日志清洗

手动清洗

1. 日志为slf4j日志，info级别
2. 用shell筛选出需要的内容
3. 用python解析成json格式

// grep 提取关键词日志
grep -E "A|B" > result.txt

// python 解析
import sys
import re
import json

fileName = sys.argv[1]

def read():
    print(fileName)
    with open(fileName, "r") as f:
        with open("data_2019-06-16_result.json", "w") as f1:
            count = 1
            for line in f.readlines():
                splits = re.split(",|\s", line)
                adict = {field.split(":")[0]: field.split(":")[1] if len(field.split(":"))==2 else 0 for field in splits}
                f1.write(json.dumps(adict))
                f1.write("\n")
                if count %10000 == 0:
                    print "this is the " + str(count) + " line..."
                count += 1

if __name__ == "__main__":
        read()