yt_dlp/extractor/mgtv.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import base64
   5 import time
   6 import uuid
   7
   8 from .common import InfoExtractor
   9 from ..compat import (
  10     compat_HTTPError,
  11     compat_str,
  12 )
  13 from ..utils import (
  14     ExtractorError,
  15     int_or_none,
  16 )
  17
  18
  19 class MGTVIE(InfoExtractor):
  20     _VALID_URL = r'https?://(?:w(?:ww)?\.)?mgtv\.com/(v|b)/(?:[^/]+/)*(?P<id>\d+)\.html'
  21     IE_DESC = '芒果TV'
  22
  23     _TESTS = [{
  24         'url': 'http://www.mgtv.com/v/1/290525/f/3116640.html',
  25         'info_dict': {
  26             'id': '3116640',
  27             'ext': 'mp4',
  28             'title': '我是歌手 第四季',
  29             'description': '我是歌手第四季双年巅峰会',
  30             'duration': 7461,
  31             'thumbnail': r're:^https?://.*\.jpg$',
  32         },
  33     }, {
  34         'url': 'http://www.mgtv.com/b/301817/3826653.html',
  35         'only_matching': True,
  36     }, {
  37         'url': 'https://w.mgtv.com/b/301817/3826653.html',
  38         'only_matching': True,
  39     }]
  40
  41     def _real_extract(self, url):
  42         video_id = self._match_id(url)
  43         tk2 = base64.urlsafe_b64encode(b'did=%s|pno=1030|ver=0.3.0301|clit=%d' % (compat_str(uuid.uuid4()).encode(), time.time()))[::-1]
  44         try:
  45             api_data = self._download_json(
  46                 'https://pcweb.api.mgtv.com/player/video', video_id, query={
  47                     'tk2': tk2,
  48                     'video_id': video_id,
  49                 }, headers=self.geo_verification_headers())['data']
  50         except ExtractorError as e:
  51             if isinstance(e.cause, compat_HTTPError) and e.cause.code == 401:
  52                 error = self._parse_json(e.cause.read().decode(), None)
  53                 if error.get('code') == 40005:
  54                     self.raise_geo_restricted(countries=self._GEO_COUNTRIES)
  55                 raise ExtractorError(error['msg'], expected=True)
  56             raise
  57         info = api_data['info']
  58         title = info['title'].strip()
  59         stream_data = self._download_json(
  60             'https://pcweb.api.mgtv.com/player/getSource', video_id, query={
  61                 'pm2': api_data['atc']['pm2'],
  62                 'tk2': tk2,
  63                 'video_id': video_id,
  64             }, headers=self.geo_verification_headers())['data']
  65         stream_domain = stream_data['stream_domain'][0]
  66
  67         formats = []
  68         for idx, stream in enumerate(stream_data['stream']):
  69             stream_path = stream.get('url')
  70             if not stream_path:
  71                 continue
  72             format_data = self._download_json(
  73                 stream_domain + stream_path, video_id,
  74                 note='Download video info for format #%d' % idx)
  75             format_url = format_data.get('info')
  76             if not format_url:
  77                 continue
  78             tbr = int_or_none(stream.get('filebitrate') or self._search_regex(
  79                 r'_(\d+)_mp4/', format_url, 'tbr', default=None))
  80             formats.append({
  81                 'format_id': compat_str(tbr or idx),
  82                 'url': format_url,
  83                 'ext': 'mp4',
  84                 'tbr': tbr,
  85                 'protocol': 'm3u8_native',
  86                 'http_headers': {
  87                     'Referer': url,
  88                 },
  89                 'format_note': stream.get('name'),
  90             })
  91         self._sort_formats(formats)
  92
  93         return {
  94             'id': video_id,
  95             'title': title,
  96             'formats': formats,
  97             'description': info.get('desc'),
  98             'duration': int_or_none(info.get('duration')),
  99             'thumbnail': info.get('thumb'),
 100         }