Python 微信爬虫完整实例【单线程与多线程】
作者:FRESHMANS 发布时间:2023-08-19 23:12:58
标签:Python,微信爬虫
本文实例讲述了Python 实现的微信爬虫。分享给大家供大家参考,具体如下:
单线程版:
import urllib.request
import urllib.parse
import urllib.error
import re,time
headers = ("User-Agent",
"Mozilla/5.0 (Windows NT 6.1; WOW64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/60.0.3107.4 Safari/537.36")
operner = urllib.request.build_opener()
operner.addheaders = [headers]
urllib.request.install_opener(operner)
list_url = []
###使用代理获取网页url内容
def use_proxy(url):
try:
# proxy = urllib.request.ProxyHandler({'http':proxy_addr})##使用代理版
# operner = urllib.request.build_opener()
# urllib.request.install_opener(operner)
headers = ("User-Agent",
"Mozilla/5.0 (Windows NT 6.1; WOW64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/60.0.3107.4 Safari/537.36")
operner = urllib.request.build_opener()
operner.addheaders = [headers]
urllib.request.install_opener(operner)
data = urllib.request.urlopen(url).read().decode('utf-8')
# print (data)
return data
except urllib.error.URLError as e:
if hasattr(e, "code"):
print(e.code)
elif hasattr(e, "reason"):
print(e.reason)
except Exception as e:
print("exception" + str(e))
time.sleep(1)
##获取要爬取的url
def get_url(key, pagestart, pageend):
try:
keycode = urllib.parse.quote(key)
for page in range(pagestart, pageend + 1):
url = "http://weixin.sogou.com/weixin?query=%s&_sug_type_=&s_from=input&_sug_=n&type=%d&page=1&ie=utf8" % (
keycode, page)
data1 = use_proxy(url)
#print("data1的内容是", data1)
listurl_pattern = '<h3>.*?("http://.*?)</h3>'
result = re.compile(listurl_pattern, re.S).findall(data1)
for i in range(len(result)):
res = result[i].replace("amp;", "").split(" ")[0].replace("\"", "")
list_url.append(res)
#print(list_url)
return list_url
except urllib.error.URLError as e:
if hasattr(e, "code"):
print(e.code)
elif hasattr(e, "reason"):
print(e.reason)
except Exception as e:
print("exception:", e)
##通过获取的url爬行内容数据并处理
def get_url_content(list_url):
fh1=open("D:\\python-script\\1.html", 'wb')
html1 = '''<!DOCTYPE html>\n<html xmlns="http://www.w3.org/1999/xhmtl">\n<head>\n<meta http-equiv="Content-Type" content="text/html; charset=utf-8">\n<title>微信文章</title></head>\n<body>'''
fh1.write(html1.encode("utf-8"))
fh1.close()
fh = open("D:\\python-script\\1.html", 'ab')
for url in list_url:
data_content = use_proxy(url)
#print (data_content)
#sys.exit()
title_pattern = '<h2.*>.*?</h2>'
result_title = re.compile(title_pattern, re.S).findall(data_content)
##标题(str)
res_title = result_title[0].replace("<h2 class=\"rich_media_title\" id=\"activity-name\">", "").replace("</h2>",
"").strip()
content_pattern = 'id="js_content">(.*?)<div class="rich_media_tool" id="js_sg_bar">'
content = re.compile(content_pattern, re.S).findall(data_content)
try:
fh.write(res_title.encode("utf-8"))
for i in content:
fh.write(i.strip().encode("utf-8"))
except UnicodeEncodeError as e:
continue
fh.write("</body></html>".encode("utf-8"))
if __name__ == '__main__':
pagestart = 1
pageend = 2
key = "人工智能"
get_url(key, pagestart, pageend)
get_url_content(list_url)
多线程版:
import urllib.request
import urllib.parse
import urllib.error
import re,time
import queue
import threading
headers = ("User-Agent","Mozilla/5.0 (Windows NT 6.1; WOW64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/60.0.3107.4 Safari/537.36")
operner = urllib.request.build_opener()
operner.addheaders = [headers]
urllib.request.install_opener(operner)
urlque = queue.Queue()
list_url = []
###使用代理获取网页url内容
def use_proxy(url):
try:
# proxy = urllib.request.ProxyHandler({'http':proxy_addr})
# operner = urllib.request.build_opener()
# urllib.request.install_opener(operner)
headers = ("User-Agent",
"Mozilla/5.0 (Windows NT 6.1; WOW64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/60.0.3107.4 Safari/537.36")
operner = urllib.request.build_opener()
operner.addheaders = [headers]
urllib.request.install_opener(operner)
data = urllib.request.urlopen(url).read().decode('utf-8')
#print (data)
return data
except urllib.error.URLError as e:
if hasattr(e,"code"):
print (e.code)
elif hasattr(e,"reason"):
print (e.reason)
except Exception as e:
print ("exception"+str(e))
time.sleep(1)
###获取文章的url连接,并将连接加入到队列
class get_url(threading.Thread):
