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     urljoin,
  13 )
  14
  15
  16 class HBOIE(InfoExtractor):
  17     IE_NAME = 'hbo'
  18     _VALID_URL = r'https?://(?:www\.)?hbo\.com/(?:video|embed)(?:/[^/]+)*/(?P<id>[^/?#]+)'
  19     _TEST = {
  20         'url': 'https://www.hbo.com/video/game-of-thrones/seasons/season-8/videos/trailer',
  21         'md5': '8126210656f433c452a21367f9ad85b3',
  22         'info_dict': {
  23             'id': '22113301',
  24             'ext': 'mp4',
  25             'title': 'Game of Thrones - Trailer',
  26         },
  27         'expected_warnings': ['Unknown MIME type application/mp4 in DASH manifest'],
  28     }
  29     _FORMATS_INFO = {
  30         'pro7': {
  31             'width': 1280,
  32             'height': 720,
  33         },
  34         '1920': {
  35             'width': 1280,
  36             'height': 720,
  37         },
  38         'pro6': {
  39             'width': 768,
  40             'height': 432,
  41         },
  42         '640': {
  43             'width': 768,
  44             'height': 432,
  45         },
  46         'pro5': {
  47             'width': 640,
  48             'height': 360,
  49         },
  50         'highwifi': {
  51             'width': 640,
  52             'height': 360,
  53         },
  54         'high3g': {
  55             'width': 640,
  56             'height': 360,
  57         },
  58         'medwifi': {
  59             'width': 400,
  60             'height': 224,
  61         },
  62         'med3g': {
  63             'width': 400,
  64             'height': 224,
  65         },
  66     }
  67
  68     def _real_extract(self, url):
  69         display_id = self._match_id(url)
  70         webpage = self._download_webpage(url, display_id)
  71         location_path = self._parse_json(self._html_search_regex(
  72             r'data-state="({.+?})"', webpage, 'state'), display_id)['video']['locationUrl']
  73         video_data = self._download_xml(urljoin(url, location_path), display_id)
  74         video_id = xpath_text(video_data, 'id', fatal=True)
  75         episode_title = title = xpath_text(video_data, 'title', fatal=True)
  76         series = xpath_text(video_data, 'program')
  77         if series:
  78             title = '%s - %s' % (series, title)
  79
  80         formats = []
  81         for source in xpath_element(video_data, 'videos', 'sources', True):
  82             if source.tag == 'size':
  83                 path = xpath_text(source, './/path')
  84                 if not path:
  85                     continue
  86                 width = source.attrib.get('width')
  87                 format_info = self._FORMATS_INFO.get(width, {})
  88                 height = format_info.get('height')
  89                 fmt = {
  90                     'url': path,
  91                     'format_id': 'http%s' % ('-%dp' % height if height else ''),
  92                     'width': format_info.get('width'),
  93                     'height': height,
  94                 }
  95                 rtmp = re.search(r'^(?P<url>rtmpe?://[^/]+/(?P<app>.+))/(?P<playpath>mp4:.+)$', path)
  96                 if rtmp:
  97                     fmt.update({
  98                         'url': rtmp.group('url'),
  99                         'play_path': rtmp.group('playpath'),
 100                         'app': rtmp.group('app'),
 101                         'ext': 'flv',
 102                         'format_id': fmt['format_id'].replace('http', 'rtmp'),
 103                     })
 104                 formats.append(fmt)
 105             else:
 106                 video_url = source.text
 107                 if not video_url:
 108                     continue
 109                 if source.tag == 'tarball':
 110                     formats.extend(self._extract_m3u8_formats(
 111                         video_url.replace('.tar', '/base_index_w8.m3u8'),
 112                         video_id, 'mp4', 'm3u8_native', m3u8_id='hls', fatal=False))
 113                 elif source.tag == 'hls':
 114                     m3u8_formats = self._extract_m3u8_formats(
 115                         video_url.replace('.tar', '/base_index.m3u8'),
 116                         video_id, 'mp4', 'm3u8_native', m3u8_id='hls', fatal=False)
 117                     for f in m3u8_formats:
 118                         if f.get('vcodec') == 'none' and not f.get('tbr'):
 119                             f['tbr'] = int_or_none(self._search_regex(
 120                                 r'-(\d+)k/', f['url'], 'tbr', default=None))
 121                     formats.extend(m3u8_formats)
 122                 elif source.tag == 'dash':
 123                     formats.extend(self._extract_mpd_formats(
 124                         video_url.replace('.tar', '/manifest.mpd'),
 125                         video_id, mpd_id='dash', fatal=False))
 126                 else:
 127                     format_info = self._FORMATS_INFO.get(source.tag, {})
 128                     formats.append({
 129                         'format_id': 'http-%s' % source.tag,
 130                         'url': video_url,
 131                         'width': format_info.get('width'),
 132                         'height': format_info.get('height'),
 133                     })
 134         self._sort_formats(formats)
 135
 136         thumbnails = []
 137         card_sizes = xpath_element(video_data, 'titleCardSizes')
 138         if card_sizes is not None:
 139             for size in card_sizes:
 140                 path = xpath_text(size, 'path')
 141                 if not path:
 142                     continue
 143                 width = int_or_none(size.get('width'))
 144                 thumbnails.append({
 145                     'id': width,
 146                     'url': path,
 147                     'width': width,
 148                 })
 149
 150         subtitles = None
 151         caption_url = xpath_text(video_data, 'captionUrl')
 152         if caption_url:
 153             subtitles = {
 154                 'en': [{
 155                     'url': caption_url,
 156                     'ext': 'ttml'
 157                 }],
 158             }
 159
 160         return {
 161             'id': video_id,
 162             'title': title,
 163             'duration': parse_duration(xpath_text(video_data, 'duration/tv14')),
 164             'series': series,
 165             'episode': episode_title,
 166             'formats': formats,
 167             'thumbnails': thumbnails,
 168             'subtitles': subtitles,
 169         }