[youtube] Fix extraction.
[youtube-dl] / youtube_dl / extractor / tbs.py
1 # coding: utf-8
2 from __future__ import unicode_literals
3
4 import re
5
6 from .turner import TurnerBaseIE
7 from ..compat import (
8     compat_urllib_parse_urlparse,
9     compat_parse_qs,
10 )
11 from ..utils import (
12     float_or_none,
13     int_or_none,
14     strip_or_none,
15 )
16
17
18 class TBSIE(TurnerBaseIE):
19     _VALID_URL = r'https?://(?:www\.)?(?P<site>tbs|tntdrama)\.com(?P<path>/(?:movies|shows/[^/]+/(?:clips|season-\d+/episode-\d+))/(?P<id>[^/?#]+))'
20     _TESTS = [{
21         'url': 'http://www.tntdrama.com/shows/the-alienist/clips/monster',
22         'info_dict': {
23             'id': '8d384cde33b89f3a43ce5329de42903ed5099887',
24             'ext': 'mp4',
25             'title': 'Monster',
26             'description': 'Get a first look at the theatrical trailer for TNT’s highly anticipated new psychological thriller The Alienist, which premieres January 22 on TNT.',
27             'timestamp': 1508175329,
28             'upload_date': '20171016',
29         },
30         'params': {
31             # m3u8 download
32             'skip_download': True,
33         }
34     }, {
35         'url': 'http://www.tbs.com/shows/search-party/season-1/episode-1/explicit-the-mysterious-disappearance-of-the-girl-no-one-knew',
36         'only_matching': True,
37     }, {
38         'url': 'http://www.tntdrama.com/movies/star-wars-a-new-hope',
39         'only_matching': True,
40     }]
41
42     def _real_extract(self, url):
43         site, path, display_id = re.match(self._VALID_URL, url).groups()
44         webpage = self._download_webpage(url, display_id)
45         drupal_settings = self._parse_json(self._search_regex(
46             r'<script[^>]+?data-drupal-selector="drupal-settings-json"[^>]*?>({.+?})</script>',
47             webpage, 'drupal setting'), display_id)
48         video_data = next(v for v in drupal_settings['turner_playlist'] if v.get('url') == path)
49
50         media_id = video_data['mediaID']
51         title = video_data['title']
52         tokenizer_query = compat_parse_qs(compat_urllib_parse_urlparse(
53             drupal_settings['ngtv_token_url']).query)
54
55         info = self._extract_ngtv_info(
56             media_id, tokenizer_query, {
57                 'url': url,
58                 'site_name': site[:3].upper(),
59                 'auth_required': video_data.get('authRequired') == '1',
60             })
61
62         thumbnails = []
63         for image_id, image in video_data.get('images', {}).items():
64             image_url = image.get('url')
65             if not image_url or image.get('type') != 'video':
66                 continue
67             i = {
68                 'id': image_id,
69                 'url': image_url,
70             }
71             mobj = re.search(r'(\d+)x(\d+)', image_url)
72             if mobj:
73                 i.update({
74                     'width': int(mobj.group(1)),
75                     'height': int(mobj.group(2)),
76                 })
77             thumbnails.append(i)
78
79         info.update({
80             'id': media_id,
81             'title': title,
82             'description': strip_or_none(video_data.get('descriptionNoTags') or video_data.get('shortDescriptionNoTags')),
83             'duration': float_or_none(video_data.get('duration')) or info.get('duration'),
84             'timestamp': int_or_none(video_data.get('created')),
85             'season_number': int_or_none(video_data.get('season')),
86             'episode_number': int_or_none(video_data.get('episode')),
87             'thumbnails': thumbnails,
88         })
89         return info