]> git.bitcoin.ninja Git - youtube-dl/blob - youtube_dl/extractor/hbo.py
[brightcove:new] extract subtitles and strip video title
[youtube-dl] / 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 HBOIE(InfoExtractor):
16     _VALID_URL = r'https?://(?:www\.)?hbo\.com/video/video\.html\?.*vid=(?P<id>[0-9]+)'
17     _TEST = {
18         'url': 'http://www.hbo.com/video/video.html?autoplay=true&g=u&vid=1437839',
19         'md5': '1c33253f0c7782142c993c0ba62a8753',
20         'info_dict': {
21             'id': '1437839',
22             'ext': 'mp4',
23             'title': 'Ep. 64 Clip: Encryption',
24         }
25     }
26     _FORMATS_INFO = {
27         '1920': {
28             'width': 1280,
29             'height': 720,
30         },
31         '640': {
32             'width': 768,
33             'height': 432,
34         },
35         'highwifi': {
36             'width': 640,
37             'height': 360,
38         },
39         'high3g': {
40             'width': 640,
41             'height': 360,
42         },
43         'medwifi': {
44             'width': 400,
45             'height': 224,
46         },
47         'med3g': {
48             'width': 400,
49             'height': 224,
50         },
51     }
52
53     def _real_extract(self, url):
54         video_id = self._match_id(url)
55         video_data = self._download_xml(
56             'http://render.lv3.hbo.com/data/content/global/videos/data/%s.xml' % video_id, video_id)
57         title = xpath_text(video_data, 'title', 'title', True)
58
59         formats = []
60         for source in xpath_element(video_data, 'videos', 'sources', True):
61             if source.tag == 'size':
62                 path = xpath_text(source, './/path')
63                 if not path:
64                     continue
65                 width = source.attrib.get('width')
66                 format_info = self._FORMATS_INFO.get(width, {})
67                 height = format_info.get('height')
68                 fmt = {
69                     'url': path,
70                     'format_id': 'http%s' % ('-%dp' % height if height else ''),
71                     'width': format_info.get('width'),
72                     'height': height,
73                 }
74                 rtmp = re.search(r'^(?P<url>rtmpe?://[^/]+/(?P<app>.+))/(?P<playpath>mp4:.+)$', path)
75                 if rtmp:
76                     fmt.update({
77                         'url': rtmp.group('url'),
78                         'play_path': rtmp.group('playpath'),
79                         'app': rtmp.group('app'),
80                         'ext': 'flv',
81                         'format_id': fmt['format_id'].replace('http', 'rtmp'),
82                     })
83                 formats.append(fmt)
84             else:
85                 video_url = source.text
86                 if not video_url:
87                     continue
88                 if source.tag == 'tarball':
89                     formats.extend(self._extract_m3u8_formats(
90                         video_url.replace('.tar', '/base_index_w8.m3u8'),
91                         video_id, 'mp4', 'm3u8_native', m3u8_id='hls', fatal=False))
92                 else:
93                     format_info = self._FORMATS_INFO.get(source.tag, {})
94                     formats.append({
95                         'format_id': 'http-%s' % source.tag,
96                         'url': video_url,
97                         'width': format_info.get('width'),
98                         'height': format_info.get('height'),
99                     })
100         self._sort_formats(formats, ('width', 'height', 'tbr', 'format_id'))
101
102         thumbnails = []
103         card_sizes = xpath_element(video_data, 'titleCardSizes')
104         if card_sizes is not None:
105             for size in card_sizes:
106                 path = xpath_text(size, 'path')
107                 if not path:
108                     continue
109                 width = int_or_none(size.get('width'))
110                 thumbnails.append({
111                     'id': width,
112                     'url': path,
113                     'width': width,
114                 })
115
116         return {
117             'id': video_id,
118             'title': title,
119             'duration': parse_duration(xpath_element(video_data, 'duration/tv14')),
120             'formats': formats,
121             'thumbnails': thumbnails,
122         }