Raphaël G. Git Repositories - youtubedl/blob - youtube_dl/extractor/screenjunkies.py

   1 from __future__ import unicode_literals
   2
   3 import re
   4
   5 from .common import InfoExtractor
   6 from ..compat import compat_str
   7 from ..utils import (
   8     int_or_none,
   9     parse_age_limit,
  10 )
  11
  12
  13 class ScreenJunkiesIE(InfoExtractor):
  14     _VALID_URL = r'http://www.screenjunkies.com/video/(?P<display_id>[^/]+?)(?:-(?P<id>\d+))?(?:[/?#&]|$)'
  15     _TESTS = [{
  16         'url': 'http://www.screenjunkies.com/video/best-quentin-tarantino-movie-2841915',
  17         'md5': '5c2b686bec3d43de42bde9ec047536b0',
  18         'info_dict': {
  19             'id': '2841915',
  20             'display_id': 'best-quentin-tarantino-movie',
  21             'ext': 'mp4',
  22             'title': 'Best Quentin Tarantino Movie',
  23             'thumbnail': 're:^https?://.*\.jpg',
  24             'duration': 3671,
  25             'age_limit': 13,
  26             'tags': list,
  27         },
  28     }, {
  29         'url': 'http://www.screenjunkies.com/video/honest-trailers-the-dark-knight',
  30         'info_dict': {
  31             'id': '2348808',
  32             'display_id': 'honest-trailers-the-dark-knight',
  33             'ext': 'mp4',
  34             'title': "Honest Trailers: 'The Dark Knight'",
  35             'thumbnail': 're:^https?://.*\.jpg',
  36             'age_limit': 10,
  37             'tags': list,
  38         },
  39     }, {
  40         # requires subscription but worked around
  41         'url': 'http://www.screenjunkies.com/video/knocking-dead-ep-1-the-show-so-far-3003285',
  42         'info_dict': {
  43             'id': '3003285',
  44             'display_id': 'knocking-dead-ep-1-the-show-so-far',
  45             'ext': 'mp4',
  46             'title': 'Knocking Dead Ep 1: State of The Dead Recap',
  47             'thumbnail': 're:^https?://.*\.jpg',
  48             'duration': 3307,
  49             'age_limit': 13,
  50             'tags': list,
  51         },
  52     }]
  53
  54     _DEFAULT_BITRATES = (48, 150, 496, 864, 2240)
  55
  56     def _real_extract(self, url):
  57         mobj = re.match(self._VALID_URL, url)
  58         video_id = mobj.group('id')
  59         display_id = mobj.group('display_id')
  60
  61         if not video_id:
  62             webpage = self._download_webpage(url, display_id)
  63             video_id = self._search_regex(
  64                 (r'src=["\']/embed/(\d+)', r'data-video-content-id=["\'](\d+)'),
  65                 webpage, 'video id')
  66
  67         webpage = self._download_webpage(
  68             'http://www.screenjunkies.com/embed/%s' % video_id,
  69             display_id, 'Downloading video embed page')
  70         embed_vars = self._parse_json(
  71             self._search_regex(
  72                 r'(?s)embedVars\s*=\s*({.+?})\s*</script>', webpage, 'embed vars'),
  73             display_id)
  74
  75         title = embed_vars['contentName']
  76
  77         formats = []
  78         bitrates = []
  79         for f in embed_vars.get('media', []):
  80             if not f.get('uri') or f.get('mediaPurpose') != 'play':
  81                 continue
  82             bitrate = int_or_none(f.get('bitRate'))
  83             if bitrate:
  84                 bitrates.append(bitrate)
  85             formats.append({
  86                 'url': f['uri'],
  87                 'format_id': 'http-%d' % bitrate if bitrate else 'http',
  88                 'width': int_or_none(f.get('width')),
  89                 'height': int_or_none(f.get('height')),
  90                 'tbr': bitrate,
  91                 'format': 'mp4',
  92             })
  93
  94         if not bitrates:
  95             # When subscriptionLevel > 0, i.e. plus subscription is required
  96             # media list will be empty. However, hds and hls uris are still
  97             # available. We can grab them assuming bitrates to be default.
  98             bitrates = self._DEFAULT_BITRATES
  99
 100         auth_token = embed_vars.get('AuthToken')
 101
 102         def construct_manifest_url(base_url, ext):
 103             pieces = [base_url]
 104             pieces.extend([compat_str(b) for b in bitrates])
 105             pieces.append('_kbps.mp4.%s?%s' % (ext, auth_token))
 106             return ','.join(pieces)
 107
 108         if bitrates and auth_token:
 109             hds_url = embed_vars.get('hdsUri')
 110             if hds_url:
 111                 f4m_formats = self._extract_f4m_formats(
 112                     construct_manifest_url(hds_url, 'f4m'),
 113                     display_id, f4m_id='hds', fatal=False)
 114                 if len(f4m_formats) == len(bitrates):
 115                     for f, bitrate in zip(f4m_formats, bitrates):
 116                         if not f.get('tbr'):
 117                             f['format_id'] = 'hds-%d' % bitrate
 118                             f['tbr'] = bitrate
 119                 # TODO: fix f4m downloader to handle manifests without bitrates if possible
 120                 # formats.extend(f4m_formats)
 121
 122             hls_url = embed_vars.get('hlsUri')
 123             if hls_url:
 124                 formats.extend(self._extract_m3u8_formats(
 125                     construct_manifest_url(hls_url, 'm3u8'),
 126                     display_id, 'mp4', entry_protocol='m3u8_native', m3u8_id='hls', fatal=False))
 127         self._sort_formats(formats)
 128
 129         return {
 130             'id': video_id,
 131             'display_id': display_id,
 132             'title': title,
 133             'thumbnail': embed_vars.get('thumbUri'),
 134             'duration': int_or_none(embed_vars.get('videoLengthInSeconds')) or None,
 135             'age_limit': parse_age_limit(embed_vars.get('audienceRating')),
 136             'tags': embed_vars.get('tags', '').split(','),
 137             'formats': formats,
 138         }