Python3利用BeautifulSoup4抓取站点小说全文的代码

再写一个用BeautifulSoup抓站的工具，体会BeautifulSoup的强大。

根据小说索引页获取小说全部章节内容并在本地整合为小说全文。不过不是智能的，不同的站点对代码需要做相应的修改。

#!/usr/bin/env python

import os
import sys
import re
import time
import chardet
import urllib.request as ur
from urllib.parse import urljoin,urlparse
from bs4 import BeautifulSoup
from threading import Thread

class Download(Thread):                          #为每个章节分配多线程
    def __init__(self,filepath,info):
        Thread.__init__(self)
        self.filepath = filepath
        (self.link,self.chapter) = info

    def run(self):
        print('开始下载： '+self.chapter)
        section(self.filepath,self.chapter,self.link)
        print('完成下载： '+self.chapter)

def getData(url):                          #主要用于判断页面编码，但是发现BeautifulSoup自带判定能力，故废弃此函数
    charsets = 'utf8'
    response = ur.urlopen(url,timeout = 10)
    html = response.read()
    charinfo = chardet.detect(html)
    charsets = charinfo['encoding']
    data = html.decode(charsets)
    return data

def merge(tmpFiles,targetFile):             #将下载的章节合并
    for tmpFile in tmpFiles:
        with open(targetFile,'a+') as wfile:
            wfile.write(open(tmpFile,'r').read())
        os.remove(tmpFile)

def content(link):                         #获取章节页面的小说内容。对于不同的站点，在此函数内修改获取章节内容的代码
    html = ur.urlopen(link,timeout = 10)
    soup =BeautifulSoup(html)
    contents = soup.find(id = 'readtext').p.span.text.replace('  ','
')   #BeautifulSoup会自动将&nbsp;转换为空格，<br/>转换为特殊符号
    return contents

def section(filepath,chapter,link):         #下载章节内容
    while True:                #反复请求页面
        try:
            with open(filepath,'w') as nfile:
                nfile.write(chapter+'
'+content(link)+'
')
            break
        except:
            pass
        
def index(url):                         #获取章节索引
    indexs = []
    while True:                   #反复请求页面
        try:
            html = ur.urlopen(url,timeout = 10)
            #html = html.read().decode('gb2312')
            #html = getData(url)
            soup = BeautifulSoup(html,from_encoding = 'gbk')#BeautifulSoup能自动识别编码，但是会将gbk页面识别为gb2312页面，可能导致页面内部分数据获取失败，故显式指定
            break
        except:
            pass
    title = soup.find(name = 'div',attrs = {'class':'booktext'}).text
    indexDiv = soup.find(name = 'div',attrs = {'class':'booktext'})
    indexUl = [ul for ul in indexDiv.find_all('ul') if ul][1:]
    for ul in indexUl:
        indexList = [li.a for li in ul.find_all('li') if li]
        index = [(urljoin(url,a.get('href')),a.text) for a in indexList if a]
        indexs +=index
    return indexs

def novel(url):
    tmpFiles = []
    tasks = []
    try:
        indexs = index(url)
        tmpDir = os.path.join(os.getcwd(),'tmp')
        if not os.path.exists(tmpDir):             #创建章节片段存放的临时目录
            os.mkdir(tmpDir)
        for i,info in enumerate(indexs):
            tmpFile = os.path.join(tmpDir,str(i))
            tmpFiles.append(tmpFile)
            task = Download(tmpFile,info)            #开启新线程下载章节内容
            task.setDaemon(True)
            task.start()
            tasks.append(task)
            if len(tasks) >= 20:                  #将线程总数控制在20个以内，如果线程过多会导致程序崩溃
                while len([task for task in tasks if task.isAlive()]):
                    print( '进度： {} / {}'.format(i+1-len([task for task in tasks if task.isAlive()]),len(indexs)))  #显示下载进度
                    time.sleep(2)
                tasks = []
            if i == len(indexs) - 1:
                while len([task for task in tasks if task.isAlive()]):
                    print( '进度： {} / {}'.format(len(indexs) - len([task for task in tasks if task.isAlive()]),len(indexs)))
                    time.sleep(2)
        print( '进度： {} / {}'.format(len(indexs),len(indexs)))
        print('开始整合......')
        merge(tmpFiles,os.path.join(os.getcwd(),title+'.txt'))
        print('下载成功！')
    except Exception as ex:
        print(ex)
        print('下载失败！')
        sys.exit()
def main(argv):
    try:
        novel(argv[0])
    except KeyboardInterrupt as kbi:            #使用<C-c>中断下载后仍然能将已下载的章节合并
        tmpDir = os.path.join(os.getcwd(),'tmp')
        if os.path.exists(tmpDir):
            tmpFiles = [os.path.join(tmpDir,tfile) for tfile in os.listdir(tmpDir) if os.path.isfile(os.path.join(tmpDir,tfile))]
            print('开始整合不完整的下载......')
            try:
                merge(tmpFiles,os.path.join(os.getcwd(),'不完整文档.txt'))
                if os.path.exists(os.path.join(os.getcwd(),'不完整文档.txt')):
                    print('部分章节下载成功！')
                else:
                    print('下载失败！')
            except:
                print('下载失败！')
                sys.exit()
            os.rmdir(tmpDir)
        else:
            print('下载失败！')
            sys.exit()
    if os.path.exists(os.path.join(os.getcwd(),'tmp')):    
        os.rmdir(os.path.join(os.getcwd(),'tmp'))

if __name__ == "__main__":
    if len(sys.argv) > 1:
        main(sys.argv[1:])
    #http://www.lueqiu.com/

截图：