From 9a235dd6f9e7deac0711878aaf125a1fa894138a Mon Sep 17 00:00:00 2001 From: subrat-lima Date: Sun, 18 Aug 2024 19:37:00 +0530 Subject: [PATCH 1/3] updated NDTV InfoExtractor changes made 1. updated the extractor 2. added test cases --- yt_dlp/extractor/ndtv.py | 94 +++++++++++++++++++++++++++++++++------- 1 file changed, 78 insertions(+), 16 deletions(-) diff --git a/yt_dlp/extractor/ndtv.py b/yt_dlp/extractor/ndtv.py index c328bd4d23..6e2869ee82 100644 --- a/yt_dlp/extractor/ndtv.py +++ b/yt_dlp/extractor/ndtv.py @@ -1,11 +1,10 @@ -import urllib.parse from .common import InfoExtractor -from ..utils import parse_duration, remove_end, unified_strdate, urljoin +from ..utils import clean_html, parse_duration, remove_end, unified_strdate class NDTVIE(InfoExtractor): - _WORKING = False + _WORKING = True _VALID_URL = r'https?://(?:[^/]+\.)?ndtv\.com/(?:[^/]+/)*videos?/?(?:[^/]+/)*[^/?^&]+-(?P\d+)' _TESTS = [ @@ -16,14 +15,13 @@ class NDTVIE(InfoExtractor): 'id': '468818', 'ext': 'mp4', 'title': 'प्राइम टाइम: सिस्टम बीमार, स्कूल बदहाल', - 'description': 'md5:f410512f1b49672e5695dea16ef2731d', + 'description': 'md5:de6008f2439005344c1c0c686032b43a', 'upload_date': '20170928', 'duration': 2218, 'thumbnail': r're:https?://.*\.jpg', }, }, { - # __filename is url 'url': 'http://movies.ndtv.com/videos/cracker-free-diwali-wishes-from-karan-johar-kriti-sanon-other-stars-470304', 'md5': 'f1d709352305b44443515ac56b45aa46', 'info_dict': { @@ -36,6 +34,71 @@ class NDTVIE(InfoExtractor): 'thumbnail': r're:https?://.*\.jpg', }, }, + { + 'url': 'https://www.ndtv.com/video/ndtv-jai-jawan-dnd-john-abraham-is-busy-watching-soldiers-fitness-routine-826286', + 'md5': '19d9a96480a8dac0e03a42efce27350f', + 'info_dict': { + 'id': '826286', + 'ext': 'mp4', + 'title': 'NDTV Jai Jawan: DND, John Abraham Is Busy Watching Soldiers\' Fitness Routine', + 'description': 'md5:ad700713b10702555555e6a01a25442a', + 'upload_date': '20240818', + 'duration': 220, + 'thumbnail': 'https://c.ndtvimg.com/2024-08/be8djnjc_jai-jawan_640x480_18_August_24.jpg', + }, + }, + { + 'url': 'https://www.ndtv.com/video/cancer-can-be-prevented-cured-with-lifestyle-changes-expert-to-ndtv-827194', + 'md5': '72734ed9226e200454100ffe7dfae2b6', + 'info_dict': { + 'id': '827194', + 'ext': 'mp4', + 'title': '"Cancer Can Be Prevented, Cured With Lifestyle Changes": Expert To NDTV', + 'description': 'md5:cbecc6372e33370014748db9ad2b5df0', + 'upload_date': '20240818', + 'duration': 931, + 'thumbnail': 'https://c.ndtvimg.com/2024-08/hh48aduk_dr-sameer-kaul_640x480_18_August_24.jpg', + }, + }, + { + 'url': 'https://sports.ndtv.com/cricket/videos/ex-india-captain-ms-dhoni-spotted-at-airport-820044', + 'md5': '9f3442f49aa660623e8925c0fc7606a4', + 'info_dict': { + 'id': '820044', + 'ext': 'mp4', + 'title': 'Ex-India Captain MS Dhoni Spotted At Airport', + 'description': 'md5:7be70dbb543bb647f83dbd6e7b8f16a0', + 'upload_date': '20240730', + 'duration': 45, + 'thumbnail': 'https://c.ndtvimg.com/2024-07/itrg8ilo_msdhoni_640x480_31_July_24.jpg', + }, + }, + { + 'url': 'https://doctor.ndtv.com/videos/attack-the-heart-attack-a-public-awareness-initiative-by-medtronics-822292', + 'md5': 'a13ad8bbb45eec4405a45f9bd7ed0356', + 'info_dict': { + 'id': '822292', + 'ext': 'mp4', + 'title': 'Attack The Heart Attack - A Public