[extractor/Hidive] Fix subtitles and age-restriction (#5828)

Authored by: chexxor
Closes #408
This commit is contained in:
Alex Berg 2023-02-11 21:47:52 -06:00 committed by GitHub
parent b85faf6ffb
commit 7708df8da0
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23

View File

@ -1,5 +1,3 @@
import re
from .common import InfoExtractor from .common import InfoExtractor
from ..utils import ( from ..utils import (
ExtractorError, ExtractorError,
@ -39,15 +37,27 @@ def _perform_login(self, username, password):
form = self._search_regex( form = self._search_regex(
r'(?s)<form[^>]+action="/account/login"[^>]*>(.+?)</form>', r'(?s)<form[^>]+action="/account/login"[^>]*>(.+?)</form>',
webpage, 'login form', default=None) webpage, 'login form', default=None)
if not form: # logged in if not form:
return return
data = self._hidden_inputs(form) data = self._hidden_inputs(form)
data.update({ data.update({
'Email': username, 'Email': username,
'Password': password, 'Password': password,
}) })
self._download_webpage( login_webpage = self._download_webpage(
self._LOGIN_URL, None, 'Logging in', data=urlencode_postdata(data)) self._LOGIN_URL, None, 'Logging in', data=urlencode_postdata(data))
# If the user has multiple profiles on their account, select one. For now pick the first profile.
profile_id = self._search_regex(r'<button [^>]+?data-profile-id="(\w+)"', login_webpage, 'profile_id')
if profile_id is None:
return # If only one profile, Hidive auto-selects it
profile_id_hash = self._search_regex(r'\<button [^>]+?data-hash="(\w+)"', login_webpage, 'profile_id_hash')
self._request_webpage(
'https://www.hidive.com/ajax/chooseprofile', None,
data=urlencode_postdata({
'profileId': profile_id,
'hash': profile_id_hash,
'returnUrl': '/dashboard'
}))
def _call_api(self, video_id, title, key, data={}, **kwargs): def _call_api(self, video_id, title, key, data={}, **kwargs):
data = { data = {
@ -60,26 +70,6 @@ def _call_api(self, video_id, title, key, data={}, **kwargs):
'https://www.hidive.com/play/settings', video_id, 'https://www.hidive.com/play/settings', video_id,
data=urlencode_postdata(data), **kwargs) or {} data=urlencode_postdata(data), **kwargs) or {}
def _extract_subtitles_from_rendition(self, rendition, subtitles, parsed_urls):
for cc_file in rendition.get('ccFiles', []):
cc_url = url_or_none(try_get(cc_file, lambda x: x[2]))
# name is used since we cant distinguish subs with same language code
cc_lang = try_get(cc_file, (lambda x: x[1].replace(' ', '-').lower(), lambda x: x[0]), str)
if cc_url not in parsed_urls and cc_lang:
parsed_urls.add(cc_url)
subtitles.setdefault(cc_lang, []).append({'url': cc_url})
def _get_subtitles(self, url, video_id, title, key, parsed_urls):
webpage = self._download_webpage(url, video_id, fatal=False) or ''
subtitles = {}
for caption in set(re.findall(r'data-captions=\"([^\"]+)\"', webpage)):
renditions = self._call_api(
video_id, title, key, {'Captions': caption}, fatal=False,
note=f'Downloading {caption} subtitle information').get('renditions') or {}
for rendition_id, rendition in renditions.items():
self._extract_subtitles_from_rendition(rendition, subtitles, parsed_urls)
return subtitles
def _real_extract(self, url): def _real_extract(self, url):
video_id, title, key = self._match_valid_url(url).group('id', 'title', 'key') video_id, title, key = self._match_valid_url(url).group('id', 'title', 'key')
settings = self._call_api(video_id, title, key) settings = self._call_api(video_id, title, key)
@ -104,10 +94,20 @@ def _real_extract(self, url):
f['format_note'] = f'{version}, {extra}' f['format_note'] = f'{version}, {extra}'
formats.extend(frmt) formats.extend(frmt)
subtitles = {}
for rendition_id, rendition in settings['renditions'].items():
audio, version, extra = rendition_id.split('_')
for cc_file in rendition.get('ccFiles') or []:
cc_url = url_or_none(try_get(cc_file, lambda x: x[2]))
cc_lang = try_get(cc_file, (lambda x: x[1].replace(' ', '-').lower(), lambda x: x[0]), str)
if cc_url not in parsed_urls and cc_lang:
parsed_urls.add(cc_url)
subtitles.setdefault(cc_lang, []).append({'url': cc_url})
return { return {
'id': video_id, 'id': video_id,
'title': video_id, 'title': video_id,
'subtitles': self.extract_subtitles(url, video_id, title, key, parsed_urls), 'subtitles': subtitles,
'formats': formats, 'formats': formats,
'series': title, 'series': title,
'season_number': int_or_none( 'season_number': int_or_none(