def __init__(self,key,pagestart,pageend,urlque):
threading.Thread.__init__(self)
self.pagestart = pagestart
self.pageend = pageend
self.key = key
self.urlque = urlque
def run(self):
try:
keycode = urllib.parse.quote(self.key)
for page in range(self.pagestart,self.pageend+1):
url = "http://weixin.sogou.com/weixin?query=%s&_sug_type_=&s_from=input&_sug_=n&type=%d&page=1&ie=utf8" % (keycode,page)
data = use_proxy(url)
print ("data1的内容是",data)
listurl_pattern = '<h3>.*?("http://.*?)</h3>'
result = re.compile(listurl_pattern,re.S).findall(data)
print (result)
if len(result) == 0:
print ("没有可用的url")
sys.exit()
for i in range(len(result)):
res = result[i].replace("amp;","").split(" ")[0].replace("\"" ,"")
#list_url.append(res) #加入列表
self.urlque.put(res) ##加入队列
self.urlque.task_done()
#return list_url
except urllib.error.URLError as e:
if hasattr(e, "code"):
print(e.code)
elif hasattr(e, "reason"):
print(e.reason)
except Exception as e:
print ("exception:",e)
##根据url获取文章内容
class get_url_content(threading.Thread):
def __init__(self,urlque):
threading.Thread.__init__(self)
self.urlque = urlque
def run(self):
fh1 = open("D:\\python-script\\1.html", 'wb')
html1 = '''<!DOCTYPE html>\n<html xmlns="http://www.w3.org/1999/xhmtl">\n<head>\n<meta http-equiv="Content-Type" content="text/html; charset=utf-8">\n<title>微信文章</title></head>\n<body>'''
fh1.write(html1.encode("utf-8"))
fh1.close()
fh = open("D:\\python-script\\1.html", 'ab')
while True:
try:
url = self.urlque.get()
data_content = use_proxy(url)
title_pattern = '<h2.*>.*?</h2>'
result_title = re.compile(title_pattern, re.S).findall(data_content)
##标题
res_title = result_title[0].replace("<h2 class=\"rich_media_title\" id=\"activity-name\">", "").replace("</h2>","").strip()
content_pattern = 'id="js_content">(.*?)<div class="rich_media_tool" id="js_sg_bar">'
content = re.compile(content_pattern, re.S).findall(data_content)
#c = '<p style="max-width: 100%;box-sizing: border-box;min-height: 1em;text-indent: 2em;word-wrap: break-word !important;">'
# for i in content:
# ##内容
# c_content=i.replace(c, "").replace("<br /></p>", "").replace("</p>", "")
fh.write(res_title.encode("utf-8"))
for i in content:
fh.write(i.strip().encode("utf-8"))
except UnicodeEncodeError as e:
continue
fh.close()
class contrl(threading.Thread):
def __init__(self,urlqueue):
threading.Thread.__init__(self)
self.urlqueue = urlqueue
while True:
print ("程序正在执行")
if self.urlqueue.empty():
time.sleep(3)
print ("程序执行完毕")
exit()
if __name__ == '__main__':
pagestart = 1
pageend = 2
key = "人工智能"
get_url = get_url(key,pagestart,pageend,urlque)
get_url.start()
get_content = get_url_content(urlque)
get_content.start()
cntrol = contrl(urlque)
cntrol.start()
希望本文所述对大家Python程序设计有所帮助。
来源:https://www.cnblogs.com/FRESHMANS/p/8125594.html


猜你喜欢
- 前言题目: Survey of Personalization Techniques for FederatedLearning会议: 20
- 为什么是Python先来聊聊为什么做数据分析一定要用Python或R语言。编程语言这么多种,Java, PHP都很成熟,但是为什么在最近热火
- 在用python的bottle框架开发时,前端使用ajax跨域访问时,js代码老是进入不了success,而是进入了error,而返回的状态
- Get Started Tutorial for Python in Visual Studio Code一、安装PythonPython简
- 效果图:搜索分类2种情况,一般的是当用户输入完,点击确定的按钮在向后发送请求,还有一种就是的我一边输入,一边向后台发送请求,但是会产生一个性
- 实例076:做函数题目:编写一个函数,输入n为偶数时,调用函数求1/2+1/4+...+1/n,当输入n为奇数时,调用函数1/1+1/3+.
- 控制资源访问前文提到threading库在多线程时,对同一资源的访问容易导致破坏与丢失数据。为了保证安全的访问一个资源对象,我们需要创建锁。
- 相比SQL Server 2000提供的FOR XML查询,SQL Server 2005版本对现有功能增强的基础上增加了不少新功能,最为吸
- 本文实例分析了JS获取年月日时分秒的方法。分享给大家供大家参考,具体如下:var d = new Date();var time = d.g
- 今天看了一个优化案例觉的挺有代表性,这里记录下来做一个标记,来纪念一下随便的字段定义的问题。回忆一下,在表的设计中很多人习惯的把表的结构设计
- mmdetection简介 目标检测工具箱mmdetection是香
- Python提供了两个内置函数从标准输入读入一行文本,默认的标准输入是键盘。如下:1.raw_input2.inputraw_input函数
- 昨天在这个上面找了好久的错,嘤嘤嘤~很多时候我们在爬取数据存储的时候都需要将当前时间作为一个依据,在python里面没有时间类型可以直接拿来
- 自定义图片生成词云图的多种方法有时候我们会根据具体的场景来结合图片展示词云,比如我分析的是美团评论,那么最好的展示方法就是利用美团的logo
- 方法一、SELECT SUM(正确数)+SUM(错误数) AS 总记录数,SUM(正确数),SUM(错误数) FROM ( SELECT C
- Matplotlib介绍Matplotlib 是一款用于数据可视化的 Python 软件包,支持跨平台运行,它能够根据 NumPy 
- 视图代码lis = []#设置一个空列表用来存放发送的验证码,用来验证def yzm1(): res1 = &qu
- 1.保存变量先创建(在tf.Session()之前)saversaver = tf.train.Saver(tf.global_variab
- js浮点数计算有时是不准确的,比如7*0.8 == 7*8/10的值为false,因为7*0.8=5.6000000000000005,乘出
- 内容介绍将日常工作中遇到的数数据冲突和样本源的方法进行总结,其中主要包括实际业务数据冲突、样本选取问题、数据共线性 等思路,并且长期更新。实