'commit'

Apps/TeLaiDian/Crawler.py

@@ -10,6 +10,7 @@ from Apps.TeLaiDian.Kit import (
     setup_logger, read_image, save_image, detect_warm_popup_xczs_cv
 )
 from Apps.TeLaiDian.ReadImageKit import ReadImageKit
+from Apps.TeLaiDian.FirstPageKit import run_ocr_rect
 from Apps.TeLaiDian.Service import TeLaiDianService
 from Apps.TeLaiDian.Config.Setting import (
     SCROLL_DISTANCE_RATIO, WAIT_AFTER_SCROLL, MAX_STATIONS_COUNT,
@@ -157,10 +158,8 @@ class TeLaiDianCrawler(BaseCrawler):
         last_md5 = None
         
         while processed_count < max_to_crawl:
-            # 1. 截图并分析
             screenshot_path = take_screenshot(d, f"tld_list_{int(time.time())}.jpg")
             
-            # 检测是否滚动到底部
             curr_md5 = get_image_content_md5(screenshot_path, top_ratio=SAFE_EXCLUDE_RATIO, bottom_ratio=BOTTOM_SAFE_EXCLUDE_RATIO)
             if last_md5 == curr_md5:
                 logger.info("内容无变化，判定已到底部")
@@ -168,9 +167,8 @@ class TeLaiDianCrawler(BaseCrawler):
                 break
             last_md5 = curr_md5
             
-            stations = await self.read_image_kit.analyze_station_list(screenshot_path)
-            if not stations:
-                # 检查是否意外退出了小程序
+            stations_page = await run_ocr_rect(screenshot_path)
+            if not stations_page:
                 is_wrong_page = await self.check_wrong_page(d, screenshot_path)
                 if is_wrong_page:
                     logger.error("检测到已退出详情列表页（可能回到了搜索页），尝试重新进入...")
@@ -183,21 +181,23 @@ class TeLaiDianCrawler(BaseCrawler):
                 await asyncio.sleep(WAIT_AFTER_SCROLL)
                 continue
             
-            for station in stations:
+            for st in stations_page:
                 if processed_count >= max_to_crawl:
                     break
                 
-                name = station.get("name")
-                status_text = station.get("status_text") or ""
-                if station.get("is_closed") is True or ("歇业中" in status_text):
-                    logger.info(f"列表标记为歇业中，跳过场站: {name}")
-                    continue
-                point = station.get("point")
+                name = st.get("station_name") or st.get("name")
+                rect = st.get("rect") or []
+                click_point = st.get("click_point") or st.get("click") or []
+                distance = st.get("distance_text") or st.get("distance")
+                piles = st.get("busy_list") or []
                 
-                if not name or not point:
+                if not name or not isinstance(rect, (list, tuple)) or len(rect) < 4:
+                    continue
+                if not isinstance(click_point, (list, tuple)) or len(click_point) < 2:
                     continue
                 
-                # [优化] 使用 Redis 进行跨运行去重
+                point = [int(click_point[0]), int(click_point[1])]
+                
                 cleaned_name = clean_station_name(name)
                 redis_key = f"crawled:tld:{cleaned_name}"
                 if await self.redis_kit.get_data(redis_key):
@@ -206,7 +206,7 @@ class TeLaiDianCrawler(BaseCrawler):
                 
                 current_idx = processed_count + 1
                 remaining = max_to_crawl - current_idx
-                logger.info(f"--- [进度: {current_idx}/{max_to_crawl}, 剩余: {remaining}] 处理场站: {name} (坐标: {point}, 距离: {station.get('distance')}) ---")
+                logger.info(f"--- [进度: {current_idx}/{max_to_crawl}, 剩余: {remaining}] 处理场站: {name} (坐标: {point}, 距离: {distance}) ---")
                 
                 # 点击进入详情
                 d.click(point[0], point[1])