1.微博登录
2.爬取数据
3.mysql存储
4.pyechart本地展示
5.用vue搭建网站web展示
1.微博登录
新浪微博的登录不是简单的post就能解决的，他的登录有加密，所以我们要使用一定的解密算法才能正常登录微博，得到我们想要的数据。
先不要慌，第一步当然是import我们需要的库
from urllib import request,parse import http.cookiejar import base64 import json import rsa import binascii from PIL import Image from bs4 import BeautifulSoup import pymysql
要登录，那肯定要先拿到验证码
def GetCode(self):
url="https://login.sina.com.cn/cgi/pin.php?r=694905&s=0&p=gz-52086a8a846fadcdacf4fb058324aa387858"#验证码地址
img=self.opener.open(url)
f=open(‘vv.png‘,‘wb‘)
f.write(img.read())
f.close()
im=Image.open(‘vv.png‘)
im.show()
拿到了验证码还不够，登录之前还有个预登录，拿到我们后面加密需要的参数
def prelogin(self):
url="https://login.sina.com.cn/sso/prelogin.php?entry=weibo&callback=sinaSSOController.preloginCallBack&su=NDc5MTkyNzQyJTQwcXEuY29t&rsakt=mod&checkpin=1&client=ssologin.js(v1.4.19)&_=1529471491358"
a=self.opener.open(url).read().decode(‘utf-8‘)
a=a[a.find("(")+1:a.find(")")]
b=json.loads(a)
return b
‘’‘
sinaSSOController.preloginCallBack({"retcode":0,"servertime":1540617565,"pcid":"gz-65c55b3534f8a1df1330b4708fb6d1b57273","nonce":"ONED4A","pubkey":"EB2A38568661887FA180BDDB5CABD5F21C7BFD59C090CB2D245A87AC253062882729293E5506350508E7F9AA3BB77F4333231490F915F6D63C55FE2F08A49B353F444AD3993CACC02DB784ABBB8E42A9B1BBFFFB38BE18D78E87A0E41B9B8F73A928EE0CCEE1F6739884B9777E4FE9E88A1BBE495927AC4A799B3181D6442443","rsakv":"1330428213","is_openlock":0,"showpin":0,"exectime":10})’‘’
然后用这些参数对登录参数进行加密，说实在话具体的加密细节我也不记得了，当时做的时候拿了个记事本把所有东西记下来然后分析，也查了很多博客的资料才做好。
def GetMixUser(self,username,password):
username_struct={‘username‘:username}
rsaPubkey=int(self.pre[‘pubkey‘],16)
key=rsa.PublicKey(rsaPubkey, 65537)
message = str(self.pre[‘servertime‘]) + ‘\t‘ + self.pre[‘nonce‘] + ‘\n‘ + password
passwd = rsa.encrypt(message.encode(‘utf-8‘), key)
upass = binascii.b2a_hex(passwd)
uname=base64.b64encode(parse.urlencode(username_struct).encode(‘utf-8‘)).decode(‘utf-8‘)[12:]
mix={‘uname‘:uname,‘upass‘:upass.decode(‘utf-8‘)}
return mix
拿到加密后的登录参数，可以提交了
def login(self,username,password,code):
mix=self.GetMixUser(username,password)
uname=mix[‘uname‘]
upass=mix[‘upass‘]
url="https://login.sina.com.cn/sso/login.php?client=ssologin.js(v1.4.19)"
print("登录中……")
postData={
"door":code,
"encoding":"utf-8",
"entry":"weibo",
"from":"null",
"gateway":1,
"nonce":self.pre[‘nonce‘],
"prelt":72,
"pwencode":"rsa2",
"qrcode_flag":False,
"returntype":"META",
"savestate":7,
"servertime":self.pre[‘servertime‘],
"service":"miniblog",
"rsakv":self.pre[‘rsakv‘],
"su":uname,
"sp":upass,
"url":"https://weibo.com/ajaxlogin.php?framelogin=1&callback=parent.sinaSSOController.feedBackUrlCallBack",
"useticket":1,
"vsnf":1
}
postData=parse.urlencode(postData).encode(‘utf-8‘)
result=self.opener.open(url,postData).read().decode(‘gbk‘)
url1=result[result.find("replace")+9:result.find(‘)‘)-1]
result=self.opener.open(url1).read().decode("gbk")
if(result.find("身份")!=-1):
return False
result=result[result.find(‘location‘)+18:]
url2=result[:result.find(‘)‘)-1]
self.opener.open(url2).read().decode("gbk")
return True
2.爬取信息
先得到用户follow的列表
def GetUserList(self,uid,pageNum):
