[filenames]

file_name=XXXXXX.csv

'''

@create on : 20190311

@Update : 20190311

@description: 该模块可以直接获取最直观的报表

'''

import pandas as pd

import configparser

import os

import json

# 获取项目根目录

dirpath = os.path.dirname(os.path.realpath(__file__))

# 拼接时候注意一下，会从第一个带有斜杠的地址开始拼接

sencondpath = os.path.join(dirpath, "log_file")

config = configparser.ConfigParser()

config.read("config.ini")

filename = config.get("filenames", "file_name")

# 改config.ini中的文件名自动拼接

finalpath = os.path.join(sencondpath, filename)

# 读入的CSV数据对象

log_df = pd.read_csv(finalpath, encoding="utf-8")

print(log_df)

# 半成品矩阵

def mergedf():

    df_right = log_df['message']

    df_left = log_df['@timestamp']

    result_df = pd.concat([df_left, df_right], axis=1)

    return result_df

def oprate_df():

    # 计算有多少符合数据旗标

    flag = 0

    df_size = log_df.__len__()

    urlParams, jrtt_reports, convert_ids = [], [], []

    try:

        goal_df = mergedf()

        for line in range(df_size):

            data_row = json.loads(log_df.loc[line, 'message'])

            print(log_df.loc[line, '@timestamp'])

            if data_row["data"]["jrtt_report"] is not None and data_row["data"]["convert_id"] is not None:

                flag = flag + 1

            line = line + 1

            urlParams.append(data_row["data"]["urlparams"])

            jrtt_reports.append(data_row["data"]["jrtt_report"])

            convert_ids.append(data_row["data"]["convert_id"])

        print(flag)

    except Exception as e:

        print("日志文件解析出错" + str(e))

    try:

        goal_df.insert(0, 'uelParmas', urlParams)

        goal_df.insert(0, 'jrtt_repot', jrtt_reports)

        goal_df.insert(0, 'convert_id', convert_ids)

    except Exception as e:

        print("矩阵组合出错！"+str(e))

    #print(goal_df)

    return goal_df

if __name__ == '__main__':

    total_df = oprate_df()

    excelFile = "D:/anylysis/dataResult/workResult.xlsx"

    writer = pd.ExcelWriter(excelFile)

    total_df.to_excel(writer, 'FinalResult')

    writer.save()