From 2f4b57594673035a59d72f7667588da848820034 Mon Sep 17 00:00:00 2001 From: HobbyistDev <105957301+HobbyistDev@users.noreply.github.com> Date: Sat, 3 Feb 2024 05:56:29 +0900 Subject: [PATCH 01/10] [ie/zetland] Add extractor (#9116) Closes #9024 Authored by: HobbyistDev --- yt_dlp/extractor/_extractors.py | 1 + yt_dlp/extractor/zetland.py | 71 +++++++++++++++++++++++++++++++++ 2 files changed, 72 insertions(+) create mode 100644 yt_dlp/extractor/zetland.py diff --git a/yt_dlp/extractor/_extractors.py b/yt_dlp/extractor/_extractors.py index 4c86040990aa..7726fe359728 100644 --- a/yt_dlp/extractor/_extractors.py +++ b/yt_dlp/extractor/_extractors.py @@ -2496,6 +2496,7 @@ Zee5SeriesIE, ) from .zeenews import ZeeNewsIE +from .zetland import ZetlandDKArticleIE from .zhihu import ZhihuIE from .zingmp3 import ( ZingMp3IE, diff --git a/yt_dlp/extractor/zetland.py b/yt_dlp/extractor/zetland.py new file mode 100644 index 000000000000..055a643b3ce5 --- /dev/null +++ b/yt_dlp/extractor/zetland.py @@ -0,0 +1,71 @@ +from .common import InfoExtractor +from ..utils import merge_dicts, unified_timestamp, url_or_none +from ..utils.traversal import traverse_obj + + +class ZetlandDKArticleIE(InfoExtractor): + _VALID_URL = r'https?://www\.zetland\.dk/\w+/(?P(?P\w{8})-(?P\w{8})-(?:\w{5}))' + _TESTS = [{ + 'url': 'https://www.zetland.dk/historie/sO9aq2MY-a81VP3BY-66e69?utm_source=instagram&utm_medium=linkibio&utm_campaign=artikel', + 'info_dict': { + 'id': 'sO9aq2MY-a81VP3BY-66e69', + 'ext': 'mp3', + 'modified_date': '20240118', + 'title': 'Afsnit 1: “Det føltes som en kidnapning.” ', + 'upload_date': '20240116', + 'uploader_id': 'a81VP3BY', + 'modified_timestamp': 1705568739, + 'release_timestamp': 1705377592, + 'uploader_url': 'https://www.zetland.dk/skribent/a81VP3BY', + 'uploader': 'Helle Fuusager', + 'release_date': '20240116', + 'thumbnail': r're:https://zetland\.imgix\.net/2aafe500-b14e-11ee-bf83-65d5e1283a57/Zetland_Image_1\.jpg', + 'description': 'md5:9619d426772c133f5abb26db27f26a01', + 'timestamp': 1705377592, + 'series_id': '62d54630-e87b-4ab1-a255-8de58dbe1b14', + } + + }] + + def _real_extract(self, url): + display_id, uploader_id = self._match_valid_url(url).group('id', 'uploader_id') + webpage = self._download_webpage(url, display_id) + + next_js_data = self._search_nextjs_data(webpage, display_id)['props']['pageProps'] + story_data = traverse_obj(next_js_data, ('initialState', 'consume', 'story', 'story')) + + formats = [] + for audio_url in traverse_obj(story_data, ('story_content', 'meta', 'audioFiles', ..., {url_or_none})): + formats.append({ + 'url': audio_url, + 'vcodec': 'none', + }) + + return merge_dicts({ + 'id': display_id, + 'formats': formats, + 'uploader_id': uploader_id + }, traverse_obj(story_data, { + 'title': ((('story_content', 'content', 'title'), 'title'), {str}), + 'uploader': ('sharer', 'name'), + 'uploader_id': ('sharer', 'sharer_id'), + 'description': ('story_content', 'content', 'socialDescription'), + 'series_id': ('story_content', 'meta', 'seriesId'), + 'release_timestamp': ('published_at', {unified_timestamp}), + 'modified_timestamp': ('revised_at', {unified_timestamp}), + }, get_all=False), traverse_obj(next_js_data, ('metaInfo', { + 'title': ((('meta', 'title'), ('ld', 'headline'), ('og', 'og:title'), ('og', 'twitter:title')), {str}), + 'description': ((('meta', 'description'), ('ld', 'description'), ('og', 'og:description'), ('og', 'twitter:description')), {str}), + 'uploader': ((('meta', 'author'), ('ld', 'author', 'name')), {str}), + 'uploader_url': ('ld', 'author', 'url', {url_or_none}), + 'thumbnail': ((('ld', 'image'), ('og', 'og:image'), ('og', 'twitter:image')), {url_or_none}), + 'modified_timestamp': ('ld', 'dateModified', {unified_timestamp}), + 'release_timestamp': ('ld', 'datePublished', {unified_timestamp}), + 'timestamp': ('ld', 'dateCreated', {unified_timestamp}), + }), get_all=False), { + 'title': self._html_search_meta(['title', 'og:title', 'twitter:title'], webpage), + 'description': self._html_search_meta(['description', 'og:description', 'twitter:description'], webpage), + 'thumbnail': self._html_search_meta(['og:image', 'twitter:image'], webpage), + 'uploader': self._html_search_meta(['author'], webpage), + 'release_timestamp': unified_timestamp(self._html_search_meta(['article:published_time'], webpage)), + }, self._search_json_ld(webpage, display_id, fatal=False)) From a0d50aabc5462aee302bd3f2663d3a3554875789 Mon Sep 17 00:00:00 2001 From: HobbyistDev <105957301+HobbyistDev@users.noreply.github.com> Date: Sat, 3 Feb 2024 05:57:53 +0900 Subject: [PATCH 02/10] [ie/orf:on] Add extractor (#9113) Closes #8903 Authored by: HobbyistDev --- yt_dlp/extractor/_extractors.py | 1 + yt_dlp/extractor/orf.py | 64 +++++++++++++++++++++++++++++++++ 2 files changed, 65 insertions(+) diff --git a/yt_dlp/extractor/_extractors.py b/yt_dlp/extractor/_extractors.py index 7726fe359728..04318a716da5 100644 --- a/yt_dlp/extractor/_extractors.py +++ b/yt_dlp/extractor/_extractors.py @@ -1394,6 +1394,7 @@ from .orf import ( ORFTVthekIE, ORFFM4StoryIE, + ORFONIE, ORFRadioIE, ORFPodcastIE, ORFIPTVIE, diff --git a/yt_dlp/extractor/orf.py b/yt_dlp/extractor/orf.py index 9a48ae1b3e49..1b2a79a625a3 100644 --- a/yt_dlp/extractor/orf.py +++ b/yt_dlp/extractor/orf.py @@ -1,3 +1,4 @@ +import base64 import functools import re @@ -565,3 +566,66 @@ def _real_extract(self, url): }) return self.playlist_result(entries) + + +class ORFONIE(InfoExtractor): + IE_NAME = 'orf:on' + _VALID_URL = r'https?://on\.orf\.at/video/(?P\d{8})/(?P[\w-]+)' + _TESTS = [{ + 'url': 'https://on.orf.at/video/14210000/school-of-champions-48', + 'info_dict': { + 'id': '14210000', + 'ext': 'mp4', + 'duration': 2651.08, + 'thumbnail': 'https://api-tvthek.orf.at/assets/segments/0167/98/thumb_16697671_segments_highlight_teaser.jpeg', + 'title': 'School of Champions (4/8)', + 'description': 'md5:d09ad279fc2e8502611e7648484b6afd', + 'media_type': 'episode', + 'timestamp': 1706472362, + 'upload_date': '20240128', + } + }] + + def _extract_video(self, video_id, display_id): + encrypted_id = base64.b64encode(f'3dSlfek03nsLKdj4Jsd{video_id}'.encode()).decode() + api_json = self._download_json( + f'https://api-tvthek.orf.at/api/v4.3/public/episode/encrypted/{encrypted_id}', display_id) + + formats, subtitles = [], {} + for manifest_type in traverse_obj(api_json, ('sources', {dict.keys}, ...)): + for manifest_url in traverse_obj(api_json, ('sources', manifest_type, ..., 'src', {url_or_none})): + if