改动：
新增功能 ：可选择只看天猫或淘宝
代码模块化封装，参数配置或输入单独在一个py文件管理，主函数功能只留出参数传入在setting配置的py文件里。
main.py代码：
# -*- coding: utf-8 -*-
import urllib
import urllib2
import requests import re
import time
import random
import os
from math import log
from math import log10
from math import sqrt
import sys import setting #"pageSize":44,"totalPage":100,"currentPage":3,"totalCount":29561 '''在Python自己IDE上要注释掉以下两行''' reload(sys)
sys.setdefaultencoding('utf8') # python2.x的的defaultencoding是ascii class counter(object):
#计数器
def __init__(self):
self.count = 0
self.try_time = 0 self.try_find = 0
self.fail_time = 0
self.url_list = []
self.new_flag = True
self.results = []
self.priSu = 0
self.descSu = 0
self.tm_tb = '' def print_counter(self):
print 'try_time:', self.try_time, " get_count:" , self.count, " fail_time:",self.fail_time ,"try_find_time:",self.try_find def recommend_rate(price, description, delivery, service, comments):
#描述为绝对值
av_p = counter1.priSu / counter1.count
av_d = counter1.descSu / counter1.count
rate = (description/av_d)**20 \
*(description + delivery + service) \
*(av_p/(price))**0.1 \
+log((comments+5),1000) #print 'all count=',counter1.count
#print "avrage price=",av_p,';',av_p/(price),';price',price,';comments=',comments,';descrip=',description
#print 'rate=',rate,'(price)yinzi',(av_p/(price))**0.1,'descrip_yinzi',(description/av_d)**20,'comments_factor=',log((comments+50),100)
return rate def product_rank(list):
for x in list:
#0开始为 x0商品名 、x1图片链接、x2商品链接、x3价格、x4评论数、 x5店铺名、 x6快递情况、x7描述相符情况3个、x8服务情况
rate = recommend_rate(x[3],x[7],x[6],x[8],x[4])
x.append(rate) def get_user_rate(item_url):
#暂时未使用该功能
'''获取卖家信用情况；未登录情况不能访问，或者需要在头部文件中加入cookie。。。；'''
html = urllib2.urlopen(item_url)
#"//rate.taobao.com/user-rate-282f910f3b70f2128abd0ee9170e6428.htm"
regrex_rate = '"(//.*?user\-rate.*?)"'
codes = re.findall(regrex_rate,html.read())
html.close() user_rate_url= 'http:'+codes[0]
print 'uu', user_rate_url user_rate_html = urllib2.urlopen(user_rate_url)
print user_rate_html.read()
#title = "4.78589分"
desc_regex = u'title="(4.[0-9]{5}).*?'
de_pat = re.compile(desc_regex) descs = re.findall(de_pat,user_rate_html.read())
print len(descs)
item_url = 'https://item.taobao.com/item.htm?id=530635294653&ns=1&abbucket=0#detail'
#get_user_rate(item_url)
'''获取卖家信用情况；未登录情况不能访问。。。暂时 无用''' def makeNewdir(savePath):
while os.path.exists(savePath):
savePath = savePath + str(random.randrange(1,10))
#print "the path exist,we'll make a new one"
try:
os.makedirs(savePath)
print 'ok,file_path we reserve results: %s'%savePath
print '保存的路径为：'.decode('utf-8') except :
print "failed to make file path\nplease restart program"
print '创建文件夹失败，请重新启动程序'.decode('utf-8')
return savePath def get_praised_good(url, file_open, keyword, counts, descripHrequ, servHrequ, descripNrequ):
#从给定的淘宝链接中 获取符合条件的商品list
html = req_s.get(url)
code = html.content
html.close() regrex2 = ur'raw_title":"(.*?)","pic_url":"(.*?)","detail_url":"(.*?)","view_price":"(.*?)".*?"comment_count":"(.*?)".*?"nick":"(.*?)".*?"delivery":\[(.*?),(.*?),(.*?)\],"description":\[(.*?),(.*?),(.*?)\],"service":\[(.*?),(.*?),(.*?)\]'
