""" 离散备料计划维护数据清理工具 功能：自动登录 ERP 系统，根据负责人姓名批量清理指定的备料计划物料。 优化点： 1. 数据库预取：从 $O(n)$ 次数据库查询优化为 $O(1)$ 内存匹配（HashSet）。 2. 日志规范：使用 logging 模块替代 print。 3. 代码整洁：移除方法内导入，增加通用定位辅助函数。 """ import os import re import time import logging from typing import Union, List, Optional, Callable, Dict, Any from datetime import datetime from playwright.sync_api import sync_playwright, TimeoutError # 统一顶部导入 from utils.auth import login, logout from db.production_order_query import ( read_production_ids, query_production_order_numbers, ) from db.materials_to_delete import get_materials_to_delete_by_managers # --- 日志配置 --- logging.basicConfig( level=logging.INFO, format="%(asctime)s [%(levelname)s] %(message)s", datefmt="%Y-%m-%d %H:%M:%S", ) logger = logging.getLogger(__name__) class DiscreteMaterialPlanCleaner: """离散备料计划维护数据清理器""" def __init__( self, username, password, manager_names: Union[str, List[str], None] = None, headless=False, verbose=True, dryrun=False, progress_callback: Optional[Callable[[int, int, str], None]] = None, ): self.username = username self.password = password self.headless = headless self.verbose = verbose self.dryrun = dryrun self.progress_callback = progress_callback # 参数规范化：支持 str、List[str]、None if manager_names is None: self.manager_names = None # 表示全部 elif isinstance(manager_names, str): self.manager_names = [manager_names] if manager_names.strip() else None else: self.manager_names = manager_names if manager_names else None # 核心优化：使用 set 存储待删除编码，查询复杂度为 $O(1)$ self.to_delete_set = set() # 统计信息 self.stats = { "total_orders": 0, "processed_orders": 0, "total_materials": 0, # 总物料数 "processed_materials": 0, # 已处理物料数 "deleted_materials": [], # [{order_id, material_code, material_name}] "skipped_materials": [], # [{order_id, material_code, material_name, reason}] "unmatched_materials": [], # [{order_id, material_code, material_name}] 不在删除列表的物料 "errors": [], # [{order_id, error_message}] "start_time": None, "end_time": None, } def _log(self, message, level="info"): """统一日志输出控制""" if self.verbose: if level == "info": logger.info(message) elif level == "warn": logger.warning(message) elif level == "error": logger.error(message) def _report_progress(self, current: int, total: int, message: str): """报告进度""" if self.progress_callback: self.progress_callback(current, total, message) def _report_material_progress( self, order_idx: int, total_orders: int, material_idx: int, total_materials: int, order_id: str, material_name: str, action: str, ): """报告物料处理进度 进度计算逻辑： - 每个订单占 1/total_orders 的固定进度配额 - 订单内物料进度按比例分配（material_idx / total_materials） - 总体进度 = (order_idx + material_idx / total_materials) / total_orders """ if self.progress_callback: order_progress = f"订单 [{order_idx + 1}/{total_orders}]" if total_materials > 0: material_progress = f"物料 [{material_idx}/{total_materials}]" message = f"{order_progress} {material_progress} - {order_id} - {action}: {material_name}" # 计算总体进度比例（0.0 到 1.0） order_internal_ratio = material_idx / total_materials overall_ratio = (order_idx + order_internal_ratio) / total_orders # 使用固定精度整数表示进度（范围 0-10000，显示时除以 100 即为百分比） PROGRESS_SCALE = 10000 overall_current = int(overall_ratio * PROGRESS_SCALE) overall_total = PROGRESS_SCALE self.progress_callback(overall_current, overall_total, message) else: message = f"{order_progress} - {order_id} - {action}: {material_name}" self.progress_callback(order_idx + 1, total_orders, message) def _is_button_enabled(self, button_locator): """判定按钮是否可用""" try: return button_locator.is_enabled() except Exception as e: self._log(f"检查按钮状态时出错: {e}", "error") return False def _get_input_value(self, container, label_regex): """通用辅助函数：根据 Label 正则获取 Input 的值""" return ( container.locator("div") .filter(has_text=re.compile(label_regex, re.MULTILINE)) .locator("input") .first.input_value() ) def preload_data(self): """批量预取数据库数据""" if