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, }