目錄

一、兩者不同

二、爬蟲源碼

三、爬蟲內(nèi)容詳解

---

一、兩者不同

• m3u8?是一種基于 HTTP Live Streaming 文件視頻格式，它主要是存放整個(gè)視頻的基本信息和分片(Segment)組成。
• 相信大家都看過m3u8格式文件的內(nèi)容，我們直來對(duì)比一下有什么不同，然后教大家怎么用python多進(jìn)程實(shí)現(xiàn)下載并且合并。

• 非加密?的m3u8文件

• ?加密?的m3u8文件

• ?相信眼尖的小伙伴已經(jīng)看出了2個(gè)內(nèi)容的不同之處，對(duì)的，其實(shí)區(qū)別就在加密文件的第 5 行的?#EXT-X-KEY 的信息
• 這個(gè)信息就是用來視頻內(nèi)容解密的，其實(shí)里面的內(nèi)容大多是一段字符串，其實(shí)也就是解密時(shí)候的KEY值
• 那么這個(gè)怎么去解密呢，我們暫時(shí)不管，我們先來解釋一下每行的意思
• 第一行:?#EXTM3U 聲明這是一個(gè)m3u8的文件
• 第二行:?#EXT-X-VERSION?協(xié)議的版本號(hào)
• 第三行:?#EXT-X-MEDIA-SEQUENCE?每一個(gè)media URI 在 PlayList中只有唯一的序號(hào)，相鄰之間序號(hào)+1
• 第四行:?#EXT-X-KEY? 記錄了加密的方式，一般是AES-128以及加密的KEY信息
• 第五行:?#EXTINF 表示這段視頻碎片的持續(xù)時(shí)間有多久
• 第六行:?sA3LRa6g.ts 視頻片段的名稱，獲取的時(shí)候需要拼接上域名，找到文件的正確的路徑

