import requestsimport parselimport csvif __name__ == "__main__":    csv_qne  = open('去哪儿旅行.csv', mode='a', encoding='utf-8-sig', newline='')    csv_writer = csv.writer(csv_qne)    csv_writer.writerow(['标题', '短评', '浏览量', '天数', '人物', '人均费用', '玩法'])    for page in range(1,100):        # 获取所有页面详情页URL        url = f'https://travel.qunar.com/travelbook/list.htm?page={page}&order=hot_heat'        response = requests.get(url)        html_data = response.text        selector = parsel.Selector(html_data)        # 提取每一页的url 可以使用bs4、正则表达式、xpath等 在这里我们使用css选择器        url_list = selector.css('body > div.qn_mainbox > div > div.left_bar > ul > li > h2 > a::attr(href)').getall()        for detail_list in url_list :            detail_id = detail_list.replace('/youji/','')            detail_url = 'https://travel.qunar.com/travelbook/note/' + detail_id            # 获取每一详情页的所有信息            response_1 = requests.get(detail_url)            data_html_1 = response_1.text            # 提取所需要的数据 包括出发日期 天数 人均费用 人物 玩法 地点 浏览量等等            selector_1 = parsel.Selector(data_html_1)            #城市            title = selector_1.css('.b_crumb_cont *:nth-child(3)::text').get()            #短评            comment = selector_1.css('.title.white::text').get()            #浏览量            count = selector_1.css('.view_count::text').get()            # 天数            days = selector_1.css('#js_mainleft > div.b_foreword > ul > li.f_item.howlong  > p > span.data::text').get()            # 人物            people = selector_1.css('#js_mainleft > div.b_foreword > ul > li.f_item.who > p > span.data::text').get()            #人均费用            money = selector_1.css('#js_mainleft > div.b_foreword > ul > li.f_item.howmuch > p > span.data::text').get()            #玩法            play_list = selector_1.css('#js_mainleft > div.b_foreword > ul > li.f_item.how > p > span.data span::text').getall()            play = '、'.join(play_list)            print(title, comment, count, days, people, money, play)            csv_writer.writerow([title, comment, count, days, people, money, play])    csv_qne.close()    #删除缺失值    import pandas as pd    data = pd.read_csv("去哪儿旅行.csv")    res = data.dropna(how="all")    res.to_csv("去哪儿1.csv", index=False)

import pandas as pdimport numpy as np#读取文件df1 = pd.read_csv("去哪儿1.csv")df2 = pd.read_csv("去哪儿2.csv")df3 = pd.read_csv("去哪儿3.csv")df4 = pd.read_csv("去哪儿4.csv")df5 = pd.read_csv("去哪儿5.csv")#合并df = pd.concat([df1,df2,df3,df4,df5])df.drop_duplicates()  #数据去重#保存合并后的文件df.to_csv('去哪1-5.csv',encoding = 'utf-8-sig')

import pandas as pdimport retitle = pd.read_csv('去哪1-5csv')#城市名字提取title_info = title['标题']title_list = title_info.values.tolist()new_title_list = []for data in title_list:    if re.match(r'.*旅游攻略',data) != None:        new_title_list.append(data)title_str = " ".join(new_title_list)print(title_str)city_list = re.findall(r'(.*?)旅游攻略',title_str)print(city_list)# 进行词频统计import jiebacity_dic = {}for city in city_list:    city_dic[city]  = city_dic.get(city,0) + 1print(city_dic)#对前三十名城市进行排序sort_data = sorted(city_dic.items(),key=lambda a:a[1],reverse=True)[:30]print(sort_data)data1 = dict(sort_data)print(data1)for a,b in data1.items():    print(str(a) + ":"+str(b))

import wordcloudimages = Image.open("picture.jpg")maskImages = np.array(images)wc = wordcloud.WordCloud(width=1500, height=1500, background_color='white',font_path='C:\\Windows\\Fonts\\STFANGSO.ttf',mask=maskImages,scale=25)wc.generate_from_frequencies(data1) wc.to_file("city.jpg")