This commit is contained in:
HuangHai
2026-01-13 19:15:29 +08:00
parent 62495ab39d
commit 11ae4b5abf

View File

@@ -232,8 +232,9 @@ async def get_station_list(d, service, uploader, max_stations_count=MAX_STATIONS
continue
# 正常处理新场站
logger.info(f">>> 发现新场站 '{station_name}',开始处理...")
logger.info(f">>> 发现新场站 '{station_name}',开始处理... ({total_processed_count + 1}/{max_stations_count})")
new_stations_processed += 1
total_processed_count += 1
click_x, click_y = card["click_point"]
logger.info(f"准备处理场站: {station_name}, 点击坐标: ({click_x}, {click_y})")
@@ -391,6 +392,13 @@ async def get_station_list(d, service, uploader, max_stations_count=MAX_STATIONS
cleaned = Kit.clean_station_name(station_name)
await redis_kit.set_data(f"crawled:xdt:{cleaned}", "1", expire=REDIS_STATION_EXPIRE)
# 检查是否已达到最大采集数量
if total_processed_count >= max_stations_count:
logger.info(f"已达到目标采集数量 {max_stations_count},准备结束采集。")
break
if total_processed_count >= max_stations_count:
break
# 清理已完成的后台任务
done_tasks = [t for t in background_tasks if t.done()]
for t in done_tasks:
@@ -490,7 +498,7 @@ async def main(service=None, do_cleanup=True):
# await service.cleanup_old_data()
# 获取场站列表
stations = await get_station_list(d, service, uploader, max_scrolls=MAX_SCROLLS)
stations = await get_station_list(d, service, uploader, max_stations_count=MAX_STATIONS_COUNT)
if stations:
logger.info("场站列表采集完成。")