_ Git - youtube-dl/blob - youtube_dl/extractor/sixplay.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import re
   5
   6 from .common import InfoExtractor
   7 from ..compat import (
   8     compat_parse_qs,
   9     compat_str,
  10     compat_urllib_parse_urlparse,
  11 )
  12 from ..utils import (
  13     determine_ext,
  14     int_or_none,
  15     try_get,
  16     qualities,
  17 )
  18
  19
  20 class SixPlayIE(InfoExtractor):
  21     IE_NAME = '6play'
  22     _VALID_URL = r'(?:6play:|https?://(?:www\.)?(?P<domain>6play\.fr|rtlplay.be)/.+?-c_)(?P<id>[0-9]+)'
  23     _TESTS = [{
  24         'url': 'https://www.6play.fr/minute-par-minute-p_9533/le-but-qui-a-marque-lhistoire-du-football-francais-c_12041051',
  25         'md5': '31fcd112637baa0c2ab92c4fcd8baf27',
  26         'info_dict': {
  27             'id': '12041051',
  28             'ext': 'mp4',
  29             'title': 'Le but qui a marqué l\'histoire du football français !',
  30             'description': 'md5:b59e7e841d646ef1eb42a7868eb6a851',
  31         },
  32     }, {
  33         'url': 'https://www.rtlplay.be/rtl-info-13h-p_8551/les-titres-du-rtlinfo-13h-c_12045869',
  34         'only_matching': True,
  35     }]
  36
  37     def _real_extract(self, url):
  38         domain, video_id = re.search(self._VALID_URL, url).groups()
  39         service, consumer_name = {
  40             '6play.fr': ('6play', 'm6web'),
  41             'rtlplay.be': ('rtlbe_rtl_play', 'rtlbe'),
  42         }.get(domain, ('6play', 'm6web'))
  43
  44         data = self._download_json(
  45             'https://pc.middleware.6play.fr/6play/v2/platforms/m6group_web/services/%s/videos/clip_%s' % (service, video_id),
  46             video_id, headers={
  47                 'x-customer-name': consumer_name
  48             }, query={
  49                 'csa': 5,
  50                 'with': 'clips',
  51             })
  52
  53         clip_data = data['clips'][0]
  54         title = clip_data['title']
  55
  56         urls = []
  57         quality_key = qualities(['lq', 'sd', 'hq', 'hd'])
  58         formats = []
  59         subtitles = {}
  60         for asset in clip_data['assets']:
  61             asset_url = asset.get('full_physical_path')
  62             protocol = asset.get('protocol')
  63             if not asset_url or protocol == 'primetime' or asset_url in urls:
  64                 continue
  65             urls.append(asset_url)
  66             container = asset.get('video_container')
  67             ext = determine_ext(asset_url)
  68             if protocol == 'http_subtitle' or ext == 'vtt':
  69                 subtitles.setdefault('fr', []).append({'url': asset_url})
  70                 continue
  71             if container == 'm3u8' or ext == 'm3u8':
  72                 if protocol == 'usp':
  73                     if compat_parse_qs(compat_urllib_parse_urlparse(asset_url).query).get('token', [None])[0]:
  74                         urlh = self._request_webpage(
  75                             asset_url, video_id, fatal=False,
  76                             headers=self.geo_verification_headers())
  77                         if not urlh:
  78                             continue
  79                         asset_url = urlh.geturl()
  80                     asset_url = re.sub(r'/([^/]+)\.ism/[^/]*\.m3u8', r'/\1.ism/\1.m3u8', asset_url)
  81                     formats.extend(self._extract_m3u8_formats(
  82                         asset_url, video_id, 'mp4', 'm3u8_native',
  83                         m3u8_id='hls', fatal=False))
  84                     formats.extend(self._extract_f4m_formats(
  85                         asset_url.replace('.m3u8', '.f4m'),
  86                         video_id, f4m_id='hds', fatal=False))
  87                     formats.extend(self._extract_mpd_formats(
  88                         asset_url.replace('.m3u8', '.mpd'),
  89                         video_id, mpd_id='dash', fatal=False))
  90                     formats.extend(self._extract_ism_formats(
  91                         re.sub(r'/[^/]+\.m3u8', '/Manifest', asset_url),
  92                         video_id, ism_id='mss', fatal=False))
  93                 else:
  94                     formats.extend(self._extract_m3u8_formats(
  95                         asset_url, video_id, 'mp4', 'm3u8_native',
  96                         m3u8_id='hls', fatal=False))
  97             elif container == 'mp4' or ext == 'mp4':
  98                 quality = asset.get('video_quality')
  99                 formats.append({
 100                     'url': asset_url,
 101                     'format_id': quality,
 102                     'quality': quality_key(quality),
 103                     'ext': ext,
 104                 })
 105         self._sort_formats(formats)
 106
 107         def get(getter):
 108             for src in (data, clip_data):
 109                 v = try_get(src, getter, compat_str)
 110                 if v:
 111                     return v
 112
 113         return {
 114             'id': video_id,
 115             'title': title,
 116             'description': get(lambda x: x['description']),
 117             'duration': int_or_none(clip_data.get('duration')),
 118             'series': get(lambda x: x['program']['title']),
 119             'formats': formats,
 120             'subtitles': subtitles,
 121         }