schmittlauch/youtube-dl
1
0
Fork 0
Code Issues Pull requests Releases Wiki Activity

[teamcoco] Fix extraction for full episodes(closes #16573)

Browse source
This commit is contained in:
Remita Amine 2018-05-30 13:21:07 +01:00
parent a07879d6b2
commit e0d42dd4b2
3 changed files with 122 additions and 88 deletions
Download patch file Download diff file Expand all files Collapse all files

61
youtube_dl/extractor/tbs.py
View file

@ -4,6 +4,10 @@ from __future__ import unicode_literals
import re
from .turner import TurnerBaseIE
from ..compat import (
compat_urllib_parse_urlparse,
compat_parse_qs,
)
from ..utils import (
float_or_none,
int_or_none,
@ -38,48 +42,22 @@ class TBSIE(TurnerBaseIE):
def _real_extract(self, url):
site, display_id = re.match(self._VALID_URL, url).groups()
webpage = self._download_webpage(url, display_id)
video_data = self._parse_json(self._search_regex(
drupal_settings = self._parse_json(self._search_regex(
r'<script[^>]+?data-drupal-selector="drupal-settings-json"[^>]*?>({.+?})</script>',
webpage, 'drupal setting'), display_id)['turner_playlist'][0]
webpage, 'drupal setting'), display_id)
video_data = drupal_settings['turner_playlist'][0]
media_id = video_data['mediaID']
title = video_data['title']
tokenizer_query = compat_parse_qs(compat_urllib_parse_urlparse(
drupal_settings['ngtv_token_url']).query)
streams_data = self._download_json(
'http://medium.ngtv.io/media/%s/tv' % media_id,
media_id)['media']['tv']
duration = None
chapters = []
formats = []
for supported_type in ('unprotected', 'bulkaes'):
stream_data = streams_data.get(supported_type, {})
m3u8_url = stream_data.get('secureUrl') or stream_data.get('url')
if not m3u8_url:
continue
if stream_data.get('playlistProtection') == 'spe':
m3u8_url = self._add_akamai_spe_token(
'http://token.vgtf.net/token/token_spe',
m3u8_url, media_id, {
'url': url,
'site_name': site[:3].upper(),
'auth_required': video_data.get('authRequired') == '1',
})
formats.extend(self._extract_m3u8_formats(
m3u8_url, media_id, 'mp4', m3u8_id='hls', fatal=False))
duration = float_or_none(stream_data.get('totalRuntime') or video_data.get('duration'))
if not chapters:
for chapter in stream_data.get('contentSegments', []):
start_time = float_or_none(chapter.get('start'))
duration = float_or_none(chapter.get('duration'))
if start_time is None or duration is None:
continue
chapters.append({
'start_time': start_time,
'end_time': start_time + duration,
})
self._sort_formats(formats)
info = self._extract_ngtv_info(
media_id, tokenizer_query, {
'url': url,
'site_name': site[:3].upper(),
'auth_required': video_data.get('authRequired') == '1',
})
thumbnails = []
for image_id, image in video_data.get('images', {}).items():
@ -98,15 +76,14 @@ class TBSIE(TurnerBaseIE):
})
thumbnails.append(i)
return {
info.update({
'id': media_id,
'title': title,
'description': strip_or_none(video_data.get('descriptionNoTags') or video_data.get('shortDescriptionNoTags')),
'duration': duration,
'duration': float_or_none(video_data.get('duration')) or info.get('duration'),
'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')),
'cahpters': chapters,
'thumbnails': thumbnails,
'formats': formats,
}
})
return info