# -*- coding: utf-8 -*-
# @Time    : 2022/12/1 16:48
# @Author  : ZAOXG
# @File    : api记录匹配.py

import data_process_tool
import pandas as pd


def step1():
    # 合并小文件
    data_process_tool.merge_table(fr'F:\工作数据存储2022\20221201_bcrAPI对比\API采集-BCR2022相同记录\API采集', on_columns=['EID'], encoding='GB2312', encoding_errors='ignore', on_bad_lines='skip')
    data_process_tool.merge_table(fr'F:\工作数据存储2022\20221201_bcrAPI对比\2', on_columns=['EID'], encoding_errors='ignore', on_bad_lines='skip')


def step2():
    record1 = data_process_tool.read_data(root_path+'2023记录.csv')
    record1.drop_duplicates(inplace=True)
    record1.rename(columns={'EID': '原始记录'}, inplace=True)
    record2 = data_process_tool.read_data(root_path+'API采集.csv')
    record2.drop_duplicates(inplace=True)
    record3 = pd.merge(record2, record1, how='left', left_on=['EID'], right_on=['原始记录'])
    print(record3)
    error_record = record3[record3['EID'].isna()]
    error_record.to_csv(root_path+'未匹配到记录2-EID.csv', index=False)
    # error_record.to_excel(root_path+'未匹配到记录-EID.xlsx', index=False)


def step3():
    record1 = data_process_tool.read_data(root_path + '2023原始记录.csv')
    record1.drop_duplicates(subset=['EID'], inplace=True)
    # 对每个表的列名做标记
    record1_rename = {}
    for r1_name in record1.columns:
        record1_rename[r1_name] = r1_name + '(2023记录)'
    record1.rename(columns=record1_rename, inplace=True)

    record2 = data_process_tool.read_data(root_path + 'API下载记录.csv')
    record2.drop_duplicates(subset=['EID'], inplace=True)
    record2_rename = {}
    for r2_name in record2.columns:
        record2_rename[r2_name] = r2_name + '(API记录)'
    record2.rename(columns=record2_rename, inplace=True)

    # 左连接找出右表缺失字段
    record3 = pd.merge(record2, record1, how='left', left_on=['EID(API记录)'], right_on=['EID(2023记录)'])
    print(record3)
    error_record = record3[record3['EID(2023记录)'].isna()]
    error_record.to_excel(root_path + 'API下载记录有2023原始记录无.xlsx', index=False)

    record4 = pd.merge(record1, record2, how='left', left_on=['EID(2023记录)'], right_on=['EID(API记录)'])
    print(record4)
    error_record2 = record4[record4['EID(API记录)'].isna()]
    error_record2.to_excel(root_path + '2023原始记录有API下载记录无.xlsx', index=False)


if __name__ == '__main__':
    root_path = 'F:/工作数据存储2022/20221201_bcrAPI对比/合并结果/'
    # step2()
    data_process_tool.merge_table(fr'F:\工作数据存储2022\20221201_bcrAPI对比\API下载记录\API采集', to_type='csv',
                                  encoding='GB2312', encoding_errors='ignore', on_bad_lines='skip')
    # data_process_tool.merge_table(fr'F:\工作数据存储2022\20221201_bcrAPI对比\2023原始记录', encoding_errors='ignore',
    #                               on_bad_lines='skip')
    # data_process_tool.merge_table(fr'F:\工作数据存储2022\20221201_bcrAPI对比\2023原始记录\2022-11-21-下载记录', encoding_errors='ignore',
    #                               on_bad_lines='skip')
    # data_process_tool.merge_table(fr'F:\工作数据存储2022\20221201_bcrAPI对比\API下载记录\API失败记录重新下载采集',
    #                               to_type='csv',
    #                               encoding='GB2312', encoding_errors='ignore',
    #                               on_bad_lines='skip')
    # step3()