url="https://weibo.com/"+str(uid)+"/follow?page="+str(pageNum)
try:
result=self.opener.open(url).read().decode(‘utf-8‘)
html = result.replace(‘\\n‘, ‘‘).replace(‘\\t‘, ‘‘).replace(‘\\r‘, ‘‘).replace(‘\\‘, ‘‘)
html = html[html.find("<!--关注/粉丝列表-->"):html.find("<!--关欧盟隐私协议弹窗-->")]
soup = BeautifulSoup(html, "html.parser")
list_a = soup.findAll(name=‘div‘, attrs={"class": "info_name W_fb W_f14"})
name = []
uid = []
for a in list_a:
try:
b = a.find(name="a")
b = b[‘usercard‘]
b = b[3:13:]
uid.append(b)
name.append(a.text)
print("加入用户:" + a.text)
except:
print("No Data")
dic = {"name": name, "uid": uid}
return dic
except:
pass
再拿到这些用户的主页微博言论，我们得到他们发的所有博文
def GetTalks(self,uid):
rlist = []
i=0
html=""
while(True):
try:
result=self.opener.open("https://weibo.com/u/"+str(uid)+"?page="+str(i)).read().decode("utf-8")
html = result.replace("\\t", "").replace("\\n", "").replace("\\r", "").replace("\\", "")
html = html[html.find("<div class=\"WB_feed WB_feed_v3 WB_feed_v4\""):]
except:
pass
soup=BeautifulSoup(html,"html.parser")
list_a = soup.find_all(name="div", attrs={"class": "WB_text W_f14"})
i = i + 1
if list_a:
print("第" + str(i) + "页")
for a in list_a:
at=a.text
at=at.replace(" ","")
if at:
rlist.append(at)
print("内容存入："+at)
else:
break
return rlist
3.数据存储
我暂时只做了这些，然后存储到数据库吧。mysql的基础语句很简单，我也就会这么几句，不会的百度吧。。
def sqllogin(self):
db=pymysql.connect(host=‘localhost‘,user=‘root‘,db=‘weibouser‘,passwd=‘root‘,charset=‘utf8mb4‘)
return db
def sqlProcess(self,db):
while(True):
cursor=db.cursor()
cursor.execute("SELECT * FROM USERS WHERE TAG =1") #1 表示 未处理 2 表示 正在处理 3 表示完成处理
result=cursor.fetchone()
if(result):
cursor.execute("UPDATE USERS SET TAG=2 WHERE USERID=‘%s‘" % (result[2]))
talks=self.GetTalks(uid=result[2])
for i in range(1,4):
uids=""
names=""
userlist = self.GetUserList(uid=result[2], pageNum=i)
try:
uids=userlist[‘uid‘]
names=userlist[‘name‘]
except:
break
if int(result[4])!=3:
for t in range(len(uids)):
try:
if self.IfExist(db,"users","name",names[t])==False:
cursor.execute("INSERT INTO USERS (NAME,USERID,TAG,CLASS) VALUES (‘%s‘,‘%s‘,%d,%d)" % (names[t], uids[t], 1, int(result[4])+1)) # 数据库写userlist
cursor.execute("INSERT INTO FOLLOWS (USERID,FUID,FUNAME) VALUES (‘%s‘,‘%s‘,‘%s‘)" % (result[2], uids[t],names[t]))
except:
print("Error")
for talk in talks:
try:
cursor.execute("INSERT INTO USERTALKS (USERID,NAME,TALK)VALUES (‘%s‘,‘%s‘,‘%s‘)" % (result[2],result[1],talk))#数据库写评论
except:
print("Error")
cursor.execute("UPDATE USERS SET TAG=3 WHERE USERID=‘%s‘"%(result[2]))
else:
break
def AnotherProcess(self,db):
cursor=db.cursor();
cursor.execute("SELECT * FROM USERS WHERE 1");
results=cursor.fetchall()
for result in results:
sex="女"
try:
r = self.opener.open("https://weibo.com/u/"+result[2]).read().decode("utf-8")
html = r.replace("\\t", "").replace("\\n", "").replace("\\r", "").replace("\\", "")
if html.find("female") == -1:
sex="男"
except:
pass
soup = BeautifulSoup(html, "html.parser")
keywords=soup.find(attrs={"name":"keywords"})[‘content‘]
description=soup.find(attrs={"name":"description"})[‘content‘]
