python 爬取國(guó)內(nèi)小說(shuō)網(wǎng)站

更新時(shí)間：2021年06月07日 14:31:08 作者：勤勞上班的卑微小張

國(guó)內(nèi)小說(shuō)網(wǎng)站的結(jié)構(gòu)，大概都如出一轍，改改地址，就差不多了，有此需求的朋友可以參考下本文的爬蟲(chóng)寫(xiě)法

原理先行

作為一個(gè)資深的小說(shuō)愛(ài)好者，國(guó)內(nèi)很多小說(shuō)網(wǎng)站如出一轍，什么 🖊*閣啊等等，大都是 get 請(qǐng)求返回 html 內(nèi)容，而且會(huì)有標(biāo)志性的<dl><dd>等標(biāo)簽。
所以大概的原理，就是先 get 請(qǐng)求這個(gè)網(wǎng)站，然后對(duì)獲取的內(nèi)容進(jìn)行清洗，寫(xiě)進(jìn)文本里面，變成一個(gè) txt，導(dǎo)入手機(jī)，方便看小說(shuō)。

實(shí)踐篇

之前踩過(guò)一個(gè)坑，一開(kāi)始我看了幾頁(yè)小說(shuō)，大概小說(shuō)的內(nèi)容網(wǎng)站是https://www.xxx.com/小說(shuō)編號(hào)/章節(jié)編號(hào).html，一開(kāi)始看前幾章，我發(fā)現(xiàn)章節(jié)編號(hào)是連續(xù)的，于是我一開(kāi)始想的就是記住起始章節(jié)編號(hào)，然后在循環(huán)的時(shí)候章節(jié)編號(hào)自增就行，后面發(fā)現(xiàn)草率了，可能看個(gè) 100 章之后，章節(jié)列表會(huì)出現(xiàn)斷層現(xiàn)象，這個(gè)具體為啥還真不知道，按理說(shuō)小說(shuō)編號(hào)固定，可以算是一個(gè)數(shù)據(jù)表，那里面的章節(jié)編號(hào)不就是一個(gè)自增 id 就完了嘛？有懂王可以科普一下！
所以這里要先獲取小說(shuō)的目錄列表，并把目錄列表洗成一個(gè)數(shù)組方便我們后期查找！getList.py文件：

定義一個(gè)請(qǐng)求書(shū)簽的方法

# 請(qǐng)求書(shū)簽地址
def req():
    url = "https://www.24kwx.com/book/4/4020/"
    strHtml = requests.get(url)
    return strHtml.text

將獲取到的內(nèi)容提取出（id:唯一值/或第 X 章小說(shuō)）(name:小說(shuō)的章節(jié)名稱(chēng))（key:小說(shuō)的章節(jié) id）

# 定義一個(gè)章節(jié)對(duì)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)

# 轉(zhuǎn)換成書(shū)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):
            # 獲取id
            id = re.findall(idrule,i)[0]
            lsKeyList = re.findall(keyrule,i)
            # 如果有章節(jié)
            if len(lsKeyList) > 0 :
                key = int(lsKeyList[0])
                lsname = re.findall(r'章(.+?)</a>',i)
            else :
                key = key + 1
            # 獲取名字
            # 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：如果你從別的語(yǔ)言轉(zhuǎn) py，第一次寫(xiě)object對(duì)象可能會(huì)比較懵，沒(méi)錯(cuò)因?yàn)樗膐bject是一個(gè)class，這里我創(chuàng)建的對(duì)象就是{id,key,name}但是你寫(xiě)入 txt 的時(shí)候還是要getString，所以后面想想我直接寫(xiě)個(gè){id:xxx,name:xxx,key:xxx}的字符串不就完了，還弄啥class,后面還是想想給兄弟盟留點(diǎn)看點(diǎn)，就留著了

最后寫(xiě)入 txt 文件

# 寫(xiě)入到文本
def writeList(list):
    f = open("xsList.txt",'w',encoding='utf-8')
    # 這里不能寫(xiě)list，要先轉(zhuǎn)字符串 TypeError: write() argument must be str, not list
    f.write('\n'.join(list))
    print('寫(xiě)入成功')

