diff options
Diffstat (limited to 'hypervideo_dl/extractor/viewlift.py')
| -rw-r--r-- | hypervideo_dl/extractor/viewlift.py | 189 |
1 files changed, 143 insertions, 46 deletions
diff --git a/hypervideo_dl/extractor/viewlift.py b/hypervideo_dl/extractor/viewlift.py index c3b2e86..4627f66 100644 --- a/hypervideo_dl/extractor/viewlift.py +++ b/hypervideo_dl/extractor/viewlift.py @@ -9,6 +9,7 @@ from ..utils import ( ExtractorError, int_or_none, parse_age_limit, + traverse_obj, ) @@ -32,26 +33,33 @@ class ViewLiftBaseIE(InfoExtractor): } _TOKENS = {} - def _call_api(self, site, path, video_id, query): - token = self._TOKENS.get(site) - if not token: - token_query = {'site': site} - email, password = self._get_login_info(netrc_machine=site) - if email: - resp = self._download_json( - self._API_BASE + 'identity/signin', video_id, - 'Logging in', query=token_query, data=json.dumps({ - 'email': email, - 'password': password, - }).encode()) - else: - resp = self._download_json( - self._API_BASE + 'identity/anonymous-token', video_id, - 'Downloading authorization token', query=token_query) - self._TOKENS[site] = token = resp['authorizationToken'] - return self._download_json( - self._API_BASE + path, video_id, - headers={'Authorization': token}, query=query) + def _fetch_token(self, site, url): + if self._TOKENS.get(site): + return + + cookies = self._get_cookies(url) + if cookies and cookies.get('token'): + self._TOKENS[site] = self._search_regex(r'22authorizationToken\%22:\%22([^\%]+)\%22', cookies['token'].value, 'token') + if not self._TOKENS.get(site): + self.raise_login_required('Cookies (not necessarily logged in) are needed to download from this website', method='cookies') + + def _call_api(self, site, path, video_id, url, query): + self._fetch_token(site, url) + try: + return self._download_json( + self._API_BASE + path, video_id, headers={'Authorization': self._TOKENS.get(site)}, query=query) + except ExtractorError as e: + if isinstance(e.cause, compat_HTTPError) and e.cause.code == 403: + webpage = e.cause.read().decode() + try: + error_message = traverse_obj(json.loads(webpage), 'errorMessage', 'message') + except json.JSONDecodeError: + raise ExtractorError(f'{site} said: {webpage}', cause=e.cause) + if error_message: + if 'has not purchased' in error_message: + self.raise_login_required(method='cookies') + raise ExtractorError(error_message, expected=True) + raise class ViewLiftEmbedIE(ViewLiftBaseIE): @@ -96,27 +104,24 @@ class ViewLiftEmbedIE(ViewLiftBaseIE): site = domain.split('.')[-2] if site in self._SITE_MAP: site = self._SITE_MAP[site] - try: - content_data = self._call_api( - site, 'entitlement/video/status', film_id, { - 'id': film_id - })['video'] - except ExtractorError as e: - if isinstance(e.cause, compat_HTTPError) and e.cause.code == 403: - error_message = self._parse_json(e.cause.read().decode(), film_id).get('errorMessage') - if error_message == 'User does not have a valid subscription or has not purchased this content.': - self.raise_login_required() - raise ExtractorError(error_message, expected=True) - raise + + content_data = self._call_api( + site, 'entitlement/video/status', film_id, url, { + 'id': film_id + })['video'] gist = content_data['gist'] title = gist['title'] video_assets = content_data['streamingInfo']['videoAssets'] - formats = [] - mpeg_video_assets = video_assets.get('mpeg') or [] - for video_asset in mpeg_video_assets: + hls_url = video_assets.get('hls') + formats, subtitles = [], {} + if hls_url: + formats, subtitles = self._extract_m3u8_formats_and_subtitles( + hls_url, film_id, 'mp4', 'm3u8_native', m3u8_id='hls', fatal=False) + + for video_asset in video_assets.get('mpeg') or []: video_asset_url = video_asset.get('url') - if not video_asset: + if not video_asset_url: continue bitrate = int_or_none(video_asset.get('bitrate')) height = int_or_none(self._search_regex( @@ -130,13 +135,17 @@ class ViewLiftEmbedIE(ViewLiftBaseIE): 'vcodec': video_asset.get('codec'), }) - hls_url = video_assets.get('hls') - if hls_url: - formats.extend(self._extract_m3u8_formats( - hls_url, film_id, 'mp4', 'm3u8_native', m3u8_id='hls', fatal=False)) - self._sort_formats(formats) + subs = {} + for sub in traverse_obj(content_data, ('contentDetails', 'closedCaptions')) or []: + sub_url = sub.get('url') + if not sub_url: + continue + subs.setdefault(sub.get('language', 'English'), []).append({ + 'url': sub_url, + }) - info = { + self._sort_formats(formats) + return { 'id': film_id, 'title': title, 'description': gist.get('description'), @@ -145,14 +154,15 @@ class ViewLiftEmbedIE(ViewLiftBaseIE): 'age_limit': parse_age_limit(content_data.get('parentalRating')), 'timestamp': int_or_none(gist.get('publishDate'), 1000), 'formats': formats, + 'subtitles': self._merge_subtitles(subs, subtitles), + 'categories': traverse_obj(content_data, ('categories', ..., 'title')), + 'tags': traverse_obj(content_data, ('tags', ..., 'title')), } - for k in ('categories', 'tags'): - info[k] = [v['title'] for v in content_data.get(k, []) if v.get('title')] - return info class ViewLiftIE(ViewLiftBaseIE): IE_NAME = 'viewlift' + _API_BASE = 'https://prod-api-cached-2.viewlift.com/' _VALID_URL = r'https?://(?:www\.)