若泽数据-造数据作业用Python造数据

#coding=UTF-8

import random 	// 引入random函数
import time			// 引入time 函数

// url_paths的数组

url_paths=[
                "class/112.html",
                "class/128.html",
                "class/145.html",
                "class/146.html",
                "class/131.html",
                "class/130.html",
                "learn/821",
                "course/list"
                 ]

// ip

ip_slices=[132,156,124,10,29,167,143,187,30,46,552,63,72,98,168]

// 从哪里跳转过来的

http_referers = [
    "https://www.baidu.com/s?wd={query}",
        "https://www.sogou.com/web?query={query}",
        "http://cn.bing.com/search?q={query}",
        "http://search.yahoo.com/search?p={query}",
]

// 搜索关键字

search_keywords = [
        "SparkSQL实战",
    "Hadoop 基础",
        "Storm实战",
        "Spark_Streaming实战",
        "大数据面试",
        "Flink 实战"，
]

// 状态码

status_codes = ["200","404","500","300"]

// 随机产生url的方法

def sample_url():
        return random.sample(url_paths,1)[0]

// 随机产生ip的方法
def sample_ip():
slice=random.sample(ip_slices,4)
return “.”.join([str(item) for item in slice])

// 随机产生跳转网站的方法

def sample_referer():
        if random.uniform(0,1)>0.2:
                return "-"
        refer_str=random.sample(http_referers,1)
        query_str=random.sample(search_keywords,1)
        return refer_str[0].format(query=query_str[0])

// 随机产生状态码的方法

def sample_status_code():
        return random.sample(status_codes,1)[0]

// 日志生成方法，默认生成10条日志

def generate_log(count=10):
        time_str = time.strftime("%Y-%m-%d %H:%M:%S",time.localtime())

        f = open("/Users/liujingmao/data/access.log","w+")

        while count>=1:
                query_log = "{ip}\t{localtime}\t\"GET /{url} HTTP/1.1\"\t{status_code}\t{referer}".format(url=sample_url(),ip=sample_ip(),referer=sample_referer(),status_code=sample_status_code(),localtime=time_str)
                print(query_log)
                f.write(query_log+"\n")
                count = count-1

// 主函数

if __name__=='__main__':
        generate_log(100)   // 产生100条日志

结果 如下：python /Users/liujingmao/data/generate_log.py