BiliBili 多线程下载

初学 Python 练手之作,让我们快乐地下载 B 站视频

功能介绍

通过接口获取视频地址用多线程进行下载
下载的文件大小决定线程数，小文件一个线程就够了，大文件最多会开启10个线程，也就是10倍速下载！
代码分两个文件，一个是主文件，一个是写多线程的

主文件 bilidown.py

# -*- coding:utf-8 -*-

import re
import json
from multithreading import thread
import requests
from bs4 import BeautifulSoup
 
headers = {
    'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/83.0.4103.61 Safari/537.36',
}
 
url = input("请输入视频地址：")
with open('SESSDATA.txt', 'r') as f:
    SESSDATA = f.read()
if SESSDATA == '0':
    print('检测到您未设置SESSDATA,最高只能下载480p画质哦！')
 
# 获取视频名称
r = requests.get(url)
r.encoding = 'utf-8'
soup = BeautifulSoup(r.text, 'html.parser')
name = soup.title.text.split('_')[0]
print(name)
 
# 获取BV号
if '?' in url:
    url = url.split('?')[0]
    print(url)
bvid = re.search(r'BV.*', url).group()
print('BV号：' + bvid + "1111")
# 获取cid
cid_json = requests.get('https://api.bilibili.com/x/player/pagelist?bvid={}&jsonp=jsonp'.format(bvid)).text
print(cid_json)
cid = re.search(r'{"cid":(\d+)', cid_json).group()[7:]
print('CID:' + cid)
# 获取视频的av号
rsp = requests.get(url, headers=headers)
aid = re.search(r'"aid":(.*?),"', rsp.text).group()[6:-2]
print('AV号：' + aid)
 
 
# 抓取视频真实地址,清晰度
qn = 16  # 先设置一个默认低清晰度
headers['Referer'] = url
api_url = 'https://api.bilibili.com/x/player/playurl?cid={}&avid={}&qn={}&otype=json&requestFrom=bilibili-helper'.format(
    cid, aid, qn)
qn_dict = {}  # 用来存放清晰度选择参数
rsp = requests.get(api_url, headers=headers).content
rsp = json.loads(rsp)
qn_accept_description = rsp.get('data').get('accept_description')
qn_accept_quality = rsp.get('data').get('accept_quality')
print('下载视频清晰度选择')
for i, j, xuhao in zip(qn_accept_description, qn_accept_quality, range(len(qn_accept_quality))):
    print(str(xuhao + 1) + '：' + i)
    qn_dict[str(xuhao + 1)] = j
xuhao = input('请选择(输入清晰度前的标号)：')
qn = qn_dict[xuhao]
print('清晰度参数qn：' + str(qn))
 
 
api_url = 'https://api.bilibili.com/x/player/playurl?cid={}&avid={}&qn={}&otype=json&requestFrom=bilibili-helper'.format(
    cid, aid, qn)
# print('api_url='+api_url)
cookies = {}
cookies['SESSDATA'] = SESSDATA  # 这里输入你的SESSDATA
rsp = requests.get(api_url, headers=headers, cookies=cookies).content  # 这里代cookies才能得到会员或者登录后才能下载的视频的链接
rsp = json.loads(rsp)
real_url = rsp.get('data').get('durl')[0].get('url')
print('成功获取视频直链！')
print('正在开启多线程极速下载……')
thread(real_url, url, name + '.flv')  # 多线程下载
 
# 把上面那行删掉，把下面注释去掉就是单线程下载
# content = requests.get(real_url, headers=headers).content
# with open('1.flv', 'wb') as f:
#    f.write(content)

多线程 multithreading.py

# -*- coding:utf-8 -*-

import requests
import threading
import datetime
import time
#改headers参数和url就好了
def thread(url, Referer, file_name):
  # print(r.status_code, r.headers)
  headers = {
      'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/83.0.4103.61 Safari/537.36',
      'Referer': Referer
  }
  r = requests.get(url, headers=headers, stream=True, timeout=30)
  all_thread = 1
  # 获取视频大小
  file_size = int(r.headers['content-length'])
  # 如果获取到文件大小，创建一个和需要下载文件一样大小的文件
  if file_size:
    fp = open(file_name, 'wb')
    fp.truncate(file_size)
    print('视频大小：' + str(int(file_size / 1024 / 1024)) + "MB")
    fp.close()
  # 每个线程每次下载大小为5M
  size = 5242880
  # 当前文件大小需大于5M
  if file_size > size:
    # 获取总线程数
    all_thread = int(file_size / size)
    # 设最大线程数为10，如总线程数大于10
    # 线程数为10
    if all_thread > 10:
      all_thread = 10
  part = file_size // all_thread
  threads = []
  starttime = datetime.datetime.now().replace(microsecond=0)
  for i in range(all_thread):
    # 获取每个线程开始时的文件位置
    start = part * i
    # 获取每个文件结束位置
    if i == all_thread - 1:
      end = file_size
    else:
      end = start + part
    if i > 0:
      start += 1
    headers = headers.copy()
    headers['Range'] = "bytes=%s-%s" % (start, end)
    t = threading.Thread(target=Handler, name='线程-' + str(i),
               kwargs={'start': start, 'end': end, 'url': url, 'filename': file_name, 'headers': headers})
    t.setDaemon(True)
    threads.append(t)
  # 线程开始
  for t in threads:
    time.sleep(0.2)
    t.start()
  # 等待所有线程结束
  print('正在下载……')
  for t in threads:
    t.join()
  endtime = datetime.datetime.now().replace(microsecond=0)
  print('下载完成！用时：%s' % (endtime - starttime))
def Handler(start, end, url, filename, headers={}):
  tt_name = threading.current_thread().getName()
  print(tt_name + ' 已启动')
  r = requests.get(url, headers=headers, stream=True)
  total_size = end - start
  downsize = 0
  startTime = time.time()
  with open(filename, 'r+b') as fp:
    fp.seek(start)
    var = fp.tell()
    for chunk in r.iter_content(204800):
      if chunk:
        fp.write(chunk)
        downsize += len(chunk)
        line = tt_name + '-downloading %d KB/s - %.2f MB， 共 %.2f MB'
        line = line % (
          downsize / 1024 / (time.time() - startTime), downsize / 1024 / 1024,
          total_size / 1024 / 1024)
        print(line, end='\r')
if __name__ == '__main__':
  url = input('输入视频链接（请输入视频原链）：')
  thread(url)