点击上方“开发者技术前线”，选择“星标”
12：00 在看 真爱
来源：Python绿色通道
先上效果图吧,no pic say bird!
古人云：知己知彼，百战不殆. 好好去了解一下妹子们的内心想法，早日脱单!
这次我在一个某知名婚恋网站，抓取了一些数据，对她们的内心读白进行分析.
我这次筛选条件：女性，年龄20-30,学历本科,就这些条件.
3000条妹子内心读白词云如下:
放心好了，她们都是追求精神满足而非物质，大部分都是要找生命中的另一半，那她们的另一半会是你吗？
完整代码
# coding=utf-8from selenium import webdriverimport timefrom lxml import etreeimport sysreload(sys)sys.setdefaultencoding('utf-8')"""PEP8 Python编程规范https://www.douban.com/note/134971609/"""# 获取浏览器驱动driver = webdriver.Firefox()driver.maximize_window()webUrl = 'http://www.lovewzly.com/jiaoyou.html'driver.get(webUrl)# 等15秒，我来手动做一下筛选条件。，女性，21-30左右，学历本科，\# 本来想通过js代码，来自动执行，但无奈对js真的不熟，也没有太多时间去整了，凑合看看.time.sleep(15)"""下拉滚动条，从1开始到3结束 分2次加载完每页数据"""while True: for i in range(1, 20): height = 1000 * i # 每次滑动20000像素 strword = "window.scrollBy(0," + str(height) + ")" driver.execute_script(strword) time.sleep(3) s = etree.HTML(driver.page_source) selectors = s.xpath('//*[@id="hibox"]/table/tbody/tr/td/div') with open('内心读白.txt', 'a') as f: for selector in selectors: img = selector.xpath('./div[1]/img/@src') nick = selector.xpath('./div[2]/p[1]/span/text()') age = selector.xpath('./div[2]/p[2]/span[1]/text()') height = selector.xpath('./div[2]/p[2]/span[2]/text()') address = selector.xpath('./div[2]/p[2]/span[3]/text()') heart = selector.xpath('./div[2]/p[3]/text()') img = img[0] if len(img) > 0 else '' nick = nick[0] if len(nick) > 0 else '' age = age[0] if len(age) > 0 else '' height = height[0] if len(height) > 0 else '' address = address[0] if len(address) > 0 else '' heart = heart[0] if len(heart) > 0 else '' print nick, age, height, address, heart, img f.write(heart)
from selenium import webdriver
import time
from lxml import etree
import sys
reload(sys)
sys.setdefaultencoding('utf-8')
"""
PEP8 Python编程规范
https://www.douban.com/note/134971609/
"""
# 获取浏览器驱动
driver = webdriver.Firefox()
driver.maximize_window()
webUrl = 'http://www.lovewzly.com/jiaoyou.html'
driver.get(webUrl)
# 等15秒，我来手动做一下筛选条件。，女性，21-30左右，学历本科，\
# 本来想通过js代码，来自动执行，但无奈对js真的不熟，也没有太多时间去整了，凑合看看.
time.sleep(15)
"""
下拉滚动条，从1开始到3结束 分2次加载完每页数据
"""
while True:
for i in range(1, 20):
height = 1000 * i # 每次滑动20000像素
strword = "window.scrollBy(0," + str(height) + ")"
driver.execute_script(strword)
time.sleep(3)
s = etree.HTML(driver.page_source)
selectors = s.xpath('//*[@id="hibox"]/table/tbody/tr/td/div')
with open('内心读白.txt', 'a') as f:
for selector in selectors:
img = selector.xpath('./div[1]/img/@src')
nick = selector.xpath('./div[2]/p[1]/span/text()')
age = selector.xpath('./div[2]/p[2]/span[1]/text()')
height = selector.xpath('./div[2]/p[2]/span[2]/text()')
address = selector.xpath('./div[2]/p[2]/span[3]/text()')
heart = selector.xpath('./div[2]/p[3]/text()')
img = img[0] if len(img) > 0 else ''
nick = nick[0] if len(nick) > 0 else ''
age = age[0] if len(age) > 0 else ''
height = height[0] if len(height) > 0 else ''
address = address[0] if len(address) > 0 else ''
heart = heart[0] if len(heart) > 0 else ''
print nick, age, height, address, heart, img
f.write(heart)
生成词云的代码之前的文章里面有，这里就不展示了，自行前往查阅!
