使用pyspider框架抓取貓途鷹旅遊資訊
阿新 • • 發佈:2018-11-21
這裡通過pyspider框架
可以直接:pip3 install pyspider 下載框架
pyspider all 執行 可以看到 run 0.0.0.0:5000
直接在瀏覽器輸入localhost:5000 進入 建立新專案
#!/usr/bin/env python # -*- encoding: utf-8 -*- # Created on 2018-08-03 09:59:05 # Project: TripAdcisor from pyspider.libs.base_handler import * #import ssl import pymongo #ssl._create_default_https_context=ssl._create_unverified_context class Handler(BaseHandler): crawl_config = { } conn = pymongo.MongoClient('localhost') db = conn.trip myset = db.london @every(minutes=24 * 60) #on_start 啟動目標主網站,validate_cert = False跳過證書檢測,callback回撥函式 def on_start(self): self.crawl('https://www.tripadvisor.cn/Attractions-g186338-Activities-c47-London_England.html', callback=self.index_page,validate_cert = False) @config(age=10 * 24 * 60 * 60) #通過內建pyquery 獲取目標網頁的連線遍歷 分別訪問,並回調 def index_page(self, response): for each in response.doc('.listing_title > a').items(): self.crawl(each.attr.href, callback=self.detail_page,validate_cert = False) #獲取下一頁連結地址訪問並回調自身 next = response.doc('#FILTERED_LIST > div.al_border.deckTools.btm > div > div > a').attr.href print(next) self.crawl(next,callback= self.index_page,validate_cert=False) @config(priority=2) #獲得每個連結的的詳細資訊返回 def detail_page(self, response): url = response.url name = response.doc('.heading_title').text() rating = response.doc('div.rating > .more').text() location = response.doc('#taplc_attraction_detail_listing_0 > div.section.location > div.detail_section.address').text()[2:] phone = response.doc('.phone > div').text() durtion = response.doc('.hours > .duration').text() return { 'name':name, 'rating':rating, 'location':location, 'phone':phone, 'durtion':durtion, 'url':url } #重寫on_result方法 儲存至資料庫 def on_result(self,result): if result: self.save_to_mongo(result) def save_to_mongo(self,result): if self.myset.insert(result): print('save ok!',result)
執行時將status 調至running 或者debug run即可