一、主题式网络爬虫设计方案(15分)
1.主题式网络爬虫名称
2.主题式网络爬虫爬取的内容与数据特征分析
3.主题式网络爬虫设计方案概述(包括实现思路与技术难点)
二、主题页面的结构特征分析(15分)
1.主题页面的结构特征
2.Htmls页面解析
3.节点(标签)查找方法与遍历方法
(必要时画出节点树结构)
titleImgUrl = dataTag.find('div', class_='img').find('img')['src']
三、网络爬虫程序设计(60分)
爬虫程序主体要包括以下各部分,要附源代码及较详细注释,并在每部分程序后面提供输出结果的截图。
①爬虫调度器模块
主要负责其他模块的协调工作
文件相对地址(文件名):奥迪/SpiderMan.py
#coding:utf-8 from DataOutput import DataOutput from UrlManager import UrlManager from HtmlParser import HtmlParser from SeabornShow import SeabornShow from HtmlDownloader import HtmlDownloader import seaborn as sns class SpiderMan(object): def __init__(self): self.manager = UrlManager() self.downloader = HtmlDownloader() self.parser = HtmlParser() self.output = DataOutput() self.show = SeabornShow() def crawl(self,root_url): self.manager.add_new_url(root_url) while (self.manager.has_new_url() and self.manager.old_url_size() < 100): try: new_url = self.manager.get_new_url() print("》》开始下载页面内容") html = self.downloader.download(new_url) print("》》开始接解析下载的页面") new_urls,data = self.parser.parser(new_url,html) self.output.store_data(data) except: print("crawl failed") print("》》对解析的数据进行mysql数据库持久化操作") self.output.output_mysql() # 数据帧格式数据 df = self.output.mysql_to_pandas() print("》》散点图展示奥迪油耗跟价格的关系") self.show.show(df) if __name__ == "__main__": spider_man = SpiderMan() aodi = "http://car.bitauto.com/tree_chexing/mb_9/" # 奥迪列表页地址: spider_man.crawl(aodi)
② Url管理模块
维护爬取的url,跟未爬取的url地址
文件相对地址(文件名):奥迪/UrlManager.py
#coding:utf-8 ''' url管理器 ''' class UrlManager(object): def __init__(self): self.new_urls = set() self.old_urls = set() def has_new_url(self): ''' 判断是否有url未被爬取 :return: ''' return self.new_url_size() != 0 def get_new_url(self): ''' 获取url :return: ''' if self.has_new_url(): new_url = self.new_urls.pop() self.old_urls.add(new_url) return new_url else: return None def add_new_url(self,url): ''' 增加url :param url: :return: ''' if url is None: return ''' 增加时去重跟判断以处理的url避免重复处理出现死循环 ''' if url not in self.new_urls and url not in self.old_urls: self.new_urls.add(url) def add_new_urls(self,urls): ''' 增加一组url :param urls: :return: ''' if urls is None or len(urls)==0: return for url in urls: self.add_new_url(url) def new_url_size(self): return len(self.new_urls) def old_url_size(self): return len(self.old_urls)
③数据库操作工具
主要负责数据库的连接,管理增删改查
import pymysql as ps import pandas as pd class MysqlHelper: def __init__(self, host, user, password, database, charset): self.host = host self.user = user self.password = password self.database = database self.charset = charset self.db = None self.curs = None # 数据库连接 def open(self): self.db = ps.connect(host=self.host, user=self.user, password=self.password,database=self.database) self.curs = self.db.cursor() # 数据库关闭 def close(self): self.curs.close() self.db.close() # 数据增删改 def aud(self, sql, params): self.open() try: row = self.curs.execute(sql, params) self.db.commit() self.close() return row except : print('cud出现错误') self.db.rollback() self.close() return 0 # 解析为pandas def findPandas(self,sql): self.open() try: df = pd.read_sql(sql=sql,con=self.db) return df except: print('解析为pandas出现错误')
④数据库实体对象
包含数据库的个个字段之间的映射