self.manager_names is None: self._log("正在从数据库提取所有负责人的待删除物料清单...") else: names_str = "、".join(self.manager_names) self._log(f"正在从数据库提取负责人 [{names_str}] 的待删除物料清单...") raw_list = get_materials_to_delete_by_managers(self.manager_names) self.to_delete_set = set(raw_list) self._log(f"预加载完成，共计 {len(self.to_delete_set)} 条不合规物料编码。") def get_production_order_numbers(self, production_id_file): """读取文件并查询生产订单号""" production_ids = read_production_ids(production_id_file) order_ids = query_production_order_numbers(production_ids) self._log( f"读取到 {len(production_ids)} 个总排号 -> 匹配到 {len(order_ids)} 个生产订单号" ) return order_ids def process_order( self, inner_frame, order_id, order_index, page1, total_orders: int = 1 ): """清理单个订单的数据 Args: inner_frame: 内层 iframe order_id: 订单 ID order_index: 订单索引（从 0 开始） page1: 页面对象 total_orders: 总订单数（用于进度报告） """ # 报告订单进度 # self._report_progress(order_index + 1, total_orders, f"正在打开订单: {order_id}") # 1. 查询订单 textbox = inner_frame.get_by_role("textbox", name="生产订单号") textbox.fill(order_id) inner_frame.locator(".search-component-searchBtn").click() # 2. 等待加载（改进：增加 60s 安全超时，防止死锁） loading_locator = inner_frame.locator("div").filter(has_text="加载中").nth(1) try: loading_locator.wait_for(state="visible", timeout=3000) loading_locator.wait_for(state="hidden", timeout=60000) except TimeoutError: pass # 3. 进入备料计划详情 inner_frame.locator("#hot-key-head_list").get_by_text("更多").click() with page1.expect_popup() as page2_info: inner_frame.get_by_text("备料计划").click() page2 = page2_info.value # 4. 穿透嵌套 Iframe detail_main_frame = page2.locator("#forwardFrame").content_frame detail_inner_frame = detail_main_frame.locator("#mainiframe").content_frame # 5. 提取订单状态信息 plan_code_locator = detail_inner_frame.get_by_text( re.compile(r"^离散备料计划维护：") ) plan_code_locator.wait_for(state="visible", timeout=30000) detail_count_text = detail_inner_frame.get_by_text( re.compile(r"^详细信息 \(\d+\)$") ).inner_text() detail_count = int(re.search(r"\((\d+)\)", detail_count_text).group(1)) status_text = detail_inner_frame.get_by_text( re.compile(r"^备料状态:.+$") ).inner_text() detail_status = re.search( r"备料状态:(.+)$", status_text.replace("\n", "") ).group(1) # 6. 执行清理逻辑 try: if detail_status == "审批通过": if detail_count > 0: # 更新总物料数统计 self.stats["total_materials"] += detail_count # --- 点击修改并等待状态切换 (保留原逻辑) --- detail_inner_frame.get_by_role("button", name="修改").click() # 关键判断：等待保存按钮出现，确认进入编辑模式 save_button_locator = detail_inner_frame.get_by_role( "button", name="保存" ) save_button_locator.wait_for(state="visible", timeout=30000) self._log("已进入编辑模式（保存按钮已就绪）") # --------------------------------------- detail_inner_frame.get_by_text("展开").first.click() child_form = detail_inner_frame.locator(".card-table-side-box") button_wrapper = child_form.locator(".button-wrapper") delete_row_btn = button_wrapper.get_by_role("button", name="删行") next_btn = button_wrapper.locator(".icon-jiantouyou") collapse_btn = button_wrapper.locator(".icon-celashouqi") last_row_number = None material_idx = 0 # 物料计数器 # page2.pause() # 调试用，正式运行时可删除 while True: material_idx += 1 self.stats["processed_materials"] += 1 # 稳定性检查：等待行号更新 current_row = self._get_input_value(child_form, r"^行号$") row_num_int = int(current_row) if current_row == last_row_number: time.sleep(0.5) material_code = self._get_input_value(child_form, r"^材料编码") material_name = self._get_input_value(child_form, r"^材料名称") pending_qty = self._get_input_value( child_form, r"^累计待发数量$" ) # 报告物料进度 self._report_material_progress( order_index, total_orders, material_idx, detail_count, order_id, material_name, "检查", ) if material_code in self.to_delete_set: self._log( f"发现匹配物料: {material_name} ({material_code})" ) if (not pending_qty) and not ( row_num_int >= 7000 and row_num_int < 8000 ): # 报告删除进度 self._report_material_progress( order_index, total_orders, material_idx, detail_count, order_id, material_name, "删除", ) # 