Python 康德乐大药房网站爬虫，使用bs4获取json，导入mysql

故事开端

目标地址

https://www.baiji.com.cn  康德乐大药房

程序流程

"爬取数量是否达标"这块没有写代码，因为我发现单页数量够多了，每页爬一半数据大概有8000条

获取商品价格接口

我发现商品价格在soup结构里面没有生成，是空值，应该是动态请求加载的。

就讲一下这部分接口我是怎么获取的吧，使用软件 Fiddler 进行抓包，如果你不知道这是什么 Fiddler使用方式

随便打开一个网站

可以看到请求的地址，这是一个 get 请求，包含了三个参数分别是 jsonp+时间戳、act 方法、goods_id 商品id，返回体是一个json，显示商品价格（注意！商品可能缺货会返回 {"price":"特价中"}，所以要做判断处理，可以看 spider.py 里的 product() 函数）

    # ^ 获取价格
    goods_id = re.search(r'\d+', href).group(0)
    url = "https://www.baiji.com.cn/domain/goods_info.php?fn=jsonp{}&act=goods_info&goods_id={}".format(int(time.time()), goods_id)

    headers_2 = {
        'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/99.0.4844.51 Safari/537.36 Edg/99.0.1150.39',
        'Referer': "https://www.baiji.com.cn/goods-{}.html".format(goods_id)
    }

    cookie = {
        'Cookie': 'NTKF_T2D_CLIENTID=guest59D35491-8F98-DD91-B9BE-A1ADEB5C6CBF; real_ipd=112.10.183.174; UserId=0; kdlusername=0; kdl_username=0; ECS_ID=887e4472a5a3613c59df1c0e60cc2a93a40d401e; nTalk_CACHE_DATA={uid:kf_9261_ISME9754_guest59D35491-8F98-DD,tid:1647709025522701}; ECS[history]=7526%2C7522%2C14684%2C15717%2C11560; arp_scroll_position=300'
    }
    res = requests.get(url, headers=headers_2, cookies=cookie, timeout=None)
    res.encoding = "utf-8"
    MyJson = loads_jsonp(res.text)
    try:
        price = float(MyJson['price'])
    except:
        price = 0
        isStockOut = True

可以看到我们先处理了 jsonp 转化为 json 对象，对取得的 json 的 price 键对应值进行 float 类型转换，如果转换失败会抛出异常，我们使用 try: ? expect: ? 的方式进行判断 

代码

代码写的很难看，但是能用，Wow！It just works！

work.py

import os
import json
import threading
import requests
import re
import time
from spider import Spider
from bs4 import BeautifulSoup


class MyThread(threading.Thread):
    """重写多线程，使其能够返回值"""

    def __init__(self, target=None, args=()):
        super(MyThread, self).__init__()
        self.func = target
        self.args = args

    def run(self):
        self.result = self.func(*self.args)

    def get_result(self):
        try:
            return self.result  # 如果子线程不使用join方法，此处可能会报没有self.result的错误
        except Exception:
            return None


host = "https://www.baiji.com.cn"

headers = {
    'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/99.0.4844.51 Safari/537.36 Edg/99.0.1150.39'
}

res = requests.get(host, headers=headers)
res.encoding = "utf-8"

soup = BeautifulSoup(res.text, "html.parser")

item_list = soup.find_all(class_='tcon')

items = []

for item in item_list:
    a = item.find_all('a')
    for i in a:
        href = i['href'].strip('// http:')
        title = i['title']
        if re.search('category', href) != None:
            items.append((href, title))

# ^ only for test 暴力开启多线程
infos = []
threads = []
start = time.perf_counter()

for item in items:
    t = MyThread(target=Spider, args=(item,))
    threads.append(t)
    t.start()

print('启动！')

# ^ 等待所有子线程结束,主线程再运行
for t in threads:
    t.join()
    result = t.get_result()
    infos.extend(result)

with open('data.json', 'w', encoding='utf-8') as f:
    f.write(json.dumps(infos, indent=4, ensure_ascii=False))

exit()

 

spider.py

import time
import os
import json
import requests
import re
from bs4 import BeautifulSoup

COUNT = 0
host = "https://"
host_ = "https://www.baiji.com.cn/"

headers = {
    'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/99.0.4844.51 Safari/537.36 Edg/99.0.1150.39'
}

path = os.path.dirname(__file__)


# ^ 主入口
def Spider(item):
    # temp = []
    # isPrescription = False
    category = item[1]
    classes = CatchClasses(item[0])
    infos = GoPage(classes, category)
    return infos


# ^ 获取所有剂型
def CatchClasses(url):
    res = requests.get(host+url, headers=headers)
    res.encoding = "utf-8"
    soup = BeautifulSoup(res.text, "html.parser")
    temp = soup.find(class_="det").find_all('a')
    classes = [a['href'] for a in temp]
    titles = [a['title'] for a in temp]
    return [classes, titles]


# ^ 进行爬虫
def GoPage(classes, category):
    isPrescribed = True
    length = len(classes[0])
    href = classes[0]
    titles = classes[1]
    infos = []
    for index, h_ in enumerate(href):
        Collect(h_, '-0.html', titles, index, category, infos)
# ---------------------------------------------------------------------------- #
        Collect(h_, '-1.html', titles, index, category, infos)
    print('已处理', COUNT, '个商品')
    return infos


def Collect(h_, add, titles, index, category, infos):
    h = h_.replace('.html', add)
    url = host_ + h
    res = requests.get(url, headers=headers)
    res.encoding = "utf-8"
    # ^ 获取该类别共几条商品记录
    target = re.findall(r'共计<strong class="red">\d+</strong>', res.text)
    target = re.search(r'\d+', target[0]).group(0)
    target = int(target)
    # ^ 如果商品记录为 0 直接跳过
    if target != 0:
        # ^ 减少爬的数据量
        target = int(target/2+1)
        print(titles[index], "需要获取的数据条目", target)

        isPrescribed = False
        products = GetProducts(h)
        print("非处方页面共有", len(products), "个商品", time.ctime())
        for product in products:
            info = Product(product, index)
            info['category'] = category
            info['isPrescribed'] = isPrescribed
            info['class'] = titles[index]
            infos.append(info)