cursor.execute("INSERT INTO USERDETAILS (NAME,DESCRIPTION,KEYWORDS,SEX)VALUES(‘{}‘,‘{}‘,‘{}‘,‘{}‘)".format(result[1],description,keywords,sex))
4.echart本地展示
import pymysql
from pyecharts import Graph
class DrawCharts:
raw_nodes = []
links = []
def __init__(self):
self.CreatRelation()
self.DrawCharts()
def CreatRelation(self):
db=pymysql.connect(host=‘localhost‘,user=‘root‘,db=‘weibouser‘,passwd=‘root‘,charset=‘utf8mb4‘)
cursor=db.cursor()
cursor.execute("SELECT * FROM USERS WHERE 1");
users=cursor.fetchall()
for user in users:#上海大学
cursor.execute("SELECT * FROM FOLLOWS WHERE USERID=‘%s‘"%(user[2]));
results=cursor.fetchall()
if results:
for result in results:
links.append({"source":user[1],"target":result[4]})
print(user[1]+"->"+result[4])
for i in range(3):
cursor.execute("SELECT * FROM USERS WHERE CLASS=%d"%(i+1))
results=cursor.fetchall()
for result in results:
self.raw_nodes.append({"name": result[1], "symbolSize": 30-i*13,"category":i})
db.close()
def DrawCharts(self):
graph = Graph("微博关注关系图", width=1200, height=600)
categories=[
{
"name":"一级深度",
"itemStyle":
{
"normal":{
"color":‘#CC0033‘,
"borderColor": "#5182ab",
"borderWidth": 1.8,
"show":"True"
}
}
},
{
"name": "二级深度",
"itemStyle":
{
"normal": {
"color": ‘#f44242‘,
"borderColor": "#5182ab",
"borderWidth": 1.8,
"show":"True"
}
}
},
{
"name": "三级深度",
"itemStyle":
{
"normal": {
"color": ‘#663366‘,
"borderColor": "#5182ab",
"borderWidth": 1.8,
"show":"True"
}
}
}
]
graph.add("",self.raw_nodes,self.links,label_pos="right",graph_repulsion=10, is_legend_show=False,categories=categories,
label_text_color=None,is_label_show=True)
graph.render()
效果就不发了，运行起来比较卡，因为节点太多了，当时老师说怎么怎么加速，也不说细节，我当时一脸懵逼，也就没管了，大不了减少节点数量嘛。
5.web展示
前端两段关键代码。。
<style scoped>
.layout-con{
height: 100%;
width: 100%;
}
.menu-item span{
display: inline-block;
overflow: hidden;
width: 69px;
text-overflow: ellipsis;
white-space: nowrap;
vertical-align: bottom;
transition: width .2s ease .2s;
}
.menu-item i{
transform: translateX(0px);
transition: font-size .2s ease, transform .2s ease;
vertical-align: middle;
font-size: 16px;
}
.collapsed-menu span{
width: 0px;
transition: width .2s ease;
}
.collapsed-menu i{
transform: translateX(5px);
transition: font-size .2s ease .2s, transform .2s ease .2s;
vertical-align: middle;
font-size: 22px;
}
</style>
<template>
<div class="layout">
<Layout :style="{minHeight: ‘100vh‘}">
<Sider collapsible :collapsed-width="78" v-model="isCollapsed">
<Menu active-name="1-1" theme="dark" width="auto" :class="menuitemClasses">
<MenuItem name="1-1" @click.native="tab_search_onclick()">
<Icon type="search"></Icon>
<span>搜索</span>
</MenuItem>
<MenuItem name="1-2" @click.native="tab_follow_onclick()">
<Icon type="person"></Icon>
<span>follow关系</span>
</MenuItem>
<MenuItem name="1-3" @click.native="tab_words_onclick()">
<Icon type="cloud"></Icon>
<span >热词云图</span>
</MenuItem>
</Menu>
</Sider>
<Layout>
<Header :style="{background: ‘#fff‘, boxShadow: ‘0 2px 3px 2px rgba(0,0,0,.1)‘,height:‘100px‘}">
<img src="../assets/logo.png" style="height: 100%"/>
</Header>
<Content :style="{padding: ‘0 16px 16px‘}">
<Card style="margin-top: 20px">
<p slot="title">{{Content_Title}}</p>
<div style="height: 650px">
<div v-show="v_search">
<div style="margin-bottom: 30px;margin-top: 30px">
<Input v-model="input_word" icon="ios-clock-outline" placeholder="请输入微博用户名称!" style="width: 250px;margin-right: 10px"></Input>
<Button type="primary" shape="circle" icon="ios-search" @click="fs_search">查询</Button>
</div>
<div style="width: 50%;margin: auto">
<Card :bordered="true" style="height: 350px;text-align: left">
<p slot="title">{{Search_Card_Title}}</p>
<p v-for="item in items" style="height: 70px">{{item.message}}</p>
</Card>
</div>
</div>
<div v-show="v_follows">
<div style="width: 80%;margin: auto">
<Card :bordered="true" style="height: 580px;width: 100%;margin-top: 50px">
<p slot="title">{{Follow_Card_Title}}</p>
<div id="Follow_Chart" style="height: 500px;width:800px;margin: auto;"></div>
</Card>
</div>
</div>
<div v-show="v_words">
<div style="width: 50%;margin: auto">
<Card :bordered="true" style="height: 330px;margin-top:50px ">
<p slot="title">{{Words_Card_Title}}</p>
<div id="Words_Chart" style="height: 250px;width:500px;margin: auto"></div>
</Card>
</div>
</div>
</div>
</Card>
</Content>
</Layout>
</Layout>
</div>
</template>
<script>
let echarts = require(‘echarts‘);
require(‘echarts-wordcloud‘);
require("echarts/lib/chart/graph/forceLayout");
export default {
data () {
return {
isCollapsed: false,
items:[],
Search_Card_Title:‘搜索结果‘,
Content_Title:"资料搜索",
Follow_Card_Title:"关系图绘制结果",
Words_Card_Title:"词云绘制结果",
input_word:‘‘,
v_search:true,
v_follows:false,
v_words:false
};
},
computed: {
menuitemClasses: function () {
return [
‘menu-item‘,
this.isCollapsed ? ‘collapsed-menu‘ : ‘‘
]
}
},
methods:{
fs_search(){
let axios=this.$axios;
let word=this.input_word;
function getDetails(){
return axios.get("http://localhost:9090/details?name="+word);
}
function getUsers(){
return axios.get("http://localhost:9090/words?name="+word);
}
function getNodes() {
return axios.get("http://localhost:9090/nodes?name="+word);
}
function getLinks(){
return axios.get("http://localhost:9090/links?name="+word);
}
axios.all([getDetails(),getUsers(),getNodes(),getLinks()]).then(axios.spread((details,users,nodes,links)=>{
this.draw_relations(nodes.data,links.data);
this.set_details(details.data);
this.draw_words(users.data);
this.$Notice.success({
title: ‘绘制成功‘,
desc: ‘您搜索的用户‘+word+‘关系图和热词已经成功绘制！‘
});
})).catch(res=>{
this.$Notice.error({
title: ‘绘制失败‘,
desc: ‘您搜索的用户‘+word+‘关系图或热词绘制失败，请检查网络或输入信息！‘
});
})
},
tab_search_onclick:function(){
this.v_search=true;
this.v_follows=false;
this.v_words=false;
this.Content_Title="资料搜索"
},
tab_follow_onclick:function(){
this.v_follows=true;
this.v_words=false;
this.v_search=false;
this.Content_Title="follow关系";
},
draw_words:function(words_data){
let mychart=echarts.init(document.getElementById(‘Words_Chart‘));
mychart.setOption({
tootip: {
show: true
},
series: {
name: ‘词云图‘,
type: ‘wordCloud‘,
textStyle: {
normal: {
color: function () {
return ‘rgb(‘ + [
Math.round(Math.random() * 160),
Math.round(Math.random() * 160),
Math.round(Math.random() * 160)
].join(‘,‘) + ‘)‘
}
}
},
textRotation: [0, 90, -45, 45],
autoSize: true,
data:
words_data
}
})
},
set_details(details){
this.items.push({"message":"微博名："+details["name"]});
this.items.push({"message":"微博描述："+details["description"]});
this.items.push({"message":"关键词："+details["keywords"]});
this.items.push({"message":"用户性别："+details["sex"]});
},
draw_relations:function(nodes,links){
let mychart=echarts.init(document.getElementById("Follow_Chart"));
let option={
title: { // 图表标题
text: "关注关系图", // 标题文本
left : ‘3%‘, // 标题距离左侧边距
top : ‘3%‘, // 标题距顶部边距
textStyle : { // 标题样式
color : ‘#000‘, // 标题字体颜色
fontSize : ‘30‘, // 标题字体大小
}
},
series: [{
type: "graph", // 系列类型:关系图
top: ‘10%‘, // 图表距离容器顶部的距离
roam: true, // 是否开启鼠标缩放和平移漫游。默认不开启。如果只想要开启缩放或者平移，可以设置成 ‘scale‘ 或者 ‘move‘。设置成 true 为都开启
focusNodeAdjacency: true, // 是否在鼠标移到节点上的时候突出显示节点以及节点的边和邻接节点。[ default: false ]
force: { // 力引导布局相关的配置项，力引导布局是模拟弹簧电荷模型在每两个节点之间添加一个斥力，每条边的两个节点之间添加一个引力，每次迭代节点会在各个斥力和引力的作用下移动位置，多次迭代后节点会静止在一个受力平衡的位置，达到整个模型的能量最小化。
// 力引导布局的结果有良好的对称性和局部聚合性，也比较美观。
repulsion: 1000, // [ default: 50 ]节点之间的斥力因子(关系对象之间的距离)。支持设置成数组表达斥力的范围，此时不同大小的值会线性映射到不同的斥力。值越大则斥力越大
// [ default: 30 ]边的两个节点之间的距离(关系对象连接线两端对象的距离,会根据关系对象值得大小来判断距离的大小)，
// 这个距离也会受 repulsion。支持设置成数组表达边长的范围，此时不同大小的值会线性映射到不同的长度。值越小则长度越长。如下示例:
// 值最大的边长度会趋向于 10，值最小的边长度会趋向于 50 edgeLength: [10, 50]
},
layout: "force", // 图的布局。[ default: ‘none‘ ]
// ‘none‘ 不采用任何布局，使用节点中提供的 x， y 作为节点的位置。
// ‘circular‘ 采用环形布局;‘force‘ 采用力引导布局.
// 标记的图形
//symbol: "path://M19.300,3.300 L253.300,3.300 C262.136,3.300 269.300,10.463 269.300,19.300 L269.300,21.300 C269.300,30.137 262.136,37.300 253.300,37.300 L19.300,37.300 C10.463,37.300 3.300,30.137 3.300,21.300 L3.300,19.300 C3.300,10.463 10.463,3.300 19.300,3.300 Z",
symbol: ‘circle‘,
lineStyle: { // 关系边的公用线条样式。其中 lineStyle.color 支持设置为‘source‘或者‘target‘特殊值，此时边会自动取源节点或目标节点的颜色作为自己的颜色。
normal: {
color: ‘#000‘, // 线的颜色[ default: ‘#aaa‘ ]
width: 1, // 线宽[ default: 1 ]
type: ‘solid‘, // 线的类型[ default: solid ] ‘dashed‘ ‘dotted‘
opacity: 0.5, // 图形透明度。支持从 0 到 1 的数字，为 0 时不绘制该图形。[ default: 0.5 ]
curveness: 0 // 边的曲度，支持从 0 到 1 的值，值越大曲度越大。[ default: 0 ]
}
},
label: { // 关系对象上的标签
normal: {
show: true, // 是否显示标签
position: "inside", // 标签位置:‘top‘‘left‘‘right‘‘bottom‘‘inside‘‘insideLeft‘‘insideRight‘‘insideTop‘‘insideBottom‘‘insideTopLeft‘‘insideBottomLeft‘‘insideTopRight‘‘insideBottomRight‘
textStyle: { // 文本样式
fontSize: 16
}
}
},
data: nodes,
links: links
}],
animationEasingUpdate: "cubicOut", // 数据更新动画的缓动效果。[ default: cubicOut ] "quinticInOut"
// 数据更新动画的时长。[ default: 300 ]
};
mychart.setOption(option);
},
tab_words_onclick:function(){
this.Content_Title="用户词云";
this.v_words=true;
this.v_follows=false;
this.v_search=false;
},
},
name:"HelloWorld"
}
</script>
后端关键代码
//用户的具体信息domainpackage com.example.demo;
import org.hibernate.engine.loading.internal.CollectionLoadContext;
import javax.persistence.*;
@Entity
@Table(name = "userdetails")
public class UserDetails {
@Id
@GeneratedValue(strategy = GenerationType.AUTO)
private int num;
@Column(name="description")
public String description;
