程式碼已同步到GitHub中，以後會陸陸續續整理之前的程式碼，放在GitHub共享，歡迎圍觀。
qingmm的GitHub

百科詞條資料整體較為規範，至少在網頁上看起來是這樣。但實際抓取時可以發現正文內容不論標題還是內容都在同一級下，無法直接獲取到某一級標題下的所有相對應的內容，因此需要巧妙地設計程式碼來自動識別多級標題，自動將與標題相對應的內容存放在該標題下。
目前網路上抓取百度百科詞條資料的程式碼大都是來自於同一個視訊教程，雖然將功能分割，寫了五個程式碼檔案，但仍不能滿足實際需求，教學作用大於實際作用。因此專門研究了下百科詞條的html頁面的程式碼規則，最終通過一定的規則實現了對頁面內容的自動規則化獲取。

輸入為詞條名稱，輸出為json檔案，資料按照字典dict格式儲存。
下面是程式碼：

#-*-coding:utf-8-*-
import re
import requests
import bs4
from bs4 import BeautifulSoup
import json
import codecs
import sys
import os

path = sys.path[0] + os.sep
headers = {
    "Accept":"text/html,application/xhtml+xml,application/xml;q=0.9,image/webp,image/apng,*/*;q=0.8"
 
,
    "Accept-Encoding":"gzip, deflate, br",
    "Accept-Language":"zh-CN,zh;q=0.9,en;q=0.8,zh-HK;q=0.7,en-US;q=0.6",
    "Connection":"keep-alive",
    "Cookie":"BAIDUID=12D740BD92DEA90B607F5B827987F30E:FG=1; BIDUPSID=12D740BD92DEA90B607F5B827987F30E; PSTM=1534166632; BKWPF=3; BDUSS=lleW52cG9MalVYcUhKeWJSYllpMlgzQXpnN2lORml-UXh3b1BqRGpqSnBtcVJiQVFBQUFBJCQAAAAAAAAAAAEAAAARJts6wu3D98flt-cAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAGkNfVtpDX1bT1; PSINO=1; H_PS_PSSID=1447_21105_20882_26350_26924_20927; BDORZ=B490B5EBF6F3CD402E515D22BCDA1598; pgv_pvi=419963904; pgv_si=s2644193280; Hm_lvt_55b574651fcae74b0a9f1cf9c8d7c93a=1534920932,1535362634,1535362645,1535362662; Hm_lpvt_55b574651fcae74b0a9f1cf9c8d7c93a=1535362662"
 
,
    "User-Agent":"Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/67.0.3396.99 Safari/537.36", 
    "Host": 'baike.baidu.com',
    "Upgrade-Insecure-Requests":"1"
}

def craw(url, item):
    html = requests.get(url, headers=headers).content
    data = dict()
    data['url'] = url
    data['name'] = item
    soup = BeautifulSoup(html, 'lxml')
    title = soup.find('h2').get_text()
    data['abstract'] = soup.find('div', class_='lemma-summary').get_text().strip().replace('\n', '').replace('\t', '')
    basic_info = soup.find('div', class_='basic-info')
    dts = basic_info.find_all('dt', class_='name')
    dds = basic_info.find_all('dd', class_='value')
    data['basic_info'] = dict()
    for i in range(len(dts)):
        name = dts[i].get_text().strip().replace('\n', '').replace('\t', '')
        value = dds[i].get_text().strip().replace('\n', '').replace('\t', '')
        data['basic_info'][name] = value
    paras = soup.find_all('div', class_=['para-title', 'para'])
    content = dict()
    # move cursor to div: para-title level-2
    for i in range(len(paras)):
        if 'level-2' in paras[i]['class']:
            paras = paras[i:]
            break
    level3_flag = False
    # traversal content, caution: there is level-3 para, so the code will be more complicate 
    for para in paras:
        if 'level-2' in para['class']:
            prefix = para.span.get_text().strip().replace('\n', '')
            name = para.h2.get_text().strip().replace('\n', '').replace(prefix, '')
            print 'name', name
            content[name] = ''
            level3_flag = False
        elif 'level-3' in para['class']:
            if not level3_flag:
                content[name] = dict()
            prefix = para.span.get_text().strip().replace('\n', '')
            children = para.h3.get_text().strip().replace('\n', '').replace(prefix, '')
            print 'children', children
            content[name][children] = ''
            level3_flag = True
        else:
            text = para.get_text().strip().replace('\n', '').replace('\t', '')
            if level3_flag:
                content[name][children] += text
            else:
                content[name] += text
    data['content'] = content
    f = codecs.open(path + 'baike.json', 'w', 'utf-8')
    json.dump(data, f, ensure_ascii=False)
    f.write('\n')
    f.close()

if __name__ == '__main__':
    baseurl = 'http://baike.baidu.com/item/'
    # items = ['Python', u'北京市', u'朝陽區']
    items = [u'北京市']
    for item in items:
        url = baseurl + item
        print url
        craw(url, item)

以上，歡迎交流。