免費學(xué)習(xí)推薦：python視頻教程

三種數(shù)據(jù)抓取的方法

1. 正則表達式（re庫）
2. BeautifulSoup（bs4）
3. lxml

*利用之前構(gòu)建的下載網(wǎng)頁函數(shù)，獲取目標(biāo)網(wǎng)頁的html，我們以https://guojiadiqu.bmcx.com/AFG__guojiayudiqu/為例，獲取html。

from get_html import download  url = 'https://guojiadiqu.bmcx.com/AFG__guojiayudiqu/'page_content = download(url)

*假設(shè)我們需要爬取該網(wǎng)頁中的國家名稱和概況，我們依次使用這三種數(shù)據(jù)抓取的方法實現(xiàn)數(shù)據(jù)抓取。
1.正則表達式

from get_html import downloadimport re  url = 'https://guojiadiqu.bmcx.com/AFG__guojiayudiqu/'page_content = download(url)country = re.findall('class="h2dabiaoti">(.*?)</h2>', page_content) #注意返回的是listsurvey_data = re.findall('<tr><td bgcolor="#FFFFFF" id="wzneirong">(.*?)</td></tr>', page_content)survey_info_list = re.findall('<p>　　(.*?)</p>', survey_data[0])survey_info = ''.join(survey_info_list)print(country[0],survey_info)

2.BeautifulSoup（bs4）

from get_html import downloadfrom bs4 import BeautifulSoup  url = 'https://guojiadiqu.bmcx.com/AFG__guojiayudiqu/'html = download(url)#創(chuàng)建 beautifulsoup 對象soup = BeautifulSoup(html,"html.parser")#搜索country = soup.find(attrs={'class':'h2dabiaoti'}).text survey_info = soup.find(attrs={'id':'wzneirong'}).textprint(country,survey_info)

3.lxml

from get_html import downloadfrom lxml import etree #解析樹url = 'https://guojiadiqu.bmcx.com/AFG__guojiayudiqu/'page_content = download(url)selector = etree.HTML(page_content)#可進行xpath解析country_select = selector.xpath('//*[@id="main_content"]/h2') #返回列表for country in country_select:     print(country.text)survey_select = selector.xpath('//*[@id="wzneirong"]/p')for survey_content in survey_select:     print(survey_content.text,end='')

運行結(jié)果：

最后，引用《用python寫網(wǎng)絡(luò)爬蟲》中對三種方法的性能對比，如下圖：

僅供參考。

相關(guān)免費學(xué)習(xí)推薦：python教程（視頻）