X-Git-Url: http://git.bitcoin.ninja/index.cgi?a=blobdiff_plain;f=youtube_dl%2Fextractor%2Fvevo.py;h=152fef42e2aa052acea07f909eee66cf842e87c5;hb=c6ca11f1b31a292413cab03012654fd0021814a8;hp=ff854e74b2cf04bbdc8b553adf10aa4fdb1853f8;hpb=2975fe1a7be39b66d14fc7ef68ea7db948f0e40f;p=youtube-dl diff --git a/youtube_dl/extractor/vevo.py b/youtube_dl/extractor/vevo.py index ff854e74b..152fef42e 100644 --- a/youtube_dl/extractor/vevo.py +++ b/youtube_dl/extractor/vevo.py @@ -7,6 +7,8 @@ from ..compat import compat_etree_fromstring from ..utils import ( ExtractorError, int_or_none, + sanitized_Request, + parse_iso8601, ) @@ -24,51 +26,39 @@ class VevoIE(InfoExtractor): _TESTS = [{ 'url': 'http://www.vevo.com/watch/hurts/somebody-to-die-for/GB1101300280', - 'md5': '2dbc7e9fd4f1c60436c9aa73a5406193', + 'md5': '95ee28ee45e70130e3ab02b0f579ae23', 'info_dict': { - 'id': 'Pt1kc_FniKM', + 'id': 'GB1101300280', 'ext': 'mp4', - 'title': 'Hurts - Somebody to Die For', - 'description': 'md5:13e925b89af6b01c7e417332bd23c4bf', - 'uploader_id': 'HurtsVEVO', - 'uploader': 'HurtsVEVO', + 'title': 'Somebody to Die For', 'upload_date': '20130624', - 'duration': 230, + 'uploader': 'Hurts', + 'timestamp': 1372057200, }, - 'add_ie': ['Youtube'], }, { 'note': 'v3 SMIL format', 'url': 'http://www.vevo.com/watch/cassadee-pope/i-wish-i-could-break-your-heart/USUV71302923', - 'md5': '13d5204f520af905eeffa675040b8e76', + 'md5': 'f6ab09b034f8c22969020b042e5ac7fc', 'info_dict': { - 'id': 'ByGmQn1uxJw', + 'id': 'USUV71302923', 'ext': 'mp4', - 'title': 'Cassadee Pope - I Wish I Could Break Your Heart', - 'description': 'md5:5e9721c92ef117a6f69d00e9b42ceba7', - 'uploader_id': 'CassadeeVEVO', - 'uploader': 'CassadeeVEVO', + 'title': 'I Wish I Could Break Your Heart', 'upload_date': '20140219', - 'duration': 226, - 'age_limit': 0, + 'uploader': 'Cassadee Pope', + 'timestamp': 1392796919, }, - 'add_ie': ['Youtube'], }, { 'note': 'Age-limited video', 'url': 'https://www.vevo.com/watch/justin-timberlake/tunnel-vision-explicit/USRV81300282', 'info_dict': { - 'id': '07FYdnEawAQ', + 'id': 'USRV81300282', 'ext': 'mp4', - 'age_limit': 18, - 'title': 'Justin Timberlake - Tunnel Vision (Explicit)', - 'description': 'md5:64249768eec3bc4276236606ea996373', - 'uploader_id': 'justintimberlakeVEVO', - 'uploader': 'justintimberlakeVEVO', + 'title': 'Tunnel Vision (Explicit)', 'upload_date': '20130703', + 'age_limit': 18, + 'uploader': 'Justin Timberlake', + 'timestamp': 1372888800, }, - 'params': { - 'skip_download': 'true', - }, - 'add_ie': ['Youtube'], }, { 'note': 'No video_info', 'url': 'http://www.vevo.com/watch/k-camp-1/Till-I-Die/USUV71503000', @@ -76,8 +66,11 @@ class VevoIE(InfoExtractor): 'info_dict': { 'id': 'USUV71503000', 'ext': 'mp4', - 'title': 'Till I Die - K Camp ft. T.I.', - 'duration': 193, + 'title': 'Till I Die', + 'upload_date': '20151207', + 'age_limit': 18, + 'uploader': 'K Camp', + 'timestamp': 1449468000, }, }] _SMIL_BASE_URL = 'http://smil.lvl3.vevo.com' @@ -96,7 +89,7 @@ class VevoIE(InfoExtractor): 18: 'dash', } _VERSIONS = { - 0: 'youtube', + 0: 'youtube', # only in AuthenticateVideo videoVersions 1: 'level3', 2: 'akamai', 3: 'level3', @@ -138,67 +131,155 @@ class VevoIE(InfoExtractor): }) return formats + def _initialize_api(self, video_id): + req = sanitized_Request( + 'http://www.vevo.com/auth', data=b'') + webpage = self._download_webpage( + req, None, + note='Retrieving oauth token', + errnote='Unable to retrieve oauth token') + + if 