用python爬取去哪兒游記攻略為十月假期做準備,,,爬蟲之路,永無止境!
熱熱鬧鬧的開學季又來了,小伙伴們又可以在一起玩耍了,不對是在一起學習了,哈哈,
再過幾周就是國慶假期,想想還是很激動的,我決定給大家做個游記爬蟲,大家早做準備,,嘿嘿
很多人學習python,不知道從何學起, 很多人學習python,掌握了基本語法過后,不知道在哪里尋找案例上手, 很多已經做案例的人,卻不知道如何去學習更加高深的知識, 那么針對這三類人,我給大家提供一個好的學習平臺,免費領取視頻教程,電子書籍,以及課程的源代碼! QQ群:701698587 歡迎加入,一起討論 一起學習!
代碼操作展示:
今天目標地址:
https://travel.qunar.com/place/
開發環境:
windows10
python3.6
開發工具:
pycharm
庫:
tkinter、re、os、lxml、threading、xlwt、xlrd
1.首先先將全國所有的城市名稱和id拿到
2.右擊檢查,進行抓包,找到資料所在的包
3.發送請求,獲取回應,決議回應
# 發送請求,獲取回應,決議回應 response = session.get(self.start_url, headers=self.headers).html # 提取所有目的地(城市)的url city_url_list = response.xpath( '//*[@id="js_destination_recommend"]/div[2]/div[1]/div[2]/dl/dd/div/ul/li/a/@href') city_id_list = [''.join(re.findall(r'-cs(.*?)-', i)) for i in city_url_list] # 提取所有的城市名稱 city_name_list = response.xpath( '//*[@id="js_destination_recommend"]/div[2]/div[1]/div[2]/dl/dd/div/ul/li/a/text()')
4.隨機點一個城市,進入該城市,查看游記攻略,本文選的是上海
5.進行抓包,查找需要的資訊
# 提取游記作者 author_list = html.xpath('//span[@]/a/text()') # 出發時間 date_list = html.xpath('//span[@]/text()') # 游玩時間 days_list = html.xpath('//span[@]/text()') # 閱讀量 read_list = html.xpath('//span[@]/span/text()') # 點贊量 like_count_list = html.xpath('//span[@]/span/text()') # 評論量 icon_list = html.xpath('//span[@]/span/text()') # 游記地址 text_url_list = html.xpath('//h3[@]/a/@href')
6.進行翻頁抓包,第二頁為異步加載
7.在上一張圖中點擊Headers,獲取游記攻略的請求地址
這樣思路是不是清晰一點,來觀察一下請求地址
https://travel.qunar.com/place/api/html/books/dist/299878?sortField=0&pageSize=10&page=2
其中299878是上海城市的id page是頁數
原始碼展示:
# !/usr/bin/nev python # -*-coding:utf8-*- import tkinter as tk import re from lxml import etree import tkinter.messagebox as msgbox from requests_html import HTMLSession from threading import Thread import os, xlwt, xlrd from xlutils.copy import copy session = HTMLSession() class TKSpider(object): def __init__(self): # 定義回圈條件 self.is_running = True # 定義起始的頁碼 self.start_page = 1 """定義可視化視窗,并設定視窗和主題大小布局""" self.window = tk.Tk() self.window.title('爬蟲資料采集') self.window.geometry('1000x800') """創建label_user按鈕,與說明書""" self.label_user = tk.Label(self.window, text='請選擇城市輸入序號:', font=('Arial', 12), width=30, height=2) self.label_user.pack() """創建label_user關聯輸入""" self.entry_user = tk.Entry(self.window, show=None, font=('Arial', 14)) self.entry_user.pack(after=self.label_user) """創建Text富文本框,用于按鈕操作結果的展示""" # 定義富文本框滑動條 scroll = tk.Scrollbar() # 放到視窗的右側, 填充Y豎直方向 scroll.pack(side=tk.RIGHT, fill=tk.Y) self.text1 = tk.Text(self.window, font=('Arial', 12), width=75, height=20) self.text1.pack() # 兩個控制元件關聯 scroll.config(command=self.text1.yview) self.text1.config(yscrollcommand=scroll.set) """定義按鈕1,系結觸發事件方法""" self.button_1 = tk.Button(self.window, text='下載游記', font=('Arial', 12), width=10, height=1, command=self.parse_hit_click_1) self.button_1.pack(before=self.text1) """定義按鈕2,系結觸發事件方法""" self.button_2 = tk.Button(self.window, text='清除', font=('Arial', 12), width=10, height=1, command=self.parse_hit_click_2) self.button_2.pack(anchor="e") """定義富文本文字的顯示""" self.start_url = 'https://travel.qunar.com/place/' self.headers = { 'user-agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/92.0.4515.159 Safari/537.36' } # 發送請求,獲取回應,決議回應 response = session.get(self.start_url, headers=self.headers).html # 提取所有目的地(城市)的url city_url_list = response.xpath( '//*[@id="js_destination_recommend"]/div[2]/div[1]/div[2]/dl/dd/div/ul/li/a/@href') city_id_list = [''.join(re.findall(r'-cs(.