python-虎扑爬虫

Posted tech-me

tags:

篇首语:本文由小常识网(cha138.com)小编为大家整理,主要介绍了python-虎扑爬虫相关的知识,希望对你有一定的参考价值。

  Python作为一个高级编程语言,不知从何时起就在圈子里流行起来了。个人也是图个鲜,跟上时代步伐学习了一下。“鲁迅”说过:不能学以致用,就是耍流氓。我用python对虎扑论坛作了一个爬虫。脚本写的糙了点,权作初学者交流使用,同时也方便以后查阅。本来是准备写一个虎扑的分析帖子,可后来动力不足就没有写成了。不过,作为一个马刺球迷很荣幸我们的组织是热度前三。

  准备工作:安装Python、安装MySQL、虚拟机【选择性,后期将每日放在服务器上执行定时任务使用】

    1、安装python:选择3.*,过程忽略

    2、安装MySQL:选择5.6版本及以上,过程忽略

    3、虚拟机:linux系列,过程忽略

  需求描述

    爬取虎扑论坛帖子,了解帖子内容、作者、热度等。

  写脚本

    一共分为三部分:part1通过对当前链接进行分析,提取帖子作者、阅读的信息;part2取得帖子本身的内容;part3对发帖人进行数据提取,为后期分析提供思路。具体的脚本如下。需要注意的是:编码、编码、编码。谢谢!

    注:由于虎扑的反爬虫导致可细分论坛的可读取页面数为10(突破防御失败,谢谢!)这种情况下,我的处理方式是将脚本放入服务器中每日爬取进行累积。

    Part1:爬取帖子的名称、作者、创建时间、阅读/回复、作者链接等,并放入本地MySQL数据库

# -*- coding: utf-8 -*-
from bs4 import BeautifulSoup
import requests
import json
import time
import pymysql
import importlib,sys
importlib.reload(sys)


forum_note_sum=[]  #variavle:save the content of tiezi
list_d=[‘原创‘,‘翻译‘,‘讨论‘]  #内容判断条件,如果帖子标题内容为此,取另一个值
type = sys.getfilesystemencoding()
#num:the record number of one page;get tiezi of author and others
def parent_li_web(num):
    forum_note_record = {}
    try:
        parent_tiezi=bs_obj.find(‘ul‘,class_=‘for-list‘).find_all(‘li‘)[num]
        div_one = parent_tiezi.find(‘div‘, class_=‘titlelink box‘)
        div_two = parent_tiezi.find(‘div‘, class_=‘author box‘)
        span_three = parent_tiezi.find(‘span‘, class_=‘ansour box‘).string.strip()
        div_four = parent_tiezi.find(‘div‘, class_=‘endreply box‘)
        subname=div_one.a.string
        sublink=‘https://bbs.hupu.com‘+div_one.a[‘href‘]
        team_tmp=theme_tmp
        for i in list_d:
            if i==subname:
                subname=div_one.find_all(‘a‘)[1].string
                sublink=‘https://bbs.hupu.com‘+div_one.find_all(‘a‘)[1][‘href‘]
                # print (i,subname,sublink)

        forum_note_record.update({
            ‘subname‘:subname,
            ‘subname_link‘:sublink,
            ‘author‘:div_two.a.string,
            ‘author_link‘:div_two.a[‘href‘],
            ‘author_create_time‘:div_two.find(‘a‘,style=‘color:#808080;cursor: initial; ‘).string,
            ‘read_reply_number‘:span_three,
            ‘last_reply_writer‘:div_four.span.string,
            ‘last_reply_time‘:div_four.a.string,
            ‘team_tmp‘:team_tmp
        })
        forum_note_sum.append(forum_note_record)
    except:
        return None

