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