python 爬取国内小说网站
作者:勤劳上班的卑微小张 时间:2022-11-30 16:02:04
目录
原理先行
实践篇
完整代码
原理先行
作为一个资深的小说爱好者,国内很多小说网站如出一辙,什么 🖊*阁啊等等,大都是 get 请求返回 html 内容,而且会有标志性的<dl><dd>等标签。
所以大概的原理,就是先 get 请求这个网站,然后对获取的内容进行清洗,写进文本里面,变成一个 txt,导入手机,方便看小说。
实践篇
之前踩过一个坑,一开始我看了几页小说,大概小说的内容网站是https://www.xxx.com/小说编号/章节编号.html,一开始看前几章,我发现章节编号是连续的, 于是我一开始想的就是记住起始章节编号,然后在循环的时候章节编号自增就行,后面发现草率了,可能看个 100 章之后,章节列表会出现断层现象,这个具体为啥 还真不知道,按理说小说编号固定,可以算是一个数据表,那里面的章节编号不就是一个自增 id 就完了嘛?有懂王可以科普一下!
所以这里要先获取小说的目录列表,并把目录列表洗成一个数组方便我们后期查找!getList.py文件:
定义一个请求书签的方法
# 请求书签地址
def req():
url = "https://www.24kwx.com/book/4/4020/"
strHtml = requests.get(url)
return strHtml.text
将获取到的内容提取出(id:唯一值/或第 X 章小说)(name:小说的章节名称)(key:小说的章节 id)
# 定义一个章节对象
class Xs(object):
def __init__(self,id,key,name):
self._id = id
self._key = key
self._name = name
@property
def id(self):
self._id
@property
def key(self):
self._key
@property
def name(self):
self._name
def getString(self):
return 'id:%s,name:%s,key:%s' %(self._id,self._name,self._key)
# 转换成书列表
def tranceList():
key = 0
name = ""
xsList = []
idrule = r'/4020/(.+?).html'
keyrule = r'第(.+?)章'
html = req()
html = re.split("</dt>",html)[2]
html = re.split("</dl>",html)[0]
htmlList = re.split("</dd>",html)
for i in htmlList:
i = i.strip()
if(i):
# 获取id
id = re.findall(idrule,i)[0]
lsKeyList = re.findall(keyrule,i)
# 如果有章节
if len(lsKeyList) > 0 :
key = int(lsKeyList[0])
lsname = re.findall(r'章(.+?)</a>',i)
else :
key = key + 1
# 获取名字
# lsname = re.findall(r'.html">(.+?)</a>',i)[0]
# name = re.sub(',',' ', lsname, flags=re.IGNORECASE)
name = re.findall(r'.html">(.+?)</a>',i)[0]
xsobj = Xs(id,key,name)
xsList.append(xsobj.getString())
writeList(xsList)
注意一下我:如果你从别的语言转 py,第一次写object对象可能会比较懵,没错因为他的object是一个class,这里我创建的对象就是{id,key,name}但是你写入 txt 的时候还是要getString,所以后面想想我直接写个{id:xxx,name:xxx,key:xxx}的字符串不就完了,还弄啥class,后面还是想想给兄弟盟留点看点,就留着了
最后写入 txt 文件
# 写入到文本
def writeList(list):
f = open("xsList.txt",'w',encoding='utf-8')
# 这里不能写list,要先转字符串 TypeError: write() argument must be str, not list
f.write('\n'.join(list))
print('写入成功')
# 大概写完的txt是这样的
id:3798160,name:第1章 孙子,我是你爷爷,key:1
id:3798161,name:第2章 孙子,等等我!,key:2
id:3798162,name:第3章 天上掉下个亲爷爷,key:3
id:3798163,name:第4章 超级大客户,key:4
id:3798164,name:第5章 一张退婚证明,key:5
ok ! Last one
这里已经写好了小说的目录,那我们就要读取小说的内容,同理
先写个请求
# 请求内容地址
def req(id):
url = "https://www.24kwx.com/book/4/4020/"+id+".html"
strHtml = requests.get(url)
return strHtml.text
读取我们刚刚保存的目录
def getList():
f = open("xsList.txt",'r', encoding='utf-8')
# 这里按行读取,读取完后line是个数组
line = f.readlines()
f.close()
return line
定义好一个清洗数据的规则
contextRule = r'<div class="content">(.+?)<script>downByJs();</script>'
titleRule = r'<h1>(.+?)</h1>'
def getcontext(objstr):
xsobj = re.split(",",objstr)
id = re.split("id:",xsobj[0])[1]
name = re.split("name:",xsobj[1])[1]
html = req(id)
lstitle = re.findall(titleRule,html)
title = lstitle[0] if len(lstitle) > 0 else name
context = re.split('<div id="content" class="showtxt">',html)[1]
context = re.split('</div>',context)[0]
context = re.sub(' |\r|\n','',context)
textList = re.split('<br />',context)
textList.insert(0,title)
for item in textList :
writeTxt(item)
print('%s--写入成功'%(title))
再写入文件