#每一个匹配项 返回 15个 字符串
#x[0]开始为 x0商品名 、x1图片链接、x2商品链接、x3价格、x4评论数、 x5店铺名、 x6快递情况3个、x9描述相符情况3个、x12服务情况3个
pat = re.compile(regrex2)
meet_code = re.findall(regrex2, code)#
if not len(meet_code):
counter1.new_flag = False
print 'no more new met products' for x in meet_code:
# if counter1.count>=counts :
# print "have get enough pruducts"
# break
counter1.try_find += 1 description_higher = int(x[10])*float(x[11])/100
service_higher = int(x[13])*float(x[14])/100
try:
x4 = int(x[4]) #description_count
except:
x4 = 0 #如果 只要淘宝 非天猫
if counter1.tm_tb == 'taobao':
if counter1.tm_tb not in x[2].split('.'):
break if (description_higher>=descripHrequ) and (service_higher>=servHrequ) and x4>=descripNrequ:
if re.findall(keyword,x[0]) : # 中文keyword在结果中匹配问题暂时没有解决，，直接加在搜索词里吧
x0 = x[0].replace(' ','').replace('/','')
detail_url = 'http:' + x[2].decode('unicode-escape').encode('utf-8')
x1 = 'http:'+ x[1].decode('unicode-escape').encode('utf-8')
#print type(x)
if detail_url in counter1.url_list or x4 == 0:
counter1.new_flag = False
print 'no more new met products'
print counter1.url_list
print detail_url
break
counter1.url_list.append(detail_url)
counter1.try_time += 1
counter1.count += 1 x11 = float(x[11])/100
x9 = float(x[9])/100
x12 = float(x[12])/100
x6 = float(x[6])/100
x3 = float(x[3])
counter1.priSu += x3
counter1.descSu += x9
x5 = unicode(x[5],'utf-8') result_list = []
result_list.append(x0)
result_list.append(x1)
result_list.append(detail_url)
result_list.append(x3)
result_list.append(x4)
result_list.append(x5)
result_list.append(x6)
result_list.append(x9)
result_list.append(x12)
#0开始为 x0商品名 、x1图片链接、x2商品链接、x3价格、x4评论数、 x5店铺名、 x6快递情况、x7描述相符情况、x8服务情况
counter1.results.append(result_list) def save_downpic(lis,file_open,savePath):
'''从商品list下载图片到reserve_file_path，并写入信息至fileopen'''
#0开始为 x0商品名 、x1图片链接、x2商品链接、x3价格、x4评论数、 x5店铺名、 x6快递情况、x7描述相符情况、x8服务情况、x9:rate
len_list = len(lis)
print 'we find:',len_list,'products'
cc = 0
for x in lis:
try :
urllib.urlretrieve(x[1], savePath+'\\%s___'%cc +unicode(x[0],'utf-8')+'.jpg') txt_name = savePath+'\\'+ '%s__'%cc+ 'custome_description_%s __'%x[7] +'__comments_%s_'%x[4]+ '___price_%srmb___'%x[3] +x[5] +'.txt' file_o = open(txt_name, 'a')
file_o.write(x[2])
file_o.close() print '\nget_one_possible_fine_goods:\n','good_name:',x[0].decode('utf-8')
print 'rate=',x[9]
print 'price:',x[3],x[5].decode('utf-8')
print 'custome_description:',x[7],'--','described_number:',x[4],' service:',x[8]
print x[2].decode('utf-8'),'\ngood_pic_url:',x[1].decode('utf-8') print txt_name
print cc+1,"th" file_open.write(u'%s__'%cc \
+ str(x[0]) \
+ '\nprice:' \
+ str(x[3]) \
+ '￥,\n' \
+ str(x[2]) + ' \n' + str(x[5]) + '\ncustomer_description:' + str(x[7]) + 'described_number:' + str(x[4])+'\n\n\n') print 'get one -^-'
except :
print "failed to down picture or creat txt"
counter1.fail_time += 1
cc += 1
time.sleep(0.5) def get_market_totalCount(url):
html = urllib2.urlopen(url)
code = html.read()
reg = '"pageSize":[0-9]*?,"totalPage":[0-9]*?,"currentPage":[0-9]*?,"totalCount":([0-9]*?)}'
