You cannot select more than 25 topics Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.
yt-dlp/yt_dlp/extractor/tubitv.py

178 lines
7.3 KiB
Python

import re
from .common import InfoExtractor
from ..networking import Request
from ..utils import (
ExtractorError,
int_or_none,
js_to_json,
traverse_obj,
url_or_none,
urlencode_postdata,
)
class TubiTvIE(InfoExtractor):
_VALID_URL = r'https?://(?:www\.)?tubitv\.com/(?P<type>video|movies|tv-shows)/(?P<id>\d+)'
_LOGIN_URL = 'http://tubitv.com/login'
_NETRC_MACHINE = 'tubitv'
_TESTS = [{
'url': 'https://tubitv.com/movies/100004539/the-39-steps',
'info_dict': {
'id': '100004539',
'ext': 'mp4',
'title': 'The 39 Steps',
'description': 'md5:bb2f2dd337f0dc58c06cb509943f54c8',
'uploader_id': 'abc2558d54505d4f0f32be94f2e7108c',
'release_year': 1935,
'thumbnail': r're:^https?://.+\.(jpe?g|png)$',
'duration': 5187,
},
'params': {'skip_download': 'm3u8'},
}, {
'url': 'https://tubitv.com/tv-shows/554628/s01-e01-rise-of-the-snakes',
'info_dict': {
'id': '554628',
'ext': 'mp4',
'title': 'S01:E01 - Rise of the Snakes',
'description': 'md5:ba136f586de53af0372811e783a3f57d',
'episode': 'Rise of the Snakes',
'episode_number': 1,
'season': 'Season 1',
'season_number': 1,
'uploader_id': '2a9273e728c510d22aa5c57d0646810b',
'release_year': 2011,
'thumbnail': r're:^https?://.+\.(jpe?g|png)$',
'duration': 1376,
},
'params': {'skip_download': 'm3u8'},
}, {
'url': 'http://tubitv.com/video/283829/the_comedian_at_the_friday',
'md5': '43ac06be9326f41912dc64ccf7a80320',
'info_dict': {
'id': '283829',
'ext': 'mp4',
'title': 'The Comedian at The Friday',
'description': 'A stand up comedian is forced to look at the decisions in his life while on a one week trip to the west coast.',
'uploader_id': 'bc168bee0d18dd1cb3b86c68706ab434',
},
'skip': 'Content Unavailable'
}, {
'url': 'http://tubitv.com/tv-shows/321886/s01_e01_on_nom_stories',
'only_matching': True,
}, {
'url': 'https://tubitv.com/movies/560057/penitentiary?start=true',
'info_dict': {
'id': '560057',
'ext': 'mp4',
'title': 'Penitentiary',
'description': 'md5:8d2fc793a93cc1575ff426fdcb8dd3f9',
'uploader_id': 'd8fed30d4f24fcb22ec294421b9defc2',
'release_year': 1979,
},
'skip': 'Content Unavailable'
}]
# DRM formats are included only to raise appropriate error
_UNPLAYABLE_FORMATS = ('hlsv6_widevine', 'hlsv6_widevine_nonclearlead', 'hlsv6_playready_psshv0',
'hlsv6_fairplay', 'dash_widevine', 'dash_widevine_nonclearlead')
def _perform_login(self, username, password):
self.report_login()
form_data = {
'username': username,
'password': password,
}
payload = urlencode_postdata(form_data)
request = Request(self._LOGIN_URL, payload)
request.headers['Content-Type'] = 'application/x-www-form-urlencoded'
login_page = self._download_webpage(
request, None, False, 'Wrong login info')
if not re.search(r'id="tubi-logout"', login_page):
raise ExtractorError(
'Login failed (invalid username/password)', expected=True)
def _real_extract(self, url):
video_id, video_type = self._match_valid_url(url).group('id', 'type')
webpage = self._download_webpage(f'https://tubitv.com/{video_type}/{video_id}/', video_id)
video_data = self._search_json(
r'window\.__data\s*=', webpage, 'data', video_id,
transform_source=js_to_json)['video']['byId'][video_id]
formats = []
drm_formats = False
for resource in traverse_obj(video_data, ('video_resources', lambda _, v: url_or_none(v['manifest']['url']))):
resource_type = resource.get('type')
manifest_url = resource['manifest']['url']
if resource_type == 'dash':
formats.extend(self._extract_mpd_formats(manifest_url, video_id, mpd_id=resource_type, fatal=False))
elif resource_type in ('hlsv3', 'hlsv6'):
formats.extend(self._extract_m3u8_formats(manifest_url, video_id, 'mp4', m3u8_id=resource_type, fatal=False))
elif resource_type in self._UNPLAYABLE_FORMATS:
drm_formats = True
else:
self.report_warning(f'Skipping unknown resource type "{resource_type}"')
if not formats and drm_formats:
self.report_drm(video_id)
elif not formats and not video_data.get('policy_match'): # policy_match is False if content was removed
raise ExtractorError('This content is currently unavailable', expected=True)
subtitles = {}
for sub in traverse_obj(video_data, ('subtitles', lambda _, v: url_or_none(v['url']))):
subtitles.setdefault(sub.get('lang', 'English'), []).append({
'url': self._proto_relative_url(sub['url']),
})
title = traverse_obj(video_data, ('title', {str}))
season_number, episode_number, episode_title = self._search_regex(
r'^S(\d+):E(\d+) - (.+)', title, 'episode info', fatal=False, group=(1, 2, 3), default=(None, None, None))
return {
'id': video_id,
'title': title,
'formats': formats,
'subtitles': subtitles,
'season_number': int_or_none(season_number),
'episode_number': int_or_none(episode_number),
'episode': episode_title,
**traverse_obj(video_data, {
'description': ('description', {str}),
'duration': ('duration', {int_or_none}),
'uploader_id': ('publisher_id', {str}),
'release_year': ('year', {int_or_none}),
'thumbnails': ('thumbnails', ..., {url_or_none}, {'url': {self._proto_relative_url}}),
}),
}
class TubiTvShowIE(InfoExtractor):
_WORKING = False
_VALID_URL = r'https?://(?:www\.)?tubitv\.com/series/[0-9]+/(?P<show_name>[^/?#]+)'
_TESTS = [{
'url': 'https://tubitv.com/series/3936/the-joy-of-painting-with-bob-ross?start=true',
'playlist_mincount': 390,
'info_dict': {
'id': 'the-joy-of-painting-with-bob-ross',
}
}]
def _entries(self, show_url, show_name):
show_webpage = self._download_webpage(show_url, show_name)
show_json = self._parse_json(self._search_regex(
r'window\.__data\s*=\s*({[^<]+});\s*</script>',
show_webpage, 'data'), show_name, transform_source=js_to_json)['video']
for episode_id in show_json['fullContentById'].keys():
if traverse_obj(show_json, ('byId', episode_id, 'type')) == 's':
continue
yield self.url_result(
f'https://tubitv.com/tv-shows/{episode_id}/',
ie=TubiTvIE.ie_key(), video_id=episode_id)
def _real_extract(self, url):
show_name = self._match_valid_url(url).group('show_name')
return self.playlist_result(self._entries(url, show_name), playlist_id=show_name)