我之前写了一个抓取妹子资料的文章，主要是使用selenium来模拟网页操作，然后使用动态加载，再用xpath来提取网页的资料，但这种方式效率不高。用Python来找合适的妹子（一）
所以今天我再补一个高效获取数据的办法.由于并没有什么模拟的操作，一切都可以人工来控制，所以也不需要打开网页就能获取数据!
但我们需要分析这个网页，打开网页 http://www.lovewzly.com/jiaoyou.html 后，按F12,进入Network项中
url在筛选条件后，只有page在发生变化，而且是一页页的累加，而且我们把这个url在浏览器中打开，会得到一批json字符串,所以我可以直接操作这里面的json数据，然后进行存储即可!
代码结构图:
操作流程:
headers 一定要构建反盗链以及模拟浏览器操作，先这样写，可以避免后续问题!
条件拼装
然后记得数据转json格式
然后对json数据进行提取，
把提取到的数据放到文件或者存储起来
主要学习到的技术:
学习requests+urllib
操作execl
文件操作
字符串
异常处理
另外其它基础
请求数据
def craw_data(self): '''数据抓取''' headers = { 'Referer': 'http://www.lovewzly.com/jiaoyou.html', 'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; WOW64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/53.0.2785.104 Safari/537.36 Core/1.53.4620.400 QQBrowser/9.7.13014.400' } page = 1 while True: query_data = { 'page':page, 'gender':self.gender, 'starage':self.stargage, 'endage':self.endgage, 'stratheight':self.startheight, 'endheight':self.endheight, 'marry':self.marry, 'salary':self.salary, } url = 'http://www.lovewzly.com/api/user/pc/list/search?'+urllib.urlencode(query_data) print url req = urllib2.Request(url, headers=headers) response = urllib2.urlopen(req).read() # print response self.parse_data(response) page += 1
'''数据抓取'''
headers = {
'Referer': 'http://www.lovewzly.com/jiaoyou.html',
'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; WOW64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/53.0.2785.104 Safari/537.36 Core/1.53.4620.400 QQBrowser/9.7.13014.400'
}
page = 1
while True:
query_data = {
'page':page,
'gender':self.gender,
'starage':self.stargage,
'endage':self.endgage,
'stratheight':self.startheight,
'endheight':self.endheight,
'marry':self.marry,
'salary':self.salary,
}
url = 'http://www.lovewzly.com/api/user/pc/list/search?'+urllib.urlencode(query_data)
print url
req = urllib2.Request(url, headers=headers)
response = urllib2.urlopen(req).read()
# print response
self.parse_data(response)
page += 1
字段提取
def parse_data(self,response): '''数据解析''' persons = json.loads(response).get('data').get('list') if persons is None: print '数据已经请求完毕' return for person in persons: nick = person.get('username') gender = person.get('gender') age = 2018 - int(person.get('birthdayyear')) address = person.get('city') heart = person.get('monolog') height = person.get('height') img_url = person.get('avatar') education = person.get('education') print nick,age,height,address,heart,education self.store_info(nick,age,height,address,heart,education,img_url) self.store_info_execl(nick,age,height,address,heart,education,img_url)
'''数据解析'''
persons = json.loads(response).get('data').get('list')
if persons is None:
print '数据已经请求完毕'
return
for person in persons:
nick = person.get('username')
gender = person.get('gender')
age = 2018 - int(person.get('birthdayyear'))
address = person.get('city')
heart = person.get('monolog')
height = person.get('height')
img_url = person.get('avatar')
education = person.get('education')
print nick,age,height,address,heart,education
self.store_info(nick,age,height,address,heart,education,img_url)
self.store_info_execl(nick,age,height,address,heart,education,img_url)
文件存放
def store_info(self, nick,age,height,address,heart,education,img_url): ''' 存照片,与他们的内心独白 ''' if age < 22: tag = '22岁以下' elif 22 <= age < 28: tag = '22-28岁' elif 28 <= age < 32: tag = '28-32岁' elif 32 <= age: tag = '32岁以上' filename = u'{}岁_身高{}_学历{}_{}_{}.jpg'.format(age,height,education, address, nick) try: # 补全文件目录 image_path = u'E:/store/pic/{}'.format(tag) # 判断文件夹是否存在。 if not os.path.exists(image_path): os.makedirs(image_path) print image_path + ' 创建成功' # 注意这里是写入图片，要用二进制格式写入。 with open(image_path + '/' + filename, 'wb') as f: f.write(urllib.urlopen(img_url).read()) txt_path = u'E:/store/txt' txt_name = u'内心独白.txt' # 判断文件夹是否存在。 if not os.path.exists(txt_path): os.makedirs(txt_path) print txt_path + ' 创建成功' # 写入txt文本 with open(txt_path + '/' + txt_name, 'a') as f: f.write(heart) except Exception as e: e.message
'''
存照片,与他们的内心独白
'''
if age < 22:
tag = '22岁以下'
elif 22 <= age < 28:
tag = '22-28岁'
elif 28 <= age < 32:
tag = '28-32岁'
elif 32 <= age:
tag = '32岁以上'
filename = u'{}岁_身高{}_学历{}_{}_{}.jpg'.format(age,height,education, address, nick)
try:
# 补全文件目录
image_path = u'E:/store/pic/{}'.format(tag)
# 判断文件夹是否存在。
if not os.path.exists(image_path):
os.makedirs(image_path)
print image_path + ' 创建成功'
# 注意这里是写入图片，要用二进制格式写入。
with open(image_path + '/' + filename, 'wb') as f:
f.write(urllib.urlopen(img_url).read())
txt_path = u'E:/store/txt'
txt_name = u'内心独白.txt'
# 判断文件夹是否存在。
if not os.path.exists(txt_path):
os.makedirs(txt_path)
print txt_path + ' 创建成功'
# 写入txt文本
with open(txt_path + '/' + txt_name, 'a') as f:
f.write(heart)
except Exception as e:
e.message
execl操作
def store_info_execl(self,nick,age,height,address,heart,education,img_url): person = [] person.append(self.count) #正好是数据条 person.append(nick) person.append(u'女' if self.gender == 2 else u'男') person.append(age) person.append(height) person.append(address) person.append(education) person.append(heart) person.append(img_url) for j in range(len(person)): self.sheetInfo.write(self.count, j, person[j]) self.f.save(u'我主良缘.xlsx') self.count += 1 print '插入了{}条数据'.format(self.count)
person = []
person.append(self.count) #正好是数据条
person.append(nick)
person.append(u'女' if self.gender == 2 else u'男')
person.append(age)
person.append(height)
person.append(address)
person.append(education)
person.append(heart)
person.append(img_url)
for j in range(len(person)):
self.sheetInfo.write(self.count, j, person[j])
self.f.save(u'我主良缘.xlsx')
self.count += 1
print '插入了{}条数据'.format(self.count)
源码地址：
https://github.com/pythonchannel/python27/blob/master/test/meizhi.py
历史阅读开发者技术前线 ，汇集技术前线快讯和关注行业趋势，大厂干货，是开发者经历和成长的优秀指南。
喜欢就点个好看吧
1480
扫一扫
被折叠的 条评论
为什么被折叠?
到【灌水乐园】发言
