python + BeautifulSoup + selenium 实现爬取中医智库的古籍分类的数据

爬取内容为

该图片下的七个分类, 然后对应的每个种类的书本信息(摘要和目录)

效果为

代码如下

import requests
from bs4 import BeautifulSoup
import re
import time
from selenium import webdriver
from selenium.common.exceptions import ElementNotInteractableException, NoSuchElementException
import pymongo

browser = webdriver.Chrome()
browser2 = webdriver.Chrome()

headers = {
    'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) '
                  'Chrome/80.0.3987.163 Safari/537.36 '
}
# 第一层的七个分类
allMap = {}
# 热门
hotMap = {}  # range(0,6)
# 基础典籍
basicBooks = {}  # range(6,21)
# 养生保健
health = {}  # range(21,22)
# 孤本
orphan = {}  # range(22,23)
# 临症各科
clinicalDepartments = {}  # range(23, 35)    23 + 12
# 争鸣流派
chirp = {}  # range(35, 44)  +9
# 历代医家
doctors = {}  # range(44, 82)  +13*3-1

def toMongoDB(Keyword, name, bookCount, nameUrl, bookName, bookUrl, bookAuthor, bookDescription, digest, catlog):
    myClient = pymongo.MongoClient("mongodb://localhost:27017/")
    myDb = myClient["doctor"]
    myCollection = myDb[str(Keyword) + "_" + str(name)]
    myDictionary = {
        "Keyword": name,
        "bookCount": bookCount,
        "KeywordUrl": nameUrl,
        "Book": [
            bookName, bookUrl, bookAuthor, bookDescription, digest, catlog
        ]
    }
    result = myCollection.insert_one(myDictionary)
    print(result)
    pass

def splitIntoSevenMap():
    for i, (k, v) in enumerate(allMap.items()):
        if i in range(0, 6):
            hotMap[k] = v
            # print(k, v)
        if i in range(6, 21):
            basicBooks[k] = v
            # print(k, v)
        if i in range(21, 22):
            health[k] = v
            # print(k, v)
        if i in range(22, 23):
            orphan[k] = v
            # print(k, v)
        if i in range(23, 35):
            clinicalDepartments[k] = v
            # print(k, v)
        if i in range(35, 44):
            chirp[k] = v
            # print(k, v)
        if i in range(44, 82):
            doctors[k] = v
            # print(k, v)
    pass

def getSevenPart(html1, html2):
    domain = 'https://www.zk120.com'

    soup = BeautifulSoup(html1, 'html.parser')
    # ResultSets = soup.find_all('h2', class_='group_title')
    ResultSets = soup.find_all('h2', class_='group-hot-title')
    # print(ResultSets)  # 得到所有的标题 非热门的
    for ResultSet in ResultSets:
        tt = re.match('''<h2 class="group(.*)title">(.*)</h2>''', str(ResultSet))
    liSoup = soup.find_all('li')
    for i in liSoup:
        lis = BeautifulSoup(str(i), 'html.parser')
        allMap[lis.find('a').text] = domain + lis.find('a').get('href')

    soup = BeautifulSoup(html2, 'html.parser')
    ResultSets = soup.find_all('h2', class_='group_title')
    for ResultSet in ResultSets:
        tt = re.match('''<h2 class="group(.*)title">(.*)</h2>''', str(ResultSet))
    liSoup = soup.find_all('li')

    for i in liSoup:
        lis = BeautifulSoup(str(i), 'html.parser')
        allMap[lis.find('a').text] = domain + lis.find('a').get('href')

    # for k, v in allMap.items():
    #     print(str(k) + "  " + str(v))

def getFirstLevel(pageUrl):
    res = requests.get(pageUrl, headers=headers)
    res.encoding = "utf8"
    print(res.url)
    # 打印出页面的所有代码
    # print(res.text)
    soup = BeautifulSoup(res.text, 'html.parser')
    ResultSets = soup.find_all('section', class_='ice_bg space_pr space_pl')
    # 生成七类
    getSevenPart(str(ResultSets[0]), str(ResultSets[1]))
    # 分类放入map中
    splitIntoSevenMap()

