Dockerfile 部署

browser_login/fetch_issue_receipt_incremental.py

@@ -98,48 +98,60 @@ def fetch_issue_receipt_incremental():
     page = get_page(port=9222)
     
     try:
-        log("INFO", f"正在回到主页起点: {HOME_URL}")
+        log("INFO", f"正在直接跳转到发料单明细页面...")
-        page.get(HOME_URL)
+        page.get("https://yunmes.tftykj.cn/WorkOrdersQuery")
         page.wait.load_start()
         time.sleep(2)
         
-        menus = [
-            ("第一层: 业务统计报表", 'xpath://*[@id="app"]/div/div[1]/div[1]/div[2]/div/div[1]/div/div[10]/div/p'),
-            ("第二层: 生产业务报表(推测)", 'xpath:/html/body/div[7]/div/div[1]/div/div[9]/div/p'),
-            ("第三层: 发料单报表", 'xpath:/html/body/div[8]/div/div[1]/div/div[6]/div/p')
-        ]
-        
-        log("INFO", "模拟点击左侧导航菜单...")
-        for name, xpath in menus:
-            ele = page.ele(xpath, timeout=5)
-            if ele:
-                try: ele.click()
-                except: page.run_js("arguments[0].click();", ele)
-                time.sleep(1.5)
-            else:
-                log("ERR", f"找不到菜单元素: {name}")
-                return
-                
-        log("OK", "✅ 成功点开发料单报表界面！")
-        
-        # 隐藏菜单
-        blank_xpath = 'xpath://*[@id="app"]/div/div[1]/div[2]/div[1]/div[2]/div[2]/div/div[1]/div'
-        blank_ele = page.ele(blank_xpath, timeout=3)
-        if blank_ele:
-            try: blank_ele.click()
-            except: page.run_js("arguments[0].click();", blank_ele)
-            time.sleep(0.5)
-        
         log("INFO", f"开启底层数据拦截网: {API_TARGET}")
         page.listen.start(API_TARGET)
         
+        # 为了能够获取当月的数据，强制设置时间为当月第一天到最后一天，并清理其他条件
+        import datetime, calendar
+        now = datetime.datetime.now()
+        first_day = datetime.date(now.year, now.month, 1).strftime('%Y-%m-%d')
+        last_day = datetime.date(now.year, now.month, calendar.monthrange(now.year, now.month)[1]).strftime('%Y-%m-%d')
+        
+        log("INFO", f"正在自动设置查询时间范围: {first_day} 至 {last_day}")
+        
+        page.run_js(f"""
+            try {{
+                var dates = document.querySelectorAll('.datebox-f, .datetimebox-f, .el-date-editor input');
+                if (dates.length >= 2) {{
+                    dates[0].value = '{first_day}';
+                    dates[1].value = '{last_day}';
+                    dates[0].dispatchEvent(new Event('input', {{ bubbles: true }}));
+                    dates[0].dispatchEvent(new Event('change', {{ bubbles: true }}));
+                    dates[1].dispatchEvent(new Event('input', {{ bubbles: true }}));
+                    dates[1].dispatchEvent(new Event('change', {{ bubbles: true }}));
+                }}
+            }} catch(e) {{ console.log(e); }}
+        """)
+        time.sleep(1)
+        
+        # 因为是直接打开 URL，数据通常不会自动加载，所以尝试点击查询按钮
         packet = page.listen.wait(timeout=10)
         if not packet:
-            query_btn_xpath = 'xpath://*[@id="app"]/div/div[1]/div[2]/div[2]/div[1]/div[1]/div/button[1]/span'
+            # 兼容多种查询按钮的查找方式
-            query_btn = page.ele(query_btn_xpath, timeout=3)
+            log("INFO", "尝试寻找并点击页面上的【查询】按钮...")
+            query_btn = page.ele('text=查询', timeout=3)
+            if not query_btn:
+                query_btn = page.ele('xpath://button[contains(., "查询")]', timeout=3)
+            
             if query_btn:
                 try: query_btn.click()
                 except: page.run_js("arguments[0].click();", query_btn)
+            else:
+                log("WARN", "常规选择器找不到查询按钮，尝试使用全局 JS 强行寻找...")
+                page.run_js("""
+                    var btns = document.querySelectorAll('button, a, .l-btn, .el-button');
+                    for(var i=0; i<btns.length; i++) {
+                        if(btns[i].innerText && btns[i].innerText.indexOf('查询') !== -1) {
+                            btns[i].click();
+                            break;
+                        }
+                    }
+                """)
             packet = page.listen.wait(timeout=15)
                 
         if not packet:
@@ -191,15 +203,8 @@ def fetch_issue_receipt_incremental():
                     log("INFO", f"🔍 正在检查本页第一条数据: 发料单 {first_item.get('workOrdersNumber')} 行号 {first_item.get('lineNumber')} 物料 {first_item.get('materialCode')}")
                     
                 for raw_item in items:
-                    # 1. 检查是否存在
+                    # 1. 如果不存在，提取并插入
-                    if item_exists(cursor, raw_item):
+                    if not item_exists(cursor, raw_item):
-                        # 发料单的新数据都在最前面。当我们遇到一条已经在数据库里的数据时，
-                        # 说明这之前的数据都是新的，这之后的数据肯定都抓过了，直接停止。
-                        log("INFO", f"🛑 在第 {current_page} 页发现本地已存在的记录 (发料单: {raw_item.get('workOrdersNumber')} 行号: {raw_item.get('lineNumber')} 物料: {raw_item.get('materialCode')})，增量扫描结束！")
-                        should_stop = True
-                        break
-                    
-                    # 2. 如果不存在，提取并插入
                         item = _extract_fields(raw_item)
                         
