網絡爬蟲（爬取網站圖片，自動保存本地）

事先申明一點，這個人品沒有什么問題，只是朋友發一段python源碼，再這里分享大家。

import requests
from lxml import html
import os
from multiprocessing.dummy import Pool as ThreadPool

def header(referer):
    headers = {
        'Host': 'i.meizitu.net',
        'Pragma': 'no-cache',
        'Accept-Encoding': 'gzip, deflate',
        'Accept-Language': 'zh-CN,zh;q=0.8,en;q=0.6',
        'Cache-Control': 'no-cache',
        'Connection': 'keep-alive',
        'User-Agent': 'Mozilla/5.0 (Macintosh; Intel Mac OS X 10_12_5) AppleWebKit/537.36 (KHTML, like Gecko) '
                      'Chrome/59.0.3071.115 Safari/537.36',
        'Accept': 'image/webp,image/apng,image/*,*/*;q=0.8',
        'Referer': '{}'.format(referer),
    }
    return headers

# 獲取主頁列表
def getPage(pageNum):
    baseUrl = 'http://www.mzitu.com/page/{}'.format(pageNum)
    selector = html.fromstring(requests.get(baseUrl).content)
    urls = []
    for i in selector.xpath('//ul[@id="pins"]/li/a/@href'):
        urls.append(i)
        print(i)
    return urls


# 圖片鏈接列表， 標題
# url是詳情頁鏈接
def getPiclink(url):
    sel = html.fromstring(requests.get(url).content)
    # 圖片總數
    total = sel.xpath('//div[@class="pagenavi"]/a[last()-1]/span/text()')[0]
    # 標題
    title = sel.xpath('//h2[@class="main-title"]/text()')[0]
    # 文件夾格式
    dirName = u"【{}P】{}".format(total, title)
    # 新建文件夾
    os.mkdir(dirName)

    n = 1
    for i in range(int(total)):
        # 每一頁
        try:
            link = '{}/{}'.format(url, i+1)
            s = html.fromstring(requests.get(link).content)
            # 圖片地址在src標簽中
            jpgLink = s.xpath('//div[@class="main-image"]/p/a/img/@src')[0]
            # print(jpgLink)
            # 文件寫入的名稱：當前路徑／文件夾／文件名
            filename = '%s/%s/%s.jpg' % (os.path.abspath('.'), dirName, n)
            print(u'開始下載圖片:%s 第%s張' % (dirName, n))
            with open(filename, "wb+") as jpg:
                jpg.write(requests.get(jpgLink, headers=header(jpgLink)).content)
            n += 1
        except:
            pass


if __name__ == '__main__':
    pageNum = input(u'請輸入頁碼：')
    p = getPage(pageNum)
    with ThreadPool(4) as pool:
        pool.map(getPiclink, p)    

至於爬出出來的效果圖，我就不發布了