2023年12月15日发(作者：)

2 Spark-1 865 Spark 14 Spark-2 867 Spark 23 Spark-10 1074 Spark 105 Spark-11 1014 Spark 111 admin 822 admin 00 hdfs 863 hdfs 0In [30]: new_df = _values(["first","second"],ascending=[True, True])[["id", "hostname"]].reset_index(drop=True)In [31]: new_dfOut[31]:id hostname0 865 Spark-11 867 Spark-22 1074 Spark-103 1014 Spark-114 822 admin5 863 hdfsIn [32]: new__dict(orient="records")Out[32]:[{'id': 865, 'hostname': 'Spark-1'},{'id': 867, 'hostname': 'Spark-2'},{'id': 1074, 'hostname': 'Spark-10'},{'id': 1014, 'hostname': 'Spark-11'},{'id': 822, 'hostname': 'admin'},{'id': 863, 'hostname': 'hdfs'}]修改前的答案：In [3]: aOut[3]:[{'id': 865, 'hostname': 'Spark-1'},{'id': 1074, 'hostname': 'Spark-10'},{'id': 1014, 'hostname': 'Spark-11'},{'id': 867, 'hostname': 'Spark-2'}]In [4]: (key = lambda x:int(x["hostname"].split("-")[1]))In [5]: a

Out[5]:[{'id': 865, 'hostname': 'Spark-1'},{'id': 867, 'hostname': 'Spark-2'},{'id': 1074, 'hostname': 'Spark-10'},{'id': 1014, 'hostname': 'Spark-11'}]