if __name__==‘__main__‘:
    # all_spurs_note
    begin_time=time.time()
    print(‘---------脚本执行时间为:{}------------‘.format(time.strftime(‘%Y-%m-%d %H:%M:%S‘,time.localtime())))
    team_list = [‘rockets‘, ‘warriors‘, ‘cavaliers‘, ‘spurs‘, ‘lakers‘, ‘celtics‘, ‘thunder‘, ‘clippers‘,
                 ‘timberwolves‘, ‘mavericks‘, ‘knicks‘, ‘bulls‘, ‘nets‘, ‘sixers‘, ‘jazz‘, ‘pacers‘, ‘blazers‘, ‘heat‘,
                 ‘suns‘, ‘grizzlies‘, ‘wizards‘, ‘pelicans‘, ‘bucks‘, ‘kings‘, ‘raptors‘, ‘nuggets‘, ‘hawks‘, ‘hornets‘,
                 ‘pistons‘, ‘magic‘]
    for li in team_list:
        forum_note_sum_code=[]
        theme_tmp=li
        for i in range(1,11,1):   #由于虎扑反爬,只能爬到10页;后续可放入linux中定时执行
            url = ‘https://bbs.hupu.com/{}-{}‘.format(li,i)
            print (url)
            wb_string = requests.get(url)
            bs_obj = BeautifulSoup(wb_string.content, ‘html.parser‘)

            with open(‘web_spider_original.txt‘,‘w‘,encoding=‘utf8‘) as f:
                f.write(str(bs_obj))
                f.write(‘\r‘*10+‘-----我是分割线-----‘+‘\r‘*10)

            for j in range(1,61,1):   #每个页面数据有60个帖子
                parent_li_web(j)
            with open(‘hupu_spider_spurs_load.txt‘, ‘w‘, encoding=‘utf8‘) as f:
                for item in forum_note_sum:
                    json.dump(item,f,ensure_ascii=False)
                    f.write(‘\r‘)

     #insert into mysql
    conn=pymysql.connect(host=‘localhost‘,user=‘root‘,passwd=‘1234‘,db=‘spider‘,port=3306,charset=‘utf8‘)
    cur=conn.cursor()
    cur.execute(‘delete from hupuforum_spurs_note_daytmp‘)
    with open(‘hupu_spider_spurs_load.txt‘,‘r‘,encoding=‘utf8‘) as f:
        for item in f:
            item=json.loads(item) #how convert string to dict
            # print(type(item))
            cur.execute(‘insert into hupuforum_spurs_note_daytmp(subname,subname_link,author,author_link,author_create_time,read_reply_number,last_reply_writer,last_reply_time,theme_title) values(%s,%s,%s,%s,%s,%s,%s,%s,%s)‘,(item[‘subname‘],item[‘subname_link‘],item[‘author‘],item[‘author_link‘],item[‘author_create_time‘],item[‘read_reply_number‘],item[‘last_reply_writer‘],item[‘last_reply_time‘],item[‘team_tmp‘]))
    conn.commit()
    cur.close()
    conn.close()
    print(‘Finished!本次执行消耗时间为:{}秒‘.format(time.time()-begin_time))

  Part2:增加贴子内容并更新部分字段

# coding=utf8
import time
import requests
from bs4 import BeautifulSoup
import pymysql
import signal

begin_time=time.strftime(%Y-%m-%d %H:%M:%S,time.localtime())
conn=pymysql.connect(host=localhost,port=3306,user=root,passwd=1234,db=spider,charset=utf8)
cur=conn.cursor()
sub_cur = conn.cursor()
cur.execute(INSERT INTO hupuforum_spurs_note SELECT * FROM hupuforum_spurs_note_daytmp WHERE subname_link NOT IN (SELECT a.subname_link FROM hupuforum_spurs_note a);)
cur.execute(update hupuforum_spurs_note a,hupuforum_spurs_note_daytmp b set a.read_reply_number=b.read_reply_number,a.last_reply_writer=b.last_reply_writer,a.last_reply_time=b.last_reply_time where a.subname_link=b.subname_link )
# conn.commit()
cur.execute(use spider;)
conn.commit()
cur.execute(select subname_link from hupuforum_spurs_note where sub_text is null;)
for url in cur.fetchall():
    url=list(url)
    # print(url)
    try:
        wb_page = requests.get(url[0],timeout=2)   #实际执行中,存在网页假死状态,设置超时
        bs_obj = BeautifulSoup(wb_page.content, html.parser)
        tmp_text = bs_obj.select(#tpc > div > div.floor_box > table.case > tbody > tr > td > div.quote-content)
        sub_text=tmp_text[0].get_text(strip=True)
        sub_text=sub_text.replace(\‘,)

        sql="""update hupuforum_spurs_note set sub_text=\‘{}\‘ where subname_link={};""".format((sub_text[:1000]),str(url).replace([,‘‘).replace(],‘‘))

