Python爬蟲實現(xiàn)“盜取”微信好友信息的方法分析

# 微信好友信息爬取+數(shù)據(jù)可視化
# encoding=utf-8
__author__ = 'Jonny'
__location__ = '濟南'
__date__ = '2018-06-02'
from wxpy import *
import re
import jieba
import numpy
import pandas as pd
import matplotlib.pyplot as plt
from scipy.misc import imread
from wordcloud import WordCloud,ImageColorGenerator
from matplotlib.patches import Polygon
from matplotlib.colors import rgb2hex
from mpl_toolkits.basemap import B
# 微信登錄
def wx_login():
  try:
    #初始化機器人，掃碼登錄
    bot = Bot()
    #獲取好友列表
    frinds = bot.friends()
    #wxpy.api.chats.chats.Chats對象是多個聊天對象的合集，
    # 可用于搜索或統(tǒng)計，可以搜索和統(tǒng)計的信息包括sex(性別)、province(省份)、city(城市)和signature(個性簽名)等
    print(type(frinds))
    #輸出好友列表
    for i in frinds:
      print(i)
  except Exception as e:
    print(e.args)
    wx_login()
  return frinds
# 數(shù)據(jù)可視化
#統(tǒng)計男女性別信息
def wx_friend_sex_infor(friends):
  sex_dict = {'male':0,'female':0,'other':0}
  for friend in friends:
    if friend.sex == 1:
      sex_dict['male'] += 1
    elif friend.sex == 2:
      sex_dict['female'] += 1
    else:
      print(friend,'性別未標記！')
      sex_dict['other'] += 1
  print(sex_dict)
  wx_show_sex_infor(sex_dict)
# pie(x, explode=None, labels=None,
#   colors=('b', 'g', 'r', 'c', 'm', 'y', 'k', 'w'),
#   autopct=None, pctdistance=0.6, shadow=False,
#   labeldistance=1.1, startangle=None, radius=None,
#   counterclock=True, wedgeprops=None, textprops=None,
#   center = (0, 0), frame = False )
# 參數(shù)說明
# x    (每一塊)的比例，如果sum(x) > 1會使用sum(x)歸一化
# labels (每一塊)餅圖外側顯示的說明文字
# explode (每一塊)離開中心距離
# startangle 起始繪制角度,默認圖是從x軸正方向逆時針畫起,如設定=90則從y軸正方向畫起
# shadow 是否陰影
# labeldistance label繪制位置,相對于半徑的比例, 如<1則繪制在餅圖內(nèi)側
# autopct 控制餅圖內(nèi)百分比設置,可以使用format字符串或者format function
#     '%1.1f'指小數(shù)點前后位數(shù)(沒有用空格補齊)
# pctdistance 類似于labeldistance,指定autopct的位置刻度
# radius 控制餅圖半徑
# 返回值:
# 如果沒有設置autopct,返回(patches, texts)
# 如果設置autopct,返回(patches, texts, autotexts)
def wx_show_sex_infor(data):
  labers = ['男性','女性','未標記']
  data = [data['male'],data['female'],data['other']]
  plt.pie(data=data,labels=labers,autopct='%.2f',shadow=True)
  plt.show()
  plt.savefig('sex.jpg')
  plt.close()
def wx_friend_location_infor(friends):
  loction_dict = {'北京': 0, '上海': 0, '天津': 0, '重慶': 0,
           '河北': 0, '山西': 0, '吉林': 0, '遼寧': 0, '黑龍江': 0,
           '陜西': 0, '甘肅': 0, '青海': 0, '山東': 0, '福建': 0,
           '浙江': 0, '臺灣': 0, '河南': 0, '湖北': 0, '湖南': 0,
           '江西': 0, '江蘇': 0, '安徽': 0, '廣東': 0, '海南': 0,
           '四川': 0, '貴州': 0, '云南': 0,
           '內(nèi)蒙古': 0, '新疆': 0, '寧夏': 0, '廣西': 0, '西藏': 0,
           '香港': 0, '澳門': 0}
  for friend in friends:
    if friend.province in loction_dict.keys():
      loction_dict[friend.province] += 1
  #轉成JSON格式：
  loction_list = []
  for key,value in loction_dict.items():
    loction_list.append({'name':key,'sum':value})
  print(loction_list)
def wx_show_location_infor():
  pass
#顯示好友個簽信息
def wx_show_signature(friends):
  #統(tǒng)計好友簽名
  for friend in friends:
    #對數(shù)據(jù)進行清洗，排除標點信息的干擾
    pattern = re.compile(r'[一-龥]+')
    filterdata = re.findall(pattern,friend.signature)
    with open('signature.txt','a',encoding='utf-8',newline='') as f:
      f.write(str(friend)+''.join(filterdata)+'\n')
  f.close()
  # 讀取文件數(shù)據(jù)
  with open('signature.txt','r',encoding='utf-8',newline='') as f:
    content = f.read()
  f.close()
  segment = jieba.lcut(content)
  words_df = pd.DataFrame({'segment':segment})
  #讀取stopwords
  stopwords = pd.read_csv('stopwords.txt',index_col=False,quoting=3,sep=' ',names=['stopword'],encoding='gb18030')
  words_df = words_df[~words_df.segment.isin(stopwords.stopword)]
  print(words_df)
  words_stat = words_df.groupby(by=['segment'])['segment'].agg({'計數(shù)':numpy.size})
  words_stat = words_stat.reset_index().sort_values(by=['計數(shù)'],ascending=False)
  #設置詞云屬性
  color_mask = imread('background.jpg')
  wordcloud = WordCloud(font_path='simhei.ttf',    #設置字體可以顯示中文
             background_color= 'white',  #背景顏色是白色
             max_words=1000,        #設置詞云顯示的最大詞數(shù)
             mask=color_mask,       #設置背景圖片
             max_font_size=400,      #設置詞云中字體的最大值
             random_state=42,
             width=500,height=430,margin=2,#設置圖片默認大小
  )
  # 生成詞云, 可以用generate輸入全部文本,也可以我們計算好詞頻后使用generate_from_frequencies函數(shù)
  word_frequence = {x[0]: x[1] for x in words_stat.head(100).values}
  print(word_frequence)
  word_frequence_dict = {}
  for key in word_frequence:
    word_frequence_dict[key] = word_frequence[key]
  wordcloud.generate_from_frequencies(word_frequence_dict)
  # 從背景圖片生成顏色值
  image_colors = ImageColorGenerator(color_mask)
  # 重新上色
  wordcloud.recolor(color_func=image_colors)
  # 保存圖片
  wordcloud.to_file('output.png')
  plt.imshow(wordcloud)
  plt.axis("off")
  plt.show()
  plt.close()
if __name__ == '__main__':
  friends = wx_login()
  print('~~~~~~~~~~~~~~~~~~~~1~~~~~~~~~~~~~~~~~~~~~~~~~~~~')
  wx_friend_sex_infor(friends)
  print('~~~~~~~~~~~~~~~~~~~~~2~~~~~~~~~~~~~~~~~~~~~~~~~~~')
  wx_friend_location_infor(friends)
  print('~~~~~~~~~~~~~~~~~~~~~~3~~~~~~~~~~~~~~~~~~~~~~~~~~')
  wx_show_signature(friends)
  print('~~~~~~~~~~~~~~~~~~~~~~~4~~~~~~~~~~~~~~~~~~~~~~~~~')