MaxCompute中这个报错怎么解决呢？ _问答-阿里云开发者社区

问题一：Users/cero/Softwares/anaconda3/envs/odpsUDF/bin/python /Users/cero/Softwares/anaconda3/envs/odpsUDF/bin/pyou -I sys.stdin udaf_waste_point.udaf_waste_point Traceback (most recent call last): File "/Users/cero/Softwares/anaconda3/envs/odpsUDF/bin/pyou", line 51, in main() File "/Users/cero/Softwares/anaconda3/envs/odpsUDF/bin/pyou", line 46, in main udf_runner = runners.get_default_runner(clz, delim, null_indicator, stdin) File "/Users/cero/Softwares/anaconda3/envs/odpsUDF/lib/python3.7/site-packages/odps/udf/tools/runners.py", line 35, in get_default_runner in_types, out_types = parse_proto(proto) File "/Users/cero/Softwares/anaconda3/envs/odpsUDF/lib/python3.7/site-packages/odps/udf/tools/runners.py", line 93, in parse_proto return _get_in_types(tokens[0].strip()), _get_types(tokens[1].strip()) File "/Users/cero/Softwares/anaconda3/envs/odpsUDF/lib/python3.7/site-packages/odps/udf/tools/runners.py", line 68, in _get_types raise Exception('type not in '+ ','.join(_allowed_data_types)) Exception: type not in bigint,string,datetime,double,boolean

Process finished with exit code 1
MaxCompute中这个报错怎么解决呢？
问题二：执行这个语句绑定的MaxCompute引擎和你本地测试udf绑定的引擎是一样的，本地测试还是报相同错误。# coding:utf-8 from odps.udf import annotate from odps.udf import BaseUDAF import numpy as np import pandas as pd

@annotate('*->array') class udaf_waste_point(BaseUDAF): def new_buffer(self): """ 初始化 [第一列未预约点位数, 第二列未预约点位数...] :return: """ return []

def iterate(self, buffer, *args):
    """
    每行中未预约点位标记为1，然后按列求和
    :param buffer:
    :param args:原始一行点位
    :return:
    """
    # 初始buffer为空列表，将args中的未预约点位标记为1，直接装入buffer
    if len(buffer) == 0:
        for i in range(len(args)):
            buffer.append(1 if i == 'cano' else 0)
        pass
    # buffer非空，将args中的未预约点位标记为1，遍历buffer求和
    for i in range(len(args)):
        buffer[i] += 1 if args[i] == 'cano' else 0

def merge(self, buffer, pbuffer):
    # 初始buffer为空列表，将pbuffer直接装入buffer
    if len(buffer) == 0:
        buffer.extend(pbuffer)
        pass
    # buffer非空，将pbuffer，遍历buffer求和
    for i in range(len(pbuffer)):
        buffer[i] += pbuffer[i]

def terminate(self, buffer):
    waste_point_df = self.to_waste_point_df(buffer)
    changed_df = self.change_df(waste_point_df)
    formated_df = self.format_df(changed_df)
    waste_point_array = formated_df.T.loc['cnt', :].to_list()
    print(waste_point_array)
    # return waste_point_array[0]
    return waste_point_array

@staticmethod
def to_waste_point_df(buffer):
    """
    list: [3, 2, 1, 0] =>  dataframe: [[1, 1, 1, 0],
                                       [1, 1, 0, 0],
                                       [1, 0, 0, 0]]
    :param buffer:
    :return:
    """
    data = [[1 if buffer[j] > i else 0 for j in range(len(buffer))] for i in range(max(buffer))]
    waste_point_df = pd.DataFrame(data=data)
    return waste_point_df

@staticmethod
def change_df(waste_point_df):
    """
    遍历废点表，前后状态不一致时打标
    dataframe: [[1, 1, 1, 0],  => dataframe:[[b1, 1],
                [1, 1, 0, 0],                [b1, 1],
                [1, 0, 0, 0]]                [b1, 1],
                                             [a1, 0],
                                             [b2, 1],
                                             [b2, 1],
                                             [a2, 0],
                                             [a2, 0],
                                             [b3, 1],
                                             [a4, 0],
                                             [a4, 0],
                                             [a4, 0]]
    :param waste_point_df:
    :return:
    """
    # 结果集
    res = []
    # 起始计数
    start = 0
    # 标记
    zero_flag = 'a'
    one_flag = 'b'
    for index, row in waste_point_df.iterrows():
        # 起始状态
        pre_value = 0
        for item in row:
            if item != pre_value:
                start += 1
            flag = zero_flag if item == 0 else one_flag
            tag = '{}{}'.format(flag, start)
            res.append([tag, item])
            pre_value = item
    new_df = pd.DataFrame(res, columns=['tag', 'value'])
    return new_df