                         cursor.execute('''
@@ -232,12 +237,15 @@ def fetch_issue_receipt_incremental():
                 conn.commit()
                 log("OK", f"第 {current_page} 页处理完毕，成功插入 {inserted_this_page} 条新数据。")
                 
-                if should_stop:
-                    break
             else:
                 log("ERR", f"第 {current_page} 页数据结构异常，中止。")
                 break
             
+            # 如果当页没有新数据插入，说明已经追上了旧数据，停止抓取
+            if inserted_this_page == 0 and current_page > 1:
+                log("OK", "🎉 本页未发现任何新数据，说明增量部分已全部抓取完毕，停止翻页！")
+                break
+                
             # 如果没遇到旧数据，继续点击下一页
             delay = random.uniform(1.5, 3.5)
             log("INFO", f"⏳ 停顿 {delay:.2f} 秒后点击下一页...")
@@ -245,6 +253,9 @@ def fetch_issue_receipt_incremental():
             
             next_btn = None
             for _ in range(3):
+                # 优先使用 pagination-next，如果不行再尝试 btn-next
+                next_btn = page.ele('xpath://*[contains(@class, "pagination-next")]', timeout=3)
+                if not next_btn:
                     next_btn = page.ele('xpath://button[contains(@class, "btn-next")]', timeout=3)
                 if next_btn:
                     break
@@ -259,12 +270,24 @@ def fetch_issue_receipt_incremental():
                 aria_disabled = next_btn.attr("aria-disabled")
                 is_disabled_attr = next_btn.attr("disabled") is not None
                 
-                if "disabled" in class_str or is_disabled_attr or aria_disabled == "true":
+                # 检查父元素 <li> 是否被禁用
+                parent_class_str = ""
+                try:
+                    parent_ele = next_btn.parent()
+                    parent_class_str = str(parent_ele.attr("class"))
+                except:
+                    pass
+                
+                if "disabled" in class_str or "disabled" in parent_class_str or is_disabled_attr or aria_disabled == "true":
                     log("OK", "🏁 下一页按钮已被禁用，已经翻到最后一页。")
                     break
                     
-                try: next_btn.click()
+                try: 
-                except: page.run_js("arguments[0].click();", next_btn)
+                    # 优先使用 JS 点击防止遮挡
+                    page.run_js("arguments[0].click();", next_btn)
+                except Exception as e: 
+                    log("ERR", f"JS 点击失败: {e}，尝试普通点击...")
+                    next_btn.click()
                 
                 packet = page.listen.wait(timeout=15)
                 if not packet:

web_ui/app.py

@@ -435,6 +435,44 @@ def sync_receipts():
     except Exception as e:
         return jsonify({"success": False, "message": f"系统错误: {str(e)}"}), 500
 
+@app.route('/api/sync_issue_receipts', methods=['POST'])
+def sync_issue_receipts():
+    """触发后台运行发料单明细增量抓取脚本"""
+    global is_browser_busy
+    import sys
+    
+    if is_browser_busy:
+        return jsonify({
+            "success": False, 
+            "message": f"系统忙碌：当前正在执行 '{current_task_name}'，请稍后再试。"
+        }), 409
+        
+    if str(BROWSER_LOGIN_DIR) not in sys.path:
+        sys.path.insert(0, str(BROWSER_LOGIN_DIR))
+        
+    def run_issue_receipt_sync():
+        set_browser_busy("手动发料单明细增量同步")
+        try:
+            from fetch_issue_receipt_incremental import fetch_issue_receipt_incremental
+            fetch_issue_receipt_incremental()
+        except Exception as e:
+            print(f"发料单明细增量同步失败: {e}")
+        finally:
+            release_browser()
+            
+    try:
+        threading.Thread(target=run_issue_receipt_sync, daemon=True).start()
+            
+        return jsonify({
+            "success": True, 
+            "message": "发料单明细增量抓取任务已在后台启动！请观察运行日志。",
+            "logs": "任务已在后台运行..."
+        })
+    except ImportError:
+        return jsonify({"success": False, "message": "找不到发料单增量抓取脚本或导入失败"}), 404
+    except Exception as e:
+        return jsonify({"success": False, "message": f"系统错误: {str(e)}"}), 500
+
 @app.route('/api/sync_work_orders', methods=['POST'])
 def sync_work_orders():
     """触发后台运行生产工单增量抓取脚本"""