对于政府网站下发的文件进行爬取,减少人去下载的过程

博问上有人不会,我写了一下

绝对不要加多线程多线程进去

import re

import requests
from lxml.html import etree

url = 'http://www.liyang.gov.cn/default.php?mod=article&fid=163250&s99679207_start=0'
rp = requests.get(url)
re_html = etree.HTML(rp.text)
url_xpath = '//*[@id="s99679207_content"]/table/tbody/tr/td/span[1]/span/a/@href'
title_xpath = '//*[@id="s99679207_content"]/table/tbody/tr/td/span[1]/span/a/text()'
url_list = re_html.xpath(url_xpath)
title_list = re_html.xpath(title_xpath)
title_list = title_list[::-1]
data_url_list = []
for url_end in url_list:
    new_url = f'http://www.liyang.gov.cn/{url_end}'
    print(new_url)
    rp_1 = requests.get(new_url)
    print(rp_1.text)
    try:
        re_1_html = etree.HTML(rp_1.text)
        data_url_xpth = '//tbody/tr[1]/td[2]/a'
        data_url = re_1_html.xpath(data_url_xpth)[0]
    except:
        data_list = re.findall('<a href="(.*?)" target="_blank">', rp_1.text)
        data_url = data_list[0]
    print(data_url)
    data_url = f'http://www.liyang.gov.cn/{data_url}'
    re = requests.get(data_url)
    data = re.content
    with open(f'{title_list.pop()}.pdf', 'wb') as fw:
        fw.write(data)