def accordingPartGetResults(name):
    print(name)
    # 热门
    if name == "hotMap":
        for k, v in hotMap.items():
            print(str(k) + " " + str(v))
            getPerResults(name, str(k), str(v))
    # 基础典籍
    elif name == "basicBooks":
        for k, v in basicBooks.items():
            print(str(k) + " " + str(v))
            getPerResults(name, str(k), str(v))
    # 养生保健
    elif name == "health":
        for k, v in health.items():
            print(str(k) + " " + str(v))
            getPerResults(name, str(k), str(v))
    # 孤本
    elif name == "orphan":
        for k, v in orphan.items():
            print(str(k) + " " + str(v))
            getPerResults(name, str(k), str(v))
    # 临症各科
    elif name == "clinicalDepartments":
        for k, v in clinicalDepartments.items():
            print(str(k) + " " + str(v))
            getPerResults(name, str(k), str(v))
    # 争鸣流派
    elif name == "chirp":
        for k, v in chirp.items():
            print(str(k) + " " + str(v))
            getPerResults(name, str(k), str(v))
    # 历代医家
    elif name == "doctors":
        for k, v in doctors.items():
            print(str(k) + " " + str(v))
            getPerResults(name, str(k), str(v))

def getPerResults(KeywordName, name, nameUrl):
    try:
        browser.get(nameUrl)
        js = 'window.scrollTo(0,document.body.scrollHeight)'  # 下滑到底部

        try:
            try:
                while browser.find_element_by_class_name('getMoreBtn') is not None:
                    browser.execute_script(js)
                    time.sleep(1)
                    browser.find_element_by_class_name('getMoreBtn').click()
            except ElementNotInteractableException:
                print("已经加载完毕")
        except NoSuchElementException:
            print("这个就一页")

        time.sleep(1)
        browser.execute_script(js)
        webElement = browser.find_elements_by_class_name("book_wrapper")
        bookCount = len(webElement)
        print("书本的数量为 -> " + str(bookCount))
        for i in webElement:

            try:
                bookUrl = i.find_element_by_tag_name('a').get_attribute('href')
            except NoSuchElementException:
                bookUrl = "没有给url"

            try:
                bookName = i.find_element_by_tag_name('a').text
            except NoSuchElementException:
                bookName = "没有给书名"

            try:
                bookAuthor = i.find_element_by_class_name('book-author').text
            except NoSuchElementException:
                bookAuthor = "没有给作者名字"

            try:
                bookDescription = i.find_element_by_class_name('book-description').text
            except NoSuchElementException:
                try:
                    bookDescription = i.find_element_by_class_name('book-comment').text
                except NoSuchElementException:
                    bookDescription = "没有给书的描述"

            print(str(name) + "中的----> " + str(bookName) + " " + bookUrl + " " + str(bookAuthor) + " " + str(
                bookDescription))

            browser2.get(bookUrl)
            time.sleep(1)
            try:
                digest = browser2.find_element_by_css_selector(
                     "#book_abstract > div.abstract-wrapper-div.abstract_wrapper_div.abstract-wrapper-divnew > div")

            except NoSuchElementException:
                print(str(bookName) + "的摘要是使用了第二种方式")
                digest = browser2.find_element_by_css_selector(
                    "#book_abstract > div.abstract_wrapper_div > div")
            finally:
                pass

            catalog = browser2.find_element_by_xpath('''//*[@id="book_catalog"]/table/tbody''')
            trLists = catalog.find_elements_by_tag_name('tr')
            Text = ""
            for tr in trLists:
                Text = Text + str(
                    tr.find_element_by_tag_name('td').find_element_by_tag_name('a').get_attribute('innerHTML')) + " "

            toMongoDB(KeywordName, name, bookCount, nameUrl, bookName, bookUrl, bookAuthor, bookDescription,
                      digest.text, Text)
            time.sleep(1)
    finally:
        print(str(name) + " 已经爬取完毕")
    pass

if __name__ == '__main__':
    page1 = 'https://www.zk120.com/ji/group/?nav=ys'
    getFirstLevel(page1)
    # 下面的是依次获取每种分类的具体信息 需要爬取哪个类的就取注
    # accordingPartGetResults("hotMap")
    # accordingPartGetResults("basicBooks")  # 备查宝典有几条没查完 备查宝典里的中医人名词典这里一直卡死
    # accordingPartGetResults("health")
    # accordingPartGetResults("orphan")
    # accordingPartGetResults("clinicalDepartments")
    # accordingPartGetResults("chirp")
    # accordingPartGetResults("doctors")