From 3ba6aabd254d3a4bea485baea7edf1ec1fcd695c Mon Sep 17 00:00:00 2001 From: Remita Amine Date: Sun, 20 Dec 2020 12:03:35 +0100 Subject: [PATCH] [arkena] fix extraction --- youtube_dl/extractor/arkena.py | 152 ++++++++++++++++++++------------- 1 file changed, 91 insertions(+), 61 deletions(-) diff --git a/youtube_dl/extractor/arkena.py b/youtube_dl/extractor/arkena.py index 854f58767..fd46b1c77 100644 --- a/youtube_dl/extractor/arkena.py +++ b/youtube_dl/extractor/arkena.py @@ -6,13 +6,11 @@ import re from .common import InfoExtractor from ..compat import compat_urlparse from ..utils import ( - determine_ext, ExtractorError, float_or_none, int_or_none, - mimetype2ext, parse_iso8601, - strip_jsonp, + try_get, ) @@ -20,22 +18,27 @@ class ArkenaIE(InfoExtractor): _VALID_URL = r'''(?x) https?:// (?: - video\.arkena\.com/play2/embed/player\?| + video\.(?:arkena|qbrick)\.com/play2/embed/player\?| play\.arkena\.com/(?:config|embed)/avp/v\d/player/media/(?P[^/]+)/[^/]+/(?P\d+) ) ''' _TESTS = [{ - 'url': 'https://play.arkena.com/embed/avp/v2/player/media/b41dda37-d8e7-4d3f-b1b5-9a9db578bdfe/1/129411', - 'md5': 'b96f2f71b359a8ecd05ce4e1daa72365', + 'url': 'https://video.qbrick.com/play2/embed/player?accountId=1034090&mediaId=d8ab4607-00090107-aab86310', + 'md5': '97f117754e5f3c020f5f26da4a44ebaf', 'info_dict': { - 'id': 'b41dda37-d8e7-4d3f-b1b5-9a9db578bdfe', + 'id': 'd8ab4607-00090107-aab86310', 'ext': 'mp4', - 'title': 'Big Buck Bunny', - 'description': 'Royalty free test video', - 'timestamp': 1432816365, - 'upload_date': '20150528', - 'is_live': False, + 'title': 'EM_HT20_117_roslund_v2.mp4', + 'timestamp': 1608285912, + 'upload_date': '20201218', + 'duration': 1429.162667, + 'subtitles': { + 'sv': 'count:3', + }, }, + }, { + 'url': 'https://play.arkena.com/embed/avp/v2/player/media/b41dda37-d8e7-4d3f-b1b5-9a9db578bdfe/1/129411', + 'only_matching': True, }, { 'url': 'https://play.arkena.com/config/avp/v2/player/media/b41dda37-d8e7-4d3f-b1b5-9a9db578bdfe/1/129411/?callbackMethod=jQuery1111023664739129262213_1469227693893', 'only_matching': True, @@ -72,62 +75,89 @@ class ArkenaIE(InfoExtractor): if not video_id or not account_id: raise ExtractorError('Invalid URL', expected=True) - playlist = self._download_json( - 'https://play.arkena.com/config/avp/v2/player/media/%s/0/%s/?callbackMethod=_' - % (video_id, account_id), - video_id, transform_source=strip_jsonp)['Playlist'][0] + media = self._download_json( + 'https://video.qbrick.com/api/v1/public/accounts/%s/medias/%s' % (account_id, video_id), + video_id, query={ + # https://video.qbrick.com/docs/api/examples/library-api.html + 'fields': 'asset/resources/*/renditions/*(height,id,language,links/*(href,mimeType),type,size,videos/*(audios/*(codec,sampleRate),bitrate,codec,duration,height,width),width),created,metadata/*(title,description),tags', + }) + metadata = media.get('metadata') or {} + title = metadata['title'] - media_info = playlist['MediaInfo'] - title = media_info['Title'] - media_files = playlist['MediaFiles'] - - is_live = False + duration = None formats = [] - for kind_case, kind_formats in media_files.items(): - kind = kind_case.lower() - for f in kind_formats: - f_url = f.get('Url') - if not f_url: - continue - is_live = f.get('Live') == 'true' - exts = (mimetype2ext(f.get('Type')), determine_ext(f_url, None)) - if kind == 'm3u8' or 'm3u8' in exts: - formats.extend(self._extract_m3u8_formats( - f_url, video_id, 