从一个简单的爬虫实例看python2和python3的区别

  以前写爬虫我都是用的python2.7，昨天晚上我试着用python3.6写爬虫，写的过程中遇到了几个问题。好在经过查资料和有大神指点，我这个小白学到了很多东西。下面我来把我这两天学到的东西记录一下。

  程序比较简单，爬取某网站的图片。写了一半有个问题一直解决不了，先用2.7版本写了一遍：

# -*- coding:utf-8 -*-
import urllib,re

#获取源码
def page(pg):
    url = 'https://www.pengfu.com/index_%s.html'%pg
    html = urllib.urlopen(url).read()#读取所有源代码
    return html

#title
def title(html):
    #html = page(1)
    reg = re.compile(r'<h1 class="dp-b"><a href=".*?" target="_blank">(.*?)</a>')#正则 .*?代表所有字符
    item = re.findall(reg,html)#匹配
    return item

#picture
def content(html):
    reg = r'<img src="(.*?)" width'
    item = re.findall(reg,html)
    return item

#download
def download(url,name):
    path = 'H:\image\%s.jpg'%name.decode('utf-8').encode('gbk')
    urllib.urlretrieve(url,path)

if __name__ == '__main__':
    for i in range(1, 3):
        html = page(i)
        title_list = title(html)  # 图片名称
        content_list = content(html)
        for i, z in zip(title_list, content_list):
            download(z, i)
            print(i, z)

python3和python2有些地方的差异还是比较明显的：

# -*- coding:utf-8 -*-
import urllib.request,re
from urllib import request

#获取源码
def page(pg):
    url = 'https://www.pengfu.com/index_%s.html'%pg
    html = urllib.request.urlopen(url).read()#读取所有源代码
    #print(html)
    return html
#title
def title(html):
    #html = page(1)
    #html = html.decode('utf-8')#python3.x
    reg = re.compile(r'<h1 class="dp-b"><a href=".*?" target="_blank">(.*?)</a>')#正则 .*?代表所有字符
    item = re.findall(reg,html)#匹配
    return item

#picture
def content(html):
    reg = r'<img src="(.*?)" width'
    item = re.findall(reg,html)
    return item

#download
def download(url,name):
    with request.urlopen(url) as web:
        with open('H:/pengfuimg/%s.jpg'%name,'wb') as image:
            image.write(web.read())

if __name__ == '__main__':
    for i in range(1, 6):
        html = page(i)
        html = html.decode('utf-8')
        title_l = title(html)  # 图片名称
        content_l = content(html)
        for i, z in zip(title_l, content_l):  # .itervalues():
            download(z,i)
            print(i, z)

  首先导入的库就不一样，在Python3.X中，把urllib和urllib2统一合并到urllib中。

下载功能的函数我是这样的写的，不知道还有没有其它写法。

def download(url,name):
    path = 'H:\image\%s.jpg'%name.decode('utf-8').encode('gbk')
    urllib.urlretrieve(url,path)

html = html.decode('utf-8')#python3.x，这里也不一样。python3需要.decode('utf-8')一下。