ytdl/youtube_dl/extractor/tbs.py

90 lines
3.3 KiB
Python
Raw Normal View History

2016-08-28 17:50:32 +02:00
# coding: utf-8
from __future__ import unicode_literals
import re
from .turner import TurnerBaseIE
from ..compat import (
compat_urllib_parse_urlparse,
compat_parse_qs,
)
2017-12-11 13:38:55 +01:00
from ..utils import (
float_or_none,
int_or_none,
strip_or_none,
)
2016-08-28 17:50:32 +02:00
class TBSIE(TurnerBaseIE):
2017-12-11 13:38:55 +01:00
_VALID_URL = r'https?://(?:www\.)?(?P<site>tbs|tntdrama)\.com/(?:movies|shows/[^/]+/(?:clips|season-\d+/episode-\d+))/(?P<id>[^/?#]+)'
2016-08-28 17:50:32 +02:00
_TESTS = [{
2017-12-11 13:38:55 +01:00
'url': 'http://www.tntdrama.com/shows/the-alienist/clips/monster',
2016-08-28 17:50:32 +02:00
'info_dict': {
2017-12-11 13:38:55 +01:00
'id': '8d384cde33b89f3a43ce5329de42903ed5099887',
2016-08-28 17:50:32 +02:00
'ext': 'mp4',
2017-12-11 13:38:55 +01:00
'title': 'Monster',
'description': 'Get a first look at the theatrical trailer for TNTs highly anticipated new psychological thriller The Alienist, which premieres January 22 on TNT.',
'timestamp': 1508175329,
'upload_date': '20171016',
},
2017-12-11 13:38:55 +01:00
'params': {
# m3u8 download
'skip_download': True,
}
2016-08-28 17:50:32 +02:00
}, {
2017-12-11 13:38:55 +01:00
'url': 'http://www.tbs.com/shows/search-party/season-1/episode-1/explicit-the-mysterious-disappearance-of-the-girl-no-one-knew',
'only_matching': True,
}, {
'url': 'http://www.tntdrama.com/movies/star-wars-a-new-hope',
'only_matching': True,
2016-08-28 17:50:32 +02:00
}]
def _real_extract(self, url):
2017-12-12 18:16:45 +01:00
site, display_id = re.match(self._VALID_URL, url).groups()
2016-08-28 17:50:32 +02:00
webpage = self._download_webpage(url, display_id)
drupal_settings = self._parse_json(self._search_regex(
2017-12-11 13:38:55 +01:00
r'<script[^>]+?data-drupal-selector="drupal-settings-json"[^>]*?>({.+?})</script>',
webpage, 'drupal setting'), display_id)
video_data = drupal_settings['turner_playlist'][0]
2017-12-11 13:38:55 +01:00
media_id = video_data['mediaID']
title = video_data['title']
tokenizer_query = compat_parse_qs(compat_urllib_parse_urlparse(
drupal_settings['ngtv_token_url']).query)
2017-12-11 13:38:55 +01:00
info = self._extract_ngtv_info(
media_id, tokenizer_query, {
'url': url,
'site_name': site[:3].upper(),
'auth_required': video_data.get('authRequired') == '1',
})
2017-12-11 13:38:55 +01:00
thumbnails = []
for image_id, image in video_data.get('images', {}).items():
image_url = image.get('url')
if not image_url or image.get('type') != 'video':
continue
i = {
'id': image_id,
'url': image_url,
}
mobj = re.search(r'(\d+)x(\d+)', image_url)
if mobj:
i.update({
'width': int(mobj.group(1)),
'height': int(mobj.group(2)),
})
thumbnails.append(i)
info.update({
2017-12-11 13:38:55 +01:00
'id': media_id,
'title': title,
'description': strip_or_none(video_data.get('descriptionNoTags') or video_data.get('shortDescriptionNoTags')),
'duration': float_or_none(video_data.get('duration')) or info.get('duration'),
2017-12-11 13:38:55 +01:00
'timestamp': int_or_none(video_data.get('created')),
'season_number': int_or_none(video_data.get('season')),
'episode_number': int_or_none(video_data.get('episode')),
'thumbnails': thumbnails,
})
return info