'mp4', 'm3u8_native', - m3u8_id=kind, fatal=False, live=is_live)) - elif kind == 'flash' or 'f4m' in exts: - formats.extend(self._extract_f4m_formats( - f_url, video_id, f4m_id=kind, fatal=False)) - elif kind == 'dash' or 'mpd' in exts: - formats.extend(self._extract_mpd_formats( - f_url, video_id, mpd_id=kind, fatal=False)) - elif kind == 'silverlight': - # TODO: process when ism is supported (see - # https://github.com/ytdl-org/youtube-dl/issues/8118) - continue - else: - tbr = float_or_none(f.get('Bitrate'), 1000) - formats.append({ - 'url': f_url, - 'format_id': '%s-%d' % (kind, tbr) if tbr else kind, - 'tbr': tbr, - }) + thumbnails = [] + subtitles = {} + for resource in media['asset']['resources']: + for rendition in (resource.get('renditions') or []): + rendition_type = rendition.get('type') + for i, link in enumerate(rendition.get('links') or []): + href = link.get('href') + if not href: + continue + if rendition_type == 'image': + thumbnails.append({ + 'filesize': int_or_none(rendition.get('size')), + 'height': int_or_none(rendition.get('height')), + 'id': rendition.get('id'), + 'url': href, + 'width': int_or_none(rendition.get('width')), + }) + elif rendition_type == 'subtitle': + subtitles.setdefault(rendition.get('language') or 'en', []).append({ + 'url': href, + }) + elif rendition_type == 'video': + f = { + 'filesize': int_or_none(rendition.get('size')), + 'format_id': rendition.get('id'), + 'url': href, + } + video = try_get(rendition, lambda x: x['videos'][i], dict) + if video: + if not duration: + duration = float_or_none(video.get('duration')) + f.update({ + 'height': int_or_none(video.get('height')), + 'tbr': int_or_none(video.get('bitrate'), 1000), + 'vcodec': video.get('codec'), + 'width': int_or_none(video.get('width')), + }) + audio = try_get(video, lambda x: x['audios'][0], dict) + if audio: + f.update({ + 'acodec': audio.get('codec'), + 'asr': int_or_none(audio.get('sampleRate')), + }) + formats.append(f) + elif rendition_type == 'index': + mime_type = link.get('mimeType') + if mime_type == 'application/smil+xml': + formats.extend(self._extract_smil_formats( + href, video_id, fatal=False)) + elif mime_type == 'application/x-mpegURL': + formats.extend(self._extract_m3u8_formats( + href, video_id, 'mp4', 'm3u8_native', + m3u8_id='hls', fatal=False)) + elif mime_type == 'application/hds+xml': + formats.extend(self._extract_f4m_formats( + href, video_id, f4m_id='hds', fatal=False)) + elif mime_type == 'application/dash+xml': + formats.extend(self._extract_f4m_formats( + href, video_id, f4m_id='hds', fatal=False)) + elif mime_type == 'application/vnd.ms-sstr+xml': + formats.extend(self._extract_ism_formats( + href, video_id, ism_id='mss', fatal=False)) self._sort_formats(formats) - description = media_info.get('Description') - video_id = media_info.get('VideoId') or video_id - timestamp = parse_iso8601(media_info.get('PublishDate')) - thumbnails = [{ - 'url': thumbnail['Url'], - 'width': int_or_none(thumbnail.get('Size')), - } for thumbnail in (media_info.get('Poster') or []) if thumbnail.get('Url')] - return { 'id': video_id, 'title': title, - 'description': description, - 'timestamp': timestamp, - 'is_live': is_live, + 'description': metadata.get('description'), + 'timestamp': parse_iso8601(media.get('created')), 'thumbnails': thumbnails, + 'subtitles': subtitles, + 'duration': duration, + 'tags': media.get('tags'), 'formats': formats, }