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 ..utils import (
   8     xpath_text,
   9     xpath_element,
  10     int_or_none,
  11     parse_duration,
  12 )
  13
  14
  15 class HBOBaseIE(InfoExtractor):
  16     _FORMATS_INFO = {
  17         '1920': {
  18             'width': 1280,
  19             'height': 720,
  20         },
  21         '640': {
  22             'width': 768,
  23             'height': 432,
  24         },
  25         'highwifi': {
  26             'width': 640,
  27             'height': 360,
  28         },
  29         'high3g': {
  30             'width': 640,
  31             'height': 360,
  32         },
  33         'medwifi': {
  34             'width': 400,
  35             'height': 224,
  36         },
  37         'med3g': {
  38             'width': 400,
  39             'height': 224,
  40         },
  41     }
  42
  43     def _extract_from_id(self, video_id):
  44         video_data = self._download_xml(
  45             'http://render.lv3.hbo.com/data/content/global/videos/data/%s.xml' % video_id, video_id)
  46         title = xpath_text(video_data, 'title', 'title', True)
  47
  48         formats = []
  49         for source in xpath_element(video_data, 'videos', 'sources', True):
  50             if source.tag == 'size':
  51                 path = xpath_text(source, './/path')
  52                 if not path:
  53                     continue
  54                 width = source.attrib.get('width')
  55                 format_info = self._FORMATS_INFO.get(width, {})
  56                 height = format_info.get('height')
  57                 fmt = {
  58                     'url': path,
  59                     'format_id': 'http%s' % ('-%dp' % height if height else ''),
  60                     'width': format_info.get('width'),
  61                     'height': height,
  62                 }
  63                 rtmp = re.search(r'^(?P<url>rtmpe?://[^/]+/(?P<app>.+))/(?P<playpath>mp4:.+)$', path)
  64                 if rtmp:
  65                     fmt.update({
  66                         'url': rtmp.group('url'),
  67                         'play_path': rtmp.group('playpath'),
  68                         'app': rtmp.group('app'),
  69                         'ext': 'flv',
  70                         'format_id': fmt['format_id'].replace('http', 'rtmp'),
  71                     })
  72                 formats.append(fmt)
  73             else:
  74                 video_url = source.text
  75                 if not video_url:
  76                     continue
  77                 if source.tag == 'tarball':
  78                     formats.extend(self._extract_m3u8_formats(
  79                         video_url.replace('.tar', '/base_index_w8.m3u8'),
  80                         video_id, 'mp4', 'm3u8_native', m3u8_id='hls', fatal=False))
  81                 else:
  82                     format_info = self._FORMATS_INFO.get(source.tag, {})
  83                     formats.append({
  84                         'format_id': 'http-%s' % source.tag,
  85                         'url': video_url,
  86                         'width': format_info.get('width'),
  87                         'height': format_info.get('height'),
  88                     })
  89         self._sort_formats(formats, ('width', 'height', 'tbr', 'format_id'))
  90
  91         thumbnails = []
  92         card_sizes = xpath_element(video_data, 'titleCardSizes')
  93         if card_sizes is not None:
  94             for size in card_sizes:
  95                 path = xpath_text(size, 'path')
  96                 if not path:
  97                     continue
  98                 width = int_or_none(size.get('width'))
  99                 thumbnails.append({
 100                     'id': width,
 101                     'url': path,
 102                     'width': width,
 103                 })
 104
 105         return {
 106             'id': video_id,
 107             'title': title,
 108             'duration': parse_duration(xpath_text(video_data, 'duration/tv14')),
 109             'formats': formats,
 110             'thumbnails': thumbnails,
 111         }
 112
 113
 114 class HBOIE(HBOBaseIE):
 115     _VALID_URL = r'https?://(?:www\.)?hbo\.com/video/video\.html\?.*vid=(?P<id>[0-9]+)'
 116     _TEST = {
 117         'url': 'http://www.hbo.com/video/video.html?autoplay=true&g=u&vid=1437839',
 118         'md5': '1c33253f0c7782142c993c0ba62a8753',
 119         'info_dict': {
 120             'id': '1437839',
 121             'ext': 'mp4',
 122             'title': 'Ep. 64 Clip: Encryption',
 123             'thumbnail': r're:https?://.*\.jpg$',
 124             'duration': 1072,
 125         }
 126     }
 127
 128     def _real_extract(self, url):
 129         video_id = self._match_id(url)
 130         return self._extract_from_id(video_id)
 131
 132
 133 class HBOEpisodeIE(HBOBaseIE):
 134     _VALID_URL = r'https?://(?:www\.)?hbo\.com/(?!video)([^/]+/)+video/(?P<id>[0-9a-z-]+)\.html'
 135
 136     _TESTS = [{
 137         'url': 'http://www.hbo.com/girls/episodes/5/52-i-love-you-baby/video/ep-52-inside-the-episode.html?autoplay=true',
 138         'md5': '689132b253cc0ab7434237fc3a293210',
 139         'info_dict': {
 140             'id': '1439518',
 141             'display_id': 'ep-52-inside-the-episode',
 142             'ext': 'mp4',
 143             'title': 'Ep. 52: Inside the Episode',
 144             'thumbnail': r're:https?://.*\.jpg$',
 145             'duration': 240,
 146         },
 147     }, {
 148         'url': 'http://www.hbo.com/game-of-thrones/about/video/season-5-invitation-to-the-set.html?autoplay=true',
 149         'only_matching': True,
 150     }]
 151
 152     def _real_extract(self, url):
 153         display_id = self._match_id(url)
 154
 155         webpage = self._download_webpage(url, display_id)
 156
 157         video_id = self._search_regex(
 158             r'(?P<q1>[\'"])videoId(?P=q1)\s*:\s*(?P<q2>[\'"])(?P<video_id>\d+)(?P=q2)',
 159             webpage, 'video ID', group='video_id')
 160
 161         info_dict = self._extract_from_id(video_id)
 162         info_dict['display_id'] = display_id
 163
 164         return info_dict