manifest_type == 'hls': + fmts, subs = self._extract_m3u8_formats_and_subtitles( + manifest_url, display_id, fatal=False, m3u8_id='hls') + elif manifest_type == 'dash': + fmts, subs = self._extract_mpd_formats_and_subtitles( + manifest_url, display_id, fatal=False, mpd_id='dash') + else: + continue + formats.extend(fmts) + self._merge_subtitles(subs, target=subtitles) + + return { + 'id': video_id, + 'formats': formats, + 'subtitles': subtitles, + **traverse_obj(api_json, { + 'duration': ('duration_second', {float_or_none}), + 'title': (('title', 'headline'), {str}), + 'description': (('description', 'teaser_text'), {str}), + 'media_type': ('video_type', {str}), + }, get_all=False), + } + + def _real_extract(self, url): + video_id, display_id = self._match_valid_url(url).group('id', 'slug') + webpage = self._download_webpage(url, display_id) + + return { + 'id': video_id, + 'title': self._html_search_meta(['og:title', 'twitter:title'], webpage, default=None), + 'description': self._html_search_meta( + ['description', 'og:description', 'twitter:description'], webpage, default=None), + **self._search_json_ld(webpage, display_id, fatal=False), + **self._extract_video(video_id, display_id), + } From ffa017cfc5973b265c92248546fcf5020dc43eaf Mon Sep 17 00:00:00 2001 From: c-basalt <117849907+c-basalt@users.noreply.github.com> Date: Fri, 2 Feb 2024 16:08:29 -0500 Subject: [PATCH 03/10] [ie/BiliBiliSearch] Set cookie to fix extraction (#9119) Closes #5083 Authored by: c-basalt --- yt_dlp/extractor/bilibili.py | 30 ++++++++++++++++++++++++++++++ 1 file changed, 30 insertions(+) diff --git a/yt_dlp/extractor/bilibili.py b/yt_dlp/extractor/bilibili.py index cd7df69ef028..4ed9e2af7299 100644 --- a/yt_dlp/extractor/bilibili.py +++ b/yt_dlp/extractor/bilibili.py @@ -7,6 +7,7 @@ import re import time import urllib.parse +import uuid from .common import InfoExtractor, SearchInfoExtractor from ..dependencies import Cryptodome @@ -1464,8 +1465,37 @@ class BiliBiliSearchIE(SearchInfoExtractor): IE_DESC = 'Bilibili video search' _MAX_RESULTS = 100000 _SEARCH_KEY = 'bilisearch' + _TESTS = [{ + 'url': 'bilisearch3:靡烟 出道一年，我怎么还在等你单推的女人睡觉后开播啊', + 'playlist_count': 3, + 'info_dict': { + 'id': '靡烟 出道一年，我怎么还在等你单推的女人睡觉后开播啊', + 'title': '靡烟 出道一年，我怎么还在等你单推的女人睡觉后开播啊', + }, + 'playlist': [{ + 'info_dict': { + 'id': 'BV1n44y1Q7sc', + 'ext': 'mp4', + 'title': '“出道一年，我怎么还在等你单推的女人睡觉后开播啊？”【一分钟了解靡烟miya】', + 'timestamp': 1669889987, + 'upload_date': '20221201', + 'description': 'md5:43343c0973defff527b5a4b403b4abf9', + 'tags': list, + 'uploader': '靡烟miya', + 'duration': 123.156, + 'uploader_id': '1958703906', + 'comment_count': int, + 'view_count': int, + 'like_count': int, + 'thumbnail': r're:^https?://.*\.(jpg|jpeg|png)$', + '_old_archive_ids': ['bilibili 988222410_part1'], + }, + }], + }] def _search_results(self, query): + if not self._get_cookies('https://api.bilibili.com').get('buvid3'): + self._set_cookie('.bilibili.com', 'buvid3', f'{uuid.uuid4()}infoc') for page_num in itertools.count(1): videos = self._download_json( 'https://api.bilibili.com/x/web-interface/search/type', query, From 8e765755f7f4909e1b535e61b7376b2d66e1ba6a Mon Sep 17 00:00:00 2001 From: bashonly <88596187+bashonly@users.noreply.github.com> Date: Fri, 2 Feb 2024 15:15:04 -0600 Subject: [PATCH 04/10] [ie/vimeo] Fix API headers (#9125) Closes #9124 Authored by: bashonly --- yt_dlp/extractor/vimeo.py | 9 ++++++--- 1 file changed, 6 insertions(+), 3 deletions(-) diff --git a/yt_dlp/extractor/vimeo.py b/yt_dlp/extractor/vimeo.py index e5e8144bb10b..208e11184905 100644 --- a/yt_dlp/extractor/vimeo.py +++ b/yt_dlp/extractor/vimeo.py @@ -269,7 +269,7 @@ def _extract_original_format(self, url, video_id, unlisted_hash=None): 'https://vimeo.com/_rv/viewer', video_id, note='Downloading jwt token', fatal=False) or {} if not jwt_response.get('jwt'): return - headers = {'Authorization': 'jwt %s' % jwt_response['jwt']} + headers = {'Authorization': 'jwt %s' % jwt_response['jwt'], 'Accept': 'application/json'} original_response = self._download_json( f'https://api.vimeo.com/videos/{video_id}', video_id, headers=headers, fatal=False, expected_status=(403, 404)) or {} @@ -751,6 +751,7 @@ def _extract_from_api(self, video_id, unlisted_hash=None): video = self._download_json( api_url, video_id, headers={ 'Authorization': 'jwt ' + token, + 'Accept': 'application/json', }, query={ 'fields': 'config_url,created_time,description,license,metadata.connections.comments.total,metadata.connections.likes.total,release_time,stats.plays', }) @@ -785,7 +786,7 @@ def _try_album_password(self, url): jwt = viewer['jwt'] album = self._download_json( 'https://api.vimeo.com/albums/' + album_id, - album_id, headers={'Authorization': 'jwt ' + jwt}, + album_id, headers={'Authorization': 'jwt ' + jwt, 'Accept': 'application/json'}, query={'fields': 'description,name,privacy'}) if try_get(album, lambda x: x['privacy']['view']) == 'password': password = self.get_param('videopassword') @@ -1147,10 +1148,12 @@ def _fetch_page(self, album_id, authorization, hashed_pass, page): 'https://api.vimeo.com/albums/%s/videos' % album_id, album_id, 'Downloading page %d' % api_page, query=query, headers={ 'Authorization': 'jwt ' + authorization, + 'Accept': 'application/json', })['data'] except ExtractorError as e: if isinstance(e.cause, HTTPError) and e.cause.status == 400: return + raise for video in videos: link = video.get('link') if not link: @@ -1171,7 +1174,7 @@ def _real_extract(self, url): jwt = viewer['jwt'] album = self._download_json( 'https://api.vimeo.com/albums/' + album_id, - album_id, headers={'Authorization': 'jwt ' + jwt}, + album_id, headers={'Authorization': 'jwt ' + jwt, 'Accept': 'application/json'}, query={'fields': 'description,name,privacy'}) hashed_pass = None if try_get(album, lambda x: x['privacy']['view']) == 'password': From 4253e3b7f483127bd812bdac02466f4a5b47ff34 Mon Sep 17 00:00:00 2001 From: sepro <4618135+seproDev@users.noreply.github.com> Date: Sat, 3 Feb 2024 15:59:43 +0100 Subject: [PATCH 05/10] [ie/CCMA] Extract 1080p DASH formats (#9130) Closes #5755 Authored by: seproDev --- yt_dlp/extractor/ccma.py | 6 ++++++ 1 file changed, 6 insertions(+) diff --git a/yt_dlp/extractor/ccma.py b/yt_dlp/extractor/ccma.py index 88ff82f6e6b2..ab840f3016be 100644 --- a/yt_dlp/extractor/ccma.py +++ b/yt_dlp/extractor/ccma.py @@ -1,6 +1,7 @@ from .common import InfoExtractor from ..utils import ( clean_html, + determine_ext, int_or_none, parse_duration, parse_resolution, @@ -60,6 +61,7 @@ def _real_extract(self, url): 'http://dinamics.ccma.cat/pvideo/media.jsp', media_id, query={ 'media': media_type, 'idint': media_id, + 'format': 'dm', }) formats = [] @@ -69,6 +71,10 @@ def _real_extract(self, url): format_url = url_or_none(format_.get('file')) if not format_url: continue + if determine_ext(format_url) == 'mpd': + formats.extend(self._extract_mpd_formats( + format_url, media_id, mpd_id='dash', fatal=False)) + continue label = format_.get('label') f = parse_resolution(label) f.update({ From e3ce2b385ec1f03fac9d4210c57fda77134495fc Mon Sep 17 00:00:00 2001 From: YoshichikaAAA <154937389+YoshichikaAAA@users.noreply.github.com> Date: Sun, 4 Feb 2024 03:44:17 +0900 Subject: [PATCH 06/10] [ie/radiko] Extract more metadata (#9115) Authored by: YoshichikaAAA --- yt_dlp/extractor/radiko.py | 11 +++++++++++ 1 file changed, 11 insertions(+) diff --git a/yt_dlp/extractor/radiko.py b/yt_dlp/extractor/radiko.py index c363d9ba5f7b..2b6405999520 100644 --- a/yt_dlp/extractor/radiko.py +++ b/yt_dlp/extractor/radiko.py @@ -1,5 +1,6 @@ import base64 import random +import re import urllib.parse from .common import InfoExtractor @@ -11,6 +12,7 @@ unified_timestamp, update_url_query, ) +from ..utils.traversal import traverse_obj class RadikoBaseIE(InfoExtractor): @@ -159,6 +161,12 @@ def _extract_formats(self, video_id, station, is_onair, ft, cursor, auth_token, return formats + def _extract_performers(self, prog): + performers = traverse_obj(prog, ( + 'pfm/text()', ..., {lambda x: re.split(r'[/／、　,，]', x)}, ..., {str.strip})) + # TODO: change 'artist' fields to 'artists' and return traversal list instead of str + return ', '.join(performers) or None + class RadikoIE(RadikoBaseIE): _VALID_URL = r'https?://(?:www\.)?radiko\.jp/#!/ts/(?P[A-Z0-9-]+)/(?P\d+)' @@ -186,10 +194,12 @@ def _real_extract(self, url): return { 'id': video_id, 'title': try_call(lambda: prog.find('title').text), + 'artist': self._extract_performers(prog), 'description': clean_html(try_call(lambda: prog.find('info').text)), 'uploader': try_call(lambda: station_program.find('.//name').text), 'uploader_id': station, 'timestamp': vid_int, + 'duration': try_call(lambda: unified_timestamp(radio_end, False) - unified_timestamp(radio_begin, False)), 'is_live': True, 'formats': self._extract_formats( video_id=video_id, station=station, is_onair=False, @@ -243,6 +253,7 @@ def _real_extract(self, url): return { 'id': station, 'title': title, + 'artist': self._extract_performers(prog), 'description': description, 'uploader': station_name, 'uploader_id': station, From 96d0f8c1cb8aec250c5614bfde6b5fb95f10819b Mon Sep 17 00:00:00 2001 From: Michal Date: Mon, 5 Feb 2024 00:25:13 +0100 Subject: [PATCH 07/10] [ie/eporner] Extract AV1 formats (#9028) Authored by: michal-repo --- yt_dlp/extractor/eporner.py | 13 ++++++++++++- 1 file changed, 12 insertions(+), 1 deletion(-) diff --git a/yt_dlp/extractor/eporner.py b/yt_dlp/extractor/eporner.py index aee2dee58176..b18a76c7c177 100644 --- a/yt_dlp/extractor/eporner.py +++ b/yt_dlp/extractor/eporner.py @@ -1,8 +1,10 @@ from .common import InfoExtractor from ..utils import ( - encode_base_n, ExtractorError, + encode_base_n, + get_elements_by_class, int_or_none, + join_nonempty, merge_dicts, parse_duration, str_to_int, @@ -81,6 +83,7 @@ def calc_hash(s): sources = video['sources'] formats = [] + has_av1 = bool(get_elements_by_class('download-av1', webpage)) for kind, formats_dict in sources.items(): if not isinstance(formats_dict, dict): continue @@ -106,6 +109,14 @@ def calc_hash(s): 'height': height, 'fps': fps, }) + if has_av1: + formats.append({ + 'url': src.replace('.mp4', '-av1.mp4'), + 'format_id': join_nonempty('av1', format_id), + 'height': height, + 'fps': fps, + 'vcodec': 'av1', + }) json_ld = self._search_json_ld(webpage, display_id, default={}) From e439693f729daf6fb15457baea1bca10ef5da34d Mon Sep 17 00:00:00 2001 From: c-basalt <117849907+c-basalt@users.noreply.github.com> Date: Sun, 4 Feb 2024 18:28:45 -0500 Subject: [PATCH 08/10] [ie/bilibili] Support `--no-playlist` (#9139) Addresses #8499 Authored by: c-basalt --- yt_dlp/extractor/bilibili.py | 25 +++++++++++++++++++++++++ 1 file changed, 25 insertions(+) diff --git a/yt_dlp/extractor/bilibili.py b/yt_dlp/extractor/bilibili.py index 4ed9e2af7299..c138bde3a507 100644 --- a/yt_dlp/extractor/bilibili.py +++ b/yt_dlp/extractor/bilibili.py @@ -1305,6 +1305,26 @@ class BilibiliPlaylistIE(BilibiliSpaceListBaseIE): 'upload_date': '20211127', }, 'playlist_mincount': 513, + }, { + 'url': 'https://www.bilibili.com/list/1958703906?sid=547718&oid=687146339&bvid=BV1DU4y1r7tz', + 'info_dict': { + 'id': 'BV1DU4y1r7tz', + 'ext': 'mp4', + 'title': '【直播回放】8.20晚9:30 3d发布喵 2022年8月20日21点场', + 'upload_date': '20220820', + 'description': '', + 'timestamp': 1661016330, + 'uploader_id': '1958703906', + 'uploader': '靡烟miya', + 'thumbnail': r're:^https?://.*\.(jpg|jpeg|png)$', + 'duration': 9552.903, + 'tags': list, + 'comment_count': int, + 'view_count': int, + 'like_count': int, + '_old_archive_ids': ['bilibili 687146339_part1'], + }, + 'params': {'noplaylist': True}, }, { 'url': 'https://www.bilibili.com/medialist/play/1958703906?business=space_series&business_id=547718&desc=1', 'info_dict': { @@ -1356,6 +1376,11 @@ def _extract_medialist(self, query, list_id): def _real_extract(self, url): list_id = self._match_id(url) + + bvid = traverse_obj(parse_qs(url), ('bvid', 0)) + if not self._yes_playlist(list_id, bvid): + return self.url_result(f'https://www.bilibili.com/video/{bvid}', BiliBiliIE) + webpage = self._download_webpage(url, list_id) initial_state = self._search_json(r'window\.