python爬取亚马逊简单的书籍信息

我有个需求就是抓取一些简单的书籍信息存储到mysql数据库，例如，封面图片，书名，类型，作者，简历，出版社，语种。

我比较之后，决定在亚马逊来实现我的需求。

我分析网站后发现，亚马逊有个高级搜索的功能，我就通过该搜索结果来获取书籍的详情URL。

由于亚马逊的高级搜索是用get方法的，所以通过分析，搜索结果的URL，可得到node参数是代表书籍类型的。field-binding_browse-bin是代表书籍装饰。

所以我固定了书籍装饰为平装，而书籍的类型，只能每次运行的时候，爬取一种类型的书籍难过

之后就是根据书籍详情页面利用正则表达式来匹配需要的信息了。

以下源代码，命名不是很规范。。。

import requests
import sys
import re
import pymysql

class product:
    type="历史"
    name=""
    author=""
    desciption=""
    pic1=""
    languages=""
    press=""

def getProUrl():
    urlList = []
    headers = {"User-Agent":"Mozilla/5.0 (Windows NT 10.0; WOW64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/50.0.2661.102 Safari/537.36"}
    session = requests.Session()
    furl="https://www.amazon.cn/gp/search/ref=sr_adv_b/?search-alias=stripbooks&field-binding_browse-bin=2038564051&sort=relevancerank&page="
    for i in range(1,7):
        html=""
        print(furl+str(i)) 
        html = session.post(furl+str(i)+'&node=658418051',headers = headers)
        html.encoding = 'utf-8'
        s=html.text.encode('gb2312','ignore').decode('gb2312')
        url=r'
'
        reg=re.compile(url,re.M)
        items = reg.findall(html.text)
        for i in range(0,len(items)):
            urlList.append(items[i])
    urlList=set(urlList)
    return urlList

def getProData(url):
    pro = product()
    headers = {"User-Agent":"Mozilla/5.0 (Windows NT 10.0; WOW64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/50.0.2661.102 Safari/537.36"}
    session = requests.Session()
    zurl="https://www.amazon.cn/dp/"
    html = session.get(zurl+url,headers = headers)
    html.encoding = 'utf-8'
    s=html.text.encode('gb2312','ignore').decode('gb2312')
    pro.pic1=getProPic(html)
    pro.name=getProName(html)
    pro.author=getProAuthor(html)
    pro.desciption=getProDescrip(html)
    pro.press=getProPress(html)
    pro.languages=getProLanguages(html)
    return pro

def getProPic(html):
    pic=r'id="imgBlkFront" data-a-dynamic-image="{"(.+?)".*?}"'
    reg=re.compile(pic,re.M)
    items = reg.findall(html.text)
    if len(items)==0:
        return ""
    else:
        return items[0]

def getProName(html):
    name=r'
(.+?)
    reg=re.compile(name,re.M)
    items = reg.findall(html.text)
    if len(items)==0:
        return ""
    else:
        return items[0]

def getProAuthor(html):
    author=r'.*?(.+?).*?(.+?)'
    reg=re.compile(author,re.S)
    items = reg.findall(html.text)
    au=""
    for i in range(0,len(items)):
        au=au+items[i][0]+items[i][1]
    return au

def getProDescrip(html):
    Descrip=r'
(.+?)
.{0,30}.{0,30}.{0,30}

    reg=re.compile(Descrip,re.S)
    items = reg.findall(html.text)
    if len(items)==0:
        return ""
    else:
        position = items[0].find('海报：')
        descrip=items[0]
        if position != -1:
            descrip=items[0][0:position]
        return descrip.strip()

def getProPress(html):
    press=r'
出版社:(.+?)
'
    reg=re.compile(press,re.M)
    items = reg.findall(html.text)
    if len(items)==0:
        return ""
    else:
        return items[0].strip()


def getProLanguages(html):
    languages=r'
语种：(.+?)
'
    reg=re.compile(languages,re.M)
    items = reg.findall(html.text)
    if len(items)==0:
        return ""
    else:
        return items[0].strip()

def getConnection():
    config = {
          'host':'121.**.**.**',
          'port':3306,
          'user':'root',
          'password':'******',
          'db':'home_work',
          'charset':'utf8',
          'cursorclass':pymysql.cursors.DictCursor,
          }
    connection = pymysql.connect(**config)
    return connection

urlList = getProUrl()
i = 0
for d in urlList:
    i = i + 1
    print (i)
    connection = getConnection()
    pro = getProData(d)
    try:
        with connection.cursor() as cursor:
            sql='INSERT INTO books (type,name,author,desciption,pic1,languages,press) VALUES (%s,%s,%s,%s,%s,%s,%s)'
            cursor.execute(sql,(pro.type,pro.name,pro.author,pro.desciption,pro.pic1,pro.languages,pro.press))
        connection.commit()
    finally:
        connection.close();