*?)-', i)) for i in city_url_list] # 提取所有的城市名稱 city_name_list = response.xpath( '//*[@id="js_destination_recommend"]/div[2]/div[1]/div[2]/dl/dd/div/ul/li/a/text()') self.dict_data = dict(zip(city_id_list, city_name_list)) text_info = '' for city_id, city_name in zip(city_id_list, city_name_list): text_info += city_id + ":" + city_name + '\t\t' self.text1.insert("insert", text_info) def parse_hit_click_1(self): """執行緒關聯:定義觸發事件1, 將執行結果顯示在文本框中""" Thread(target=self.parse_start_url_job).start() def parse_city_id_name_info(self): """ 富文本內容展示: :return: """ # 發送請求,獲取回應,決議回應 response = session.get(self.start_url, headers=self.headers).html # 提取所有目的地(城市)的url city_url_list = response.xpath( '//*[@id="js_destination_recommend"]/div[2]/div[1]/div[2]/dl/dd/div/ul/li/a/@href') city_id_list = [''.join(re.findall(r'-cs(.*?)-', i)) for i in city_url_list] # 提取所有的城市名稱 city_name_list = response.xpath( '//*[@id="js_destination_recommend"]/div[2]/div[1]/div[2]/dl/dd/div/ul/li/a/text()') self.dict_data = dict(zip(city_id_list, city_name_list)) text_info = '' for city_id, city_name in zip(city_id_list, city_name_list): text_info += city_id + ":" + city_name + '\t\t****' self.text1.insert("insert", text_info) def parse_start_url_job(self): # 從輸入視窗獲取輸入 city_id = self.entry_user.get() self.text1.delete("1.0", "end") # 例外捕捉,是否存在key值 try: city_name = self.dict_data[city_id] self.parse_request_yj(city_id, city_name) except: # 報錯提示 msgbox.showerror(title='錯誤', message='檢測到瞎搞,請重新輸入!') def parse_request_yj(self, city_id, city_name): """ 拼接游記地址,進行翻頁 :param city_id: 城市的id :param city_name: 城市的名稱 :return: """ while self.is_running: yj_url = f'https://travel.qunar.com/place/api/html/books/dist/{city_id}?sortField=0&pageSize=10&page={self.start_page}' response = session.get(yj_url, headers=self.headers).json()['data'].replace('\\', '') html = etree.HTML(response) # 提取游記的標題: title_list = html.xpath('//h3[@]/a/text()') # 死回圈的終止條件 if not title_list: print("該城市已經采集到最后一頁----------城市切換中---------logging!!!") break else: # 提取游記作者 author_list = html.xpath('//span[@]/a/text()') # 出發時間 date_list = html.xpath('//span[@]/text()') # 游玩時間 days_list = html.xpath('//span[@]/text()') # 閱讀量 read_list = html.xpath('//span[@]/span/text()') # 點贊量 like_count_list = html.xpath('//span[@]/span/text()') # 評論量 icon_list = html.xpath('//span[@]/span/text()') # 游記地址 text_url_list = html.xpath('//h3[@]/a/@href') """構造保存資料格式字典""" for a, b, c, d, e, f, g, h in zip(title_list, author_list, date_list, days_list, read_list, like_count_list, icon_list, text_url_list): dict_dd = { city_name: [a, b, c, d, e, f, g, h] } self.text1.insert("insert", f"{a}-----------采集完成!!!" + '\n') self.SaveExcels(dict_dd) self.start_page += 1 def SaveExcels(self, data): # 獲取表的名稱 sheet_name = [i for i in data.keys()][0] # 創建保存excel表格的檔案夾 # os.getcwd() 獲取當前檔案路徑 os_mkdir_path = os.getcwd() + '/資料/' # 判斷這個路徑是否存在,不存在就創建 if not os.path.exists(os_mkdir_path): os.mkdir(os_mkdir_path) # 判斷excel表格是否存在 作業簿檔案名稱 os_excel_path = os_mkdir_path + '資料.xls' if not os.path.exists(os_excel_path): # 不存在,創建作業簿(也就是創建excel表格) workbook = xlwt.Workbook(encoding='utf-8') """作業簿中創建新的sheet表""" # 設定表名 worksheet1 = workbook.add_sheet(sheet_name, cell_overwrite_ok=True) """設定sheet表的表頭""" sheet1_headers = ('游記標題', '發布游記的作者名稱', '出發時間', '游玩時間', '閱讀量', '點贊量', '評論量', '游記地址') # 將表頭寫入作業簿 for header_num in range(0, len(sheet1_headers)): # 設定表格長度 worksheet1.col(header_num).width = 2560 * 3 # 寫入表頭 行, 列, 內容 worksheet1.write(0, header_num, sheet1_headers[header_num]) # 回圈結束,代表表頭寫入完成,保存作業簿 workbook.save(os_excel_path) """=============================已有作業簿添加新表===============================================""" # 打開作業薄 workbook = xlrd.open_workbook(os_excel_path) # 獲取作業薄中所有表的名稱 sheets_list = workbook.sheet_names() # 如果表名稱:字典的key值不在作業簿的表名串列中 if sheet_name not in sheets_list: # 復制先有作業簿物件 work = copy(workbook) # 通過復制過來的作業簿物件,創建新表 -- 保留原有表結構 sh = work.add_sheet(sheet_name) # 給新表設定表頭 excel_headers_tuple = ('游記標題', '發布游記的作者名稱', '出發時間', '游玩時間', '閱讀量', '點贊量', '評論量', '游記地址') for head_num in range(0, len(excel_headers_tuple)): sh.col(head_num).width = 2560 * 3 # 行,列, 內容, 樣式 sh.write(0, head_num, excel_headers_tuple[head_num]) work.save(os_excel_path) """=========================================================================================""" # 判斷作業簿是否存在 if os.path.exists(os_excel_path): # 打開作業簿 workbook = xlrd.open_workbook(os_excel_path) # 獲取作業薄中所有表的個數 sheets = workbook.sheet_names() for i in range(len(sheets)): for name in data.keys(): worksheet = workbook.sheet_by_name(sheets[i]) # 獲取作業薄中所有表中的表名與資料名對比 if worksheet.name == name: # 獲取表中已存在的行數 rows_old = worksheet.nrows # 將xlrd物件拷貝轉化為xlwt物件 new_workbook = copy(workbook) # 獲取轉化后的作業薄中的第i張表 new_worksheet = new_workbook.get_sheet(i) for num in range(0, len(data[name])): new_worksheet.write(rows_old, num, data[name][num]) new_workbook.save(os_excel_path) def parse_hit_click_2(self): """定義觸發事件2,洗掉文本框中內容""" self.entry_user.delete(0, "end") # self.entry_passwd.delete(0, "end") self.text1.delete("1.0", "end") self.parse_city_id_name_info() def center(self): """創建視窗居中函式方法""" ws = self.window.winfo_screenwidth() hs = self.window.winfo_screenheight() x = int((ws / 2) - (800 / 2)) y = int((hs / 2) - (600 / 2)) self.window.geometry('{}x{}+{}+{}'.format(800, 600, x, y)) def run_loop(self): """禁止修改表單大小規格""" self.window.resizable(False, False) """視窗居中""" self.center() """視窗維持--持久化""" self.window.mainloop() if __name__ == '__main__': t = TKSpider() t.run_loop()
轉載請註明出處,本文鏈接:https://www.uj5u.com/houduan/302651.html
標籤:其他
上一篇:Spring Boot 2.5.5 發布,2.6.0 也要來了!
下一篇:java基礎之AQS