__INITIAL_STATE__\s*=', webpage, 'initial state', list_id) if traverse_obj(initial_state, ('error', 'code', {int_or_none})) != 200: From 07256b9fee23960799024b95d5972abc7174aa81 Mon Sep 17 00:00:00 2001 From: SirElderling <148036781+SirElderling@users.noreply.github.com> Date: Mon, 5 Feb 2024 00:35:52 +0000 Subject: [PATCH 09/10] [ie/nytimes] Overhaul extractors (#9075) Closes #2899, Closes #8605 Authored by: SirElderling --- yt_dlp/extractor/_extractors.py | 1 + yt_dlp/extractor/nytimes.py | 450 +++++++++++++++++++++----------- 2 files changed, 303 insertions(+), 148 deletions(-) diff --git a/yt_dlp/extractor/_extractors.py b/yt_dlp/extractor/_extractors.py index 04318a716da5..36335286c3a0 100644 --- a/yt_dlp/extractor/_extractors.py +++ b/yt_dlp/extractor/_extractors.py @@ -1352,6 +1352,7 @@ NYTimesIE, NYTimesArticleIE, NYTimesCookingIE, + NYTimesCookingRecipeIE, ) from .nuvid import NuvidIE from .nzherald import NZHeraldIE diff --git a/yt_dlp/extractor/nytimes.py b/yt_dlp/extractor/nytimes.py index 2e21edbb4120..354eb02c3485 100644 --- a/yt_dlp/extractor/nytimes.py +++ b/yt_dlp/extractor/nytimes.py @@ -1,50 +1,92 @@ -import hmac -import hashlib -import base64 +import json +import uuid from .common import InfoExtractor from ..utils import ( + ExtractorError, + clean_html, determine_ext, + extract_attributes, float_or_none, + get_elements_html_by_class, int_or_none, - js_to_json, + merge_dicts, mimetype2ext, parse_iso8601, + remove_end, remove_start, + str_or_none, + traverse_obj, + url_or_none, ) class NYTimesBaseIE(InfoExtractor): - _SECRET = b'pX(2MbU2);4N{7J8)>YwKRJ+/pQ3JkiU2Q^V>mFYv6g6gYvt6v' - - def _extract_video_from_id(self, video_id): - # Authorization generation algorithm is reverse engineered from `signer` in - # http://graphics8.nytimes.com/video/vhs/vhs-2.x.min.js - path = '/svc/video/api/v3/video/' + video_id - hm = hmac.new(self._SECRET, (path + ':vhs').encode(), hashlib.sha512).hexdigest() - video_data = self._download_json('http://www.nytimes.com' + path, video_id, 'Downloading video JSON', headers={ - 'Authorization': 'NYTV ' + base64.b64encode(hm.encode()).decode(), - 'X-NYTV': 'vhs', - }, fatal=False) - if not video_data: - video_data = self._download_json( - 'http://www.nytimes.com/svc/video/api/v2/video/' + video_id, - video_id, 'Downloading video JSON') - - title = video_data['headline'] - - def get_file_size(file_size): - if isinstance(file_size, int): - return file_size - elif isinstance(file_size, dict): - return int(file_size.get('value', 0)) - else: - return None - + _DNS_NAMESPACE = uuid.UUID('36dd619a-56dc-595b-9e09-37f4152c7b5d') + _TOKEN = 'MIIBIjANBgkqhkiG9w0BAQEFAAOCAQ8AMIIBCgKCAQEAuNIzKBOFB77aT/jN/FQ+/QVKWq5V1ka1AYmCR9hstz1pGNPH5ajOU9gAqta0T89iPnhjwla+3oec/Z3kGjxbpv6miQXufHFq3u2RC6HyU458cLat5kVPSOQCe3VVB5NRpOlRuwKHqn0txfxnwSSj8mqzstR997d3gKB//RO9zE16y3PoWlDQXkASngNJEWvL19iob/xwAkfEWCjyRILWFY0JYX3AvLMSbq7wsqOCE5srJpo7rRU32zsByhsp1D5W9OYqqwDmflsgCEQy2vqTsJjrJohuNg+urMXNNZ7Y3naMoqttsGDrWVxtPBafKMI8pM2ReNZBbGQsQXRzQNo7+QIDAQAB' + _GRAPHQL_API = 'https://samizdat-graphql.nytimes.com/graphql/v2' + _GRAPHQL_QUERY = '''query VideoQuery($id: String!) { + video(id: $id) { + ... on Video { + bylines { + renderedRepresentation + } + duration + promotionalHeadline + promotionalMedia { + ... on Image { + crops { + name + renditions { + name + width + height + url + } + } + } + } + renditions { + type + width + height + url + bitrate + } + summary + } + } +}''' + + def _call_api(self, media_id): + # reference: `id-to-uri.js` + video_uuid = uuid.uuid5(self._DNS_NAMESPACE, 'video') + media_uuid = uuid.uuid5(video_uuid, media_id) + + return traverse_obj(self._download_json( + self._GRAPHQL_API, media_id, 'Downloading JSON from GraphQL API', data=json.dumps({ + 'query': self._GRAPHQL_QUERY, + 'variables': {'id': f'nyt://video/{media_uuid}'}, + }, separators=(',', ':')).encode(), headers={ + 'Content-Type': 'application/json', + 'Nyt-App-Type': 'vhs', + 'Nyt-App-Version': 'v3.52.21', + 'Nyt-Token': self._TOKEN, + 'Origin': 'https://nytimes.com', + }, fatal=False), ('data', 'video', {dict})) or {} + + def _extract_thumbnails(self, thumbs): + return traverse_obj(thumbs, (lambda _, v: url_or_none(v['url']), { + 'url': 'url', + 'width': ('width', {int_or_none}), + 'height': ('height', {int_or_none}), + }), default=None) + + def _extract_formats_and_subtitles(self, video_id, content_media_json): urls = [] formats = [] subtitles = {} - for video in video_data.get('renditions', []): + for video in traverse_obj(content_media_json, ('renditions', ..., {dict})): video_url = video.get('url') format_id = video.get('type') if not video_url or format_id == 'thumbs' or video_url in urls: @@ -56,11 +98,9 @@ def get_file_size(file_size): video_url, video_id, 'mp4', 'm3u8_native', m3u8_id=format_id or 'hls', fatal=False) formats.extend(m3u8_fmts) - subtitles = self._merge_subtitles(subtitles, m3u8_subs) + self._merge_subtitles(m3u8_subs, target=subtitles) elif ext == 'mpd': - continue - # formats.extend(self._extract_mpd_formats( - # video_url, video_id, format_id or 'dash', fatal=False)) + continue # all mpd urls give 404 errors else: formats.append({ 'url': video_url, @@ -68,55 +108,49 @@ def get_file_size(file_size): 'vcodec': video.get('videoencoding') or video.get('video_codec'), 'width': int_or_none(video.get('width')), 'height': int_or_none(video.get('height')), - 'filesize': get_file_size(video.get('file_size') or video.get('fileSize')), + 'filesize': traverse_obj(video, ( + ('file_size', 'fileSize'), (None, ('value')), {int_or_none}), get_all=False), 'tbr': int_or_none(video.get('bitrate'), 1000) or None, 'ext': ext, }) - thumbnails = [] - for image in video_data.get('images', []): - image_url = image.get('url') - if not image_url: - continue - thumbnails.append({ - 'url': 'http://www.nytimes.com/' + image_url, - 'width': int_or_none(image.get('width')), - 'height': int_or_none(image.get('height')), - }) + return formats, subtitles - publication_date = video_data.get('publication_date') - timestamp = parse_iso8601(publication_date[:-8]) if publication_date else None + def _extract_video(self, media_id): + data = self._call_api(media_id) + formats, subtitles = self._extract_formats_and_subtitles(media_id, data) return { - 'id': video_id, - 'title': title, - 'description': video_data.get('summary'), - 'timestamp': timestamp, - 'uploader': video_data.get('byline'), - 'duration': float_or_none(video_data.get('duration'), 1000), + 'id': media_id, + 'title': data.get('promotionalHeadline'), + 'description': data.get('summary'), + 'duration': float_or_none(data.get('duration'), scale=1000), + 'creator': ', '.join(traverse_obj(data, ( # TODO: change to 'creators' + 'bylines', ..., 'renderedRepresentation', {lambda x: remove_start(x, 'By ')}))), 'formats': formats, 'subtitles': subtitles, - 'thumbnails': thumbnails, + 'thumbnails': self._extract_thumbnails( + traverse_obj(data, ('promotionalMedia', 'crops', ..., 'renditions', ...))), } class NYTimesIE(NYTimesBaseIE): _VALID_URL = r'https?://(?:(?:www\.)?nytimes\.com/video/(?:[^/]+/)+?|graphics8\.nytimes\.com/bcvideo/\d+(?:\.\d+)?/iframe/embed\.html\?videoId=)(?P\d+)' _EMBED_REGEX = [r']+src=(["\'])(?P(?:https?:)?//graphics8\.nytimes\.com/bcvideo/[^/]+/iframe/embed\.html.+?)