?(?P<domain>%s)(?P<path>(?:/(?:films/title|show|(?:news/)?videos?|watch))?/(?P<id>[^?#]+))' % ViewLiftBaseIE._DOMAINS_REGEX _TESTS = [{ 'url': 'http://www.snagfilms.com/films/title/lost_for_life', @@ -222,24 +232,111 @@ class ViewLiftIE(ViewLiftBaseIE): }, { 'url': 'https://www.marquee.tv/watch/sadlerswells-sacredmonsters', 'only_matching': True, + }, { # Free film with langauge code + 'url': 'https://www.hoichoi.tv/bn/films/title/shuyopoka', + 'info_dict': { + 'id': '7a7a9d33-1f4c-4771-9173-ee4fb6dbf196', + 'ext': 'mp4', + 'title': 'Shuyopoka', + 'description': 'md5:e28f2fb8680096a69c944d37c1fa5ffc', + 'thumbnail': r're:^https?://.*\.jpg$', + 'upload_date': '20211006', + 'series': None + }, + 'params': {'skip_download': True}, + }, { # Free film + 'url': 'https://www.hoichoi.tv/films/title/dadu-no1', + 'info_dict': { + 'id': '0000015b-b009-d126-a1db-b81ff3780000', + 'ext': 'mp4', + 'title': 'Dadu No.1', + 'description': 'md5:605cba408e51a79dafcb824bdeded51e', + 'thumbnail': r're:^https?://.*\.jpg$', + 'upload_date': '20210827', + 'series': None + }, + 'params': {'skip_download': True}, + }, { # Free episode + 'url': 'https://www.hoichoi.tv/webseries/case-jaundice-s01-e01', + 'info_dict': { + 'id': 'f779e07c-30c8-459c-8612-5a834ab5e5ba', + 'ext': 'mp4', + 'title': 'Humans Vs. Corona', + 'description': 'md5:ca30a682b4528d02a3eb6d0427dd0f87', + 'thumbnail': r're:^https?://.*\.jpg$', + 'upload_date': '20210830', + 'series': 'Case Jaundice' + }, + 'params': {'skip_download': True}, + }, { # Free video + 'url': 'https://www.hoichoi.tv/videos/1549072415320-six-episode-02-hindi', + 'info_dict': { + 'id': 'b41fa1ce-aca6-47b6-b208-283ff0a2de30', + 'ext': 'mp4', + 'title': 'Woman in red - Hindi', + 'description': 'md5:9d21edc1827d32f8633eb67c2054fc31', + 'thumbnail': r're:^https?://.*\.jpg$', + 'upload_date': '20211006', + 'series': 'Six (Hindi)' + }, + 'params': {'skip_download': True}, + }, { # Free episode + 'url': 'https://www.hoichoi.tv/shows/watch-asian-paints-moner-thikana-online-season-1-episode-1', + 'info_dict': { + 'id': '1f45d185-8500-455c-b88d-13252307c3eb', + 'ext': 'mp4', + 'title': 'Jisshu Sengupta', + 'description': 'md5:ef6ffae01a3d83438597367400f824ed', + 'thumbnail': r're:^https?://.*\.jpg$', + 'upload_date': '20211004', + 'series': 'Asian Paints Moner Thikana' + }, + 'params': {'skip_download': True}, + }, { # Free series + 'url': 'https://www.hoichoi.tv/shows/watch-moner-thikana-bengali-web-series-online', + 'playlist_mincount': 5, + 'info_dict': { + 'id': 'watch-moner-thikana-bengali-web-series-online', + }, + }, { # Premium series + 'url': 'https://www.hoichoi.tv/shows/watch-byomkesh-bengali-web-series-online', + 'playlist_mincount': 14, + 'info_dict': { + 'id': 'watch-byomkesh-bengali-web-series-online', + }, + }, { # Premium movie + 'url': 'https://www.hoichoi.tv/movies/detective-2020', + 'only_matching': True }] @classmethod def suitable(cls, url): return False if ViewLiftEmbedIE.suitable(url) else super(ViewLiftIE, cls).suitable(url) + def _show_entries(self, domain, seasons): + for season in seasons: + for episode in season.get('episodes') or []: + path = traverse_obj(episode, ('gist', 'permalink')) + if path: + yield self.url_result(f'https://www.{domain}{path}', ie=self.ie_key()) + def _real_extract(self, url): domain, path, display_id = self._match_valid_url(url).groups() site = domain.split('.')[-2] if site in self._SITE_MAP: site = self._SITE_MAP[site] modules = self._call_api( - site, 'content/pages', display_id, { + site, 'content/pages', display_id, url, { 'includeContent': 'true', 'moduleOffset': 1, 'path': path, 'site': site, })['modules'] + + seasons = next((m['contentData'][0]['seasons'] for m in modules if m.get('moduleType') == 'ShowDetailModule'), None) + if seasons: + return self.playlist_result(self._show_entries(domain, seasons), display_id) + film_id = next(m['contentData'][0]['gist']['id'] for m in modules if m.get('moduleType') == 'VideoDetailModule') return { '_type': 'url_transparent', |