@staticmethod
def format_df(res_df):
    """
    打标后的df按废点数量格式化输出1~10废点数量
    dataframe:[[b1, 1],   => dataframe:[[1], [1], [1], [0], [0], [0], [0], [0], [0], [0]]
               [b1, 1],
               [b1, 1],
               [a1, 0],
               [b2, 1],
               [b2, 1],
               [a2, 0],
               [a2, 0],
               [b3, 1],
               [a4, 0],
               [a4, 0],
               [a4, 0]]
    :param res_df:
    :return:
    """
    df = res_df.groupby('tag').agg({'value': sum}).reset_index().groupby('value').agg(
        {'tag': np.size}).reset_index()
    df = df.loc[(df['value'] > 0) & (df['value'] < 10)]
    data = []
    # 取1~10废点数量
    for i in range(10):
        tmp_list = df.loc[(df['value'] == i + 1), 'tag'].to_list()
        if tmp_list:
            data.append([tmp_list[0]])
        else:
            data.append([0])

    formated_df = pd.DataFrame(data=data, columns=['cnt'])
    return formated_df                                                      测试数据是一个变长一维数组                                              问题三：我是udaf不需要evaluate，在返回值类型bigint已经测试通过了  ，帮忙看下？ select  store_id
    ,udaf_waste_point(8_1_point, 8_2_point , 8_3_point, 9_1_point, 9_2_point , 9_3_point, 10_1_point, 10_2_point , 10_3_point, 11_1_point, 11_2_point , 11_3_point, 12_1_point, 12_2_point , 12_3_point, 13_1_point, 13_2_point , 13_3_point, 14_1_point, 14_2_point , 14_3_point, 15_1_point, 15_2_point , 15_3_point, 16_1_point, 16_2_point , 16_3_point, 17_1_point, 17_2_point , 17_3_point, 18_1_point, 18_2_point , 18_3_point, 19_1_point, 19_2_point , 19_3_point, 20_1_point, 20_2_point , 20_3_point, 21_1_point, 21_2_point , 21_3_point, 22_1_point, 22_2_point , 22_3_point)