totalCount = int(re.findall(reg,code)[0]) return totalCount #"pageSize":44,"totalPage":100,"currentPage":3,"totalCount":29561 def get_all_praised_goods(serchProd,counts,savePath ,keyword, price_min=0,price_max=0,descripHrequ =0,servHrequ=0 ,descripNrequ=0):
#边里搜索结果每一页
#initial url and page number
initial_url = 'https://s.taobao.com/search?q='+serchProd + '&_input_charset=utf-8' if counter1.tm_tb == 'tmall':
initial_url = initial_url + '&filter_tianmao=tmall' if price_min:
if price_min < price_max :
initial_url = initial_url+'&filter=reserve_price%5B'+'%s'%price_min+'%2C' +'%s'%price_max
initial_url = initial_url +'&cd=false&%5D&s=' #tian_mall = 'https://list.tmall.com/search_product.htm?q=' print "initial_url",initial_url+'' page_n = 0
reserve_file = savePath+r'\found_goods.txt'
file_open = open(reserve_file,'a') file_open.write('****************************\n')
file_open.write(time.ctime())
file_open.write('\n****************************\n') total = get_market_totalCount(initial_url+'')
print "totalcount",total
if total>counts*10:
total = sqrt(total) while counter1.new_flag and counter1.try_find<total : url_1 = initial_url + str(44*page_n)
#print initial_url
print 'url_1:', url_1
#print 'ss',initial_url+'%s'%(44*page_n)
page_n += 1 get_praised_good(url_1,file_open,keyword,counts,descripHrequ,servHrequ ,descripNrequ)
print "let web network rest for 1s lest make traffic jams "
time.sleep(1)
# except:
print page_n, "pages have been searched"
if total < counts :
print "check keyword,maybe too restrict"
break print url_1
product_rank(counter1.results) counter1.results.sort(key = lambda x : x[9], reverse=True)
counter1.results = counter1.results[:counts] counter1.print_counter() save_downpic(counter1.results,file_open,savePath) #
for a in counter1.results:
for b in a :
file_open.write(unicode(str(b),'utf-8'))
file_open.write('\t')
file_open.write('\n\n') file_open.close()
counter1.print_counter() counter1 = counter() market_totalcounts = 0 req_s = requests.Session()
req_s.adapters.DEFAULT_RETRIES = 3
req_s.keep_alive = True def main():
print "说明:".decode('utf-8')
print '本程序用于在淘宝上搜索商品时主动通过 价格范围、商品描述、服务态度、评论数来筛选商品;\n筛选出来的商品图片下载保存到磁盘（默认桌面新建find_worty_goods文件夹）并建立同序号开头的txt文件，图片显示商品，其旁的txt文件名显示价格等关键信息，txt里保存商品的淘宝链接'.decode('utf-8') if setting.userDefine: #自己输入 配置参数-筛选要求
setting.inputPara()
#否则 使用setting中的配置参数 serchProd = setting.serchProd #淘宝搜索词
keyword = setting.keyword #raw_input().decode("gbk").encode("utf-8") #个人限定词，商品名字必须包含，防止淘宝推荐了其他相关词 (正则表达式). 为任意表示不作限制
price_min = setting.price_min #价格区间
price_max = setting.price_max
descripHrequ = setting.descripHrequ # % 默认高于average, 输出结果大于此值
servHrequ = setting.servHrequ # % 默认高于average, 输出结果大于此值
descripNrequ = setting.descripNrequ
counts = setting.counts #要求选出多少个商品
counter1.tm_tb = setting.tm_tb #不区分天猫淘宝则，字符串为空，，只要天猫 则 ='tmall' ,只要淘宝 = 'taobao' #savePath = r"C:\Users\Administrator\Desktop\Python scrapy\find_worthy_goods\results"#结果保存路径
savePath = u"results%s"%serchProd #结果保存路径
savePath = makeNewdir(savePath) get_all_praised_goods(serchProd, counts, savePath, keyword, price_min, price_max ,descripHrequ ,servHrequ ,descripNrequ) if __name__ == "__main__" :
main() #保存图片，以文件名为商品图片名字，并以序号开头