Awareness Initiative By Medtronics', + 'description': 'md5:31b876e23795b4d7ef91797f88b22730', + 'upload_date': '20240805', + 'duration': 1304, + 'thumbnail': 'https://c.ndtvimg.com/2024-08/kh67tomo_heartattack_640x480_05_August_24.jpg', + }, + }, + { + 'url': 'https://food.ndtv.com/video-watch-ndtv-food-awards-2024-on-ndtv-24x7-on-saturday-9th-march-2024-767098', + 'md5': '78ddea4f057115e84952563749c6b94c', + 'info_dict': { + 'id': '767098', + 'ext': 'mp4', + 'title': 'Watch NDTV Food Awards 2024 On NDTV 24x7 On Saturday 9th March, 2024', + 'description': 'md5:4a8831a038ab50dc5ea76b1fad4b51d0', + 'upload_date': '20240307', + 'duration': 35, + 'thumbnail': 'https://c.ndtvimg.com/2024-03/2eianfs_food-awards_640x480_08_March_24.jpg', + }, + }, { 'url': 'https://www.ndtv.com/video/news/news/delhi-s-air-quality-status-report-after-diwali-is-very-poor-470372', 'only_matching': True, @@ -74,19 +137,15 @@ def _real_extract(self, url): video_id = self._match_id(url) webpage = self._download_webpage(url, video_id) - # '__title' does not contain extra words such as sub-site name, "Video" etc. - title = urllib.parse.unquote_plus( - self._search_regex(r"__title\s*=\s*'([^']+)'", webpage, 'title', default=None) - or self._og_search_title(webpage)) + title = clean_html(self._html_search_meta('taboola:title', webpage, 'title', default=None) or self._html_search_meta('name', webpage, 'title', default=None) or self._og_search_title(webpage)) - filename = self._search_regex( - r"(?:__)?filename\s*[:=]\s*'([^']+)'", webpage, 'video filename') # in "movies" sub-site pages, filename is URL - video_url = urljoin('https://ndtvod.bc-ssl.cdn.bitgravity.com/23372/ndtv/', filename.lstrip('/')) + # in "food" sub-site pages, url string contains escape characters + video_url = (self._html_search_meta('contentUrl', webpage, 'video-url', default=None) + or self._search_regex(r'\"media_mp4\"\s*:\s*\"([^\"]+)\"', webpage, 'video-url')).replace('\\/', '/') # "doctor" sub-site has MM:SS format - duration = parse_duration(self._search_regex( - r"(?:__)?duration\s*[:=]\s*'([^']+)'", webpage, 'duration', fatal=False)) + duration = parse_duration(self._html_search_meta('video:duration', webpage, 'duration', default=None) or self._search_regex(r'\"dur\"\s*:\s*\"([^\"]+)\"', webpage, 'duration')) # "sports", "doctor", "swirlster" sub-sites don't have 'publish-date' upload_date = unified_strdate(self._html_search_meta( @@ -94,14 +153,17 @@ def _real_extract(self, url): 'uploadDate', webpage, 'upload date', default=None) or self._search_regex( r'datePublished"\s*:\s*"([^"]+)"', webpage, 'upload date', fatal=False)) - description = remove_end(self._og_search_description(webpage), ' (Read more)') + description = clean_html(self._html_search_meta('description', webpage, 'description', default=None) + or remove_end(self._og_search_description(webpage), ' (Read more)')) + + thumbnail = (self._html_search_meta('thumbnailUrl', webpage, 'thumbnail', default=None) or self._search_regex(r'\"image\"\s*:\s*\"([^\"]+)\"', webpage, 'thumbnail')).replace('\\/', '/') return { 'id': video_id, 'url': video_url, 'title': title, 'description': description, - 'thumbnail': self._og_search_thumbnail(webpage), + 'thumbnail': thumbnail, 'duration': duration, 'upload_date': upload_date, } From