二、爬蟲源碼

#!/usr/bin/env python # encoding: utf-8 ''' #------------------------------------------------------------------- # CONFIDENTIAL --- CUSTOM STUDIOS #------------------------------------------------------------------- # # @Project Name : 多進(jìn)程M3U8視頻下載助手 # # @File Name : main.py # # @Programmer : Felix # # @Start Date : 2020/7/30 14:42 # # @Last Update : 2020/7/30 14:42 # #------------------------------------------------------------------- ''' import requests, os, platform, time from Crypto.Cipher import AES import multiprocessing from retrying import retryclass M3u8:'''This is a main Class, the file contains all documents.One document contains paragraphs that have several sentencesIt loads the original file and converts the original file to new contentThen the new content will be saved by this class'''def __init__(self):'''Initial the custom file by self'''self.encrypt = Falseself.headers = {"User-Agent": "Mozilla/5.0 (Windows NT 6.1; Win64; x64; rv:78.0) Gecko/20100101 Firefox/78.0"}def hello(self):'''This is a welcome speech:return: self'''print("*" * 50)print(' ' * 15 + 'm3u8鏈接下載小助手')print(' ' * 5 + '作者: Felix Date: 2020-05-20 13:14')print(' ' * 10 + '適用于非加密 | 加密鏈接')print("*" * 50)return selfdef checkUrl(self, url):'''Determine if it is a available link of m3u8:return: bool'''if '.m3u8' not in url:return Falseelif not url.startswith('http'):return Falseelse:return Truedef parse(self, url):'''Analyze a link of m3u8:param url: string, the link need to analyze:return: list'''container = list()response = self.request(url).text.split('\n')for ts in response:if '.ts' in ts:container.append(ts)if '#EXT-X-KEY:' in ts:self.encrypt = Truereturn containerdef getEncryptKey(self, url):'''Access to the secret key:param url: string, Access to the secret key by the url:return: string'''encryptKey = self.request("{}/key.key".format(url)).contentreturn encryptKeydef aesDecode(self, data, key):'''Decode the data:param data: stream, the data need to decode:param key: secret key:return: decode the data'''crypt = AES.new(key, AES.MODE_CBC, key)plain_text = crypt.decrypt(data)return plain_text.rstrip(b'\0')def download(self, queue, sort, file, downPath, url):'''Download the debris of video:param queue: the queue:param sort: which number debris:param file: the link of debris:param downPath: the path to save debris:param url: the link of m3u8:return: None'''queue.put(file)baseUrl = '/'.join(url.split("/")[:-1])if self.encrypt:self.encryptKey = self.getEncryptKey(baseUrl)if not file.startswith("http"):file = baseUrl + '/' +filedebrisName = "{}/{}.ts".format(downPath, sort)if not os.path.exists(debrisName):response = self.request(file)with open(debrisName, "wb") as f:if self.encrypt:data = self.aesDecode(response.content, self.encryptKey)f.write(data)f.flush()else:f.write(response.content)f.flush()def progressBar(self, queue, count):'''Show progress bar:param queue: the queue:param count: the number count of debris:return: None'''print('---一共{}個(gè)碎片...'.format(count))offset = 0while True:offset += 1file = queue.get()rate = offset * 100 / countprint("\r%s下載成功，當(dāng)前進(jìn)度%0.2f%%, 第%s/%s個(gè)" % (file, rate, offset, count))if offset >= count:break@retry(stop_max_attempt_number=3)def request(self, url, params):'''Send a request:param url: the url of request:param params: the params of request:return: the result of request'''response = requests.get(url, params=params, headers=self.headers, timeout=10)assert response.status_code == 200return responsedef run(self):'''program entry, Input basic information'''downPath = str(input("碎片的保存路徑, 默認(rèn)./Download：")) or "./Download"savePath = str(input("視頻的保存路徑, 默認(rèn)./Complete：")) or "./Complete"clearDebris = bool(input("是否清除碎片, 默認(rèn)True：")) or TruesaveSuffix = str(input("視頻格式, 默認(rèn)ts：")) or "ts"while True:url = str(input("請(qǐng)輸入合法的m3u8鏈接："))if self.checkUrl(url):break# create a not available folderif not os.path.exists(downPath):os.mkdir(downPath)if not os.path.exists(savePath):os.mkdir(savePath)# start analyze a link of m3u8print('---正在分析鏈接...')container = self.parse(url)print('---鏈接分析成功...')# run processing to do somethingprint('---進(jìn)程開始運(yùn)行...')po = multiprocessing.Pool(30)queue = multiprocessing.Manager().Queue()size = 0for file in container:sort = str(size).zfill(5)po.apply_async(self.download, args=(queue, sort, file, downPath, url,))size += 1po.close()self.progressBar(queue, len(container))print('---進(jìn)程運(yùn)行結(jié)束...')# handler debrissys = platform.system()saveName = time.strftime("%Y%m%d_%H%M%S", time.localtime())print('---文件合并清除...')if sys == "Windows":os.system("copy /b {}/*.ts {}/{}.{}".format(downPath, savePath, saveName, saveSuffix))if clearDebris:os.system("rmdir /s/q {}".format(downPath))else:os.system("cat {}/*.ts>{}/{}.{}".format(downPath, savePath, saveName, saveSuffix))if clearDebris:os.system("rm -rf {}".format(downPath))print('---合并清除完成...')print('---任務(wù)下載完成...')print('---歡迎再次使用...')if __name__ == "__main__":M3u8().hello().run()

三、爬蟲內(nèi)容詳解

• 初始化m3u8下載類

if __name__ == "__main__":M3u8().hello().run()

• hello方法

def hello(self):'''This is a welcome speech:return: self'''print("*" * 50)print(' ' * 15 + 'm3u8鏈接下載小助手')print(' ' * 5 + '作者: Felix Date: 2020-05-20 13:14')print(' ' * 10 + '適用于非加密 | 加密鏈接')print("*" * 50)return self

• run方法

• hello方法其實(shí)就是歡迎語，介紹了一些基本信息
• 如果鏈?zhǔn)秸{(diào)用的話，必須返回 self，初學(xué)者需要注意

def run(self):'''program entry, Input basic information'''downPath = str(input("碎片的保存路徑, 默認(rèn)./Download：")) or "./Download"savePath = str(input("視頻的保存路徑, 默認(rèn)./Complete：")) or "./Complete"clearDebris = bool(input("是否清除碎片, 默認(rèn)True：")) or TruesaveSuffix = str(input("視頻格式, 默認(rèn)ts：")) or "ts"while True:url = str(input("請(qǐng)輸入合法的m3u8鏈接："))if self.checkUrl(url):break# create a not available folderif not os.path.exists(downPath):os.mkdir(downPath)if not os.path.exists(savePath):os.mkdir(savePath)

• ?就是提示一些保存碎片的路徑，合并完成后是否需要進(jìn)行碎片清除
• 保存的視頻格式，默認(rèn)是ts，因?yàn)閠s一般的視頻軟件都可以打開，如果不放心可以輸入mp4
• 合法的連接這里調(diào)用了一個(gè)方法，checkUrl 其實(shí)就是檢測(cè)下是否是合格的m3u8鏈接
• 然后創(chuàng)建了一些不存在的文件夾

def checkUrl(self, url):'''Determine if it is a available link of m3u8:return: bool'''if '.m3u8' not in url:return Falseelif not url.startswith('http'):return Falseelse:return True

