m3u8_xz库

关于m3u8文件的下载网上有好多方法，但是没有一个拿来即用的，因此在这分享一个库，这个库也很简单

# 安装
pip install m3u8_xz

因为这个库是国内的网友写的，注释一看就懂， 没事可以研究一下源码。

示例

from m3u8_XZ import m3u8


# use m3u8 url 通过url
m3u8_url = 'https://svipsvip.ffzy-online5.com/20240721/30372_37f1038c/2000k/hls/mixed.m3u8'
obj = m3u8(m3u8_url, folder='test')
# use local file 通过本地文件
# m3u8(m3u8_file='fileName.m3u8', folder='test')
# 异步启动方式 old
# obj.run()
# 2024年5月23日 新增启动方式
obj.run(thread_num=20)

其他格式的文件下载

这个方法可以下载各种二进制文件，考虑到文件过大，内存爆掉的现象使用流的方式下载

# coding: utf-8
from pathlib import Path
from typing import Tuple
from copy import deepcopy

import requests
import urllib3

urllib3.disable_warnings()



def ExceptionHandler(*default):
    """ decorator for exception handling

    Parameters
    ----------
    *default:
        the default value returned when an exception occurs
    """

    def outer(func):

        def inner(*args, **kwargs):
            try:
                return func(*args, **kwargs)
            except BaseException as e:
                value = deepcopy(default)
                if len(value) == 0:
                    return None
                elif len(value) == 1:
                    return value[0]

                return value

        return inner

    return outer


class StreamDownload:
    def __init__(
            self,
            url: str,
            folder: str,
            name: str,
            headers: dict = None,
            chunk_size: int = 1024,
            print_callback=None
    ):
        """
        StreamDownload类从url下载文件。
        :param url:
        :param folder: 保存文件夹
        :param name: 文件名
        :param headers:
        :param chunk_size:
        :param print_callback: 打印回调函数
        """
        self.url = url
        self.folder = folder
        self.name = name
        self.chunk_size = chunk_size
        self.headers = headers or {}
        self._isStop = False
        self.print_callback = print_callback

    @ExceptionHandler(0)
    def content_length(self):
        StreamDownload.log(f'url: {self.url}, 正在获取内容长度，请稍候...')
        response = requests.head(self.url, verify=False, headers=self.headers)
        response.raise_for_status()
        return int(response.headers.get('content-length'))

    def get_start_size(self) -> Tuple[Path, Path, int]:
        name = self.folder / self.name
        name.parent.mkdir(parents=True, exist_ok=True)
        temp_name = name.with_suffix('.temp')
        if not name.exists() and not temp_name.exists():
            StreamDownload.log('文件不存在，创建临时文件')
            temp_name.touch()
        if name.exists():
            start_size = name.stat().st_size
            StreamDownload.log('文件已存在')
        elif temp_name.exists():
            start_size = temp_name.stat().st_size
        else:
            start_size = 0
        return name, temp_name, start_size

    def run(self):
        """
        开始下载
        :return:
        """
        name: Path
        temp_name: Path
        start_size: int

        response_length = self.content_length()
        if response_length == 0:
            return

        StreamDownload.log(f'内容长度: {response_length} kb')
        name, temp_name, start_size = self.get_start_size()
        progress = start_size / self.chunk_size
        if response_length == start_size:
            return

        headers = self.headers.copy()
        headers['Range'] = f'bytes={start_size}-'
        response = requests.get(self.url, headers=headers, stream=True, verify=False)
        response.raise_for_status()

        with open(temp_name, 'ab') as fp:
            for index, content in enumerate(response.iter_content(chunk_size=self.chunk_size)):
                if self._isStop:
                    break
                if content:
                    fp.write(content)
                loaded_num = (index + 1 + progress) * self.chunk_size
                load_count = response_length
                StreamDownload.log(
                    '\r进度: {:.2f}%'.format(loaded_num / load_count * 100),
                    end='', flush=True
                )
                if self.print_callback:
                    self.print_callback(loaded_num=loaded_num, load_count=load_count)
        temp_name.rename(name)

    @staticmethod
    def log(*args, **kwargs):
        print(*args, **kwargs)

    def stop(self):
        """
        停止下载
        :return:
        """
        self._isStop = True

if __name__ == '__main__':
	mp4_url = (
    'http://v16m-default.akamaized.net/d4979b2fc814c79cdce8747e082e6bc2/669fcbbc/video/tos/alisg/tos-alisg-v-0000/osPoAgeQBuamDMGMxek8GdDr0nPebyAcbpCvrg'
    '/?a=2011&bti=MzhALjBg&ch=0&cr=0&dr=0&net=5&cd=0%7C0%7C0%7C0&br=3432&bt=1716&cs=0&ds=4&ft=XE5bCqT0mmjPD12fdGK73wU7C1JcMeF~O5&mime_type=video_mp4&qs=0&rc=NTx'
    'lMzg2ODxlaWllOzs3O0BpM2h2aDo6Znk3ZzMzODYzNEAuLjReYl5fNV4xNV9gMzUuYSMwYzNkcjQwa3FgLS1kMC1zcw%3D%3D&vvpl=1&l=20240723090214D6FEEF98890A2308D775&btag=e000a8000')
	stream_download = StreamDownload(mp4_url, 'test1', 'test.mp4')
	stream_download.run()