首先我是从淘宝进去，爬取了按销量排序的所有（100页）女装的列表信息按综合、销量分别爬取淘宝女装列表信息，然后导出前100商品的 link，爬取其详细信息。这些商品有淘宝的，也有天猫的，这两个平台有些区别，处理的时候要注意。比如，有的说“面料”、有的说“材质成分”，其实是一个意思，等等。可以取不同的链接做一下测试。
import re
from collections import OrderedDict
from bs4 import BeautifulSoup
from pyquery import PyQuery as pq #获取整个网页的源代码
from config import * #可引用congif的所有变量
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 #获取整个网页的源代码
import pandas as pd
# 测试 淘宝+天猫，可完整输出及保存
browser = webdriver.Firefox()
wait = WebDriverWait(browser,10)
####### 天猫上半部分详情 #############
def get_tianmao_header(url):
browser.get(url)
# wait.until(EC.presence_of_all_elements_located((By.CSS_SELECTOR,'#mainsrp-itemlist .items .item'))) #加载所有宝贝
html=browser.page_source
doc = pq(html)
# print(doc)
info = OrderedDict() # 存放该商品所具有的全部信息
items = doc('#page')
# info['店铺名'] = items.find('.slogo').find('.slogo-shopname').text()
# info['ID'] = items.find('#LineZing').attr['itemid']
info['宝贝'] = items.find('.tb-detail-hd').find('h1').text()
info['促销价'] = items.find('#J_PromoPrice').find('.tm-promo-price').find('.tm-price').text()
info['原价'] = items.find('#J_StrPriceModBox').find('.tm-price').text()
# '月销量' :items.find('.tm-ind-panel').find('.tm-ind-item tm-ind-sellCount').find('.tm-indcon').find('.tm-count').text(),
info['月销量'] = items.find('.tm-ind-panel').find('.tm-indcon').find('.tm-count').text().split(' ',2)[0]
info['累计评价'] = items.find('#J_ItemRates').find('.tm-indcon').find('.tm-count').text()
# print(info)
return info
######## 淘宝上半部分详情 ###############
def get_taobao_header(url):
browser.get(url)
# wait.until(EC.presence_of_all_elements_located((By.CSS_SELECTOR,'#mainsrp-itemlist .items .item'))) #加载所有宝贝
html=browser.page_source
doc = pq(html)
# print(doc)
info = OrderedDict() # 存放该商品所具有的全部信息
items = doc('#page')
# info['店铺名'] = items.find('.tb-shop-seller').find('.tb-seller-name').text()
# info['ID'] = items.find('#J_Pine').attr['data-itemid']
info['宝贝'] = items.find('#J_Title').find('h3').text()
info['原价'] = items.find('#J_StrPrice').find('.tb-rmb-num').text()
info['促销价'] = items.find('#J_PromoPriceNum').text()
# '月销量' :items.find('.tm-ind-panel').find('.tm-ind-item tm-ind-sellCount').find('.tm-indcon').find('.tm-count').text(),
info['月销量'] = items.find('#J_SellCounter').text()
info['累计评价'] = items.find('#J_RateCounter').text()
# print(info)
return info
####################### 详情 ############################
# 抓取所有商品详情
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%">材质成分: 其他100%</li>,<li ...</li>]
attrs_name = []
attrs_value = []
'''''
[\s] 匹配空格，[\s]*,后面有 *，则可以为空
* : 匹配前面的子表达式任意次
'''
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', '材质成分', ...]
# print('attrs_value=',attrs_value) # attrs_value= ['薄', '其他100%', ...]
allattrs = OrderedDict() # 存放该产品详情页面所具有的属性
for k in range(0, len(attrs_name)):
allattrs[attrs_name[k]] = attrs_value[k]