class AODIItems(object): def __init__(self,name,titleImgUrl,referencePrice,guidePrice,displacement,oilConsumption): self.name = name self.titleImgUrl = titleImgUrl self.referencePrice = referencePrice self.guidePrice = guidePrice self.displacement = displacement self.oilConsumption = oilConsumption
1.数据爬取与采集
负责下载url管理器中提供的未爬url链接并在html
文件相对地址(文件名):奥迪/HtmlDownloader.py
#coding:utf-8 import requests import chardet ''' html下载器 ''' class HtmlDownloader(object): def download(self,url): try: if url is None: return sessions = requests.session() sessions.headers[ 'User-Agent'] = 'Mozilla/5.0 (Macintosh; Intel Mac OS X 10_9_2) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/34.0.1847.131 Safari/537.36' r = sessions.get(url) if (r.status_code == 200): r.encoding = chardet.detect(r.content)["encoding"] return r.text return None except: print("downloader failed") if __name__ == "__main__": pass
2.对数据进行清洗和处理
解析下载器的html页面,并解析出有效数据,也可以解析跟进的url链接
内嵌一个小的详情页爬虫,包括DetailsParser.py,DetailsDownloader.py
文件相对地址(文件名):奥迪/HtmlParser.py
#coding:utf-8 import re import urlparser import urllib import urllib3 from bs4 import BeautifulSoup ''' 奥迪详情html解释器 ''' class DetailsParser(object): def parser(self,page_url,html_cont): try: if page_url is None and html_cont is None: return soup = BeautifulSoup(html_cont, "html.parser") new_datas = self._get_new_data(page_url, soup) return new_datas except: print("DetailsParser failed") ''' 获取奥迪详细信息 ''' def _get_new_data(self, page_url, soup): contTag = soup.find('h5',id='factory-price').find_parent().find_parent() data = {} # 指导价格 data['guidePrice'] = soup.find('h5',id='factory-price').find('span',class_='price').string # 排量 data['displacement'] = contTag.find_next_sibling().select('li:nth-of-type(1)')[0].find('span',class_='data').string # 油耗 data['oilConsumption'] = contTag.find_next_sibling().select('li:nth-of-type(2)')[0].find('a',class_='data').string return data if __name__ == "__main__": pass
详情页内容下载与解析
文件相对地址(文件名):奥迪/DetailsDownloader.py
#coding:utf-8 import requests import chardet class DetailsDownloader(object): def download(self, url): try: if url is None: return sessions = requests.session() sessions.headers[ 'User-Agent'] = 'Mozilla/5.0 (Macintosh; Intel Mac OS X 10_9_2) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/34.0.1847.131 Safari/537.36' r = sessions.get(url) if (r.status_code == 200): r.encoding = chardet.detect(r.content)["encoding"] return r.text return None except: print("DetailsDownloader failed") if __name__ == "__main__": pass
文件相对地址(文件名):奥迪/DetailsParser.py
#coding:utf-8 import re import urlparser import urllib import urllib3 from bs4 import BeautifulSoup ''' 奥迪详情html解释器 ''' class DetailsParser(object): def parser(self,page_url,html_cont): try: if page_url is None and html_cont is None: return soup = BeautifulSoup(html_cont, "html.parser") new_datas = self._get_new_data(page_url, soup) return new_datas except: print("DetailsParser failed") ''' 获取奥迪详细信息 ''' def _get_new_data(self, page_url, soup): contTag = soup.find('h5',id='factory-price').find_parent().find_parent() data = {} # 指导价格 data['guidePrice'] = soup.find('h5',id='factory-price').find('span',class_='price').string # 排量 data['displacement'] = contTag.find_next_sibling().select('li:nth-of-type(1)')[0].find('span',class_='data').string # 油耗 data['oilConsumption'] = contTag.find_next_sibling().select('li:nth-of-type(2)')[0].find('a',class_='data').string return data if __name__ == "__main__": pass