\1>'] - _TESTS = [{ 'url': 'http://www.nytimes.com/video/opinion/100000002847155/verbatim-what-is-a-photocopier.html?playlistId=100000001150263', - 'md5': 'd665342765db043f7e225cff19df0f2d', + 'md5': 'a553aa344014e3723d33893d89d4defc', 'info_dict': { 'id': '100000002847155', - 'ext': 'mov', + 'ext': 'mp4', 'title': 'Verbatim: What Is a Photocopier?', 'description': 'md5:93603dada88ddbda9395632fdc5da260', - 'timestamp': 1398631707, - 'upload_date': '20140427', - 'uploader': 'Brett Weiner', + 'timestamp': 1398631707, # FIXME + 'upload_date': '20140427', # FIXME + 'creator': 'Brett Weiner', + 'thumbnail': r're:https?://\w+\.nyt.com/images/.+\.jpg', 'duration': 419, - } + }, }, { 'url': 'http://www.nytimes.com/video/travel/100000003550828/36-hours-in-dubai.html', 'only_matching': True, @@ -125,138 +159,258 @@ class NYTimesIE(NYTimesBaseIE): def _real_extract(self, url): video_id = self._match_id(url) - return self._extract_video_from_id(video_id) + return self._extract_video(video_id) class NYTimesArticleIE(NYTimesBaseIE): - _VALID_URL = r'https?://(?:www\.)?nytimes\.com/(.(?[^.]+)(?:\.html)?' + _VALID_URL = r'https?://(?:www\.)?nytimes\.com/\d{4}/\d{2}/\d{2}/(?!books|podcasts)[^/?#]+/(?:\w+/)?(?P[^./?#]+)(?:\.html)?' _TESTS = [{ 'url': 'http://www.nytimes.com/2015/04/14/business/owner-of-gravity-payments-a-credit-card-processor-is-setting-a-new-minimum-wage-70000-a-year.html?_r=0', - 'md5': 'e2076d58b4da18e6a001d53fd56db3c9', + 'md5': '3eb5ddb1d6f86254fe4f233826778737', 'info_dict': { 'id': '100000003628438', - 'ext': 'mov', - 'title': 'New Minimum Wage: $70,000 a Year', - 'description': 'Dan Price, C.E.O. of Gravity Payments, surprised his 120-person staff by announcing that he planned over the next three years to raise the salary of every employee to $70,000 a year.', - 'timestamp': 1429033037, + 'ext': 'mp4', + 'title': 'One Company’s New Minimum Wage: $70,000 a Year', + 'description': 'md5:89ba9ab67ca767bb92bf823d1f138433', + 'timestamp': 1429047468, 'upload_date': '20150414', 'uploader': 'Matthew Williams', - } + 'creator': 'Patricia Cohen', + 'thumbnail': r're:https?://\w+\.nyt.com/images/.*\.jpg', + 'duration': 119.0, + }, }, { - 'url': 'http://www.nytimes.com/2016/10/14/podcasts/revelations-from-the-final-weeks.html', - 'md5': 'e0d52040cafb07662acf3c9132db3575', + # article with audio and no video + 'url': 'https://www.nytimes.com/2023/09/29/health/mosquitoes-genetic-engineering.html', + 'md5': '2365b3555c8aa7f4dd34ca735ad02e6a', 'info_dict': { - 'id': '100000004709062', - 'title': 'The Run-Up: ‘He Was Like an Octopus’', + 'id': '100000009110381', 'ext': 'mp3', - 'description': 'md5:fb5c6b93b12efc51649b4847fe066ee4', - 'series': 'The Run-Up', - 'episode': '‘He Was Like an Octopus’', - 'episode_number': 20, - 'duration': 2130, - } + 'title': 'The Gamble: Can Genetically Modified Mosquitoes End Disease?', + 'description': 'md5:9ff8b47acbaf7f3ca8c732f5c815be2e', + 'timestamp': 1695960700, + 'upload_date': '20230929', + 'creator': 'Stephanie Nolen, Natalija Gormalova', + 'thumbnail': r're:https?://\w+\.nyt.com/images/.*\.jpg', + 'duration': 1322, + }, }, { - 'url': 'http://www.nytimes.com/2016/10/16/books/review/inside-the-new-york-times-book-review-the-rise-of-hitler.html', + 'url': 'https://www.nytimes.com/2023/11/29/business/dealbook/kamala-harris-biden-voters.html', + 'md5': '3eb5ddb1d6f86254fe4f233826778737', 'info_dict': { - 'id': '100000004709479', - 'title': 'The Rise of Hitler', - 'ext': 'mp3', - 'description': 'md5:bce877fd9e3444990cb141875fab0028', - 'creator': 'Pamela Paul', - 'duration': 3475, + 'id': '100000009202270', + 'ext': 'mp4', + 'title': 'Kamala Harris Defends Biden Policies, but Says ‘More Work’ Needed to Reach Voters', + 'description': 'md5:de4212a7e19bb89e4fb14210ca915f1f', + 'timestamp': 1701290997, + 'upload_date': '20231129', + 'uploader': 'By The New York Times', + 'creator': 'Katie Rogers', + 'thumbnail': r're:https?://\w+\.nyt.com/images/.*\.jpg', + 'duration': 97.631, }, 'params': { - 'skip_download': True, + 'skip_download': 'm3u8', + }, + }, { + # multiple videos in the same article + 'url': 'https://www.nytimes.com/2023/12/02/business/air-traffic-controllers-safety.html', + 'info_dict': { + 'id': 'air-traffic-controllers-safety', + 'title': 'Drunk and Asleep on the Job: Air Traffic Controllers Pushed to the Brink', + 'description': 'md5:549e5a5e935bf7d048be53ba3d2c863d', + 'upload_date': '20231202', + 'creator': 'Emily Steel, Sydney Ember', + 'timestamp': 1701511264, }, + 'playlist_count': 3, }, { - 'url': 'http://www.nytimes.com/news/minute/2014/03/17/times-minute-whats-next-in-crimea/?_php=true&_type=blogs&_php=true&_type=blogs&_r=1', + 'url': 'https://www.nytimes.com/2023/12/02/business/media/netflix-squid-game-challenge.html', 'only_matching': True, }] - def _extract_podcast_from_json(self, json, page_id, webpage): - podcast_audio = self._parse_json( - json, page_id, transform_source=js_to_json) + def _extract_content_from_block(self, block): + details = traverse_obj(block, { + 'id': ('sourceId', {str}), + 'uploader': ('bylines', ..., 'renderedRepresentation', {str}), + 'duration': (None, (('duration', {lambda x: float_or_none(x, scale=1000)}), ('length', {int_or_none}))), + 'timestamp': ('firstPublished', {parse_iso8601}), + 'series': ('podcastSeries', {str}), + }, get_all=False) + + formats, subtitles = self._extract_formats_and_subtitles(details.get('id'), block) + # audio articles will have an url and no formats + url = traverse_obj(block, ('fileUrl', {url_or_none})) + if not formats and url: + formats.append({'url': url, 'vcodec': 'none'}) - audio_data = podcast_audio['data'] - track = audio_data['track'] - - episode_title = track['title'] - video_url = track['source'] + return { + **details, + 'thumbnails': self._extract_thumbnails(traverse_obj( + block, ('promotionalMedia', 'crops', ..., 'renditions', ...))), + 'formats': formats, + 'subtitles': subtitles + } - description = track.get('description') or self._html_search_meta( - ['og:description', 'twitter:description'], webpage) + def _real_extract(self, url): + page_id = self._match_id(url) + webpage = self._download_webpage(url, page_id) + art_json = self._search_json( + r'window\.__preloadedData\s*=', webpage, 'media details', page_id, + transform_source=lambda x: x.replace('undefined', 'null'))['initialData']['data']['article'] + + blocks = traverse_obj(art_json, ( + 'sprinkledBody', 'content', ..., ('ledeMedia', None), + lambda _, v: v['__typename'] in ('Video', 'Audio'))) + if not blocks: + raise ExtractorError('Unable to extract any media blocks from webpage') + + common_info = { + 'title': remove_end(self._html_extract_title(webpage), ' - The New York Times'), + 'description': traverse_obj(art_json, ( + 'sprinkledBody', 'content', ..., 'summary', 'content', ..., 'text', {str}), + get_all=False) or self._html_search_meta(['og:description', 'twitter:description'], webpage), + 'timestamp': traverse_obj(art_json, ('firstPublished', {parse_iso8601})), + 'creator': ', '.join( + traverse_obj(art_json, ('bylines', ..., 'creators', ..., 'displayName'))), # TODO: change to 'creators' (list) + 'thumbnails': self._extract_thumbnails(traverse_obj( + art_json, ('promotionalMedia', 'assetCrops', ..., 'renditions', ...))), + } - podcast_title = audio_data.get('podcast', {}).get('title') - title = ('%s: %s' % (podcast_title, episode_title) - if podcast_title else episode_title) + entries = [] + for block in blocks: + entries.append(merge_dicts(self._extract_content_from_block(block), common_info)) - episode = audio_data.get('podcast', {}).get('episode') or '' - episode_number = int_or_none(self._search_regex( - r'[Ee]pisode\s+(\d+)', episode, 'episode number', default=None)) + if len(entries) > 1: + return self.playlist_result(entries, page_id, **common_info) return { - 'id': remove_start(podcast_audio.get('target'), 'FT') or page_id, - 'url': video_url, - 'title': title, - 'description': description, - 'creator': track.get('credit'), - 'series': podcast_title, - 'episode': episode_title, - 'episode_number': episode_number, - 'duration': int_or_none(track.get('duration')), + 'id': page_id, + **entries[0], } + +class NYTimesCookingIE(NYTimesBaseIE): + IE_NAME = 'NYTimesCookingGuide' + _VALID_URL = r'https?://cooking\.nytimes\.com/guides/(?P[\w-]+)' + _TESTS = [{ + 'url': 'https://cooking.nytimes.com/guides/13-how-to-cook-a-turkey', + 'info_dict': { + 'id': '13-how-to-cook-a-turkey', + 'title': 'How to Cook a Turkey', + 'description': 'md5:726cfd3f9b161bdf5c279879e8050ca0', + }, + 'playlist_count': 2, + }, { + # single video example + 'url': 'https://cooking.nytimes.com/guides/50-how-to-make-mac-and-cheese', + 'md5': '64415805fe0b8640fce6b0b9def5989a', + 'info_dict': { + 'id': '100000005835845', + 'ext': 'mp4', + 'title': 'How to Make Mac and Cheese', + 'description': 'md5:b8f2f33ec1fb7523b21367147c9594f1', + 'duration': 9.51, + 'creator': 'Alison Roman', + 'thumbnail': r're:https?://\w+\.nyt.com/images/.*\.jpg', + }, + }, { + 'url': 'https://cooking.nytimes.com/guides/20-how-to-frost-a-cake', + 'md5': '64415805fe0b8640fce6b0b9def5989a', + 'info_dict': { + 'id': '20-how-to-frost-a-cake', + 'title': 'How to Frost a Cake', + 'description': 'md5:a31fe3b98a8ce7b98aae097730c269cd', + }, + 'playlist_count': 8, + }] + def _real_extract(self, url): page_id = self._match_id(url) - webpage = self._download_webpage(url, page_id) + title = self._html_search_meta(['og:title', 'twitter:title'], webpage) + description = self._html_search_meta(['og:description', 'twitter:description'], webpage) - video_id = self._search_regex( - r'data-videoid=["\'](\d+)', webpage, 'video id', - default=None, fatal=False) - if video_id is not None: - return self._extract_video_from_id(video_id) + lead_video_id = self._search_regex( + r'data-video-player-id="(\d+)">', webpage, 'lead video') + media_ids = traverse_obj( + get_elements_html_by_class('video-item', webpage), (..., {extract_attributes}, 'data-video-id')) - podcast_data = self._search_regex( - (r'NYTD\.FlexTypes\.push\s*\(\s*({.+?})\s*\)\s*;\s*([^<]+)

', webpage, 'author', default=None), + } -class NYTimesCookingIE(NYTimesBaseIE): - _VALID_URL = r'https?://cooking\.nytimes\.com/(?:guid|recip)es/(?P\d+)' + +class NYTimesCookingRecipeIE(InfoExtractor): + _VALID_URL = r'https?://cooking\.nytimes\.com/recipes/(?P\d+)' _TESTS = [{ 'url': 'https://cooking.nytimes.com/recipes/1017817-cranberry-curd-tart', - 'md5': 'dab81fa2eaeb3f9ed47498bdcfcdc1d3', + 'md5': '579e83bbe8e61e9de67f80edba8a78a8', 'info_dict': { - 'id': '100000004756089', - 'ext': 'mov', - 'timestamp': 1479383008, - 'uploader': 'By SHAW LASH, ADAM SAEWITZ and JAMES HERRON', - 'title': 'Cranberry Tart', - 'upload_date': '20161117', - 'description': 'If you are a fan of lemon curd or the classic French tarte au citron, you will love this cranberry version.', + 'id': '1017817', + 'ext': 'mp4', + 'title': 'Cranberry Curd Tart', + 'description': 'md5:ad77a3fc321db636256d4343c5742152', + 'timestamp': 1447804800, + 'upload_date': '20151118', + 'creator': 'David Tanis', + 'thumbnail': r're:https?://\w+\.nyt.com/images/.*\.jpg', }, }, { - 'url': 'https://cooking.nytimes.com/guides/13-how-to-cook-a-turkey', - 'md5': '4b2e8c70530a89b8d905a2b572316eb8', + 'url': 'https://cooking.nytimes.com/recipes/1024781-neapolitan-checkerboard-cookies', + 'md5': '58df35998241dcf0620e99e646331b42', 'info_dict': { - 'id': '100000003951728', - 'ext': 'mov', - 'timestamp': 1445509539, - 'description': 'Turkey guide', - 'upload_date': '20151022', - 'title': 'Turkey', - } + 'id': '1024781', + 'ext': 'mp4', + 'title': 'Neapolitan Checkerboard Cookies', + 'description': 'md5:ba12394c585ababea951cb6d2fcc6631', + 'timestamp': 1701302400, + 'upload_date': '20231130', + 'creator': 'Sue Li', + 'thumbnail': r're:https?://\w+\.nyt.com/images/.*\.jpg', + }, + }, { + 'url': 'https://cooking.nytimes.com/recipes/1019516-overnight-oats', + 'md5': '2fe7965a3adc899913b8e25ada360823', + 'info_dict': { + 'id': '1019516', + 'ext': 'mp4', + 'timestamp': 1546387200, + 'description': 'md5:8856ce10239161bd2596ac335b9f9bfb', + 'upload_date': '20190102', + 'title': 'Overnight Oats', + 'creator': 'Genevieve Ko', + 'thumbnail': r're:https?://\w+\.nyt.com/images/.