爬虫起因
前面两个星期,利用周末的时间尝试和了解了一下Python爬虫,紧接着就开始用Scrapy框架做了一些小的爬虫,不过,由于最近一段时间的迷茫,和处于对职业生涯的规划。以及对市场需求的分析,我通过网上查阅资料。对比较大的前程无忧和智联招聘进行了数据爬取。
这里我们以智联招聘为例做一些讲解。
前期准备
首先我在我自己做爬虫之前就已经规划好了我需要爬取什么数据,并且创建了数据库表,并提前对网页内容有大概的了解。其次处于对数据分析的考虑,我对我比较关系的字段例如,经验,学历,薪资等都要求尽量能够爬取到。最后,通过书本以及网络资源等各种工具了解Scrapy,正则表达式,Xpath,BeautifulSoup等各种知识,为后面做好爬虫打下了基础。
实战
在本次小练习中,我们主要会用到,piplines,items,和我们自己新建的Spider类,
items是针对实体的,与数据库表中最好具有对应关系,代码如下:
import scrapy
class ZhaopinItem(scrapy.Item):
jobname = scrapy.Field()
salary = scrapy.Field()
experience = scrapy.Field()
address = scrapy.Field()
comany_name = scrapy.Field()
head_count = scrapy.Field()
education_require = scrapy.Field()
comany_size = scrapy.Field()
job_require =scrapy.Field()
release_date = scrapy.Field()
piplines在本例中主要是对items进行数据操作的。代码如下:
import pymysql
from zhaopin import settings
class ZhaopinPipeline(object):
def __init__(self, ):
self.conn = pymysql.connect(
host=settings.MYSQL_HOST,
db=settings.MYSQL_DBNAME,
user=settings.MYSQL_USER,
passwd=settings.MYSQL_PASSWORD,
charset='utf8', # 编码要加上,否则可能出现中文乱码问题
use_unicode=False)
self.cursor = self.conn.cursor()
def process_item(self, item, spider):
self.insertData(item)
return item
def insertData(self, item):
sql = "insert into shenzhen(jobname,salary,company_name,job_require,address,experience,company_size,head_count,education_require,release_date) VALUES(%s,%s,%s,%s,%s,%s,%s,%s,%s,%s);"
params = (item['jobname'],item['salary'],item['comany_name'],item['job_require'],item['address'],item['experience'],item['comany_size'],item['head_count'],item['education_require'],item['release_date'])
self.cursor.execute(sql, params)
self.conn.commit()
最后最为主要的是,数据的获取以及解析,代码如下。
from zhaopin.items import ZhaopinItem
from scrapy import Spider,Request
from bs4 import BeautifulSoup
import re
class ZhaopinSpider(Spider):
name = 'zhaopin'
allowed_domains = ['www.zhaopin.com']
start_urls = ['http://www.zhaopin.com/']
#start_urls = ['http://sou.zhaopin.com/jobs/searchresult.ashx?jl=%E4%B8%8A%E6%B5%B7&kw=java%E5%B7%A5%E7%A8%8B%E5%B8%88&sm=0&sg=720f662a0e894031b9b072246ac2f919&p=1']
def start_requests(self):
#for num in (1,60):
url='http://sou.zhaopin.com/jobs/searchresult.ashx?jl=%E6%B7%B1%E5%9C%B3&kw=java%E5%B7%A5%E7%A8%8B%E5%B8%88&sm=0&isadv=0&sg=cc9fe709f8cc4139afe2ad0808eb7983&p=42'
#.format(num)
#yield Request(url,callback=self.parse)
yield Request(url,callback=self.parse)
def parse(self, response):
#self.log('page url is ' + response.url)
wbdata = response.text
soup = BeautifulSoup(wbdata, 'lxml')
job_name = soup.select("table.newlist > tr > td.zwmc > div > a:nth-of-type(1)")
salary = soup.select("table.newlist > tr > td.zwyx")
#company_name = soup.select("table.newlist > tr > td.gsmc > div > a:nth-of-type(2)")
times = soup.select("table.newlist > tr > td.gxsj > span")
for name,salary,time in zip(job_name,salary,times):
item = ZhaopinItem()
item["jobname"] = name.get_text()
url= name.get('href')
#print("职位"+name.get_text()+"工资"+salary.get_text()+"发布日期"+time.get_text()+"连接"+url)
item["salary"] = salary.get_text()
item["release_date"] = time.get_text()
# item["comany_name"] = company _name.get_text()
#yield item
yield Request(url=url, meta={"item": item}, callback=self.parse_moive,dont_filter=True)
def parse_moive(self, response):
#item = ZhaopinItem()
jobdata = response.body
require_data = response.xpath(
'//body/div[@class="terminalpage clearfix"]/div[@class="terminalpage-left"]/div[@class="terminalpage-main clearfix"]/div[@class="tab-cont-box"]/div[1]/p').extract()
require_data_middle = ''
for i in require_data:
i_middle = re.sub(r'<.*?>', r'', i, re.S)
require_data_middle = require_data_middle + re.sub(r'\s*', r'', i_middle, re.S)
jobsoup = BeautifulSoup(jobdata, 'lxml')
item = response.meta['item']
item['job_require'] = require_data_middle
item['experience'] = jobsoup.select('div.terminalpage-left strong')[4].text.strip()
item['comany_name'] = jobsoup.select('div.fixed-inner-box h2')[0].text
item['comany_size'] = jobsoup.select('ul.terminal-ul.clearfix li strong')[8].text.strip()
item['head_count'] = jobsoup.select('div.terminalpage-left strong')[6].text.strip()
item['address'] = jobsoup.select('ul.terminal-ul.clearfix li strong')[11].text.strip()
item['education_require'] = jobsoup.select('div.terminalpage-left strong')[5].text.strip()
yield item
当然最后还需要对一些基础的配置在setting文件中进行设置,如下
ROBOTSTXT_OBEY = False
ITEM_PIPELINES = {
'zhaopin.pipelines.ZhaopinPipeline':300
}
MYSQL_HOST = '127.0.0.1'
MYSQL_DBNAME = 'zhaopin' # 数据库名
MYSQL_USER = 'root' # 数据库用户
MYSQL_PASSWORD = '123456' # 数据库密码
最后,运行成功会获得如下结果:
后记
后面如果我开发了数据分析相关的技能包,可能还会对这里的数据进行分析,到时候会将分析的一些有趣的东西分析出来,
代码请戳这里