Raphaël G. Git Repositories - youtubedl/blob - youtube_dl/extractor/hbo.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 compat_str
   8 from ..utils import (
   9     xpath_text,
  10     xpath_element,
  11     int_or_none,
  12     parse_duration,
  13 )
  14
  15
  16 class HBOBaseIE(InfoExtractor):
  17     _FORMATS_INFO = {
  18         'pro7': {
  19             'width': 1280,
  20             'height': 720,
  21         },
  22         '1920': {
  23             'width': 1280,
  24             'height': 720,
  25         },
  26         'pro6': {
  27             'width': 768,
  28             'height': 432,
  29         },
  30         '640': {
  31             'width': 768,
  32             'height': 432,
  33         },
  34         'pro5': {
  35             'width': 640,
  36             'height': 360,
  37         },
  38         'highwifi': {
  39             'width': 640,
  40             'height': 360,
  41         },
  42         'high3g': {
  43             'width': 640,
  44             'height': 360,
  45         },
  46         'medwifi': {
  47             'width': 400,
  48             'height': 224,
  49         },
  50         'med3g': {
  51             'width': 400,
  52             'height': 224,
  53         },
  54     }
  55
  56     def _extract_from_id(self, video_id):
  57         video_data = self._download_xml(
  58             'http://render.lv3.hbo.com/data/content/global/videos/data/%s.xml' % video_id, video_id)
  59         title = xpath_text(video_data, 'title', 'title', True)
  60
  61         formats = []
  62         for source in xpath_element(video_data, 'videos', 'sources', True):
  63             if source.tag == 'size':
  64                 path = xpath_text(source, './/path')
  65                 if not path:
  66                     continue
  67                 width = source.attrib.get('width')
  68                 format_info = self._FORMATS_INFO.get(width, {})
  69                 height = format_info.get('height')
  70                 fmt = {
  71                     'url': path,
  72                     'format_id': 'http%s' % ('-%dp' % height if height else ''),
  73                     'width': format_info.get('width'),
  74                     'height': height,
  75                 }
  76                 rtmp = re.search(r'^(?P<url>rtmpe?://[^/]+/(?P<app>.+))/(?P<playpath>mp4:.+)$', path)
  77                 if rtmp:
  78                     fmt.update({
  79                         'url': rtmp.group('url'),
  80                         'play_path': rtmp.group('playpath'),
  81                         'app': rtmp.group('app'),
  82                         'ext': 'flv',
  83                         'format_id': fmt['format_id'].replace('http', 'rtmp'),
  84                     })
  85                 formats.append(fmt)
  86             else:
  87                 video_url = source.text
  88                 if not video_url:
  89                     continue
  90                 if source.tag == 'tarball':
  91                     formats.extend(self._extract_m3u8_formats(
  92                         video_url.replace('.tar', '/base_index_w8.m3u8'),
  93                         video_id, 'mp4', 'm3u8_native', m3u8_id='hls', fatal=False))
  94                 elif source.tag == 'hls':
  95                     m3u8_formats = self._extract_m3u8_formats(
  96                         video_url.replace('.tar', '/base_index.m3u8'),
  97                         video_id, 'mp4', 'm3u8_native', m3u8_id='hls', fatal=False)
  98                     for f in m3u8_formats:
  99                         if f.get('vcodec') == 'none' and not f.get('tbr'):
 100                             f['tbr'] = int_or_none(self._search_regex(
 101                                 r'-(\d+)k/', f['url'], 'tbr', default=None))
 102                     formats.extend(m3u8_formats)
 103                 elif source.tag == 'dash':
 104                     formats.extend(self._extract_mpd_formats(
 105                         video_url.replace('.tar', '/manifest.mpd'),
 106                         video_id, mpd_id='dash', fatal=False))
 107                 else:
 108                     format_info = self._FORMATS_INFO.get(source.tag, {})
 109                     formats.append({
 110                         'format_id': 'http-%s' % source.tag,
 111                         'url': video_url,
 112                         'width': format_info.get('width'),
 113                         'height': format_info.get('height'),
 114                     })
 115         self._sort_formats(formats)
 116
 117         thumbnails = []
 118         card_sizes = xpath_element(video_data, 'titleCardSizes')
 119         if card_sizes is not None:
 120             for size in card_sizes:
 121                 path = xpath_text(size, 'path')
 122                 if not path:
 123                     continue
 124                 width = int_or_none(size.get('width'))
 125                 thumbnails.append({
 126                     'id': width,
 127                     'url': path,
 128                     'width': width,
 129                 })
 130
 131         return {
 132             'id': video_id,
 133             'title': title,
 134             'duration': parse_duration(xpath_text(video_data, 'duration/tv14')),
 135             'formats': formats,
 136             'thumbnails': thumbnails,
 137         }
 138
 139
 140 class HBOIE(HBOBaseIE):
 141     IE_NAME = 'hbo'
 142     _VALID_URL = r'https?://(?:www\.)?hbo\.com/video/video\.html\?.*vid=(?P<id>[0-9]+)'
 143     _TEST = {
 144         'url': 'http://www.hbo.com/video/video.html?autoplay=true&g=u&vid=1437839',
 145         'md5': '2c6a6bc1222c7e91cb3334dad1746e5a',
 146         'info_dict': {
 147             'id': '1437839',
 148             'ext': 'mp4',
 149             'title': 'Ep. 64 Clip: Encryption',
 150             'thumbnail': r're:https?://.*\.jpg$',
 151             'duration': 1072,
 152         }
 153     }
 154
 155     def _real_extract(self, url):
 156         video_id = self._match_id(url)
 157         return self._extract_from_id(video_id)
 158
 159
 160 class HBOEpisodeIE(HBOBaseIE):
 161     IE_NAME = 'hbo:episode'
 162     _VALID_URL = r'https?://(?:www\.)?hbo\.com/(?P<path>(?!video)(?:(?:[^/]+/)+video|watch-free-episodes)/(?P<id>[0-9a-z-]+))(?:\.html)?'
 163
 164     _TESTS = [{
 165         'url': 'http://www.hbo.com/girls/episodes/5/52-i-love-you-baby/video/ep-52-inside-the-episode.html?autoplay=true',
 166         'md5': '61ead79b9c0dfa8d3d4b07ef4ac556fb',
 167         'info_dict': {
 168             'id': '1439518',
 169             'display_id': 'ep-52-inside-the-episode',
 170             'ext': 'mp4',
 171             'title': 'Ep. 52: Inside the Episode',
 172             'thumbnail': r're:https?://.*\.jpg$',
 173             'duration': 240,
 174         },
 175     }, {
 176         'url': 'http://www.hbo.com/game-of-thrones/about/video/season-5-invitation-to-the-set.html?autoplay=true',
 177         'only_matching': True,
 178     }, {
 179         'url': 'http://www.hbo.com/watch-free-episodes/last-week-tonight-with-john-oliver',
 180         'only_matching': True,
 181     }]
 182
 183     def _real_extract(self, url):
 184         path, display_id = re.match(self._VALID_URL, url).groups()
 185
 186         content = self._download_json(
 187             'http://www.hbo.com/api/content/' + path, display_id)['content']
 188
 189         video_id = compat_str((content.get('parsed', {}).get(
 190             'common:FullBleedVideo', {}) or content['selectedEpisode'])['videoId'])
 191
 192         info_dict = self._extract_from_id(video_id)
 193         info_dict['display_id'] = display_id
 194
 195         return info_dict