# print('allattrs=',allattrs) # allattrs= OrderedDict([('厚薄', '薄'), ('材质成分', '其他100%'),...])
# info = OrderedDict() # 存放该商品所具有的全部信息
# info = get_headdetail2(url)
# 下面三条语句获取描述、服务、物流的评分信息
# 下面的语句用来判断该商品具有哪些属性，如果具有该属性，将属性值插入有序字典，否则，该属性值为空
# 适用场景
if '材质成分' in attrs_name:
info['材质成分'] = allattrs['材质成分']
elif '面料' in attrs_name:
info['材质成分'] = allattrs['面料']
else:
info['材质成分'] = 'NA'
# 适用对象
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'
#季节
if '年份季节' in attrs_name:
info['年份季节'] = allattrs['年份季节']
else:
info['年份季节'] = 'NA'
# 款式
if '袖长' in attrs_name:
info['袖长'] = allattrs['袖长']
else:
info['袖长'] = 'NA'
# 尺码
if '销售渠道类型' in attrs_name:
info['销售渠道类型'] = allattrs['销售渠道类型']
else:
info['销售渠道类型'] = 'NA'
# 帽顶款式
if '货号' in attrs_name:
info['货号'] = allattrs['货号']
else:
info['货号'] = 'NA'
# 帽檐款式
if '服装版型' in attrs_name:
info['服装版型'] = allattrs['服装版型']
else:
info['服装版型'] = 'NA'
# 檐形
if '衣长' in attrs_name:
info['衣长'] = allattrs['衣长']
else:
info['衣长'] = 'NA'
# 主要材质
if '领型' in attrs_name:
info['领型'] = allattrs['领型']
else:
info['领型'] = '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'
# 风格
if '图案' in attrs_name:
info['图案'] = allattrs['图案']
elif '中老年女装图案' in attrs_name:
info['图案'] = allattrs['中老年女装图案']
else:
info['图案'] = 'NA'
# 款式细节
if '服装款式细节' in attrs_name:
info['服装款式细节'] = allattrs['服装款式细节']
else:
info['服装款式细节'] = 'NA'
# 适用年龄
if '适用年龄' in attrs_name:
info['适用年龄'] = allattrs['适用年龄']
else:
info['适用年龄'] = 'NA'
# 风格
if '风格' in attrs_name:
info['风格'] = allattrs['风格']
elif '中老年风格' in attrs_name:
info['风格'] = allattrs['中老年风格']
else:
info['风格'] = 'NA'
#通勤
if '通勤' in attrs_name:
info['通勤'] = allattrs['通勤']
else:
info['通勤'] = 'NA'
if '裙长' in attrs_name:
info['裙长'] = allattrs['裙长']
else:
info['裙长'] = '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'
# 颜色分类
if '主要颜色' in attrs_name:
info['主要颜色'] = allattrs['主要颜色']
else:
info['主要颜色'] = 'NA'
if '颜色分类' in attrs_name:
info['主要颜色'] = allattrs['颜色分类']
else:
info['主要颜色'] = 'NA'
#尺码
if '尺码' in attrs_name:
info['尺码'] = allattrs['尺码']
else:
info['尺码'] = 'NA'
if '组合形式' in attrs_name:
info['组合形式'] = allattrs['组合形式']
else:
info['组合形式'] = 'NA'
if '裤长' in attrs_name:
info['裤长'] = allattrs['裤长']
else:
info['裤长'] = 'NA'
return info
import csv
def main():
# 提取 列
with open('clothes_detai.csv', 'w', newline='', encoding='utf-8') as csvfile:
# fieldnames = ['店铺ID','店铺名','链接','宝贝','原价','促销价','月销量','累计评价','材质成分','流行元素','袖长','年份季节','销售渠道类型','货号','服装版型','衣长','领型','袖型',
# '裙型','裙长','腰型','裤长','组合形式','品牌','图案','服装款式细节', '适用年龄','风格','通勤','主要颜色','尺码']
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({'店铺ID':f['shop_id'],'店铺名':f['shop']})
keys, values = [], []
# for url in urls:
for i in urls:
url = 'http:' + i
# endswith 判断字符串是否以指定的字符串结尾
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') # 淘宝 class
else:
info = get_tianmao_header(url)
res = requests.get(url)
soup = BeautifulSoup(res.text, "html.parser")
attrs = soup.select('#J_AttrUL li') # 天猫 id
# print('attrs=',attrs)
d = get_Details(attrs,info)
print(d)
# for j in f[shop_id]:
# d['店铺ID'] = j
# for s in f['shop']:
# d['店铺名'] = s
#'Shop':d['店铺名'],'Data_id':d['ID'],
writer.writerow({'Link':url,'Brand':d['品牌'],'Title':d['宝贝'], 'Price':d['原价'], 'Sale price':d['促销价'], 'Sales':d['月销量'], 'Evaluations':d['累计评价'],
'Component':d['材质成分'], 'Fashion elements':d['流行元素'], 'Sleeve':d['袖长'], 'Seasons':d['年份季节'], 'Sales channels':d['销售渠道类型'],
'Number':d['货号'],'Clothes_Style':d['服装版型'],'Long':d['衣长'],'Collar type':d['领型'], 'Sleeve type':d['袖型'],
'Skirt type':d['裙型'], 'Skirt length':d['裙长'], 'Waist':d['腰型'], 'Combining form':d['组合形式'], 'Outseam':d['裤长'],
'Design':d['图案'], 'Fashion pattern detail':d['服装款式细节'], 'Applicable age':d['适用年龄'],
'Style':d['风格'], 'Commuter':d['通勤'], 'color':d['主要颜色'], 'Size':d['尺码']})
if __name__=='__main__':
main()
