 0))  # SKC近7天商详访客
                 sku_item.append(skc_trend.get('epsGdsCtrIdx', 0))  # SKC近7天点击率
                 sku_item.append(skc_trend.get('payUvIdx', 0))  # SKC近7天支付人数
                 sku_item.append(skc_trend.get('gdsPayCtrIdx', 0))  # SKC近7天支付率

{qrpa-1.0.89 → qrpa-1.0.91}/qrpa/shein_ziniao.py

@@ -5,7 +5,7 @@
 import os
 import platform
 import shutil
-import time
+import time, datetime
 import traceback
 import uuid
 import json
@@ -74,7 +74,7 @@ class ZiniaoClient:
         """启动客户端"""
         try:
             if self.is_windows:
-                cmd = [self.client_path, '--run_type=web_driver', '--ipc_type=http', '--port=' + str(self.socket_port)]
+                cmd = [self.client_path, '--run_type=web_driver', '--show_sidb=true', '--ipc_type=http', '--port=' + str(self.socket_port)]
             elif self.is_mac:
                 cmd = ['open', '-a', self.client_path, '--args', '--run_type=web_driver', '--ipc_type=http',
                        '--port=' + str(self.socket_port)]
@@ -272,6 +272,73 @@ class ZiniaoTaskManager:
         self.browser = browser
         self.config = config
 
+    def daily_cleanup_superbrowser(self, browser_id):
+        """
+        每天删除一次SuperBrowser缓存文件夹
+
+        Args:
+            browser_id (str): 浏览器ID，如 '26986387919128'
+        """
+
+        # 获取本地AppData路径
+        local_appdata = os.getenv('LOCALAPPDATA')
+        if not local_appdata:
+            log("错误: 无法获取LOCALAPPDATA路径")
+            return False
+
+        # 构建路径
+        cache_path = os.path.join(local_appdata, 'SuperBrowser')
+        target_folder = os.path.join(cache_path, f'User Data\\Chromium_{browser_id}')
+        flag_file = os.path.join(cache_path, f'User Data\\cleanup_flag_{browser_id}.txt')
+
+        # 检查目标文件夹是否存在
+        if not os.path.exists(target_folder):
+            log(f"目标文件夹不存在: {target_folder}")
+            return False
+
+        # 获取当前日期
+        today = datetime.date.today()
+        today_str = today.strftime('%Y-%m-%d')
+
+        # 检查标志文件
+        need_cleanup = True
+
+        if os.path.exists(flag_file):
+            try:
+                # 读取标志文件中的日期
+                with open(flag_file, 'r', encoding='utf-8') as f:
+                    last_cleanup_date = f.read().strip()
+
+                # 如果是今天已经清理过，则跳过
+                if last_cleanup_date == today_str:
+                    log(f"今天({today_str})已经清理过，跳过删除操作")
+                    need_cleanup = False
+
+            except Exception as e:
+                log(f"读取标志文件时出错: {e}")
+                # 如果读取出错，继续执行清理
+
+        if need_cleanup:
+            try:
+                # 删除目标文件夹
+                log(f"正在删除文件夹: {target_folder}")
+                shutil.rmtree(target_folder)
+                log("删除成功!")
+
+                # 创建/更新标志文件
+                os.makedirs(os.path.dirname(flag_file), exist_ok=True)
+                with open(flag_file, 'w', encoding='utf-8') as f:
+                    f.write(today_str)
+
+                log(f"已创建标志文件: {flag_file}")
+                return True
+
+            except Exception as e:
+                log(f"删除文件夹时出错: {e}")
+                return False
+
+        return True
+
     def run_single_store_task(self, browser_info: Dict[str, Any],
             run_func: Callable, task_key: str,
             just_store_username: Optional[List[str]] = None,
@@ -282,6 +349,10 @@ class ZiniaoTaskManager:
         store_name = browser_info.get("browserName")
         store_username = browser_info.get("store_username")
 
+        # 删除浏览器缓存,一天一删
+        browser_id = browser_info.get("browserId")
+        self.daily_cleanup_superbrowser(browser_id)
+
         retry_count = 0
         while True:
             try:

{qrpa-1.0.89 → qrpa-1.0.91}/qrpa/time_utils.py

@@ -260,20 +260,31 @@ class TimeUtils:
 
     @staticmethod
     def get_dates_from_first_of_month_to_yesterday() -> List[str]:
-        """获取从本月第一天到昨天的日期列表"""
+        """获取从本月第一天到昨天的日期列表
+        如果今天是本月第一天，则返回上个月的日期列表
+        """
         today = datetime.today()
-        first_day_of_month = today.replace(day=1)
         yesterday = today - timedelta(days=1)
 
-        date_list = []
-        current_date = first_day_of_month
+        # 如果今天是本月第一天，取上个月
+        if today.day == 1:
+            # 找到上个月最后一天
+            last_month_last_day = today - timedelta(days=1)
+            # 上个月第一天
+            first_day_of_last_month = last_month_last_day.replace(day=1)
+            start_date = first_day_of_last_month
+            end_date = last_month_last_day
+        else:
+            start_date = today.replace(day=1)
+            end_date = yesterday
 
-        while current_date <= yesterday:
+        date_list = []
+        current_date = start_date
+        while current_date <= end_date:
             date_list.append(current_date.strftime('%Y-%m-%d'))
             current_date += timedelta(days=1)
 
         return date_list
-
     # ==================== 月份相关 ====================
 
     @staticmethod

{qrpa-1.0.89 → qrpa-1.0.91}/qrpa.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: qrpa
-Version: 1.0.89
+Version: 1.0.91
 Summary: qsir's rpa library
 Author: QSir
 Author-email: QSir <1171725650@qq.com>