Python定向爬虫——校园论坛帖子信息

引言

写这个小爬虫主要是为了爬校园论坛上的实习信息，主要采用了Requests库

源码

URLs.py

主要功能是根据一个初始url（包含page页面参数）来获得page页面从当前页面数到pageNum的url列表

import re

def getURLs(url, attr, pageNum=1):
all_links = []
try:
now_page_number = int(re.search(attr+'=(\d+)', url, re.S).group(1))
for i in range(now_page_number, pageNum + 1):
new_url = re.sub(attr+'=\d+', attr+'=%s' % i, url, re.S)
all_links.append(new_url)
return all_links
except TypeError:
print "arguments TypeError:attr should be string."

uni_2_native.py

由于论坛上爬取得到的网页上的中文都是unicode编码的形式，文本格式都为 &#XXXX;的形式，所以在爬得网站内容后还需要对其进行转换

import sys
import re
reload(sys)
sys.setdefaultencoding('utf-8')

def get_native(raw):
tostring = raw
while True:
obj = re.search('&#(.*?);', tostring, flags=re.S)
if obj is None:
break
else:
raw, code = obj.group(0), obj.group(1)
tostring = re.sub(raw, unichr(int(code)), tostring)
return tostring

存入SQLite数据库：saveInfo.py

# -*- coding: utf-8 -*-

import MySQLdb

class saveSqlite():
def __init__(self):
self.infoList = []

def saveSingle(self, author=None, title=None, date=None, url=None,reply=0, view=0):
if author is None or title is None or date is None or url is None:
print "No info saved!"
else:
singleDict = {}
singleDict['author'] = author
singleDict['title'] = title
singleDict['date'] = date
singleDict['url'] = url
singleDict['reply'] = reply
singleDict['view'] = view
self.infoList.append(singleDict)

def toMySQL(self):
conn = MySQLdb.connect(host='localhost', user='root', passwd='', port=3306, db='db_name', charset='utf8')
cursor = conn.cursor()
# sql = "select * from info"
# n = cursor.execute(sql)
# for row in cursor.fetchall():
#     for r in row:
#         print r
#     print '\n'
sql = "delete from info"
cursor.execute(sql)
conn.commit()

sql = "insert into info(title,author,url,date,reply,view) values (%s,%s,%s,%s,%s,%s)"
params = []
for each in self.infoList:
params.append((each['title'], each['author'], each['url'], each['date'], each['reply'], each['view']))
cursor.executemany(sql, params)

conn.commit()
cursor.close()
conn.close()

def show(self):
for each in self.infoList:
print "author: "+each['author']
print "title: "+each['title']
print "date: "+each['date']
print "url: "+each['url']
print "reply: "+str(each['reply'])
print "view: "+str(each['view'])
print '\n'

if __name__ == '__main__':
save = saveSqlite()
save.saveSingle('网','aaa','2008-10-10 10:10:10','www.baidu.com',1,1)
# save.show()
save.toMySQL()

主要爬虫代码

import requests
from lxml import etree
from cc98 import uni_2_native, URLs, saveInfo

# 根据自己所需要爬的网站，伪造一个header
headers ={
'Accept': '',
'Accept-Encoding': '',
'Accept-Language': '',
'Connection': '',
'Cookie': '',
'Host': '',
'Referer': '',
'Upgrade-Insecure-Requests': '',
'User-Agent': ''
}
url = 'http://www.cc98.org/list.asp?boardid=459&page=1&action='
cc98 = 'http://www.cc98.org/'

print "get infomation from cc98..."

urls = URLs.getURLs(url, "page", 50)
savetools = saveInfo.saveSqlite()

for url in urls:
r = requests.get(url, headers=headers)
html = uni_2_native.get_native(r.text)

selector = etree.HTML(html)
content_tr_list = selector.xpath('//form/table[@class="tableborder1 list-topic-table"]/tbody/tr')

for each in content_tr_list:
href = each.xpath('./td[2]/a/@href')
if len(href) == 0:
continue
else:
# print len(href)
# not very well using for, though just one element in list
# but I don't know why I cannot get the data by index
for each_href in href:
link = cc98 + each_href
title_author_time = each.xpath('./td[2]/a/@title')

# print len(title_author_time)
for info in title_author_time:
info_split = info.split('\n')
title = info_split[0][1:len(info_split[0])-1]
author = info_split[1][3:]
date = info_split[2][3:]

hot = each.xpath('./td[4]/text()')
# print len(hot)
for hot_num in hot:
reply_view = hot_num.strip().split('/')
reply, view = reply_view[0], reply_view[1]
savetools.saveSingle(author=author, title=title, date=date, url=link, reply=reply, view=view)

print "All got! Now saving to Database..."
# savetools.show()
savetools.toMySQL()
print "ALL CLEAR! Have Fun!"