Python 百度百科关键字定向爬取

输入关键字可以查找与之相关的关键字。
比较简单的spiders，望大佬指教。(●’◡’●)

import requests, re
from bs4 import BeautifulSoup

def getHTMLText(url):
    try:
        kv = {"User-Agent": "Mozilla/5.0"}
        r = requests.get(url, headers=kv)
        r.encoding = 'utf-8'
        return r.text
    except:
        return ''

def parse(ilt, html):
    soup = BeautifulSoup(html, 'html.parser')
    for div in soup.find_all('div', class_='para'):
        for a in div.find_all('a', href=re.compile(r'/item/.')):
            ilt.append([a.attrs['href'], a.string])

def printInfo(ilt):
    start_url = 'https://baike.baidu.com'
    ans = 1
    for i in ilt:
        print("key word {0:<3}:{1:<20}\t{2:<}".format(str(ans), str(i[1]), str(start_url+i[0])))
        ans += 1

def main():
    key = input('输入关键字：')
    url = 'https://baike.baidu.com/item/' + key
    infoList = []
    html = getHTMLText(url)
    parse(infoList, html)
    printInfo(infoList)

main()