from tbl where ds = '20230509' group by store_id ddl CREATE TABLE IF NOT EXISTS dwd_participant_store_schedule_table_di ( store_id STRING COMMENT '门店id', schedule_date STRING COMMENT '排单日期', max_index BIGINT COMMENT '最大点位数', start_appointment_time STRING COMMENT '开始预约时间', stop_appointment_time STRING COMMENT '结束预约时间', service_interval_minute BIGINT COMMENT '服务间隔-分钟', service_cnt_per_period BIGINT COMMENT '每个时间段服务数', row_no STRING COMMENT '行号', 8_1_point STRING COMMENT '8点第1个点位', 8_2_point STRING COMMENT '8点第2个点位', 8_3_point STRING COMMENT '8点第3个点位', 8_4_point STRING COMMENT '8点第4个点位', 8_5_point STRING COMMENT '8点第5个点位', 8_6_point STRING COMMENT '8点第6个点位', 9_1_point STRING COMMENT '9点第1个点位', 9_2_point STRING COMMENT '9点第2个点位', 9_3_point STRING COMMENT '9点第3个点位', 9_4_point STRING COMMENT '9点第4个点位', 9_5_point STRING COMMENT '9点第5个点位', 9_6_point STRING COMMENT '9点第6个点位', 10_1_point STRING COMMENT '10点第1个点位', 10_2_point STRING COMMENT '10点第2个点位', 10_3_point STRING COMMENT '10点第3个点位', 10_4_point STRING COMMENT '10点第4个点位', 10_5_point STRING COMMENT '10点第5个点位', 10_6_point STRING COMMENT '10点第6个点位', 11_1_point STRING COMMENT '11点第1个点位', 11_2_point STRING COMMENT '11点第2个点位', 11_3_point STRING COMMENT '11点第3个点位', 11_4_point STRING COMMENT '11点第4个点位', 11_5_point STRING COMMENT '11点第5个点位', 11_6_point STRING COMMENT '11点第6个点位', 12_1_point STRING COMMENT '12点第1个点位', 12_2_point STRING COMMENT '12点第2个点位', 12_3_point STRING COMMENT '12点第3个点位', 12_4_point STRING COMMENT '12点第4个点位', 12_5_point STRING COMMENT '12点第5个点位', 12_6_point STRING COMMENT '12点第6个点位', 13_1_point STRING COMMENT '13点第1个点位', 13_2_point STRING COMMENT '13点第2个点位', 13_3_point STRING COMMENT '13点第3个点位', 13_4_point STRING COMMENT '13点第4个点位', 13_5_point STRING COMMENT '13点第5个点位', 13_6_point STRING COMMENT '13点第6个点位', 14_1_point STRING COMMENT '14点第1个点位', 14_2_point STRING COMMENT '14点第2个点位', 14_3_point STRING COMMENT '14点第3个点位', 14_4_point STRING COMMENT '14点第4个点位', 14_5_point STRING COMMENT '14点第5个点位', 14_6_point STRING COMMENT '14点第6个点位', 15_1_point STRING COMMENT '15点第1个点位', 15_2_point STRING COMMENT '15点第2个点位', 15_3_point STRING COMMENT '15点第3个点位', 15_4_point STRING COMMENT '15点第4个点位', 15_5_point STRING COMMENT '15点第5个点位', 15_6_point STRING COMMENT '15点第6个点位', 16_1_point STRING COMMENT '16点第1个点位', 16_2_point STRING COMMENT '16点第2个点位', 16_3_point STRING COMMENT '16点第3个点位', 16_4_point STRING COMMENT '16点第4个点位', 16_5_point STRING COMMENT '16点第5个点位', 16_6_point STRING COMMENT '16点第6个点位', 17_1_point STRING COMMENT '17点第1个点位', 17_2_point STRING COMMENT '17点第2个点位', 17_3_point STRING COMMENT '17点第3个点位', 17_4_point STRING COMMENT '17点第4个点位', 17_5_point STRING COMMENT '17点第5个点位', 17_6_point STRING COMMENT '17点第6个点位', 18_1_point STRING COMMENT '18点第1个点位', 18_2_point STRING COMMENT '18点第2个点位', 18_3_point STRING COMMENT '18点第3个点位', 18_4_point STRING COMMENT '18点第4个点位', 18_5_point STRING COMMENT '18点第5个点位', 18_6_point STRING COMMENT '18点第6个点位', 19_1_point STRING COMMENT '19点第1个点位', 19_2_point STRING COMMENT '19点第2个点位', 19_3_point STRING COMMENT '19点第3个点位', 19_4_point STRING COMMENT '19点第4个点位', 19_5_point STRING COMMENT '19点第5个点位', 19_6_point STRING COMMENT '19点第6个点位', 20_1_point STRING COMMENT '20点第1个点位', 20_2_point STRING COMMENT '20点第2个点位', 20_3_point STRING COMMENT '20点第3个点位', 20_4_point STRING COMMENT '20点第4个点位', 20_5_point STRING COMMENT '20点第5个点位', 20_6_point STRING COMMENT '20点第6个点位', 21_1_point STRING COMMENT '21点第1个点位', 21_2_point STRING COMMENT '21点第2个点位', 21_3_point STRING COMMENT '21点第3个点位', 21_4_point STRING COMMENT '21点第4个点位', 21_5_point STRING COMMENT '21点第5个点位', 21_6_point STRING COMMENT '21点第6个点位', 22_1_point STRING COMMENT '22点第1个点位', 22_2_point STRING COMMENT '22点第2个点位', 22_3_point STRING COMMENT '22点第3个点位', 22_4_point STRING COMMENT '22点第4个点位', 22_5_point STRING COMMENT '22点第5个点位', 22_6_point STRING COMMENT '22点第6个点位', created_at STRING COMMENT '创建时间', updated_at STRING COMMENT '更新时间', version STRING COMMENT '版本' ) COMMENT '排单表' PARTITIONED BY ( ds STRING COMMENT '分区' );