# 大概寫(xiě)完的txt是這樣的
id:3798160,name:第1章 孫子，我是你爺爺,key:1
id:3798161,name:第2章 孫子，等等我！,key:2
id:3798162,name:第3章 天上掉下個(gè)親爺爺,key:3
id:3798163,name:第4章 超級(jí)大客戶(hù),key:4
id:3798164,name:第5章 一張退婚證明,key:5

ok ! Last one
這里已經(jīng)寫(xiě)好了小說(shuō)的目錄，那我們就要讀取小說(shuō)的內(nèi)容，同理

先寫(xiě)個(gè)請(qǐng)求

# 請(qǐng)求內(nèi)容地址
def req(id):
    url = "https://www.24kwx.com/book/4/4020/"+id+".html"
    strHtml = requests.get(url)
    return strHtml.text

讀取我們剛剛保存的目錄

def getList():
    f = open("xsList.txt",'r', encoding='utf-8')
    # 這里按行讀取,讀取完后line是個(gè)數(shù)組
    line = f.readlines()
    f.close()
    return line

定義好一個(gè)清洗數(shù)據(jù)的規(guī)則

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('&nbsp;|\r|\n','',context)
    textList = re.split('<br />',context)
    textList.insert(0,title)
    for item in textList :
        writeTxt(item)
    print('%s--寫(xiě)入成功'%(title))

再寫(xiě)入文件

def writeTxt(txt):
    if txt :
        f = open("nr.txt",'a',encoding="utf-8")
        f.write(txt+'\n')

最后當(dāng)然是串聯(lián)起來(lái)啦

def getTxt():
    # 默認(rèn)參數(shù)配置
    startNum = 1261 # 起始章節(jié)
    endNum = 1300 # 結(jié)束章節(jié)
    # 開(kāi)始主程序
    f = open("nr.txt",'w',encoding='utf-8')
    f.write("")
    if endNum < startNum:
        print('結(jié)束條數(shù)必須大于開(kāi)始條數(shù)')
        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")

完整代碼

getList.py

import requests
import re

# 請(qǐng)求書(shū)簽地址
def req():
    url = "https://www.24kwx.com/book/4/4020/"
    strHtml = requests.get(url)
    return strHtml.text

# 定義一個(gè)章節(jié)對(duì)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)

# 轉(zhuǎn)換成書(shū)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):
            # 獲取id
            id = re.findall(idrule,i)[0]
            lsKeyList = re.findall(keyrule,i)
            # 如果有章節(jié)
            if len(lsKeyList) > 0 :
                key = int(lsKeyList[0])
                lsname = re.findall(r'章(.+?)</a>',i)
            else :
                key = key + 1
            # 獲取名字
            # 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)

# 寫(xiě)入到文本
def writeList(list):
    f = open("xsList.txt",'w',encoding='utf-8')
    # 這里不能寫(xiě)list，要先轉(zhuǎn)字符串 TypeError: write() argument must be str, not list
    f.write('\n'.join(list))
    print('寫(xiě)入成功')


def main():
    tranceList()

if __name__ == '__main__':
    main()

writeTxt.py

import requests
import re
import time


# 請(qǐng)求內(nèi)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')
    # 這里按行讀取
    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('&nbsp;|\r|\n','',context)
    textList = re.split('<br />',context)
    textList.insert(0,title)
    for item in textList :
        writeTxt(item)
    print('%s--寫(xiě)入成功'%(title))

def writeTxt(txt):
    if txt :
        f = open("nr.txt",'a',encoding="utf-8")
        f.write(txt+'\n')

def getTxt():
    # 默認(rèn)參數(shù)配置
    startNum = 1261 # 起始章節(jié)
    endNum = 1300 # 結(jié)束章節(jié)
    # 開(kāi)始主程序
    f = open("nr.txt",'w',encoding='utf-8')
    f.write("")
    if endNum < startNum:
        print('結(jié)束條數(shù)必須大于開(kāi)始條數(shù)')
        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()

以上就是python 爬取國(guó)內(nèi)小說(shuō)網(wǎng)站的詳細(xì)內(nèi)容，更多關(guān)于python 爬取小說(shuō)網(wǎng)站的資料請(qǐng)關(guān)注腳本之家其它相關(guān)文章！

您可能感興趣的文章: