python利用xpath爬取网上数据并存储到django模型中

帮朋友制作一个网站，需要一些产品数据信息，因为是代理其他公司产品，直接爬取代理公司产品数据

1.设计数据库

from django.db import models
from uuslug import slugify
import uuid
import os

def products_directory_path(instance, filename):
 ext = filename.split('.')[-1]
 filename = '{}.{}'.format(uuid.uuid4().hex[:8], ext)
 # return the whole path to the file
 return os.path.join('images', "products", instance.title, filename)

def product_relatedimage_directory_path(instance, filename):
 ext = filename.split('.')[-1]
 filename = '{}.{}'.format(uuid.uuid4().hex[:8], ext)
 # return the whole path to the file
 return os.path.join('images', "product_relatedimage", instance.product.title, filename)

class ProductsCategory(models.Model):
 """产品分类"""
 name = models.CharField('产品分类名', max_length=80, unique=True)
 description = models.TextField('产品分类描述', blank=True, null=True)
 slug = models.SlugField('slug', max_length=80, blank=True, null=True)
 parent_category = models.ForeignKey('self', verbose_name="父级分类", blank=True, null=True, on_delete=models.CASCADE)

def save(self, *args, **kwargs):
   if not self.id or not self.slug:
     self.slug = slugify(self.name)
   super().save(*args, **kwargs)

def __str__(self):
   return self.name

class Meta:
   ordering = ['name']
   verbose_name = "产品分类"
   verbose_name_plural = verbose_name

class ProductsTag(models.Model):
 """产品标签"""
 name = models.CharField('产品标签名', max_length=30, unique=True)
 slug = models.SlugField('slug', max_length=40)

def __str__(self):
   return self.name

def save(self, *args, **kwargs):
   if not self.id or not self.slug:
     self.slug = slugify(self.name)
   super().save(*args, **kwargs)

class Meta:
   ordering = ['name']
   verbose_name = "产品标签"
   verbose_name_plural = verbose_name

class Product(models.Model):
 title = models.CharField('标题', max_length=255, unique=True)
 slug = models.SlugField('slug', max_length=255, blank=True, null=True)
 jscs = models.TextField('技术参数', blank=True, null=True)
 image = models.ImageField(upload_to=products_directory_path, verbose_name="产品图片")
 views = models.PositiveIntegerField('浏览量', default=0)
 category = models.ForeignKey('ProductsCategory', verbose_name='分类', on_delete=models.CASCADE, blank=True, null=True)
 tags = models.ManyToManyField('ProductsTag', verbose_name='标签集合', blank=True)

def save(self, *args, **kwargs):
   if not self.id or not self.slug:
     self.slug = slugify(self.title)
   super().save(*args, **kwargs)

def update_views(self):
   self.views += 1
   self.save(update_fields=['views'])

def get_pre(self):
   return Product.objects.filter(id__lt=self.id).order_by('-id').first()

def get_next(self):
   return Product.objects.filter(id__gt=self.id).order_by('id').first()

def __str__(self):
   return self.title

class Meta:
   verbose_name = "产品"
   verbose_name_plural = verbose_name

class ProductAdvantage(models.Model):
 content = models.TextField('产品优势', blank=True, null=True)
 product = models.ForeignKey(Product, on_delete=models.CASCADE, blank=True, null=True)

def __str__(self):
   return self.content

class Meta:
   verbose_name = "产品优势"
   verbose_name_plural = verbose_name

class ProductBody(models.Model):
 body = models.CharField('产品内容', max_length=256, blank=True, null=True)
 product = models.ForeignKey(Product, on_delete=models.CASCADE, blank=True, null=True)

def __str__(self):
   return self.product.title

class Meta:
   verbose_name = "产品内容"
   verbose_name_plural = verbose_name

2.脚本编写

2.1编写获取网页源代码函数

def get_one_page(url):
 try:
   headers = {
     "User-Agent": "Mozilla/5.0 (Windows NT 10.0; WOW64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/51.0.2704.103 Safari/537.36"}
   res = requests.get(url=url, headers=headers)
   res.encoding = 'utf-8'
   if res.status_code == 200:
     return res.text
   else:
     return None
 except Exception:
   return None

2.2根据base页面获取所有产品分类页面链接

if __name__ == '__main__':
 content = get_one_page(url)
 tree = etree.HTML(content)
 # 产品分类url
 catgory_urls = tree.xpath('//div[@class="fdh-01-nav"]/div/h3/a/@href')
 # 处理catgory_urls
 for url in catgory_urls:
   url = 'http://www.kexinjianji.com' + url
   print(url)