记录删除前的行号 old_row_number = current_row delete_row_btn.click() self._log(f"✅ 已点击删行，等待删除完成...") # 等待行号变化（表示删除完成且新数据已加载） max_wait_time = 10 # 最大等待10秒 start_time = time.time() delete_success = False while time.time() - start_time < max_wait_time: try: new_row_number = self._get_input_value( child_form, r"^行号$" ) if new_row_number != old_row_number: self._log( f"✓ 删除完成，行号已从 {old_row_number} 变更为 {new_row_number}" ) delete_success = True break time.sleep(0.2) # 每200ms检查一次 except Exception as e: self._log(f"获取新行号时出错: {e}", "warn") time.sleep(0.2) else: self._log( f"⚠️ 等待删除完成超时（{max_wait_time}秒）", "warn", ) # 记录删除统计 if delete_success: self.stats["deleted_materials"].append( { "order_id": order_id, "material_code": material_code, "material_name": material_name, } ) continue elif row_num_int >= 7000 and row_num_int < 8000: reason = f"行号 {row_num_int} 在 7000-8000 范围内" self._report_material_progress( order_index, total_orders, material_idx, detail_count, order_id, material_name, "跳过", ) self._log(f"⚠️ {reason}，跳过删除", "warn") self.stats["skipped_materials"].append( { "order_id": order_id, "material_code": material_code, "material_name": material_name, "reason": reason, } ) elif pending_qty: reason = f"待发数量为 {pending_qty}" self._report_material_progress( order_index, total_orders, material_idx, detail_count, order_id, material_name, "跳过", ) self._log(f"⚠️ {reason}，跳过删除", "warn") self.stats["skipped_materials"].append( { "order_id": order_id, "material_code": material_code, "material_name": material_name, "reason": reason, } ) else: reason = "不满足删除条件" self._report_material_progress( order_index, total_orders, material_idx, detail_count, order_id, material_name, "跳过", ) self._log( f"⚠️ {reason}，跳过物料 {material_name} ({material_code})", "warn", ) self.stats["skipped_materials"].append( { "order_id": order_id, "material_code": material_code, "material_name": material_name, "reason": reason, } ) else: self._log( f"ℹ️ 物料 {material_name} ({material_code}) 不在删除列表中，不做处理" ) # 记录到 unmatched_materials self.stats["unmatched_materials"].append( { "order_id": order_id, "material_code": material_code, "material_name": material_name, } ) if self._is_button_enabled(next_btn): last_row_number = current_row next_btn.click() else: break collapse_btn.click() # 执行最终保存逻辑（如业务需要） if self.dryrun: self._log("[DRYRUN] 跳过保存操作") else: save_button_locator.click() self._log("已点击保存，等待保存完成...") save_button_locator.wait_for( state="hidden", timeout=60000 ) # 等待按钮消失，超时60秒 self._log("✅ 保存成功（保存按钮已消失）") else: self._log("订单无备料计划数据，无需处理") elif detail_status == "完成": self._log("订单已完成，无需处理") else: self._log(f"订单状态为 [{detail_status}]，不符合处理条件", "warn") finally: page2.close() def setup_query_interface(self, inner_frame): """初始化查询界面配置""" inner_frame.locator(".search-name-wrapper > .iconfont").click() inner_frame.get_by_text("订单号查询").click() inner_frame.get_by_role("tab", name="全部").click() # 填充每页显示条数（5000条测试值） input_el = inner_frame.locator("#rc_select_0") input_el.fill("5000") input_el.press("Enter") def clean(self, production_id_file): """执行完整清理流程""" # 初始化统计 self.stats["start_time"] = datetime.now() # 0. 预加载数据库数据 self.preload_data() with sync_playwright() as playwright: browser, context, page, main_frame = login( playwright=playwright, username=self.username, password=self.password, headless=self.headless, ignore_https_errors=True, ) self._log("=" * 30 + " 开始清理任务 " + "=" * 30) # 进入功能页面 main_frame.locator("i").first.click() with page.expect_popup() as page1_info: main_frame.get_by_title("离散生产订单维护", exact=True).first.click() page1 = page1_info.value # 定位主 Iframe work_main_frame = page1.locator("#forwardFrame").content_frame inner_frame = work_main_frame.locator("#mainiframe").content_frame inner_frame.locator("#hot-key-head_list").wait_for( state="visible", timeout=30000 ) self.setup_query_interface(inner_frame) order_ids = self.get_production_order_numbers(production_id_file) # 设置总订单数 self.stats["total_orders"] = len(order_ids) # 遍历处理 for index, order_id in enumerate(order_ids): self._log(f"进度: [{index+1}/{len(order_ids)}] 处理单号: {order_id}") try: self.process_order( inner_frame, order_id, index, page1, len(order_ids) ) self.stats["processed_orders"] += 1 except Exception as e: self._log(f"处理单号 {order_id} 时发生异常: {e}", "error") self.stats["errors"].append( {"order_id": order_id, "error_message": str(e)} ) continue # 单个失败不影响整体执行 # 登出清理 logout(work_main_frame, verbose=self.verbose) context.close() browser.close() self._log("=" * 30 + " 任务全部完成 " + "=" * 30) # 记录结束时间 self.stats["end_time"] = datetime.now() def generate_report(self) -> str: """生成 Markdown 格式的执行报告 Returns: Markdown 格式的报告字符串 """ report_lines = [] # 标题 report_lines.append("# 执行报告") report_lines.append("") # 概述 report_lines.append("## 概述") report_lines.append("") start_time = self.stats.get("start_time") end_time = self.stats.get("end_time") duration = None if start_time and end_time: duration = end_time - start_time report_lines.append( f"- 开始时间: {start_time.strftime('%Y-%m-%d %H:%M:%S')}" ) report_lines.append(f"- 结束时间: {end_time.strftime('%Y-%m-%d %H:%M:%S')}") report_lines.append(f"- 执行时长: {duration}") report_lines.append( f"- 处理订单: {self.stats['processed_orders']}/{self.stats['total_orders']} 个" ) report_lines.append( f"- 处理物料: {self.stats['processed_materials']}/{self.stats['total_materials']} 条" ) report_lines.append(f"- 删除物料: {len(self.stats['deleted_materials'])} 条") report_lines.append(f"- 跳过物料: {len(self.stats['skipped_materials'])} 条") report_lines.append( f"- 未处理物料: {len(self.stats['unmatched_materials'])} 条" ) report_lines.append(f"- 错误数量: {len(self.stats['errors'])} 个") report_lines.append( f"- 执行模式: {'预览模式 (dryrun)' if self.dryrun else '正常执行'}" ) if self.manager_names: report_lines.append(f"- 负责人: {', '.join(self.manager_names)}") report_lines.append("") # 删除明细 if self.stats["deleted_materials"]: report_lines.append("## 删除明细") report_lines.append("") report_lines.append("| 订单号 | 物料编码 | 物料名称 |") report_lines.append("|--------|----------|----------|") for item in self.stats["deleted_materials"]: report_lines.append( f"| {item['order_id']} | {item['material_code']} | {item['material_name']} |" ) report_lines.append("") # 跳过明细 if self.stats["skipped_materials"]: report_lines.append("## 跳过明细") report_lines.append("") report_lines.append("| 订单号 | 物料编码 | 物料名称 | 跳过原因 |") report_lines.append("|--------|----------|----------|----------|") for item in self.stats["skipped_materials"]: report_lines.append( f"| {item['order_id']} | {item['material_code']} | {item['material_name']} | {item['reason']} |" ) report_lines.append("") # 未处理物料明细 if self.stats["unmatched_materials"]: report_lines.append("## 未处理物料") report_lines.append("") report_lines.append("| 订单号 | 物料编码 | 物料名称 |") report_lines.append("|--------|----------|----------|") for item in self.stats["unmatched_materials"]: report_lines.append( f"| {item['order_id']} | {item['material_code']} | {item['material_name']} |" ) report_lines.append("") # 错误明细 if self.stats["errors"]: report_lines.append("## 错误明细") report_lines.append("") for item in self.stats["errors"]: report_lines.append(f"### 订单号: `{item['order_id']}`") report_lines.append("") report_lines.append("```") report_lines.append(item['error_message']) report_lines.append("```") report_lines.append("") return "\n".join(report_lines) def main(): # 路径配置 base_dir = os.path.dirname(__file__) id_file = os.path.join(base_dir, "productionID.txt") cleaner = DiscreteMaterialPlanCleaner( username="BLDpengqiangqiang", password="your_password_here", manager_names="彭羽", # 支持字符串、列表或 None headless=False, ) cleaner.clean(id_file) input("执行完毕，按回车键退出程序...") if __name__ == "__main__": main()