記錄一次利用 python 進行日誌模塊開發過程

只記錄大致思路和我認爲其中須要記錄的地方。

正則匹配

正則匹配的模式很難記憶，即便記住了，也很難寫出無錯誤的匹配模式。可是，藉助網上一些提供實時對比的網站，如 regexr.com。
代碼示意：

import os
import re

'''
define regex mode
'''
get_imgpath_regex = re.compile(r'''(
    (\d{4}-\d{2}-\d{2}\s\d{2}:\d{2}:\d{2},\d+)      # time
    \s      # separate
    (.*)?get_imgpath\sused\stime\sis\s      # info
    (\d+.\d+)       # time
    \sargs\sis\s\(u'     # separate
    (.*?)',\)\sresult\sis\s     # img dir
    (.*?.jpg)      # img path
)''', re.VERBOSE)

get_imgpath_flag = 'get_imgpath'

def main():

    for infile in infile_list:
        # prase line using regex mode
        with open(infile, "r") as file:
            for line in file:    # 先判斷關鍵詞
                currentDict = {}
                if get_imgpath_flag in line:
                    for groups in regex['get_imgpath_regex'].findall(line):    #　再提取模式對應的內容
                        currentDict = {'date': groups[1], 'cost_time':groups[3],
                        'img_dir':groups[4], 'img_path':groups[5]}
                        # print(currentDict)
                        get_imgpath_match.append(currentDict)
    else:
                    pass

參數配置

採起的方式爲參數存儲在一個單獨的文件，如 config.json。

{
    "FLAG" : {
        "SAVE_SPILT_LOG_FILE_FLAG" : false ,
        "SAVE_MERGE_LOG_FILE_FLAG" : false ,
        "USE_CURRENT_PATH" : false
    },

    "PATH" : {
        "INPUT_LOG_FILE_PATH" : "E:\\zwk\\Code\\logger_read\\data\\pro_data" ,
        "SAVE_SPILT_MERGE_LOG_PATH" : "E:\\zwk\\Code\\logger_read\\output\\spilt_merge_log" ,
        "OUTPUT_RESULT_PATH" : "E:\\zwk\\Code\\logger_read\\output" 
    },
   
    "PARAMETERS" : {
        "windows_size" : 2 ,
        "duplicate_times" : 1 
    }
}

再對參數進行解析，

import json

def main():
    # outfile_path = os.path.normpath("output/filtered")
    global parameters
    if(config['FLAG']['USE_CURRENT_PATH']):
        pwd = os.getcwd()
        config['PATH']['INPUT_LOG_FILE_PATH'] = os.path.join(pwd, 'data')
        config['PATH']['SAVE_SPILT_MERGE_LOG_PATH'] = os.path.join(pwd, 'output')
        config['PATH']['OUTPUT_RESULT_PATH'] = os.getcwd()

    output_file_path = config['PATH']
    flag = config['FLAG']

if __name__ == '__main__':
    this_folder = os.path.dirname(os.path.abspath(__file__))
    config_file = os.path.join(this_folder, 'config.json') 
    exists_check = os.path.isfile(config_file)
    if not exists_check:
        print('Error: loss of config file, Exit !!!')
    with open(config_file, 'r') as f:
        config = json.load(f)
    # improve it, use as global variable
    parameters = config['PARAMETERS']
    main()

目錄、文件名

在這裏，示例我認爲的還不錯作法，主要是不受系統影響

this_folder = os.path.dirname(os.path.abspath(__file__))
    config_file = os.path.join(this_folder, 'config.json') 
    exists_check = os.path.isfile(config_file)
    if not exists_check:
        print('Error: loss of config file, Exit !!!')
    with open(config_file, 'r') as f:
        config = json.load(f)

    # 省略中間
    # write output to files
    os.makedirs(outfile['SAVE_SPILT_MERGE_LOG_PATH'], exist_ok=True)

```