• ?這里我簡(jiǎn)單的判斷了下鏈接是否是m3u8
• 首先鏈接要是m3u8結(jié)尾的
• 其次鏈接需要是http打頭

• 分析輸入的鏈接

# start analyze a link of m3u8 print('---正在分析鏈接...') container = self.parse(url) print('---鏈接分析成功...') def parse(self, url):'''Analyze a link of m3u8:param url: string, the link need to analyze:return: list'''container = list()response = self.request(url).text.split('\n')for ts in response:if '.ts' in ts:container.append(ts)if '#EXT-X-KEY:' in ts:self.encrypt = Truereturn container

• 請(qǐng)求鏈接，判斷是否是加密m3u8還是非加密
• 將所有碎片文件進(jìn)行返回

• 打開多進(jìn)程，開啟進(jìn)程池，加速下載速度

# run processing to do something print('---進(jìn)程開始運(yùn)行...') po = multiprocessing.Pool(30) queue = multiprocessing.Manager().Queue() size = 0 for file in container:sort = str(size).zfill(5)po.apply_async(self.download, args=(queue, sort, file, downPath, url,))size += 1po.close()

• zfill方法，其實(shí)就是在數(shù)字前填充0，因?yàn)槲蚁Ｍ螺d的文件是00001.ts，00002.ts這樣有序的，最后合并的時(shí)候才不會(huì)混亂?
• queue 是多進(jìn)程共享變量的一種方式，用來顯示下載的進(jìn)度條

• download方法

def download(self, queue, sort, file, downPath, url):'''Download the debris of video:param queue: the queue:param sort: which number debris:param file: the link of debris:param downPath: the path to save debris:param url: the link of m3u8:return: None'''queue.put(file)baseUrl = '/'.join(url.split("/")[:-1])if self.encrypt:self.encryptKey = self.getEncryptKey(baseUrl)if not file.startswith("http"):file = baseUrl + '/' +filedebrisName = "{}/{}.ts".format(downPath, sort)if not os.path.exists(debrisName):response = self.request(file)with open(debrisName, "wb") as f:if self.encrypt:data = self.aesDecode(response.content, self.encryptKey)f.write(data)f.flush()else:f.write(response.content)f.flush()

• 一開始就加入隊(duì)列，是為了防止文件之前已經(jīng)存在的情況下，導(dǎo)致長度不對(duì)

• 如果是加密m3u8就通過 getEncryptKey 去獲取KEY值

• 寫入文件的時(shí)候如果是加密的，就將文件進(jìn)行 aesDecode 方法解密，具體請(qǐng)看源碼

• 進(jìn)度條顯示

def progressBar(self, queue, count):'''Show progress bar:param queue: the queue:param count: the number count of debris:return: None'''print('---一共{}個(gè)碎片...'.format(count))offset = 0while True:offset += 1file = queue.get()rate = offset * 100 / countprint("\r%s下載成功，當(dāng)前進(jìn)度%0.2f%%, 第%s/%s個(gè)" % (file, rate, offset, count))if offset >= count:break

• 其實(shí)就是通過當(dāng)前的下載到第幾個(gè)碎片，和所有碎片的數(shù)量進(jìn)行比較
• 一旦大于等于總數(shù)的時(shí)候，就退出循環(huán)

• 文件合并，碎片清除

• 這里兼容了 window 和 linux 下的合并清除命令
• 是否清除，剛開始的選擇中可設(shè)置

# handler debris sys = platform.system() saveName = time.strftime("%Y%m%d_%H%M%S", time.localtime())print('---文件合并清除...') if sys == "Windows":os.system("copy /b {}/*.ts {}/{}.{}".format(downPath, savePath, saveName, saveSuffix))if clearDebris:os.system("rmdir /s/q {}".format(downPath)) else:os.system("cat {}/*.ts>{}/{}.{}".format(downPath, savePath, saveName, saveSuffix))if clearDebris:os.system("rm -rf {}".format(downPath)) print('---合并清除完成...') print('---任務(wù)下載完成...') print('---歡迎再次使用...')

總結(jié)

以上是生活随笔為你收集整理的使用多进程教你下载 m3u8 加密或非加密视频的全部?jī)?nèi)容，希望文章能夠幫你解決所遇到的問題。

如果覺得生活随笔網(wǎng)站內(nèi)容還不錯(cuò)，歡迎將生活随笔推薦給好友。