爬虫大作业

1.选一个自己感兴趣的主题或网站。(所有同学不能雷同)

2.用python 编写爬虫程序，从网络上爬取相关主题的数据。

3.对爬了的数据进行文本分析，生成词云。

import re
import requests
from bs4 import BeautifulSoup
from datetime import datetime
import pandas
import matplotlib.pyplot as plt
from wordcloud import WordCloud
import jieba.analyse
def getNewDetail(newurl):
    res = requests.get(newurl)
    res.encoding = 'utf-8'
    soup = BeautifulSoup(res.text, 'html.parser')
    div = soup.select_one("#lft-art")
    if(div==None):
        div = soup.select_one(".main_art")
    if (div == None):
        return None;
    #标题
    title=div.select_one("h1").text
    info=div.select_one(".info_l").text
    #作者
    a=re.match("\s*(By )*(.*) \|",info)
    if(a==None) :
        author=None;
    else:
        author=a.group(2)
    #时间
    time= datetime.strptime(re.search("Updated: (.{16})", info).group(1), "%Y-%m-%d %H:%M")
    #正文
    content=div.select_one("#Content").text
    writeNewsDetail(content)
    return {"title":title,"author":author,"time":time,"content":content}
def getAll():
    newstotal = [];
    newsurl = 'http://www.chinadaily.com.cn/china/governmentandpolicy'
    res = requests.get(newsurl)
    res.encoding = 'utf-8'
    soup = BeautifulSoup(res.text, 'html.parser')
    last = soup.select_one("#div_currpage").select("a")[-1].attrs.get('href')
    num=re.match("http://www.chinadaily.com.cn/china/governmentandpolicy/page_(\d*).html", last).group(1)
    num=int(num,10)
    for i in range(num):
        print(i)
        listurl="http://www.chinadaily.com.cn/china/governmentandpolicy/page_{}.html".format(i+1)
        newstotal.extend(getAPage(listurl))
    df = pandas.DataFrame(newstotal)
    df.to_excel('newsData.xlsx')
def getAPage(url):
    newsls = []
    res = requests.get(url)
    res.encoding = 'utf-8'
    soup = BeautifulSoup(res.text, 'html.parser')
    div=soup.select_one("#lft-art").select("div.mb10")
    for i in div:
        newurl=i.select_one("a").attrs.get('href')
        data=getNewDetail(newurl)
        if(data!=None):
            if next(re.compile(r'[\000-\010]|[\013-\014]|[\016-\037]').finditer(data["content"]), None):
                writeIllegalCharacterErrorUrl(newurl+"\n")
            else:
                newsls.append(data)
        else:
            writeAnalysisErrorUrl(newurl+"\n")
    return newsls
def writeNewsDetail(content):
    f = open('content.txt', 'a',encoding='utf-8')
    f.write(content)
    f.close()
def writeAnalysisErrorUrl(error):
    f = open('AnalysisError.txt', 'a',encoding='utf-8')
    f.write(error)
    f.close()
def writeAnalysisErrorUrl(error):
    f = open('AnalysisError.txt', 'a',encoding='utf-8')
    f.write(error)
    f.close()

def writeIllegalCharacterErrorUrl(error):
    f = open('IllegalCharacterError.txt', 'a', encoding='utf-8')
    f.write(error)
    f.close()

def getWordCloud():
    fo = open("content.txt", "r",encoding='utf-8')
    srt = fo.read()
    fo.close()
    result = jieba.analyse.textrank(srt, topK=50, withWeight=True)
    keywords = dict()
    for i in result:
        keywords[i[0]] = i[1]
    my_wordcloud = WordCloud(font_path="1.ttf").generate_from_frequencies(keywords)
    plt.imshow(my_wordcloud)
    plt.axis("off")
    plt.show()

getAll()
getWordCloud()

 

4.对文本分析结果进行解释说明。

分析出中国在国际上的热点是什么

5.写一篇完整的博客，描述上述实现过程、遇到的问题及解决办法、数据分析思想及结论。

 实现过程：从新闻列表页第一页的分页栏获得新闻列表页总页数；爬下每个列表页的每个新闻；提取每个新闻的数据并保存；最后生成词云

遇到的问题及解决办法：遇见一些数据格式与众不同的新闻页，保存这些链接；遇见一些无法存入xlsx的新闻，字符异常报错，分析出这些新闻带非打印字符，也是保存这些链接

6.最后提交爬取的全部数据、爬虫及数据分析源代码。

 

 

全部数据：https://pan.baidu.com/s/1YXSpQwZ-_itlC2NfZyqyvg