]> jfr.im git - yt-dlp.git/blob - yt_dlp/extractor/tbs.py
[cleanup, docs] Misc cleanup
[yt-dlp.git] / yt_dlp / 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|watchtnt|watchtbs|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 = self._match_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 isLive = 'watchtnt' in path or 'watchtbs' in path
49 video_data = next(v for v in drupal_settings['turner_playlist'] if isLive or v.get('url') == path)
50
51 media_id = video_data['mediaID']
52 title = video_data['title']
53 tokenizer_query = compat_parse_qs(compat_urllib_parse_urlparse(
54 drupal_settings['ngtv_token_url']).query)
55
56 info = self._extract_ngtv_info(
57 media_id, tokenizer_query, {
58 'url': url,
59 'site_name': site[:3].upper(),
60 'auth_required': video_data.get('authRequired') == '1' or isLive,
61 'is_live': isLive
62 })
63
64 thumbnails = []
65 for image_id, image in video_data.get('images', {}).items():
66 image_url = image.get('url')
67 if not image_url or image.get('type') != 'video':
68 continue
69 i = {
70 'id': image_id,
71 'url': image_url,
72 }
73 mobj = re.search(r'(\d+)x(\d+)', image_url)
74 if mobj:
75 i.update({
76 'width': int(mobj.group(1)),
77 'height': int(mobj.group(2)),
78 })
79 thumbnails.append(i)
80
81 info.update({
82 'id': media_id,
83 'title': title,
84 'description': strip_or_none(video_data.get('descriptionNoTags') or video_data.get('shortDescriptionNoTags')),
85 'duration': float_or_none(video_data.get('duration')) or info.get('duration'),
86 'timestamp': int_or_none(video_data.get('created')),
87 'season_number': int_or_none(video_data.get('season')),
88 'episode_number': int_or_none(video_data.get('episode')),
89 'thumbnails': thumbnails,
90 'is_live': isLive
91 })
92 return info