yt_dlp/extractor/threeqsdn.py

   1 from __future__ import unicode_literals
   2
   3 import re
   4
   5 from .common import InfoExtractor
   6 from ..compat import compat_HTTPError
   7 from ..utils import (
   8     determine_ext,
   9     ExtractorError,
  10     float_or_none,
  11     int_or_none,
  12     parse_iso8601,
  13 )
  14
  15
  16 class ThreeQSDNIE(InfoExtractor):
  17     IE_NAME = '3qsdn'
  18     IE_DESC = '3Q SDN'
  19     _VALID_URL = r'https?://playout\.3qsdn\.com/(?P<id>[\da-f]{8}-[\da-f]{4}-[\da-f]{4}-[\da-f]{4}-[\da-f]{12})'
  20     _TESTS = [{
  21         # https://player.3qsdn.com/demo.html
  22         'url': 'https://playout.3qsdn.com/7201c779-6b3c-11e7-a40e-002590c750be',
  23         'md5': '64a57396b16fa011b15e0ea60edce918',
  24         'info_dict': {
  25             'id': '7201c779-6b3c-11e7-a40e-002590c750be',
  26             'ext': 'mp4',
  27             'title': 'Video Ads',
  28             'is_live': False,
  29             'description': 'Video Ads Demo',
  30             'timestamp': 1500334803,
  31             'upload_date': '20170717',
  32             'duration': 888.032,
  33             'subtitles': {
  34                 'eng': 'count:1',
  35             },
  36         },
  37         'expected_warnings': ['Unknown MIME type application/mp4 in DASH manifest'],
  38     }, {
  39         # live video stream
  40         'url': 'https://playout.3qsdn.com/66e68995-11ca-11e8-9273-002590c750be',
  41         'info_dict': {
  42             'id': '66e68995-11ca-11e8-9273-002590c750be',
  43             'ext': 'mp4',
  44             'title': 're:^66e68995-11ca-11e8-9273-002590c750be [0-9]{4}-[0-9]{2}-[0-9]{2} [0-9]{2}:[0-9]{2}$',
  45             'is_live': True,
  46         },
  47         'params': {
  48             'skip_download': True,  # m3u8 downloads
  49         },
  50     }, {
  51         # live audio stream
  52         'url': 'http://playout.3qsdn.com/9edf36e0-6bf2-11e2-a16a-9acf09e2db48',
  53         'only_matching': True,
  54     }, {
  55         # live audio stream with some 404 URLs
  56         'url': 'http://playout.3qsdn.com/ac5c3186-777a-11e2-9c30-9acf09e2db48',
  57         'only_matching': True,
  58     }, {
  59         # geo restricted with 'This content is not available in your country'
  60         'url': 'http://playout.3qsdn.com/d63a3ffe-75e8-11e2-9c30-9acf09e2db48',
  61         'only_matching': True,
  62     }, {
  63         # geo restricted with 'playout.3qsdn.com/forbidden'
  64         'url': 'http://playout.3qsdn.com/8e330f26-6ae2-11e2-a16a-9acf09e2db48',
  65         'only_matching': True,
  66     }, {
  67         # live video with rtmp link
  68         'url': 'https://playout.3qsdn.com/6092bb9e-8f72-11e4-a173-002590c750be',
  69         'only_matching': True,
  70     }, {
  71         # ondemand from http://www.philharmonie.tv/veranstaltung/26/
  72         'url': 'http://playout.3qsdn.com/0280d6b9-1215-11e6-b427-0cc47a188158?protocol=http',
  73         'only_matching': True,
  74     }, {
  75         # live video stream
  76         'url': 'https://playout.3qsdn.com/d755d94b-4ab9-11e3-9162-0025907ad44f?js=true',
  77         'only_matching': True,
  78     }]
  79
  80     @staticmethod
  81     def _extract_url(webpage):
  82         mobj = re.search(
  83             r'<iframe[^>]+\b(?:data-)?src=(["\'])(?P<url>%s.*?)\1' % ThreeQSDNIE._VALID_URL, webpage)
  84         if mobj:
  85             return mobj.group('url')
  86
  87     def _real_extract(self, url):
  88         video_id = self._match_id(url)
  89
  90         try:
  91             config = self._download_json(
  92                 url.replace('://playout.3qsdn.com/', '://playout.3qsdn.com/config/'), video_id)
  93         except ExtractorError as e:
  94             if isinstance(e.cause, compat_HTTPError) and e.cause.code == 401:
  95                 self.raise_geo_restricted()
  96             raise
  97
  98         live = config.get('streamContent') == 'live'
  99         aspect = float_or_none(config.get('aspect'))
 100
 101         formats = []
 102         subtitles = {}
 103         for source_type, source in (config.get('sources') or {}).items():
 104             if not source:
 105                 continue
 106             if source_type == 'dash':
 107                 fmts, subs = self._extract_mpd_formats_and_subtitles(
 108                     source, video_id, mpd_id='mpd', fatal=False)
 109                 formats.extend(fmts)
 110                 subtitles = self._merge_subtitles(subtitles, subs)
 111             elif source_type == 'hls':
 112                 fmts, subs = self._extract_m3u8_formats_and_subtitles(
 113                     source, video_id, 'mp4', 'm3u8' if live else 'm3u8_native',
 114                     m3u8_id='hls', fatal=False)
 115                 formats.extend(fmts)
 116                 subtitles = self._merge_subtitles(subtitles, subs)
 117             elif source_type == 'progressive':
 118                 for s in source:
 119                     src = s.get('src')
 120                     if not (src and self._is_valid_url(src, video_id)):
 121                         continue
 122                     width = None
 123                     format_id = ['http']
 124                     ext = determine_ext(src)
 125                     if ext:
 126                         format_id.append(ext)
 127                     height = int_or_none(s.get('height'))
 128                     if height:
 129                         format_id.append('%dp' % height)
 130                         if aspect:
 131                             width = int(height * aspect)
 132                     formats.append({
 133                         'ext': ext,
 134                         'format_id': '-'.join(format_id),
 135                         'height': height,
 136                         'source_preference': 0,
 137                         'url': src,
 138                         'vcodec': 'none' if height == 0 else None,
 139                         'width': width,
 140                     })
 141         # It seems like this would be correctly handled by default
 142         # However, unless someone can confirm this, the old
 143         # behaviour is being kept as-is
 144         self._sort_formats(formats, ('res', 'source_preference'))
 145
 146         for subtitle in (config.get('subtitles') or []):
 147             src = subtitle.get('src')
 148             if not src:
 149                 continue
 150             subtitles.setdefault(subtitle.get('label') or 'eng', []).append({
 151                 'url': src,
 152             })
 153
 154         title = config.get('title') or video_id
 155
 156         return {
 157             'id': video_id,
 158             'title': self._live_title(title) if live else title,
 159             'thumbnail': config.get('poster') or None,
 160             'description': config.get('description') or None,
 161             'timestamp': parse_iso8601(config.get('upload_date')),
 162             'duration': float_or_none(config.get('vlength')) or None,
 163             'is_live': live,
 164             'formats': formats,
 165             'subtitles': subtitles,
 166         }