使用python绘制人人网好友关系图示例

yipeiwu_com6年前Python基础

代码依赖:networkx matplotlib

复制代码 代码如下:

 #! /bin/env python
# -*- coding: utf-8 -*-

import urllib
import urllib2
import cookielib
import re
import cPickle as p
import networkx as nx
import matplotlib.pyplot as plt

__author__ = """Reverland (lhtlyy@gmail.com)"""

# Control parameters,EDIT it here
## Login
username = 'None'
password = 'None'
## Control Graphs, Edit for better graphs as you need
label_flag = True # Whether shows labels.NOTE: configure your matplotlibrc for Chinese characters.
remove_isolated = True # Whether remove isolated nodes(less than iso_level connects)
different_size = True # Nodes for different size, bigger means more shared friends
iso_level = 10
node_size = 40 # Default node size

 
def login(username, password):
    """log in and return uid"""
    logpage = "http://www.renren.com/ajaxLogin/login"
    data = {'email': username, 'password': password}
    login_data = urllib.urlencode(data)
    cj = cookielib.CookieJar()
    opener = urllib2.build_opener(urllib2.HTTPCookieProcessor(cj))
    urllib2.install_opener(opener)
    res = opener.open(logpage, login_data)
    print "Login now ..."
    html = res.read()
    #print html

    # Get uid
    print "Getting user id of you now"
    res = urllib2.urlopen("http://www.renren.com/home")
    html = res.read()
    # print html
    uid = re.search("'ruid':'(\\d+)'", html).group(1)
    # print uid
    print "Login and got uid successfully"
    return uid

 
def getfriends(uid):
    """Get the uid's friends and return the dict with uid as key,name as value."""
    print "Get %s 's friend list" % str(uid)
    pagenum = 0
    dict1 = {}
    while True:
        targetpage = "http://friend.renren.com/GetFriendList.do?curpage=" + str(pagenum) + "&id=" + str(uid)
        res = urllib2.urlopen(targetpage)
        html = res.read()

        pattern = '<a href="http://www\\.renren\\.com/profile\\.do\\?id=(\\d+)"><img src="[\\S]*" alt="[\\S]*[\\s]\\((.*)\\)" />'

        m = re.findall(pattern, html)
        #print len(m)
        if len(m) == 0:
            break
        for i in range(0, len(m)):
            no = m[i][0]
            uname = m[i][1]
            #print uname, no
            dict1[no] = uname
        pagenum += 1
    print "Got %s 's friends list successfully." % str(uid)
    return dict1

 
def getdict(uid):
    """cache dict of uid in the disk."""
    try:
        with open(str(uid) + '.txt', 'r') as f:
            dict_uid = p.load(f)
    except:
        with open(str(uid) + '.txt', 'w') as f:
            p.dump(getfriends(uid), f)
        dict_uid = getdict(uid)
    return dict_uid

 
def getrelations(uid1, uid2):
    """receive two user id, If they are friends, return 1, otherwise 0."""
    dict_uid1 = getdict(uid1)
    if uid2 in dict_uid1:
        return 1
    else:
        return 0

 
def getgraph(username, password):
    """Get the Graph Object and return it.
You must specify a Chinese font such as `SimHei` in ~/.matplotlib/matplotlibrc"""
    uid = login(username, password)
    dict_root = getdict(uid) # Get root tree

    G = nx.Graph() # Create a Graph object
    for uid1, uname1 in dict_root.items():
        # Encode Chinese characters for matplotlib **IMPORTANT**
        # if you want to draw Chinese labels,
        uname1 = unicode(uname1, 'utf8')
        G.add_node(uname1)
        for uid2, uname2 in dict_root.items():
            uname2 = unicode(uname2, 'utf8')
            # Not necessary for networkx
            if uid2 == uid1:
                continue
            if getrelations(uid1, uid2):
                G.add_edge(uname1, uname2)

    return G

 
def draw_graph(username, password, filename='graph.txt', label_flag=True, remove_isolated=True, different_size=True, iso_level=10, node_size=40):
    """Reading data from file and draw the graph.If not exists, create the file and re-scratch data from net"""
    print "Generating graph..."
    try:
        with open(filename, 'r') as f:
            G = p.load(f)
    except:
        G = getgraph(username, password)
        with open(filename, 'w') as f:
            p.dump(G, f)
    #nx.draw(G)
    # Judge whether remove the isolated point from graph
    if remove_isolated is True:
        H = nx.empty_graph()
        for SG in nx.connected_component_subgraphs(G):
            if SG.number_of_nodes() > iso_level:
                H = nx.union(SG, H)
        G = H
    # Ajust graph for better presentation
    if different_size is True:
        L = nx.degree(G)
        G.dot_size = {}
        for k, v in L.items():
            G.dot_size[k] = v
        node_size = [G.dot_size[v] * 10 for v in G]
    pos = nx.spring_layout(G, iterations=50)
    nx.draw_networkx_edges(G, pos, alpha=0.2)
    nx.draw_networkx_nodes(G, pos, node_size=node_size, node_color='r', alpha=0.3)
    # Judge whether shows label
    if label_flag is True:
        nx.draw_networkx_labels(G, pos, alpha=0.5)
    #nx.draw_graphviz(G)
    plt.show()

    return G

if __name__ == "__main__":
    G = draw_graph(username, password)

相关文章

Pandas读写CSV文件的方法示例

Pandas读写CSV文件的方法示例

读csv 使用pandas读取 import pandas as pd import csv if name == '__main__': # header=0——表示csv文件的...

Python socket 套接字实现通信详解

首先:我们介绍一下socket什么是socket: 1. socket 在操作系统中它是处于应用层与传输层的抽象层,它是一组操作起来非常简单的接口(接收数据的),此接口接受数据之后交个...

安装python及pycharm的教程图解

安装python及pycharm的教程图解

一.安装python 进入python官网,点击依次点击红色选中部分,开始下载。。。 下载完成后,打开安装包,如下有两个选项,一个是立即安装,另一个自定义安装,如果C盘空间足够的话,直...

神经网络理论基础及Python实现详解

神经网络理论基础及Python实现详解

一、多层前向神经网络 多层前向神经网络由三部分组成:输出层、隐藏层、输出层,每层由单元组成; 输入层由训练集的实例特征向量传入,经过连接结点的权重传入下一层,前一层的输出是下一层的输入;...

基于Django统计博客文章阅读量

如何精确地记录一篇文章的阅读量是一个比较复杂的问题,不过对于我们的博客来说,没有必要记录的那么精确。因此我们使用一种简单但有效的方式来记录博客文章的阅读量:文章每被浏览一次,则其阅读量...