From 998dffb5a2343ec709b3d6bbf2bf019649080239 Mon Sep 17 00:00:00 2001 From: "J. Gonzalez" Date: Fri, 23 Feb 2024 11:07:35 -0500 Subject: [PATCH] [ie/cnbc] Overhaul extractors (#8741) Closes #5871, Closes #8378 Authored by: gonzalezjo, Noor-5, zhijinwuu, ruiminggu, seproDev Co-authored-by: Noor Mostafa <93787875+Noor-5@users.noreply.github.com> Co-authored-by: zhijinwuu Co-authored-by: ruiminggu Co-authored-by: sepro <4618135+seproDev@users.noreply.github.com> --- yt_dlp/extractor/_extractors.py | 1 - yt_dlp/extractor/cnbc.py | 141 +++++++++++++++++++------------- 2 files changed, 85 insertions(+), 57 deletions(-) diff --git a/yt_dlp/extractor/_extractors.py b/yt_dlp/extractor/_extractors.py index fc22e1571..583477b98 100644 --- a/yt_dlp/extractor/_extractors.py +++ b/yt_dlp/extractor/_extractors.py @@ -379,7 +379,6 @@ from .clubic import ClubicIE from .clyp import ClypIE from .cmt import CMTIE from .cnbc import ( - CNBCIE, CNBCVideoIE, ) from .cnn import ( diff --git a/yt_dlp/extractor/cnbc.py b/yt_dlp/extractor/cnbc.py index 7d209b6d9..b8ce2b49a 100644 --- a/yt_dlp/extractor/cnbc.py +++ b/yt_dlp/extractor/cnbc.py @@ -1,68 +1,97 @@ from .common import InfoExtractor -from ..utils import smuggle_url +from ..utils import int_or_none, parse_iso8601, str_or_none, url_or_none +from ..utils.traversal import traverse_obj -class CNBCIE(InfoExtractor): - _VALID_URL = r'https?://video\.cnbc\.com/gallery/\?video=(?P[0-9]+)' - _TEST = { - 'url': 'http://video.cnbc.com/gallery/?video=3000503714', +class CNBCVideoIE(InfoExtractor): + _VALID_URL = r'https?://(?:www\.)?cnbc\.com/video/(?:[^/?#]+/)+(?P[^./?#&]+)\.html' + + _TESTS = [{ + 'url': 'https://www.cnbc.com/video/2023/12/07/mcdonalds-just-unveiled-cosmcsits-new-spinoff-brand.html', 'info_dict': { - 'id': '3000503714', 'ext': 'mp4', - 'title': 'Fighting zombies is big business', - 'description': 'md5:0c100d8e1a7947bd2feec9a5550e519e', - 'timestamp': 1459332000, - 'upload_date': '20160330', - 'uploader': 'NBCU-CNBC', + 'id': '107344774', + 'display_id': 'mcdonalds-just-unveiled-cosmcsits-new-spinoff-brand', + 'modified_timestamp': 1702053483, + 'timestamp': 1701977810, + 'channel': 'News Videos', + 'upload_date': '20231207', + 'description': 'md5:882c001d85cb43d7579b514307b3e78b', + 'release_timestamp': 1701977375, + 'modified_date': '20231208', + 'release_date': '20231207', + 'duration': 65, + 'author': 'Sean Conlon', + 'title': 'Here\'s a first look at McDonald\'s new spinoff brand, CosMc\'s', + 'thumbnail': 'https://image.cnbcfm.com/api/v1/image/107344192-1701894812493-CosMcsskyHero_2336x1040_hero-desktop.jpg?v=1701894855', }, - 'params': { - # m3u8 download - 'skip_download': True, - }, - 'skip': 'Dead link', - } - - def _real_extract(self, url): - video_id = self._match_id(url) - return { - '_type': 'url_transparent', - 'ie_key': 'ThePlatform', - 'url': smuggle_url( - 'http://link.theplatform.com/s/gZWlPC/media/guid/2408950221/%s?mbr=true&manifest=m3u' % video_id, - {'force_smil_url': True}), - 'id': video_id, - } - - -class CNBCVideoIE(InfoExtractor): - _VALID_URL = r'https?://(?:www\.)?cnbc\.com(?P/video/(?:[^/]+/)+(?P[^./?