3.数据分析与可视化
(例如:数据柱形图、直方图、散点图、盒图、分布图、数据回归分析等)
文件相对地址(文件名):奥迪/SeabornShow.py
import seaborn as sns import pandas as pd import pymysql as ps import numpy as np import scipy.stats as sci import matplotlib.pyplot as plt import re # seaborn构建散点图 class SeabornShow(object): ''' 奥迪油耗参考价数据的展示函数 ''' def show(self,data): sns.set(style="dark") # 数据整理 for i in range(len(data)): # 油耗 取算数平均值 data.iloc[i, 5] = self.ave(self.analyticalFigures(data.iloc[i, 5])) # 参考价 取算数平均值 data.iloc[i, 3] = self.ave(self.analyticalFigures(data.iloc[i, 3])) print(data[['id','name','referencePrice','displacement']]) sns.jointplot(x='displacement',y='referencePrice',data=data) plt.rcParams['font.sans-serif'] = ['SimHei'] plt.ylabel("油耗-价格对照表") plt.show() def analyticalFigures(self,str): return re.compile(r'[\d.]+').findall(str) def ave(self,numList): result = 0 sum = 0.0 for num in numList: sum = sum + float(num) result = sum / len(numList) return result if __name__ == '__main__': seabornShow = SeabornShow() # seabornShow.show() rs = np.random.RandomState(2) df = pd.DataFrame(rs.randn(200, 2), columns=['A', 'B']) # print(df) sns.jointplot(x='A',y='B',data=df,kind='reg')
4.数据持久化
将解析器解析处理的数据持久化化到mysql数据库
文件相对地址(文件名):奥迪/DataOutput.py
#coding:utf-8 import codecs from MysqlHelper import MysqlHelper class DataOutput(object): def __init__(self): self.datas=[] self.host = "localhost" self.user = "root" self.password = "" self.database = "ai_info" self.charset = "utf-8" self.mh = MysqlHelper(self.host,self.user,self.password,self.database,self.charset) def store_data(self,data): if data is None: return self.datas = data # 在mysql数据库持久化 def output_mysql(self): TABLE_NAME = "ad_data" sql = "insert into " + TABLE_NAME + " (name, titleImgUrl, referencePrice, guidePrice, displacement, oilConsumption) values(%s,%s,%s,%s,%s,%s)" rows = 0 for data in self.datas: name = data['name'] titleImgUrl = data['titleImgUrl'] referencePrice = data['referencePrice'] guidePrice = data['guidePrice'] displacement = data['displacement'] oilConsumption = data['oilConsumption'] params = (name, titleImgUrl, referencePrice, guidePrice, displacement, oilConsumption) row = self.mh.aud(sql,params) rows = rows + row print("*******插入%s 辆车的信息成功!" % rows) ''' 取轿车信息并转化为pandas 的数据帧类型存储 ''' def mysql_to_pandas(self): TABLE_NAME = "ad_data" sql = "select * from " + TABLE_NAME return self.mh.findPandas(sql)
表ad_data
id |
Int |
自增主键 |
name |
Varchar(255) |
汽车名称 |
titleImgURl |
Varchar(255) |
镖旗图片 |
referencePrice |
Varchar(255) |
参考价 |
guidePrice |
Varchar(255) |
厂商价 |
Displacement |
Varchar(255) |
排量 |
OilConsumption |
Varchar(255) |
油耗
|
四、结论(10分)
1.经过对主题数据的分析与可视化,可以得到哪些结论?
2T的排气量的奥迪普遍比较贵
2.对本次程序设计任务完成的情况做一个简单的小结。
克服了很多困难,百度找教程,请教他人,最后与组员一起终于完成了这次的程序设计。
感觉收获到很多东西,纸上得来终觉浅”,只有去实践了才能真正的掌握牢知识!
来源:https://www.cnblogs.com/1363598499zzr/p/12010040.html