def writeTxt(txt):
if txt :
f = open("nr.txt",'a',encoding="utf-8")
f.write(txt+'\n')
最后当然是串联起来啦
def getTxt():
# 默认参数配置
startNum = 1261 # 起始章节
endNum = 1300 # 结束章节
# 开始主程序
f = open("nr.txt",'w',encoding='utf-8')
f.write("")
if endNum < startNum:
print('结束条数必须大于开始条数')
return
allList = getList()
needList = allList[startNum-1:endNum]
for item in needList:
getcontext(item)
time.sleep(0.2)
print("全部爬取完成")
完整代码
getList.py
import requests
import re
# 请求书签地址
def req():
url = "https://www.24kwx.com/book/4/4020/"
strHtml = requests.get(url)
return strHtml.text
# 定义一个章节对象
class Xs(object):
def __init__(self,id,key,name):
self._id = id
self._key = key
self._name = name
@property
def id(self):
self._id
@property
def key(self):
self._key
@property
def name(self):
self._name
def getString(self):
return 'id:%s,name:%s,key:%s' %(self._id,self._name,self._key)
# 转换成书列表
def tranceList():
key = 0
name = ""
xsList = []
idrule = r'/4020/(.+?).html'
keyrule = r'第(.+?)章'
html = req()
html = re.split("</dt>",html)[2]
html = re.split("</dl>",html)[0]
htmlList = re.split("</dd>",html)
for i in htmlList:
i = i.strip()
if(i):
# 获取id
id = re.findall(idrule,i)[0]
lsKeyList = re.findall(keyrule,i)
# 如果有章节
if len(lsKeyList) > 0 :
key = int(lsKeyList[0])
lsname = re.findall(r'章(.+?)</a>',i)
else :
key = key + 1
# 获取名字
# lsname = re.findall(r'.html">(.+?)</a>',i)[0]
# name = re.sub(',',' ', lsname, flags=re.IGNORECASE)
name = re.findall(r'.html">(.+?)</a>',i)[0]
xsobj = Xs(id,key,name)
xsList.append(xsobj.getString())
writeList(xsList)
# 写入到文本
def writeList(list):
f = open("xsList.txt",'w',encoding='utf-8')
# 这里不能写list,要先转字符串 TypeError: write() argument must be str, not list
f.write('\n'.join(list))
print('写入成功')
def main():
tranceList()
if __name__ == '__main__':
main()
writeTxt.py
import requests
import re
import time
# 请求内容地址
def req(id):
url = "https://www.24kwx.com/book/4/4020/"+id+".html"
strHtml = requests.get(url)
return strHtml.text
def getList():
f = open("xsList.txt",'r', encoding='utf-8')
# 这里按行读取
line = f.readlines()
f.close()
return line
contextRule = r'<div class="content">(.+?)<script>downByJs();</script>'
titleRule = r'<h1>(.+?)</h1>'
def getcontext(objstr):
xsobj = re.split(",",objstr)
id = re.split("id:",xsobj[0])[1]
name = re.split("name:",xsobj[1])[1]
html = req(id)
lstitle = re.findall(titleRule,html)
title = lstitle[0] if len(lstitle) > 0 else name
context = re.split('<div id="content" class="showtxt">',html)[1]
context = re.split('</div>',context)[0]
context = re.sub(' |\r|\n','',context)
textList = re.split('<br />',context)
textList.insert(0,title)
for item in textList :
writeTxt(item)
print('%s--写入成功'%(title))
def writeTxt(txt):
if txt :
f = open("nr.txt",'a',encoding="utf-8")
f.write(txt+'\n')
def getTxt():
# 默认参数配置
startNum = 1261 # 起始章节
endNum = 1300 # 结束章节
# 开始主程序
f = open("nr.txt",'w',encoding='utf-8')
f.write("")
if endNum < startNum:
print('结束条数必须大于开始条数')
return
allList = getList()
needList = allList[startNum-1:endNum]
for item in needList:
getcontext(item)
time.sleep(0.2)
print("全部爬取完成")
def main():
getTxt()
if __name__ == "__main__":
main()
来源:https://github.com/2277419213/novel-crawler