'commit'

Apps/AiTeJiYiChong/Crawler.py

@@ -83,9 +83,9 @@ async def get_station_list(d, service, max_stations_count=MAX_STATIONS_COUNT):
     total_new_processed_count = 0
     scroll_count = 0
 
-    while total_encountered_count < max_stations_count:
+    while total_new_processed_count < max_stations_count:
         scroll_count += 1
-        logger.info(f"正在处理第 {scroll_count} 次滚动 (已遇到: {total_encountered_count}/{max_stations_count}, 新采集: {total_new_processed_count})...")
+        logger.info(f"正在处理第 {scroll_count} 次滚动 (已遇到: {total_encountered_count}, 新采集: {total_new_processed_count}/{max_stations_count})...")
         
         # 1. 拍摄截图
         image_uuid = str(uuid.uuid4())
@@ -121,7 +121,7 @@ async def get_station_list(d, service, max_stations_count=MAX_STATIONS_COUNT):
         logger.info("正在调用 VL 模型识别场站信息 (整页识别)...")
         # 优先使用带绿框的图片进行识别
         img_to_process = vl_img_path if os.path.exists(vl_img_path) else screenshot_path
-        stations = await service.process_station_list_vl(img_to_process, json_metadata, device_info=device_info, max_count=max_stations_count - total_encountered_count)
+        stations = await service.process_station_list_vl(img_to_process, json_metadata, device_info=device_info, max_count=max_stations_count - total_new_processed_count)
         logger.info(f"本页识别到 {len(stations)} 个有效场站")
 
         ad_top_y_norm = 1.0 - BOTTOM_SAFE_EXCLUDE_RATIO
@@ -130,8 +130,8 @@ async def get_station_list(d, service, max_stations_count=MAX_STATIONS_COUNT):
         new_stations_processed_in_page = 0
         if json_metadata.get("cards") and stations:
             for card_idx, card in enumerate(json_metadata["cards"]):
-                # 检查是否已达到最大采集数量
-                if total_encountered_count >= max_stations_count:
+                # 检查是否已达到最大采集数量（按新采集的场站数量限制）
+                if total_new_processed_count >= max_stations_count:
                     break
 
                 # 检查索引是否越界 (VL 模型可能返回的数组长度不一致)
@@ -158,7 +158,7 @@ async def get_station_list(d, service, max_stations_count=MAX_STATIONS_COUNT):
 
                     # 正常处理新场站
                     click_x, click_y = card["click_point"]
-                    current_idx = total_encountered_count
+                    current_idx = total_new_processed_count + 1
                     remaining = max_stations_count - current_idx
                     logger.info(f"--- [进度: {current_idx}/{max_stations_count}, 剩余: {remaining}] 发现新场站 '{station_name}'，开始处理... ---")
                     new_stations_processed_in_page += 1

Apps/XinDianTu/Crawler.py

@@ -157,9 +157,9 @@ async def get_station_list(d, service, uploader, max_stations_count=MAX_STATIONS
     total_new_processed_count = 0
     scroll_count = 0
 
-    while total_encountered_count < max_stations_count:
+    while total_new_processed_count < max_stations_count:
         scroll_count += 1
-        logger.info(f"正在处理第 {scroll_count} 次滚动 (已遇到: {total_encountered_count}/{max_stations_count}, 新采集: {total_new_processed_count})...")
+        logger.info(f"正在处理第 {scroll_count} 次滚动 (已遇到: {total_encountered_count}, 新采集: {total_new_processed_count}/{max_stations_count})...")
 
         # 1. 拍摄截图
         image_uuid = str(uuid.uuid4())
@@ -204,7 +204,7 @@ async def get_station_list(d, service, uploader, max_stations_count=MAX_STATIONS
         
         # 4. 【优化】直接使用本地路径调用 VL 模型识别，避免等待上传
         logger.info("正在调用 VL 模型识别场站信息...")
-        stations = await service.process_station_list_vl(vl_img_path, json_metadata, device_info=device_info, max_count=max_stations_count - total_encountered_count)
+        stations = await service.process_station_list_vl(vl_img_path, json_metadata, device_info=device_info, max_count=max_stations_count - total_new_processed_count)
         logger.info(f"本页识别到 {len(stations)} 个场站")
 
         # 5. 匹配几何卡片与 VL 识别结果 (XinDianTu 的 parse_vl_image 已经按顺序返回了)
@@ -212,8 +212,8 @@ async def get_station_list(d, service, uploader, max_stations_count=MAX_STATIONS
         new_stations_processed_in_page = 0
         if json_metadata.get("cards") and stations:
             for idx, card in enumerate(json_metadata["cards"]):
-                # 检查是否已达到最大采集数量
-                if total_encountered_count >= max_stations_count:
+                # 检查是否已达到最大采集数量（按新采集的场站数量限制）
+                if total_new_processed_count >= max_stations_count:
                     break
 
                 # 检查索引是否越界 (VL 模型可能返回的数组长度不一致)
@@ -240,7 +240,7 @@ async def get_station_list(d, service, uploader, max_stations_count=MAX_STATIONS
                         continue
 
                     # 正常处理新场站
-                    current_idx = total_encountered_count
+                    current_idx = total_new_processed_count + 1
                     remaining = max_stations_count - current_idx
                     logger.info(f"--- [进度: {current_idx}/{max_stations_count}, 剩余: {remaining}] 发现新场站 '{station_name}'，开始处理... ---")
                     new_stations_processed_in_page += 1