*\.jpg', + }, }] def _real_extract(self, url): page_id = self._match_id(url) - webpage = self._download_webpage(url, page_id) + recipe_data = self._search_nextjs_data(webpage, page_id)['props']['pageProps']['recipe'] - video_id = self._search_regex( - r'data-video-id=["\'](\d+)', webpage, 'video id') + formats, subtitles = self._extract_m3u8_formats_and_subtitles( + recipe_data['videoSrc'], page_id, 'mp4', m3u8_id='hls') - return self._extract_video_from_id(video_id) + return { + **traverse_obj(recipe_data, { + 'id': ('id', {str_or_none}), + 'title': ('title', {str}), + 'description': ('topnote', {clean_html}), + 'timestamp': ('publishedAt', {int_or_none}), + 'creator': ('contentAttribution', 'cardByline', {str}), + }), + 'formats': formats, + 'subtitles': subtitles, + 'thumbnails': [{'url': thumb_url} for thumb_url in traverse_obj( + recipe_data, ('image', 'crops', 'recipe', ..., {url_or_none}))], + } From acaf806c15f0a802ba286c23af02a10cf4bd4731 Mon Sep 17 00:00:00 2001 From: DmitryScaletta Date: Mon, 5 Feb 2024 05:17:39 +0300 Subject: [PATCH 10/10] [ie/nuum] Add extractors (#8868) Authored by: DmitryScaletta, seproDev Co-authored-by: sepro <4618135+seproDev@users.noreply.github.com> --- yt_dlp/extractor/_extractors.py | 10 +- yt_dlp/extractor/nuum.py | 199 ++++++++++++++++++++++++++++++++ yt_dlp/extractor/wasdtv.py | 159 ------------------------- 3 files changed, 204 insertions(+), 164 deletions(-) create mode 100644 yt_dlp/extractor/nuum.py delete mode 100644 yt_dlp/extractor/wasdtv.py diff --git a/yt_dlp/extractor/_extractors.py b/yt_dlp/extractor/_extractors.py index 36335286c3a0..e7dd34c77bda 100644 --- a/yt_dlp/extractor/_extractors.py +++ b/yt_dlp/extractor/_extractors.py @@ -1354,6 +1354,11 @@ NYTimesCookingIE, NYTimesCookingRecipeIE, ) +from .nuum import ( + NuumLiveIE, + NuumTabIE, + NuumMediaIE, +) from .nuvid import NuvidIE from .nzherald import NZHeraldIE from .nzonscreen import NZOnScreenIE @@ -2315,11 +2320,6 @@ WashingtonPostIE, WashingtonPostArticleIE, ) -from .wasdtv import ( - WASDTVStreamIE, - WASDTVRecordIE, - WASDTVClipIE, -) from .wat import WatIE from .wdr import ( WDRIE, diff --git a/yt_dlp/extractor/nuum.py b/yt_dlp/extractor/nuum.py new file mode 100644 index 000000000000..3db663ded0dd --- /dev/null +++ b/yt_dlp/extractor/nuum.py @@ -0,0 +1,199 @@ +import functools + +from .common import InfoExtractor +from ..utils import ( + ExtractorError, + OnDemandPagedList, + UserNotLive, + filter_dict, + int_or_none, + parse_iso8601, + str_or_none, + url_or_none, +) +from ..utils.traversal import traverse_obj + + +class NuumBaseIE(InfoExtractor): + def _call_api(self, path, video_id, description, query={}): + response = self._download_json( + f'https://nuum.ru/api/v2/{path}', video_id, query=query, + note=f'Downloading {description} metadata', + errnote=f'Unable to download {description} metadata') + if error := response.get('error'): + raise ExtractorError(f'API returned error: {error!r}') + return response['result'] + + def _get_channel_info(self, channel_name): + return self._call_api( + 'broadcasts/public', video_id=channel_name, description='channel', + query={ + 'with_extra': 'true', + 'channel_name': channel_name, + 'with_deleted': 'true', + }) + + def _parse_video_data(self, container, extract_formats=True): + stream = traverse_obj(container, ('media_container_streams', 0, {dict})) or {} + media = traverse_obj(stream, ('stream_media', 0, {dict})) or {} + media_url = traverse_obj(media, ( + 'media_meta', ('media_archive_url', 'media_url'), {url_or_none}), get_all=False) + + video_id = str(container['media_container_id']) + is_live = media.get('media_status') == 'RUNNING' + + formats, subtitles = None, None + if extract_formats: + formats, subtitles = self._extract_m3u8_formats_and_subtitles( + media_url, video_id, 'mp4', live=is_live) + + return filter_dict({ + 'id': video_id, + 'is_live': is_live, + 'formats': formats, + 'subtitles': subtitles, + **traverse_obj(container, { + 'title': ('media_container_name', {str}), + 'description': ('media_container_description', {str}), + 'timestamp': ('created_at', {parse_iso8601}), + 'channel': ('media_container_channel', 'channel_name', {str}), + 'channel_id': ('media_container_channel', 'channel_id', {str_or_none}), + }), + **traverse_obj(stream, { + 'view_count': ('stream_total_viewers', {int_or_none}), + 'concurrent_view_count': ('stream_current_viewers', {int_or_none}), + }), + **traverse_obj(media, { + 'duration': ('media_duration', {int_or_none}), + 'thumbnail': ('media_meta', ('media_preview_archive_url', 'media_preview_url'), {url_or_none}), + }, get_all=False), + }) + + +class NuumMediaIE(NuumBaseIE): + IE_NAME = 'nuum:media' + _VALID_URL = r'https?://nuum\.ru/(?:streams|videos|clips)/(?P[\d]+)' + _TESTS = [{ + 'url': 'https://nuum.ru/streams/1592713-7-days-to-die', + 'only_matching': True, + }, { + 'url': 'https://nuum.ru/videos/1567547-toxi-hurtz', + 'md5': 'f1d9118a30403e32b702a204eb03aca3', + 'info_dict': { + 'id': '1567547', + 'ext': 'mp4', + 'title': 'Toxi$ - Hurtz', + 'description': '', + 'timestamp': 1702631651, + 'upload_date': '20231215', + 'thumbnail': r're:^https?://.+\.jpg', + 'view_count': int, + 'concurrent_view_count': int, + 'channel_id': '6911', + 'channel': 'toxis', + 'duration': 116, + }, + }, { + 'url': 'https://nuum.ru/clips/1552564-pro-misu', + 'md5': 'b248ae1565b1e55433188f11beeb0ca1', + 'info_dict': { + 'id': '1552564', + 'ext': 'mp4', + 'title': 'Про Мису 🙃', + 'timestamp': 1701971828, + 'upload_date': '20231207', + 'thumbnail': r're:^https?://.+\.jpg', + 'view_count': int, + 'concurrent_view_count': int, + 'channel_id': '3320', + 'channel': 'Misalelik', + 'duration': 41, + }, + }] + + def _real_extract(self, url): + video_id = self._match_id(url) + video_data = self._call_api(f'media-containers/{video_id}', video_id, 'media') + + return self._parse_video_data(video_data) + + +class NuumLiveIE(NuumBaseIE): + IE_NAME = 'nuum:live' + _VALID_URL = r'https?://nuum\.ru/channel/(?P[^/#?]+)/?(?:$|[#?])' + _TESTS = [{ + 'url': 'https://nuum.ru/channel/mts_live', + 'only_matching': True, + }] + + def _real_extract(self, url): + channel = self._match_id(url) + channel_info = self._get_channel_info(channel) + if traverse_obj(channel_info, ('channel', 'channel_is_live')) is False: + raise UserNotLive(video_id=channel) + + info = self._parse_video_data(channel_info['media_container']) + return { + 'webpage_url': f'https://nuum.ru/streams/{info["id"]}', + 'extractor_key': NuumMediaIE.ie_key(), + 'extractor': NuumMediaIE.IE_NAME, + **info, + } + + +class NuumTabIE(NuumBaseIE): + IE_NAME = 'nuum:tab' + _VALID_URL = r'https?://nuum\.ru/channel/(?P[^/#?]