2.3根据产品分类页面链接获取对应所有产品链接

if __name__ == '__main__':
 content = get_one_page(url)
 tree = etree.HTML(content)
 # 产品分类
 catgory = tree.xpath('//div[@class="cplb-3n-ts-03 b"]/h3/span/text()')
 print("产品分类：" + catgory[0])
 # 该分类下产品url
 urls = tree.xpath('//div[@class="cplb-3n-ts-03-list"]/dl/dt/a/@href')
 # 处理url
 for url in urls:
   url = 'http://www.kexinjianji.com' + url
   print(url)
 print("=====================================================")

两者结合起来就可以打印出所有产品链接

if __name__ == '__main__':
 content = get_one_page(url)
 tree = etree.HTML(content)
 # 产品分类url
 catgory_urls = tree.xpath('//div[@class="fdh-01-nav"]/div/h3/a/@href')
 # 处理catgory_urls
 for url in catgory_urls:
   url = 'http://www.kexinjianji.com' + url
   content = get_one_page(url)
   tree = etree.HTML(content)
   # 产品分类
   catgory = tree.xpath('//div[@class="cplb-3n-ts-03 b"]/h3/span/text()')
   print("产品分类：" + catgory[0])
   # 该分类下产品url
   urls = tree.xpath('//div[@class="cplb-3n-ts-03-list"]/dl/dt/a/@href')
   # 处理url
   for url in urls:
     url = 'http://www.kexinjianji.com' + url
     print(url)
   print("=====================================================")

2.2使用xpath解析函数返回产品链接的内容

if __name__ == '__main__':
 content = get_one_page(url)
 tree = etree.HTML(content)
 # 产品名称
 title = tree.xpath('//*[@id="wrap"]//h1/text()')
 images = tree.xpath('//div[@class="sol_tj_left"]/a/img/@src')
 # 产品图片
 images_url = 'http://www.kexinjianji.com/' + images[0]
 # 性能特点
 xntd = tree.xpath('//div[@class="w"]//div/span/text()|//div[@class="w"]//div/text()')
 # 技术参数
 jscs = tree.xpath('//table')[0]
 jscs_str = etree.tostring(jscs, encoding='utf-8').decode('utf-8')
 # 产品内容
 cpnr = tree.xpath('//div[@class="describe"]/p')
 print('产品名称:' + title[0])
 print('产品图片:' + images_url)
 for td in xntd:
   print('性能特点:' + td)
 print('技术参数:' + jscs_str)
 for cp in cpnr:
   # string(.) 获取当前标签下所有文本内容
   cp = cp.xpath('string(.)')
   print('产品内容:' + cp)
 print('============================================')

将三者结合在一起就可以获取所有产品信息

if __name__ == '__main__':
 content = get_one_page(url)
 tree = etree.HTML(content)
 # 产品分类url
 catgory_urls = tree.xpath('//div[@class="fdh-01-nav"]/div/h3/a/@href')
 # 处理catgory_urls
 for url in catgory_urls:
   url = 'http://www.kexinjianji.com' + url
   content = get_one_page(url)
   tree = etree.HTML(content)
   # 产品分类
   catgory = tree.xpath('//div[@class="cplb-3n-ts-03 b"]/h3/span/text()')
   # 该分类下产品url
   urls = tree.xpath('//div[@class="cplb-3n-ts-03-list"]/dl/dt/a/@href')
   # 处理url
   for url in urls:
     url = 'http://www.kexinjianji.com' + url
     content = get_one_page(url)
     try:
       tree = etree.HTML(content)
       # 产品名称
       title = tree.xpath('//*[@id="wrap"]//h1/text()')
       images = tree.xpath('//div[@class="sol_tj_left"]/a/img/@src')
       # 产品图片
       images_url = 'http://www.kexinjianji.com' + images[0]
       # 性能特点
       xntd = tree.xpath('//div[@class="w"]//div/span/text()|//div[@class="w"]//div/text()')
       # 技术参数
       jscs = tree.xpath('//table')[0]
       jscs_str = etree.tostring(jscs, encoding='utf-8').decode('utf-8')
       # 产品内容
       cpnr = tree.xpath('//div[@class="describe"]/p')
       print("产品分类：" + catgory[0])
       print('产品链接:' + url)
       print('产品名称:' + title[0])
       print('产品图片:' + images_url)
       for td in xntd:
         print('性能特点:' + td.strip())
       # print('技术参数:' + jscs_str)
       for cp in cpnr:
         # string(.) 获取当前标签下所有文本内容
         cp = cp.xpath('string(.)')
         print('产品内容:' + cp)
       print('============================================')
     except Exception as e:
       print(e)
       print('出错url:' + url)
       pass

