【爬蟲】獲取淘寶天貓商品的詳細引數
阿新 • • 發佈:2019-02-07
首先我是從淘寶進去,爬取了按銷量排序的所有(100頁)女裝的列表資訊按綜合、銷量分別爬取淘寶女裝列表資訊,然後匯出前100商品的 link,爬取其詳細資訊。這些商品有淘寶的,也有天貓的,這兩個平臺有些區別,處理的時候要注意。比如,有的說“面料”、有的說“材質成分”,其實是一個意思,等等。可以取不同的連結做一下測試。
import re from collections import OrderedDict from bs4 import BeautifulSoup from pyquery import PyQuery as pq #獲取整個網頁的原始碼 from config import * #可引用congif的所有變數 import pymysql import urllib import json import bs4 import requests from selenium import webdriver from selenium.webdriver.support.ui import WebDriverWait from pyquery import PyQuery as pq #獲取整個網頁的原始碼 import pandas as pd # 測試 淘寶+天貓,可完整輸出及儲存 browser = webdriver.Firefox() wait = WebDriverWait(browser,10) ####### 天貓上半部分詳情 ############# def get_tianmao_header(url): browser.get(url) # wait.until(EC.presence_of_all_elements_located((By.CSS_SELECTOR,'#mainsrp-itemlist .items .item'))) #載入所有寶貝 html=browser.page_source doc = pq(html) # print(doc) info = OrderedDict() # 存放該商品所具有的全部資訊 items = doc('#page') # info['店鋪名'] = items.find('.slogo').find('.slogo-shopname').text() # info['ID'] = items.find('#LineZing').attr['itemid'] info['寶貝'] = items.find('.tb-detail-hd').find('h1').text() info['促銷價'] = items.find('#J_PromoPrice').find('.tm-promo-price').find('.tm-price').text() info['原價'] = items.find('#J_StrPriceModBox').find('.tm-price').text() # '月銷量' :items.find('.tm-ind-panel').find('.tm-ind-item tm-ind-sellCount').find('.tm-indcon').find('.tm-count').text(), info['月銷量'] = items.find('.tm-ind-panel').find('.tm-indcon').find('.tm-count').text().split(' ',2)[0] info['累計評價'] = items.find('#J_ItemRates').find('.tm-indcon').find('.tm-count').text() # print(info) return info ######## 淘寶上半部分詳情 ############### def get_taobao_header(url): browser.get(url) # wait.until(EC.presence_of_all_elements_located((By.CSS_SELECTOR,'#mainsrp-itemlist .items .item'))) #載入所有寶貝 html=browser.page_source doc = pq(html) # print(doc) info = OrderedDict() # 存放該商品所具有的全部資訊 items = doc('#page') # info['店鋪名'] = items.find('.tb-shop-seller').find('.tb-seller-name').text() # info['ID'] = items.find('#J_Pine').attr['data-itemid'] info['寶貝'] = items.find('#J_Title').find('h3').text() info['原價'] = items.find('#J_StrPrice').find('.tb-rmb-num').text() info['促銷價'] = items.find('#J_PromoPriceNum').text() # '月銷量' :items.find('.tm-ind-panel').find('.tm-ind-item tm-ind-sellCount').find('.tm-indcon').find('.tm-count').text(), info['月銷量'] = items.find('#J_SellCounter').text() info['累計評價'] = items.find('#J_RateCounter').text() # print(info) return info ####################### 詳情 ############################ # 抓取所有商品詳情 def get_Details(attrs,info): # res = requests.get(url) # soup = BeautifulSoup(res.text, "html.parser") # # attrs = soup.select('.attributes-list li') # attrs= [<li title=" 薄">厚薄: 薄</li>, <li title=" 其他100%">材質成分: 其他100%</li>,<li ...