+)/(?Pstreams|videos|clips)' + _TESTS = [{ + 'url': 'https://nuum.ru/channel/dankon_/clips', + 'info_dict': { + 'id': 'dankon__clips', + 'title': 'Dankon_', + }, + 'playlist_mincount': 29, + }, { + 'url': 'https://nuum.ru/channel/dankon_/videos', + 'info_dict': { + 'id': 'dankon__videos', + 'title': 'Dankon_', + }, + 'playlist_mincount': 2, + }, { + 'url': 'https://nuum.ru/channel/dankon_/streams', + 'info_dict': { + 'id': 'dankon__streams', + 'title': 'Dankon_', + }, + 'playlist_mincount': 1, + }] + + _PAGE_SIZE = 50 + + def _fetch_page(self, channel_id, tab_type, tab_id, page): + CONTAINER_TYPES = { + 'clips': ['SHORT_VIDEO', 'REVIEW_VIDEO'], + 'videos': ['LONG_VIDEO'], + 'streams': ['SINGLE'], + } + + media_containers = self._call_api( + 'media-containers', video_id=tab_id, description=f'{tab_type} tab page {page + 1}', + query={ + 'limit': self._PAGE_SIZE, + 'offset': page * self._PAGE_SIZE, + 'channel_id': channel_id, + 'media_container_status': 'STOPPED', + 'media_container_type': CONTAINER_TYPES[tab_type], + }) + for container in traverse_obj(media_containers, (..., {dict})): + metadata = self._parse_video_data(container, extract_formats=False) + yield self.url_result(f'https://nuum.ru/videos/{metadata["id"]}', NuumMediaIE, **metadata) + + def _real_extract(self, url): + channel_name, tab_type = self._match_valid_url(url).group('id', 'type') + tab_id = f'{channel_name}_{tab_type}' + channel_data = self._get_channel_info(channel_name)['channel'] + + return self.playlist_result(OnDemandPagedList(functools.partial( + self._fetch_page, channel_data['channel_id'], tab_type, tab_id), self._PAGE_SIZE), + playlist_id=tab_id, playlist_title=channel_data.get('channel_name')) diff --git a/yt_dlp/extractor/wasdtv.py b/yt_dlp/extractor/wasdtv.py deleted file mode 100644 index f57c619b5f88..000000000000 --- a/yt_dlp/extractor/wasdtv.py +++ /dev/null @@ -1,159 +0,0 @@ -from .common import InfoExtractor -from ..utils import ( - ExtractorError, - int_or_none, - parse_iso8601, - traverse_obj, - try_get, -) - - -class WASDTVBaseIE(InfoExtractor): - - def _fetch(self, path, video_id, description, query={}): - response = self._download_json( - f'https://wasd.tv/api/{path}', video_id, query=query, - note=f'Downloading {description} metadata', - errnote=f'Unable to download {description} metadata') - error = response.get('error') - if error: - raise ExtractorError(f'{self.IE_NAME} returned error: {error}', expected=True) - return response.get('result') - - def _extract_thumbnails(self, thumbnails_dict): - return [{ - 'url': url, - 'preference': index, - } for index, url in enumerate( - traverse_obj(thumbnails_dict, (('small', 'medium', 'large'),))) if url] - - def _real_extract(self, url): - container = self._get_container(url) - stream = traverse_obj(container, ('media_container_streams', 0)) - media = try_get(stream, lambda x: x['stream_media'][0]) - if not media: - raise ExtractorError('Can not extract media data.', expected=True) - media_meta = media.get('media_meta') - media_url, is_live = self._get_media_url(media_meta) - video_id = media.get('media_id') or container.get('media_container_id') - formats, subtitles = self._extract_m3u8_formats_and_subtitles(media_url, video_id, 'mp4') - return { - 'id': str(video_id), - 'title': container.get('media_container_name') or self._og_search_title(self._download_webpage(url, video_id)), - 'description': container.get('media_container_description'), - 'thumbnails': self._extract_thumbnails(media_meta.get('media_preview_images')), - 'timestamp': parse_iso8601(container.get('created_at')), - 'view_count': int_or_none(stream.get('stream_current_viewers' if is_live else 'stream_total_viewers')), - 'is_live': is_live, - 'formats': formats, - 'subtitles': subtitles, - } - - def _get_container(self, url): - raise NotImplementedError('Subclass for get media container') - - def _get_media_url(self, media_meta): - raise NotImplementedError('Subclass for get media url') - - -class WASDTVStreamIE(WASDTVBaseIE): - IE_NAME = 'wasdtv:stream' - _VALID_URL = r'https?://wasd\.tv/(?P[^/#?]+)$' - _TESTS = [{ - 'url': 'https://wasd.tv/24_7', - 'info_dict': { - 'id': '559738', - 'ext': 'mp4', - 'title': 'Live 24/7 Music', - 'description': '24/7 Music', - 'timestamp': int, - 'upload_date': r're:^\d{8}$', - 'is_live': True, - 'view_count': int, - }, - }] - - def _get_container(self, url): - nickname = self._match_id(url) - channel = self._fetch(f'channels/nicknames/{nickname}', video_id=nickname, description='channel') - channel_id = channel.get('channel_id') - containers = self._fetch( - 'v2/media-containers', channel_id, 'running media containers', - query={ - 'channel_id': channel_id, - 'media_container_type': 'SINGLE', - 'media_container_status': 'RUNNING', - }) - if not containers: - raise ExtractorError(f'{nickname} is offline', expected=True) - return containers[0] - - def _get_media_url(self, media_meta): - return media_meta['media_url'], True - - -class WASDTVRecordIE(WASDTVBaseIE): - IE_NAME = 'wasdtv:record' - _VALID_URL = r'https?://wasd\.tv/[^/#?]+(?:/videos)?\?record=(?P\d+)$' - _TESTS = [{ - 'url': 'https://wasd.tv/spacemita/videos?record=907755', - 'md5': 'c9899dd85be4cc997816ff9f9ca516ce', - 'info_dict': { - 'id': '906825', - 'ext': 'mp4', - 'title': 'Музыкальный', - 'description': 'md5:f510388d929ff60ae61d4c3cab3137cc', - 'timestamp': 1645812079, - 'upload_date': '20220225', - 'thumbnail': r're:^https?://.+\.jpg', - 'is_live': False, - 'view_count': int, - }, - }, { - 'url': 'https://wasd.tv/spacemita?record=907755', - 'only_matching': True, - }] - - def _get_container(self, url): - container_id = self._match_id(url) - return self._fetch( - f'v2/media-containers/{container_id}', container_id, 'media container') - - def _get_media_url(self, media_meta): - media_archive_url = media_meta.get('media_archive_url') - if media_archive_url: - return media_archive_url, False - return media_meta['media_url'], True - - -class WASDTVClipIE(WASDTVBaseIE): - IE_NAME = 'wasdtv:clip' - _VALID_URL = r'https?://wasd\.tv/[^/#?]+/clips\?clip=(?P\d+)$' - _TESTS = [{ - 'url': 'https://wasd.tv/spacemita/clips?clip=26804', - 'md5': '818885e720143d7a4e776ff66fcff148', - 'info_dict': { - 'id': '26804', - 'ext': 'mp4', - 'title': 'Пуш флексит на голове стримера', - 'timestamp': 1646682908, - 'upload_date': '20220307', - 'thumbnail': r're:^https?://.+\.jpg', - 'view_count': int, - }, - }] - - def _real_extract(self, url): - clip_id = self._match_id(url) - clip = self._fetch(f'v2/clips/{clip_id}', video_id=clip_id, description='clip') - clip_data = clip.get('clip_data') - formats, subtitles = self._extract_m3u8_formats_and_subtitles(clip_data.get('url'), video_id=clip_id, ext='mp4') - return { - 'id': clip_id, - 'title': clip.get('clip_title') or self._og_search_title(self._download_webpage(url, clip_id, fatal=False)), - 'thumbnails': self._extract_thumbnails(clip_data.get('preview')), - 'timestamp': parse_iso8601(clip.get('created_at')), - 'view_count': int_or_none(clip.get('clip_views_count')), - 'formats': formats, - 'subtitles': subtitles, - }