3.存储到django模型

import requests
from lxml.html import etree
import os
import django
import uuid
from django.core.files.base import ContentFile

os.environ.setdefault("DJANGO_SETTINGS_MODULE", "jiaobanzhan.settings")
django.setup()

from products.models import ProductBody, ProductsCategory, Product, ProductAdvantage

url = 'http://www.kexinjianji.com/product/hzshntjbz_1/'

def get_one_page(url):
 try:
   headers = {
     "User-Agent": "Mozilla/5.0 (Windows NT 10.0; WOW64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/51.0.2704.103 Safari/537.36"}
   res = requests.get(url=url, headers=headers, timeout=10)
   res.encoding = 'utf-8'
   if res.status_code == 200:
     return res.text
   else:
     return None
 except Exception:
   print('aa')
   return None

if __name__ == '__main__':
 content = get_one_page(url)
 tree = etree.HTML(content)
 # 产品分类url
 catgory_urls = tree.xpath('//div[@class="fdh-01-nav"]/div/h3/a/@href')
 # 处理catgory_urls
 for url in catgory_urls:
   url = 'http://www.kexinjianji.com' + url
   content = get_one_page(url)
   tree = etree.HTML(content)
   # 产品分类
   p_catgory = tree.xpath('//div[@class="cplb-3n-ts-03 b"]/h3/span/text()')
   # 该分类下产品url
   urls = tree.xpath('//div[@class="cplb-3n-ts-03-list"]/dl/dt/a/@href')
   # 处理url
   for url in urls:
     url = 'http://www.kexinjianji.com' + url
     content = get_one_page(url)
     try:
       tree = etree.HTML(content)
       # 产品名称
       title = tree.xpath('//*[@id="wrap"]//h1/text()')
       images = tree.xpath('//div[@class="sol_tj_left"]/a/img/@src')
       # 产品图片
       images_url = 'http://www.kexinjianji.com' + images[0]
       # 性能特点
       xntd = tree.xpath('//div[@class="w"]//div/span/text()|//div[@class="w"]//div/text()')
       # 技术参数
       jscs = tree.xpath('//table')[0]
       jscs_str = etree.tostring(jscs, encoding='utf-8').decode('utf-8')
       # 产品内容
       cpnr = tree.xpath('//div[@class="describe"]/p')
       # 判断是否有这分类，没有则新建
       catgory = p_catgory[0]
       products_catgory = ProductsCategory.objects.filter(name=catgory).exists()
       if products_catgory:
         products_catgory = ProductsCategory.objects.get(name=catgory)
       else:
         products_catgory = ProductsCategory(name=catgory)
         products_catgory.save()
       print(products_catgory)

# 保存产品图片
       image_content = requests.get(url=images_url)
       ext = images_url.split('.')[-1] # 获取图片类型
       filename = '{}.{}'.format(uuid.uuid4().hex[:8], ext) # 随机生成图片名字
       upload_image_file = ContentFile(image_content.content, name=filename) # 将图片保存为django类型
       product = Product(title=title[0], jscs=jscs_str, image=upload_image_file, category=products_catgory)
       product.save()
       for td in xntd:
         product_advantage = ProductAdvantage()
         product_advantage.content = td
         product_advantage.product = product
         product_advantage.save()
       for cp in cpnr:
         cp = cp.xpath('string(.)')
         product_body = ProductBody()
         product_body.body = cp
         product_body.product = product
         product_body.save()
     except Exception as e:
       print(e)
       print('出错url:' + url)

最后自己手动处理出错url（页面没有获取到技术参数，技术参数是一张图片）

4.总结

1.xpath 获取标签内容时，p标签中嵌套span标签，源码如下

<div class="describe" style="position: relative;">
  <p><span>板  宽：</span>1500mm</p>
  <p><span>板  厚：</span>4.5 mm</p>
  <p><span>出料口：</span>6口</p>
  <p><span>重  量：</span>6000 kg</p>
</div>

使用xpath获取p标签内容
我想得到的效果如下
板 宽：1500mm
板 厚：4.5 mm
出料口：6口
重 量：6000 kg
使用以下xpath 只能分开获取，不是想要的效果

//div[@class="describe"]/p/span/text()|//div[@class="describe"]/p/text()

百度之后找到的解决办法，使用xpath(‘string(.)')
1.先获取所有p标签

cpnr = tree.xpath('//div[@class="describe"]/p')

2.使用**string(.)**获取所有标签所有文本

cp = cp.xpath('string(.)')

循环遍历所有p标签即可

来源：https://blog.csdn.net/cll_869241/article/details/114005783