Detailed explanation of real IP request Pandas for Python data analysis

Foreword

pandas is a data analysis package built based on Numpy that contains more advanced data structures and tools. Similar to Numpy, whose core is ndarray, pandas also revolves around the two core data structures of Series and DataFrame. Series and DataFrame correspond to one-dimensional sequence and two-dimensional table structure respectively. The conventional import method of pandas is as follows:

from pandas import Series,DataFrame
import pandas as pd

1.1. Pandas analysis steps

1. Load log data

2. Load area_ip data

3. Count the number of real_ip requests. SQL similar to the following:

SELECT inet_aton(l.real_ip),
  count(*),
  a.addr
FROM log AS l
INNER JOIN area_ip AS a
  ON a.start_ip_num <= inet_aton(l.real_ip)
  AND a.end_ip_num >= inet_aton(l.real_ip)
GROUP BY real_ip
ORDER BY count(*)
LIMIT 0, 100;

1.2. Code

cat pd_ng_log_stat.py
#!/usr/bin/env python
#-*- coding: utf-8 -*-
 
from ng_line_parser import NgLineParser
 
import pandas as pd
import socket
import struct
 
class PDNgLogStat(object):
 
  def __init__(self):
    self.ng_line_parser = NgLineParser()
 
  def _log_line_iter(self, pathes):
    """解析文件中的每一行并生成一个迭代器"""
    for path in pathes:
      with open(path, 'r') as f:
        for index, line in enumerate(f):
          self.ng_line_parser.parse(line)
          yield self.ng_line_parser.to_dict()
 
  def _ip2num(self, ip):
    """用于IP转化为数字"""
    ip_num = -1
    try:
      # 将IP转化成INT/LONG 数字
      ip_num = socket.ntohl(struct.unpack("I",socket.inet_aton(str(ip)))[0])
    except:
      pass
    finally:
      return ip_num
 
  def _get_addr_by_ip(self, ip):
    """通过给的IP获得地址"""
    ip_num = self._ip2num(ip)
 
    try:
      addr_df = self.ip_addr_df[(self.ip_addr_df.ip_start_num <= ip_num) & 
                   (ip_num <= self.ip_addr_df.ip_end_num)]
      addr = addr_df.at[addr_df.index.tolist()[0], 'addr']
      return addr
    except:
      return None
           
  def load_data(self, path):
    """通过给的文件路径加载数据生成 DataFrame"""
    self.df = pd.DataFrame(self._log_line_iter(path))
 
 
  def uv_real_ip(self, top = 100):
    """统计cdn ip量"""
    group_by_cols = ['real_ip'] # 需要分组的列,只计算和显示该列
     
    # 直接统计次数
    url_req_grp = self.df[group_by_cols].groupby(
                   self.df['real_ip'])
    return url_req_grp.agg(['count'])['real_ip'].nlargest(top, 'count')
     
  def uv_real_ip_addr(self, top = 100):
    """统计real ip 地址量"""
    cnt_df = self.uv_real_ip(top)
 
    # 添加 ip 地址 列
    cnt_df.insert(len(cnt_df.columns),
           'addr',
           cnt_df.index.map(self._get_addr_by_ip))
    return cnt_df
     
  def load_ip_addr(self, path):
    """加载IP"""
    cols = ['id', 'ip_start_num', 'ip_end_num',
        'ip_start', 'ip_end', 'addr', 'operator']
    self.ip_addr_df = pd.read_csv(path, sep='\t', names=cols, index_col='id')
    return self.ip_addr_df
 
def main():
  file_pathes = ['www.ttmark.com.access.log']
 
  pd_ng_log_stat = PDNgLogStat()
  pd_ng_log_stat.load_data(file_pathes)
 
  # 加载 ip 地址
  area_ip_path = 'area_ip.csv'
  pd_ng_log_stat.load_ip_addr(area_ip_path)
 
  # 统计 用户真实 IP 访问量 和 地址
  print pd_ng_log_stat.uv_real_ip_addr()
 
if __name__ == '__main__':
  main()

Running statistics and output results

python pd_ng_log_stat.py
 
         count  addr
real_ip            
60.191.123.80  101013 浙江省杭州市
-        32691  None
218.30.118.79  22523   北京市
......
136.243.152.18   889   德国
157.55.39.219   889   美国
66.249.65.170   888   美国
 
[100 rows x 2 columns]

Summary

The above is the entire content of this article. I hope the content of this article will be helpful to everyone’s study or work. If you have any questions, you can leave a message to communicate.