做个实习僧爬虫，自动筛选信息

主要是实习僧的反爬是通过字符编码加密进行的，要进行字体解析
话不多说 上代码 希望给你们有建议

import requests
import parsel
from urllib.parse import urljoin
from urllib.request import urlretrieve
from urllib.parse import quote
from io import BytesIO
from fontTools.ttLib import TTFont

import time



COUNT= 0

base_url = 'https://www.shixiseng.com/'
index_url = 'https://www.shixiseng.com/interns?page={page}&type=intern&keyword={keyword}&area=&months=&days=&degree=&official=&enterprise=&salary=-0&publishTime=&sortType=&city={city}&internExtend='

headers = {
    'user-agent': 'Mozilla/5.0 (Windows NT 6.2; WOW64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/27.0.1453.94 Safari/537.36',
}


def get_hex_charMap(bytesIo: BytesIO):
    """
    获得字体与字符的映射
    注意此处映射关系有规律：字形编码即为真实字符的unicode码
    """
    font = TTFont(bytesIo)
    # getBestCmap()获取字体中自定义unicode字符的{unicode码的10进制: 字形编码}
    cMap = font.getBestCmap()
    hex_charMap = {chr(key): chr(int(cMap[key][3:], 16)) for index, key in enumerate(cMap) if index > 0}
    return hex_charMap


def get_real_string(unicode_string, hex_charMap):
    """抓取的数据调用此方法获取真正的字符串"""
    if unicode_string is None:
        return ''
    final_string = ''
    for char in unicode_string:
        tmp = hex_charMap.get(char)
        if tmp:
            final_string += tmp
        else:
            final_string += char
    return final_string





def get_content(url):
    r = requests.get(url, headers=headers)
    assert r.status_code == 200
    return r.content


def get_html(url):
    content = get_content(url)
    return content.decode('utf8')


def run(keyword, city="全国"):
    global COUNT
    page = 1
    while True:
        url = index_url.format(page=page, keyword=quote(keyword), city=quote(city))
        html = get_html(url)

        sel = parsel.Selector(html)
        # 获取字体的下载链接
        font_url = sel.re_first(r'@font-face.*?src: url\((.*?)\)')
        font_url = urljoin(base_url, font_url)
        # urlretrieve(font_url, "实习僧字体.woff")
        content = get_content(font_url)
        hex_charMap = get_hex_charMap(BytesIO(content))
        # pprint(hex_charMap)

        items = sel.css('.intern-wrap.intern-item')
        for item in items:
            data = {}
            data['title'] = item.css('.intern-detail__job>p>a::text').get()
            job_chunk = item.css('.intern-detail__job>p>span::text').extract()
            data['salary_tip'] = ', '.join(job_chunk).replace('|, ', '')
            data['company'] = item.css('.intern-detail__company>p>a::attr(title)').get()
            company_chunk = item.css('.intern-detail__company>p>span::text').extract()
            data['company_tip'] = ''.join(company_chunk)
            job_benefits = item.css('.advantage-wrap>.f-l>span::text').extract()
            data['job_benefits'] = ', '.join(job_benefits)
            data['company_desc'] = item.css('.advantage-wrap>.f-r>span::text').get()
            # 替换真实字符串
            data = {key: get_real_string(value, hex_charMap) for key, value in data.items()}
            
            print(data)
            COUNT += 1

        if sel.css('.btn-next::attr(disabled)').get() is None:
            page += 1
        else:
            break


def main():
    try:
        # 输入你想抓取的职位名称，如Java,Python
        keyword = '数据'
        # 输入你想查找的范围，如北京、上海等，默认是全国
        city = ''
        run(keyword, city)
        print(f'共抓取{COUNT}条数据')
    finally:
        print('..')
        


if __name__ == '__main__':
    start = time.time()
    main()
    end = time.time()
    print(f"耗时{end-start} s")

写在最后

四@@时 学习，中国大学生在线啊，大家加油学习，可以交流

加油学习