        # print(sql)
        sub_cur.execute(sql)
        conn.commit()
        print(success)
    except IndexError as e:   #这个错误意味着页面也不存在
        sql="""update hupuforum_spurs_note set sub_text=\‘{}\‘ where subname_link={};""".format(网页不存在,str(url).replace([,‘‘).replace(],‘‘))
        sub_cur.execute(sql)
        conn.commit()
    except pymysql.err.InternalError as e:  #说明内容中包含emoj等utf8四字符内容
        sql="""update hupuforum_spurs_note set sub_text=\‘{}\‘ where subname_link={};""".format(内容格式有误,导致出错!,str(url).replace([,‘‘).replace(],‘‘))
        sub_cur.execute(sql)
        conn.commit()
    except requests.exceptions.ReadTimeout as e: #网页响应超时
        sql="""update hupuforum_spurs_note set sub_text=\‘{}\‘ where subname_link={};""".format(网页打开超时,str(url).replace([,‘‘).replace(],‘‘))
        sub_cur.execute(sql)
        conn.commit()
    else:
        sql="""update hupuforum_spurs_note set sub_text=\‘{}\‘ where subname_link={};""".format(其他类型错误,str(url).replace([,‘‘).replace(],‘‘))
        sub_cur.execute(sql)
        conn.commit()
conn.commit()
cur.close()
sub_cur.close()
conn.close()
end_time=time.strftime(%Y-%m-%d %H:%M:%S,time.localtime())
print(Finished,任务开始时间为:{},结束时间为:{}.format(begin_time,end_time))

   Part3:爬取注册用户信息

# coding=utf8
import time
import requests
from bs4 import BeautifulSoup
import pymysql

begin_time=time.strftime(%Y-%m-%d %H:%M:%S,time.localtime())
conn=pymysql.connect(host=localhost,port=3306,user=root,passwd=1234,db=spider,charset=utf8)
cur=conn.cursor()
sub_cur=conn.cursor()
cur.execute(select distinct author_link from hupuforum_spurs_note;)
for author_url in cur.fetchall():
    try:
        author_url=list(author_url)
        wb_obj=requests.get(author_url[0],timeout=2)
        bs_obj=BeautifulSoup(wb_obj.content,html.parser)
        author=bs_obj.select(#main > div.personal > div.personal_right > h3 > div)[0].string
        author_visited=bs_obj.select(#main > div.personal > div.personal_right > h3 > span)[0].string.replace(,‘‘).replace(人次访问,‘‘)
        author_info=bs_obj.select(#main > div.personal > div.personal_right > div)[0].get_text(strip=True)
        sub_cur.execute(insert into hupuforum_authors_info(author,author_link,author_visited,author_info,author_status) values(%s,%s,%s,%s,%s),(author,author_url[0],author_visited,author_info,正常))
    except IndexError as e:
        sub_cur.execute(
            insert into hupuforum_authors_info(author,author_link,author_visited,author_info,author_status) values(%s,%s,%s,%s,%s),
            (author, author_url[0], ‘‘, ‘‘, 无法访问))
    conn.commit()
conn.commit()
cur.close()
conn.close()
end_time=time.strftime(%Y-%m-%d %H:%M:%S,time.localtime())
print(Finished,任务开始时间为:{},结束时间为:{}.format(begin_time,end_time))

以上是关于python-虎扑爬虫的主要内容,如果未能解决你的问题,请参考以下文章

爬虫大作业

scrapy按顺序启动多个爬虫代码片段(python3)

scrapy主动退出爬虫的代码片段(python3)

爬虫大作业

Python 高级应用程序设计任务

python爬虫时,bs4无法读取网页标签中的文本?