├── README.md
└── m3u8.py


/README.md:
--------------------------------------------------------------------------------
1 | # m3u8-Downloader
2 | Download m3u8 to ts with multithreading
3 | 
4 | # How to use
5 | ```
6 | downloader = Downloader(50)
7 | downloader.run('http://m3u8.test.com/test.m3u8', '/home/video/')
8 | ```
9 | 


--------------------------------------------------------------------------------
/m3u8.py:
--------------------------------------------------------------------------------
 1 | #coding: utf-8
 2 | 
 3 | from gevent import monkey
 4 | monkey.patch_all()
 5 | from gevent.pool import Pool
 6 | import gevent
 7 | import requests
 8 | import urlparse
 9 | import os
10 | import time
11 | 
12 | class Downloader:
13 |     def __init__(self, pool_size, retry=3):
14 |         self.pool = Pool(pool_size)
15 |         self.session = self._get_http_session(pool_size, pool_size, retry)
16 |         self.retry = retry
17 |         self.dir = ''
18 |         self.succed = {}
19 |         self.failed = []
20 |         self.ts_total = 0
21 | 
22 |     def _get_http_session(self, pool_connections, pool_maxsize, max_retries):
23 |             session = requests.Session()
24 |             adapter = requests.adapters.HTTPAdapter(pool_connections=pool_connections, pool_maxsize=pool_maxsize, max_retries=max_retries)
25 |             session.mount('http://', adapter)
26 |             session.mount('https://', adapter)
27 |             return session
28 | 
29 |     def run(self, m3u8_url, dir=''):
30 |         self.dir = dir
31 |         if self.dir and not os.path.isdir(self.dir):
32 |             os.makedirs(self.dir)
33 | 
34 |         r = self.session.get(m3u8_url, timeout=10)
35 |         if r.ok:
36 |             body = r.content
37 |             if body:
38 |                 ts_list = [urlparse.urljoin(m3u8_url, n.strip()) for n in body.split('\n') if n and not n.startswith("#")]
39 |                 ts_list = zip(ts_list, [n for n in xrange(len(ts_list))])
40 |                 if ts_list:
41 |                     self.ts_total = len(ts_list)
42 |                     print self.ts_total
43 |                     g1 = gevent.spawn(self._join_file)
44 |                     self._download(ts_list)
45 |                     g1.join()
46 |         else:
47 |             print r.status_code
48 | 
49 |     def _download(self, ts_list):
50 |         self.pool.map(self._worker, ts_list)
51 |         if self.failed:
52 |             ts_list = self.failed
53 |             self.failed = []
54 |             self._download(ts_list)
55 | 
56 |     def _worker(self, ts_tuple):
57 |         url = ts_tuple[0]
58 |         index = ts_tuple[1]
59 |         retry = self.retry
60 |         while retry:
61 |             try:
62 |                 r = self.session.get(url, timeout=20)
63 |                 if r.ok:
64 |                     file_name = url.split('/')[-1].split('?')[0]
65 |                     print file_name
66 |                     with open(os.path.join(self.dir, file_name), 'wb') as f:
67 |                         f.write(r.content)
68 |                     self.succed[index] = file_name
69 |                     return
70 |             except:
71 |                 retry -= 1
72 |         print '[FAIL]%s' % url
73 |         self.failed.append((url, index))
74 | 
75 |     def _join_file(self):
76 |         index = 0
77 |         outfile = ''
78 |         while index < self.ts_total:
79 |             file_name = self.succed.get(index, '')
80 |             if file_name:
81 |                 infile = open(os.path.join(self.dir, file_name), 'rb')
82 |                 if not outfile:
83 |                     outfile = open(os.path.join(self.dir, file_name.split('.')[0]+'_all.'+file_name.split('.')[-1]), 'wb')
84 |                 outfile.write(infile.read())
85 |                 infile.close()
86 |                 os.remove(os.path.join(self.dir, file_name))
87 |                 index += 1
88 |             else:
89 |                 time.sleep(1)
90 |         if outfile:
91 |             outfile.close()
92 | 
93 | if __name__ == '__main__':
94 |     downloader = Downloader(50)
95 |     downloader.run('http://m3u8.test.com/test.m3u8', '/home/video/')
96 | 


--------------------------------------------------------------------------------