@Column(name="keywords")
public String keywords;
@Column(name="sex")
public String sex;
@Column(name="name")
public String name;
void setName(String name){
this.name=name;
}
void setDescription(String description){
this.description=description;
}
void setKeywords(String keywords){
this.keywords=keywords;
}
void setSex(String sex){
this.sex=sex;
}
String getName(){
return this.name;
}
String getDescription(){
return this.description;
}
String getKeywords(){
return this.keywords;
}
String getSex(){
return this.sex;
}
}
//最重要的controllerpackage com.example.demo;
import org.springframework.stereotype.Controller;
import org.springframework.web.bind.annotation.CrossOrigin;
import org.springframework.web.bind.annotation.RequestMapping;
import org.springframework.web.bind.annotation.ResponseBody;
import org.springframework.beans.factory.annotation.Autowired;
import java.util.ArrayList;
import java.util.List;
@Controller
@CrossOrigin
public class IndexController {
@Autowired
private UserRepositroy userRepositroy;
@RequestMapping(value="users")
@ResponseBody
public Object GetUser(){
List<User> userlist=userRepositroy.findAll();
return userlist;
}
@RequestMapping(value="links")
@ResponseBody
public Object GetLinks(String name){
List<Follows> followsList =followsRepositroy.findByName(name);
List<RetnLinks> links =new ArrayList<RetnLinks>();
for(int i=0;i<followsList.size();i++){
RetnLinks l=new RetnLinks();
l.setSource(followsList.get(i).getName());
l.setTarget(followsList.get(i).getFuname());
links.add(l);
}
return links;
}
@RequestMapping(value="nodes")
@ResponseBody
public Object GetNodes(String name){
List<Follows> followsList=followsRepositroy.findByName(name);
List<RetnNodes> nodes =new ArrayList<RetnNodes>();
RetnNodes r=new RetnNodes();
r.setName(name);
r.setCategory(1);
r.setSymbolSize(30);
nodes.add(r);
for(int i=0;i<followsList.size();i++){
RetnNodes rs=new RetnNodes();
rs.setName(followsList.get(i).getFuname());
rs.setCategory(2);
rs.setSymbolSize(20);
nodes.add(rs);
}
return nodes;
}
@RequestMapping(value="words")
@ResponseBody
public Object GetWords(String name){
List<User> userlist=userRepositroy.findByName(name);
String word = userlist.get(0).getKeywords();
String value=userlist.get(0).getKeyvalue();
String[] array_word= word.split(",");
String[] array_value= value.split(",");
List<RetnWords> words=new ArrayList<RetnWords>();
for(int i=0;i<50;i++){
RetnWords w=new RetnWords();
w.setName(array_word[i]);
w.setValue(Double.parseDouble(array_value[i]));
words.add(w);
}
return words;
}
@Autowired
private FollowsRepositroy followsRepositroy;
@RequestMapping(value="follows")
@ResponseBody
public Object GetFollows(String name){
List<Follows> followslist=followsRepositroy.findByName(name);
return followslist;
}
@Autowired
private DetailsRepositroy detailsRepositroy;
@RequestMapping(value="details")
@ResponseBody
public Object GetDetails(String name){
UserDetails userDetails=detailsRepositroy.findOneByName(name);
return userDetails;
}
}
其他的都是类似的。无限重复改名。
最终结果：
原文地址：https://www.cnblogs.com/aoru45/p/9860946.html
时间： 2024-11-01 23:17:00