db4b19eaffecae7d18b781c5a56bd9e47ee410ff Mon Sep 17 00:00:00 2001 From: subrat-lima Date: Sun, 18 Aug 2024 21:32:37 +0530 Subject: [PATCH 2/3] refined code as per coding guidelines --- yt_dlp/extractor/ndtv.py | 31 +++++++++++++++++++++---------- 1 file changed, 21 insertions(+), 10 deletions(-) diff --git a/yt_dlp/extractor/ndtv.py b/yt_dlp/extractor/ndtv.py index 6e2869ee82..6e33a3c67f 100644 --- a/yt_dlp/extractor/ndtv.py +++ b/yt_dlp/extractor/ndtv.py @@ -137,26 +137,37 @@ def _real_extract(self, url): video_id = self._match_id(url) webpage = self._download_webpage(url, video_id) - title = clean_html(self._html_search_meta('taboola:title', webpage, 'title', default=None) or self._html_search_meta('name', webpage, 'title', default=None) or self._og_search_title(webpage)) + title = clean_html( + self._html_search_meta('taboola:title', webpage, 'title', default=None) + or self._html_search_meta('name', webpage, 'title', default=None) + or self._og_search_title(webpage)) # in "movies" sub-site pages, filename is URL # in "food" sub-site pages, url string contains escape characters video_url = (self._html_search_meta('contentUrl', webpage, 'video-url', default=None) - or self._search_regex(r'\"media_mp4\"\s*:\s*\"([^\"]+)\"', webpage, 'video-url')).replace('\\/', '/') + or self._search_regex(r'"media_mp4"\s*:\s*"([^"]+)', webpage, 'video-url')) + if video_url is not None: + video_url = video_url.replace('\\/', '/') # "doctor" sub-site has MM:SS format - duration = parse_duration(self._html_search_meta('video:duration', webpage, 'duration', default=None) or self._search_regex(r'\"dur\"\s*:\s*\"([^\"]+)\"', webpage, 'duration')) + duration = parse_duration( + self._html_search_meta('video:duration', webpage, 'duration', default=None) + or self._search_regex(r'"dur"\s*:\s*"([^"]+)', webpage, 'duration', fatal=False, default=None)) # "sports", "doctor", "swirlster" sub-sites don't have 'publish-date' - upload_date = unified_strdate(self._html_search_meta( - 'publish-date', webpage, 'upload date', default=None) or self._html_search_meta( - 'uploadDate', webpage, 'upload date', default=None) or self._search_regex( - r'datePublished"\s*:\s*"([^"]+)"', webpage, 'upload date', fatal=False)) + upload_date = unified_strdate( + self._html_search_meta('publish-date', webpage, 'upload date', default=None) + or self._html_search_meta('uploadDate', webpage, 'upload date', default=None) + or self._search_regex(r'datePublished"\s*:\s*"([^"]+)', webpage, 'upload date', fatal=False, default=None)) - description = clean_html(self._html_search_meta('description', webpage, 'description', default=None) - or remove_end(self._og_search_description(webpage), ' (Read more)')) + description = clean_html( + self._html_search_meta('description', webpage, 'description', default=None) + or remove_end(self._og_search_description(webpage, default=None), ' (Read more)')) - thumbnail = (self._html_search_meta('thumbnailUrl', webpage, 'thumbnail', default=None) or self._search_regex(r'\"image\"\s*:\s*\"([^\"]+)\"', webpage, 'thumbnail')).replace('\\/', '/') + thumbnail = (self._html_search_meta('thumbnailUrl', webpage, 'thumbnail', default=None) + or self._search_regex(r'"image"\s*:\s*"([^\"]+)', webpage, 'thumbnail', fatal=False, default=None)) + if thumbnail is not None: + thumbnail = thumbnail.replace('\\/', '/') return { 'id': video_id, From dbb1b746d4dcbd1da7481375491f306adb0577b0 Mon Sep 17 00:00:00 2001 From: subrat-lima Date: Thu, 22 Aug 2024 13:52:33 +0530 Subject: [PATCH 3/3] updated extractor to a simpler process --- yt_dlp/extractor/ndtv.py | 82 ++++++++++++++++++++++++---------------- 1 file changed, 49 insertions(+), 33 deletions(-) diff --git a/yt_dlp/extractor/ndtv.py b/yt_dlp/extractor/ndtv.py index 6e33a3c67f..ec4c898753 100644 --- a/yt_dlp/extractor/ndtv.py +++ b/yt_dlp/extractor/ndtv.py @@ -1,6 +1,6 @@ from .common import InfoExtractor -from ..utils import clean_html, parse_duration, remove_end, unified_strdate +from ..utils import clean_html, determine_ext, parse_duration, traverse_obj, unescapeHTML, unified_strdate class NDTVIE(InfoExtractor): @@ -10,7 +10,7 @@ class NDTVIE(InfoExtractor): _TESTS = [ { 'url': 'https://khabar.ndtv.com/video/show/prime-time/prime-time-ill-system-and-poor-education-468818', - 'md5': '78efcf3880ef3fd9b83d405ca94a38eb', + 'md5': '5237500d57c8af47a519c4af9e2fd9c0', 'info_dict': { 'id': '468818', 'ext': 'mp4', @@ -19,11 +19,13 @@ class NDTVIE(InfoExtractor): 'upload_date': '20170928', 'duration': 2218, 'thumbnail': r're:https?://.*\.jpg', + 'categories': 'shows', + 'channel_id': '2', }, }, { 'url': 'http://movies.ndtv.com/videos/cracker-free-diwali-wishes-from-karan-johar-kriti-sanon-other-stars-470304', - 'md5': 'f1d709352305b44443515ac56b45aa46', + 'md5': 'b7ef90f55e63885aef59c59cf2e44e81', 'info_dict': { 'id': '470304', 'ext': 'mp4', @@ -32,6 +34,8 @@ class NDTVIE(InfoExtractor): 'upload_date': '20171019', 'duration': 137, 'thumbnail': r're:https?://.*\.jpg', + 'categories': 'entertainment', + 'channel_id': '1', }, }, { @@ -45,6 +49,8 @@ class NDTVIE(InfoExtractor): 'upload_date': '20240818', 'duration': 220, 'thumbnail': 'https://c.ndtvimg.com/2024-08/be8djnjc_jai-jawan_640x480_18_August_24.jpg', + 'categories': 'entertainment', + 'channel_id': '1', }, }, { @@ -58,6 +64,8 @@ class NDTVIE(InfoExtractor): 'upload_date': '20240818', 'duration': 931, 'thumbnail': 'https://c.ndtvimg.com/2024-08/hh48aduk_dr-sameer-kaul_640x480_18_August_24.jpg', + 'categories': 'news', + 'channel_id': '1', }, }, { @@ -71,6 +79,8 @@ class NDTVIE(InfoExtractor): 'upload_date': '20240730', 'duration': 45, 'thumbnail': 'https://c.ndtvimg.com/2024-07/itrg8ilo_msdhoni_640x480_31_July_24.jpg', + 'categories': 'fangully', + 'channel_id': '', }, }, { @@ -84,6 +94,8 @@ class NDTVIE(InfoExtractor): 'upload_date': '20240805', 'duration': 1304, 'thumbnail': 'https://c.ndtvimg.com/2024-08/kh67tomo_heartattack_640x480_05_August_24.jpg', + 'categories': 'Health', + 'channel_id': '1', }, }, { @@ -93,10 +105,12 @@ class NDTVIE(InfoExtractor): 'id': '767098', 'ext': 'mp4', 'title': 'Watch NDTV Food Awards 2024 On NDTV 24x7 On Saturday 9th March, 2024', - 'description': 'md5:4a8831a038ab50dc5ea76b1fad4b51d0', + 'description': 'md5:11948e8404b058cd0a9d174aa1e88e84', 'upload_date': '20240307', 'duration': 35, 'thumbnail': 'https://c.ndtvimg.com/2024-03/2eianfs_food-awards_640x480_08_March_24.jpg', + 'categories': 'food', + 'channel_id': '1', }, }, { @@ -137,44 +151,46 @@ def _real_extract(self, url): video_id = self._match_id(url) webpage = self._download_webpage(url, video_id) - title = clean_html( - self._html_search_meta('taboola:title', webpage, 'title', default=None) - or self._html_search_meta('name', webpage, 'title', default=None) - or self._og_search_title(webpage)) + # required for removing unwanted js expression in the json data + # found this expression in one article + # will update this json data handler more generic if more of similar patterns found + webpage = webpage.replace('true / false', 'true') - # in "movies" sub-site pages, filename is URL - # in "food" sub-site pages, url string contains escape characters - video_url = (self._html_search_meta('contentUrl', webpage, 'video-url', default=None) - or self._search_regex(r'"media_mp4"\s*:\s*"([^"]+)', webpage, 'video-url')) - if video_url is not None: - video_url = video_url.replace('\\/', '/') + data = self._search_json(r'__html5playerdata\s*=\s*', webpage, 'json_data', video_id) - # "doctor" sub-site has MM:SS format - duration = parse_duration( - self._html_search_meta('video:duration', webpage, 'duration', default=None) - or self._search_regex(r'"dur"\s*:\s*"([^"]+)', webpage, 'duration', fatal=False, default=None)) + title = clean_html(unescapeHTML(traverse_obj(data, ('title')))) + duration = parse_duration(traverse_obj(data, ('dur'))) + upload_date = unified_strdate(traverse_obj(data, ('date'))) + video_m3u8 = traverse_obj(data, ('media')) + video_mp4 = traverse_obj(data, ('media_mp4')) + video_webm = traverse_obj(data, ('media_webm')) + description = clean_html(unescapeHTML(traverse_obj(data, ('description')))) + thumbnail = traverse_obj(data, ('image')) + channel_id = traverse_obj(data, ('channel')) + categories = traverse_obj(data, ('category')) - # "sports", "doctor", "swirlster" sub-sites don't have 'publish-date' - upload_date = unified_strdate( - self._html_search_meta('publish-date', webpage, 'upload date', default=None) - or self._html_search_meta('uploadDate', webpage, 'upload date', default=None) - or self._search_regex(r'datePublished"\s*:\s*"([^"]+)', webpage, 'upload date', fatal=False, default=None)) - - description = clean_html( - self._html_search_meta('description', webpage, 'description', default=None) - or remove_end(self._og_search_description(webpage, default=None), ' (Read more)')) - - thumbnail = (self._html_search_meta('thumbnailUrl', webpage, 'thumbnail', default=None) - or self._search_regex(r'"image"\s*:\s*"([^\"]+)', webpage, 'thumbnail', fatal=False, default=None)) - if thumbnail is not None: - thumbnail = thumbnail.replace('\\/', '/') + formats = [] + if video_mp4: + formats.append({'url': video_mp4, 'ext': determine_ext(video_mp4)}) + if video_webm: + formats.append({'url': video_webm, 'ext': determine_ext(video_webm)}) + if video_m3u8: + m3u8_data = self._extract_m3u8_formats(video_m3u8, video_id) + formats.extend(m3u8_data) + # here, making the assumption that mp4 video available is always the highest quality + # will update if any deviation found + formats[0]['width'] = m3u8_data[-1]['width'] + formats[0]['height'] = m3u8_data[-1]['height'] + formats[0]['tbr'] = m3u8_data[-1]['tbr'] return { 'id': video_id, - 'url': video_url, + 'formats': formats, 'title': title, 'description': description, 'thumbnail': thumbnail, 'duration': duration, 'upload_date': upload_date, + 'categories': categories, + 'channel_id': channel_id, }