计算日志信息里各性能所消耗的时间

日志信息如图

import re
import os 
import csv
import argparse

#得到输入的参数，日志路径
def get_args():
    parser = argparse.ArgumentParser(\
            description="get cost time",\
            formatter_class=argparse.ArgumentDefaultsHelpFormatter)
    
    parser.add_argument("-d", "--log_dir",\
            required=True,\
            help="the logs directory",\
            dest="d")
    
    args = parser.parse_args()
    log_dir = args.d
    return log_dir

#获取目录里所需的文件，并保存至list
def get_files(log_dir):
    log_list = []
    for root, dirs, files in os.walk(log_dir):
        if not files:
            continue
        for file in files:
            if file.endswith(".log") :
                complete_path = os.path.join(root, file)
                log_list.append(complete_path)
        return log_list

#得到每一个需要统计的属性
def items_classify(file_name):
    pattern = re.compile("](.*)cost time:\d+.\d+\(ms\)")
    with open(file_name, "r") as f:
        contents = f.read()
        items = set(pattern.findall(contents))
        f.close()
    return contents, items

#匹配时间消耗
def items_match(contents, item):
    number_list = []
    pattern = re.compile("](.*)cost time:(\d+.\d+|\d)\(ms\)")
    titles = pattern.findall(contents)
    for i in range(len(titles)):
        if item in titles[i]:
            number_list.append(titles[i][1])
    return item, number_list

#计算min, avg, max消耗
def calculate(number_list, item):
    time_cost = 0
    num = 0
    for i in range(len(number_list)):
        time_cost = float(number_list[i]) + time_cost
        num += 1
    avg_time = time_cost/num
    max_time, min_time = compare_value(number_list)
    save_result(item, min_time, round(avg_time, 3), max_time)


def compare_value(alist):
    if alist != []:
        max_value, min_value = alist[0], alist[0]
        for x in alist:
            if float(max_value) < float(x):
                max_value = x
            if float(min_value) > float(x):
                min_value = x
        return max_value, min_value

#将结果保存至CSV
def save_result(item, min_time, avg_time, #max_time):
    csv_data = [item, min_time, avg_time, max_time]
    print "%s: min=%s, avg=%s, max=%s"%(item ,min_time, avg_time, max_time)
    with open("result1.csv", "ab") as f:
        writer = csv.writer(f)
        writer.writerow(csv_data)
        f.close()
    
def work(log_dir):
    #可根据实际情况选择单个文件或者文件夹
    #file_names = get_files(log_dir)
    #for file_name in file_names:
    #if log_dir.endswith('.log'):
    contents, items = items_classify(log_dir)
    for item in items:
        item, number_list = items_match(contents, item)    
        calculate(number_list, item)


def main():
    log_dir = get_args()
    work(log_dir)


if __name__ == "__main__":
    main()