</li>] attrs_name = [] attrs_value = [] ''' [\s] 匹配空格,[\s]*,後面有 *,則可以為空 * : 匹配前面的子表示式任意次 ''' for attr in attrs: attrs_name.append(re.search(r'(.*?):[\s]*(.*)', attr.text).group(1)) attrs_value.append(re.search(r'(.*?):[\s]*(.*)', attr.text).group(2)) # print('attrs_name=',attrs_name) # attrs_name= ['厚薄', '材質成分', ...] # print('attrs_value=',attrs_value) # attrs_value= ['薄', '其他100%', ...] allattrs = OrderedDict() # 存放該產品詳情頁面所具有的屬性 for k in range(0, len(attrs_name)): allattrs[attrs_name[k]] = attrs_value[k] # print('allattrs=',allattrs) # allattrs= OrderedDict([('厚薄', '薄'), ('材質成分', '其他100%'),...]) # info = OrderedDict() # 存放該商品所具有的全部資訊 # info = get_headdetail2(url) # 下面三條語句獲取描述、服務、物流的評分資訊 # 下面的語句用來判斷該商品具有哪些屬性,如果具有該屬性,將屬性值插入有序字典,否則,該屬性值為空 # 適用場景 if '材質成分' in attrs_name: info['材質成分'] = allattrs['材質成分'] elif '面料' in attrs_name: info['材質成分'] = allattrs['面料'] else: info['材質成分'] = 'NA' # 適用物件 if '流行元素' in attrs_name: info['流行元素'] = allattrs['流行元素'] else: info['流行元素'] = 'NA' #季節 if '年份季節' in attrs_name: info['年份季節'] = allattrs['年份季節'] else: info['年份季節'] = 'NA' # 款式 if '袖長' in attrs_name: info['袖長'] = allattrs['袖長'] else: info['袖長'] = 'NA' # 尺碼 if '銷售渠道型別' in attrs_name: info['銷售渠道型別'] = allattrs['銷售渠道型別'] else: info['銷售渠道型別'] = 'NA' # 帽頂款式 if '貨號' in attrs_name: info['貨號'] = allattrs['貨號'] else: info['貨號'] = 'NA' # 帽檐款式 if '服裝版型' in attrs_name: info['服裝版型'] = allattrs['服裝版型'] else: info['服裝版型'] = 'NA' # 檐形 if '衣長' in attrs_name: info['衣長'] = allattrs['衣長'] else: info['衣長'] = 'NA' # 主要材質 if '領型' in attrs_name: info['領型'] = allattrs['領型'] else: info['領型'] = 'NA' # 人群 if '袖型' in attrs_name: info['袖型'] = allattrs['袖型'] else: info['袖型'] = 'NA' # 品牌 if '品牌' in attrs_name: info['品牌'] = allattrs['品牌'] else: info['品牌'] = 'NA' # 風格 if '圖案' in attrs_name: info['圖案'] = allattrs['圖案'] elif '中老年女裝圖案' in attrs_name: info['圖案'] = allattrs['中老年女裝圖案'] else: info['圖案'] = 'NA' # 款式細節 if '服裝款式細節' in attrs_name: info['服裝款式細節'] = allattrs['服裝款式細節'] else: info['服裝款式細節'] = 'NA' # 適用年齡 if '適用年齡' in attrs_name: info['適用年齡'] = allattrs['適用年齡'] else: info['適用年齡'] = 'NA' # 風格 if '風格' in attrs_name: info['風格'] = allattrs['風格'] elif '中老年風格' in attrs_name: info['風格'] = allattrs['中老年風格'] else: info['風格'] = 'NA' #通勤 if '通勤' in attrs_name: info['通勤'] = allattrs['通勤'] else: info['通勤'] = 'NA' if '裙長' in attrs_name: info['裙長'] = allattrs['裙長'] else: info['裙長'] = 'NA' if '裙型' in attrs_name: info['裙型'] = allattrs['裙型'] else: info['裙型'] = 'NA' if '腰型' in attrs_name: info['腰型'] = allattrs['腰型'] else: info['腰型'] = 'NA' # 顏色分類 if '主要顏色' in attrs_name: info['主要顏色'] = allattrs['主要顏色'] else: info['主要顏色'] = 'NA' if '顏色分類' in attrs_name: info['主要顏色'] = allattrs['顏色分類'] else: info['主要顏色'] = 'NA' #尺碼 if '尺碼' in attrs_name: info['尺碼'] = allattrs['尺碼'] else: info['尺碼'] = 'NA' if '組合形式' in attrs_name: info['組合形式'] = allattrs['組合形式'] else: info['組合形式'] = 'NA' if '褲長' in attrs_name: info['褲長'] = allattrs['褲長'] else: info['褲長'] = 'NA' return info import csv def main(): # 提取 列 with open('clothes_detai.csv', 'w', newline='', encoding='utf-8') as csvfile: # fieldnames = ['店鋪ID','店鋪名','連結','寶貝','原價','促銷價','月銷量','累計評價','材質成分','流行元素','袖長','年份季節','銷售渠道型別','貨號','服裝版型','衣長','領型','袖型', # '裙型','裙長','腰型','褲長','組合形式','品牌','圖案','服裝款式細節', '適用年齡','風格','通勤','主要顏色','尺碼'] fieldnames=[ 'Link','Brand','Title','Price','Sale price','Sales','Evaluations', 'Component', 'Fashion elements','Sleeve','Seasons','Sales channels', 'Number','Clothes_Style','Long','Collar type','Sleeve type', 'Skirt type','Skirt length','Waist','Combining form','Outseam', 'Design','Fashion pattern detail','Applicable age', 'Style','Commuter','color','Size'] # 'Shop','Data_id','Shop_id','Shop','Link','Data_id', writer = csv.DictWriter(csvfile, fieldnames = fieldnames) writer.writeheader() # urls = ['//detail.tmall.com/item.htm?spm=a230r.1.14.1.ebb2eb2eGyUw1&id=549177691667&ns=1&abbucket=4', # '//item.taobao.com/item.htm?id=548443640333&ns=1&abbucket=0#detail'] f = pd.read_csv('women_clothes_sales2.csv') urls = f['link'][0:100] # sh = f['shop_id'][0:3] # s = f['shop'][0:3] # for url in urls: # print(url) # writer.writerow({'店鋪ID':f['shop_id'],'店鋪名':f['shop']}) keys, values = [], [] # for url in urls: for i in urls: url = 'http:' + i # endswith 判斷字串是否以指定的字串結尾 if url.endswith('detail'): info = get_taobao_header(url) res = requests.get(url) soup = BeautifulSoup(res.text, "html.parser") attrs = soup.select('.attributes-list li') # 淘寶 class else: info = get_tianmao_header(url) res = requests.get(url) soup = BeautifulSoup(res.text, "html.parser") attrs = soup.select('#J_AttrUL li') # 天貓 id # print('attrs=',attrs) d = get_Details(attrs,info) print(d) # for j in f[shop_id]: # d['店鋪ID'] = j # for s in f['shop']: # d['店鋪名'] = s #'Shop':d['店鋪名'],'Data_id':d['ID'], writer.writerow({'Link':url,'Brand':d['品牌'],'Title':d['寶貝'], 'Price':d['原價'], 'Sale price':d['促銷價'], 'Sales':d['月銷量'], 'Evaluations':d['累計評價'], 'Component':d['材質成分'], 'Fashion elements':d['流行元素'], 'Sleeve':d['袖長'], 'Seasons':d['年份季節'], 'Sales channels':d['銷售渠道型別'], 'Number':d['貨號'],'Clothes_Style':d['服裝版型'],'Long':d['衣長'],'Collar type':d['領型'], 'Sleeve type':d['袖型'], 'Skirt type':d['裙型'], 'Skirt length':d['裙長'], 'Waist':d['腰型'], 'Combining form':d['組合形式'], 'Outseam':d['褲長'], 'Design':d['圖案'], 'Fashion pattern detail':d['服裝款式細節'], 'Applicable age':d['適用年齡'], 'Style':d['風格'], 'Commuter':d['通勤'], 'color':d['主要顏色'], 'Size':d['尺碼']}) if __name__=='__main__': main()