#同时，输出 价格、商家名，商品描述、服务等 到 txt文本
#在商品图片看中后，便可按序号查找
#按描述、服务评价高于平均，购物体验应该可以的
setting.py
# -*- coding: utf-8 -*- userDefine = False
#筛选要求设置 serchProd='背包' #淘宝搜索词
keyword='' #raw_input().decode("gbk").encode("utf-8") #个人限定词，商品名字必须包含，防止淘宝推荐了其他相关词 (正则表达式). 为任意表示不作限制
price_min=22 #价格区间
price_max=100
descripHrequ=0 # % 默认高于average, 输出结果大于此值
servHrequ=0 # % 默认高于average, 输出结果大于此值
descripNrequ=6
counts=25 #要求选出多少个商品
tm_tb ='tmall' #不区分天猫淘宝则，字符串为空，，只要天猫 则 ='tmall' ,只要淘宝 = 'taobao' def inputPara():
''' 用户选择是否自定义要求，根据要求进行获取商品，并按推荐排序输出'''
print "please input reserch _goods_name"
global serchProd , keyword , price_min, price_max, descripHrequ , servHrequ, descripNrequ ,counts ,tm_tb serchProd=raw_input().replace(' ','') #淘宝搜索词 ,并去除中间意外输入的空格 if serchProd: print "if customise price_range ,decriptiom require .etc.\ninput Y/N \n default by : no price limit avarage than descriptiom,get 50 products \n 默认要求为：无价格限制，商品描述、快递、服务高于均值，获取50个商品。自定义要求请输入 ‘Y’ (区分大小写)".decode('utf-8')
if raw_input() == 'Y':
print "\nplease input _minimal price and _maximal price; \ndefault by 0,10000\nnext by 'enter'key input nothing means by default,the same below "
print '请输入价格范围 ；默认0-10000 ；两项用半角逗号","分隔 按回车键确认；什么也不输入代表使用默认值 '.decode('utf-8')
try:
price_min, price_max=input()
except:
print 'not input or wrong number,use default range'
price_min, price_max = 0 ,10000 #
print '是否要求 只看天猫/正品保障 还是只看淘宝 \n 只看天猫输入 tmall ,只看淘宝输入taobao，都看则回车略过'
try:
tm_tb=raw_input().decode("gbk").encode("utf-8") #个人限定词，商品名字必须包含，防止淘宝推荐了其他相关词 (正则表达式). 为任意表示不作限制
except:
tm_tb=''
#
# #
print "please input _keyword that goods name must include:\n(more than one keyword must use Regular Expression); default by no kewords"
try:
keyword=raw_input().decode("gbk").encode("utf-8") #个人限定词，商品名字必须包含，防止淘宝推荐了其他相关词 (正则表达式). 为任意表示不作限制
except:
keyword=''
# print "\nplease input _description_higher_percent_require and _service_higher__percent_require\n range:(-100,100) ; \ndefault by 0,0 I.e better than average"
print '请输入商品描述、服务高于平均值的百分比-100 ~100'.decode('utf-8')
# % 默认高于average, 输出结果大于此值
try:
descripHrequ,servHrequ=input()
except:
print 'not input or wrong number,use default range'
descripHrequ = 0 # % 默认高于average, 输出结果大于此值
servHrequ = 0 #
print "\nplease input description count limit, default more than 5\n" ,'输入最低商品评价数，默认大于5'.decode('utf-8')
try:
descripNrequ=input()
except :
print 'not input or wrong number,use default range'
descripNrequ=5
# # print "\nIF customise file reserve path, Y or N \ndefault/sample as: C:\\Users\\Administrator\\Desktop\\find_worthy_goods\\results "
# print '是否自定义保存文件目录 Y or N'.decode('utf-8')
# if raw_input()=='Y':
# print "please input path that you want to reserve; \n "
# savePath = raw_input()
# else:
# #savePath=r"C:\Users\Administrator\Desktop\find_worthy_goods\results"#结果保存路径
#
print "\nplease input how many results you want, default by 50\n" ,'您要获取的商品数目，默认50'.decode('utf-8')
try:
counts=input()
except :
counts=50
else :
counts =50
keyword = ''
tm_tb = ''
price_min ,price_max ,descripHrequ ,servHrequ ,descripNrequ = 0,0,0,0,0
else:
print "no search goods，please restart"
print '没有输入商品名称，请重新启动程序'.decode('utf-8')