#&]+)\.html)' - _TEST = { - 'url': 'https://www.cnbc.com/video/2018/07/19/trump-i-dont-necessarily-agree-with-raising-rates.html', + 'expected_warnings': ['Unable to download f4m manifest'], + }, { + 'url': 'https://www.cnbc.com/video/2023/12/08/jim-cramer-shares-his-take-on-seattles-tech-scene.html', 'info_dict': { - 'id': '7000031301', + 'author': 'Jim Cramer', + 'channel': 'Mad Money with Jim Cramer', + 'description': 'md5:72925be21b952e95eba51178dddf4e3e', + 'duration': 299.0, 'ext': 'mp4', - 'title': "Trump: I don't necessarily agree with raising rates", - 'description': 'md5:878d8f0b4ebb5bb1dda3514b91b49de3', - 'timestamp': 1531958400, - 'upload_date': '20180719', - 'uploader': 'NBCU-CNBC', + 'id': '107345451', + 'display_id': 'jim-cramer-shares-his-take-on-seattles-tech-scene', + 'thumbnail': 'https://image.cnbcfm.com/api/v1/image/107345481-1702079431MM-B-120823.jpg?v=1702079430', + 'timestamp': 1702080139, + 'title': 'Jim Cramer shares his take on Seattle\'s tech scene', + 'release_date': '20231208', + 'upload_date': '20231209', + 'modified_timestamp': 1702080139, + 'modified_date': '20231209', + 'release_timestamp': 1702073551, }, - 'params': { - 'skip_download': True, + 'expected_warnings': ['Unable to download f4m manifest'], + }, { + 'url': 'https://www.cnbc.com/video/2023/12/08/the-epicenter-of-ai-is-in-seattle-says-jim-cramer.html', + 'info_dict': { + 'author': 'Jim Cramer', + 'channel': 'Mad Money with Jim Cramer', + 'description': 'md5:72925be21b952e95eba51178dddf4e3e', + 'duration': 113.0, + 'ext': 'mp4', + 'id': '107345474', + 'display_id': 'the-epicenter-of-ai-is-in-seattle-says-jim-cramer', + 'thumbnail': 'https://image.cnbcfm.com/api/v1/image/107345486-Screenshot_2023-12-08_at_70339_PM.png?v=1702080248', + 'timestamp': 1702080535, + 'title': 'The epicenter of AI is in Seattle, says Jim Cramer', + 'release_timestamp': 1702077347, + 'modified_timestamp': 1702080535, + 'release_date': '20231208', + 'upload_date': '20231209', + 'modified_date': '20231209', }, - 'skip': 'Dead link', - } + 'expected_warnings': ['Unable to download f4m manifest'], + }] def _real_extract(self, url): - path, display_id = self._match_valid_url(url).groups() - video_id = self._download_json( - 'https://webql-redesign.cnbcfm.com/graphql', display_id, query={ - 'query': '''{ - page(path: "%s") { - vcpsId - } -}''' % path, - })['data']['page']['vcpsId'] - return self.url_result( - 'http://video.cnbc.com/gallery/?video=%d' % video_id, - CNBCIE.ie_key()) + display_id = self._match_id(url) + webpage = self._download_webpage(url, display_id) + data = self._search_json(r'window\.__s_data=', webpage, 'video data', display_id) + + player_data = traverse_obj(data, ( + 'page', 'page', 'layout', ..., 'columns', ..., 'modules', + lambda _, v: v['name'] == 'clipPlayer', 'data', {dict}), get_all=False) + + return { + 'id': display_id, + 'display_id': display_id, + 'formats': self._extract_akamai_formats(player_data['playbackURL'], display_id), + **self._search_json_ld(webpage, display_id, fatal=False), + **traverse_obj(player_data, { + 'id': ('id', {str_or_none}), + 'title': ('title', {str}), + 'description': ('description', {str}), + 'author': ('author', ..., 'name', {str}), + 'timestamp': ('datePublished', {parse_iso8601}), + 'release_timestamp': ('uploadDate', {parse_iso8601}), + 'modified_timestamp': ('dateLastPublished', {parse_iso8601}), + 'thumbnail': ('thumbnail', {url_or_none}), + 'duration': ('duration', {int_or_none}), + 'channel': ('section', 'title', {str}), + }, get_all=False), + }