# ^ 获取当前页面所有产品
def GetProducts(url):
    res = requests.get(host_+url, headers=headers)
    res.encoding = "utf-8"
    soup = BeautifulSoup(res.text, "html.parser")
    products = soup.find_all(class_='pro_boxin')
    return products


# ^ 解析jsonp对象
def loads_jsonp(_jsonp):
    try:
        return json.loads(re.match(".*?({.*}).*", _jsonp, re.S).group(1))
    except:
        raise ValueError('Invalid Input')


# ^ 获取指定商品信息
# @ info 传入一个 class = pro_boxin 的soup对象
def Product(p, category):
    global COUNT
    COUNT += 1
    a = p.find(class_="name").a
    isStockOut = False  # IMPORTANT
    productname = a['title']  # IMPORTANT
    QuantityPerunit = a.contents[0].strip().split(' ')[1]  # IMPORTANT
    describe = a.span.string  # IMPORTANT
    img = p.find('img')
    try:
        imageSrc = img['data-original'].strip('//')  # IMPORTANT
    # ^ 下载图片到本地
        imageSrc = DownloadPic(imageSrc, category)
    except:
        imageSrc = None
    # ^ 准备获取商品页面
    href = a['href']  # IMPORTANT
    # print(productname, unit, href)

    # ^ 获取价格
    goods_id = re.search(r'\d+', href).group(0)
    url = "https://www.baiji.com.cn/domain/goods_info.php?fn=jsonp{}&act=goods_info&goods_id={}".format(int(time.time()), goods_id)

    headers_2 = {
        'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/99.0.4844.51 Safari/537.36 Edg/99.0.1150.39',
        'Referer': "https://www.baiji.com.cn/goods-{}.html".format(goods_id)
    }

    cookie = {
        'Cookie': 'NTKF_T2D_CLIENTID=guest59D35491-8F98-DD91-B9BE-A1ADEB5C6CBF; real_ipd=112.10.183.174; UserId=0; kdlusername=0; kdl_username=0; ECS_ID=887e4472a5a3613c59df1c0e60cc2a93a40d401e; nTalk_CACHE_DATA={uid:kf_9261_ISME9754_guest59D35491-8F98-DD,tid:1647709025522701}; ECS[history]=7526%2C7522%2C14684%2C15717%2C11560; arp_scroll_position=300'
    }
    res = requests.get(url, headers=headers_2, cookies=cookie, timeout=None)
    res.encoding = "utf-8"
    MyJson = loads_jsonp(res.text)
    try:
        price = float(MyJson['price'])
    except:
        price = 0
        isStockOut = True
    # print(price)
    return dict(
        productname=productname,
        QuantityPerunit=QuantityPerunit,
        unitprice=price,
        isStockOut=isStockOut,
        notes=describe,
        imageSrc=imageSrc
    )


def DownloadPic(src, category):
    category = str(category+1)
    # print(src)
    fileName = src.split('/')
    fileName.reverse()
    fileName = fileName[0].strip('/')
    # print(fileName)
    pic = requests.get(host+src)

    savePath = path+'/goods_img/'+category+'/'

    isExists = os.path.exists(savePath)
    if not isExists:
        # 如果不存在则创建目录
        # 创建目录操作函数
        os.makedirs(savePath)
    with open(savePath+fileName, 'wb') as b:
        b.write(pic.content)
    return '/goods_img/'+category+'/'+fileName

 

json_to_mysql.py

import os
import json
import random
import mysql.connector

path = os.path.dirname(__file__)

db = mysql.connector.connect(
    host="localhost",
    user="root",
    passwd="sql2008",
    database="test",
    auth_plugin='mysql_native_password'
)

cursor = db.cursor()

with open(path+'/data.json', 'r', encoding='utf-8') as f:
    data = json.load(f)

print(data[0], len(data))

cursor.execute("SELECT categoryid, categoryname FROM t_categorytreeb")

res = cursor.fetchall()

categoryid = [i[0] for i in res]
categoryname = [i[1] for i in res]
print(categoryname.index('丙肝'))

cursor.execute("SELECT supplierid FROM t_suppliers")

res = cursor.fetchall()
suppliers = [i[0] for i in res]
print(suppliers)

# ^ 插入数据

sql = "INSERT INTO t_medicines(ProductName,QuantityPerunit,Unitprice,SupplierID,SubcategoryID,Photopath,notes,ytype,isPrescribed,isStockOut) VALUES (%s,%s,%s,%s,%s,%s,%s,%s,%s,%s)"
val = [
]
print()
for item in data:
    try:
        # print(item['category'], categoryname.index(item['category']))
        index = categoryname.index(item['category'].strip())
        cid = categoryid[index]
    except:
        cid = None
    val.append((
        item["productname"],
        item["QuantityPerunit"],
        item["unitprice"],
        random.choice(suppliers),
        cid,
        item["imageSrc"],
        item["notes"],
        item["class"],
        item["isPrescribed"],
        item["isStockOut"]
    ))


cursor.executemany(sql, val)

db.commit()

print(cursor.rowcount, "was inserted.")

数据集

2020_3/work.sql

2020_3/data.zip  （80MB）包括图片

故事结尾