'THIS PAGE IS CURRENTLY UNAVAILABLE IN YOUR REGION' in webpage: + raise ExtractorError( + '%s said: This page is currently unavailable in your region.' % self.IE_NAME, expected=True) + + auth_info = self._parse_json(webpage, video_id) + self._api_url_template = self.http_scheme() + '//apiv2.vevo.com/%s?token=' + auth_info['access_token'] + + def _call_api(self, path, video_id, note, errnote, fatal=True): + return self._download_json(self._api_url_template % path, video_id, note, errnote) + def _real_extract(self, url): video_id = self._match_id(url) - webpage = None - json_url = 'http://videoplayer.vevo.com/VideoService/AuthenticateVideo?isrc=%s' % video_id - response = self._download_json(json_url, video_id) - video_info = response['video'] or {} + response = self._download_json( + json_url, video_id, 'Downloading video info', 'Unable to download info') + video_info = response.get('video') or {} + video_versions = video_info.get('videoVersions') + uploader = None + timestamp = None + view_count = None + formats = [] if not video_info: - ytid = response.get('errorInfo', {}).get('ytid') - if ytid: - return self.url_result(ytid, 'Youtube', ytid) - if response.get('statusCode') != 909: + ytid = response.get('errorInfo', {}).get('ytid') + if ytid: + self.report_warning( + 'Video is geoblocked, trying with the YouTube video %s' % ytid) + return self.url_result(ytid, 'Youtube', ytid) + if 'statusMessage' in response: raise ExtractorError('%s said: %s' % ( self.IE_NAME, response['statusMessage']), expected=True) raise ExtractorError('Unable to extract videos') - if url.startswith('vevo:'): - raise ExtractorError( - 'Please specify full Vevo URL for downloading', expected=True) - webpage = self._download_webpage(url, video_id) + self._initialize_api(video_id) + video_info = self._call_api( + 'video/%s' % video_id, video_id, 'Downloading api video info', + 'Failed to download video info') - title = video_info.get('title') or self._og_search_title(webpage) + video_versions = self._call_api( + 'video/%s/streams' % video_id, video_id, + 'Downloading video versions info', + 'Failed to download video versions info') - smil_parsed = False - formats = [] - for video_version in video_info['videoVersions']: - version = self._VERSIONS.get(video_version['version']) - if version == 'youtube': - return self.url_result( - video_version['id'], 'Youtube', video_version['id']) - else: - source_type = self._SOURCE_TYPES.get(video_version['sourceType']) - renditions = compat_etree_fromstring(video_version['data']) - if source_type == 'http': - for rend in renditions.findall('rendition'): - attr = rend.attrib - formats.append({ - 'url': attr['url'], - 'format_id': '%s-%s' % (version, attr['name']), - 'height': int_or_none(attr.get('frameheight')), - 'width': int_or_none(attr.get('frameWidth')), - 'tbr': int_or_none(attr.get('totalBitrate')), - 'vbr': int_or_none(attr.get('videoBitrate')), - 'abr': int_or_none(attr.get('audioBitrate')), - 'vcodec': attr.get('videoCodec'), - 'acodec': attr.get('audioCodec'), - }) - elif source_type == 'hls': + timestamp = parse_iso8601(video_info.get('releaseDate')) + artists = video_info.get('artists') + if artists: + uploader = artists[0]['name'] + view_count = int_or_none(video_info.get('views', {}).get('total')) + + for video_version in video_versions: + version = self._VERSIONS.get(video_version['version']) + version_url = video_version.get('url') + if not version_url: + continue + + if '.ism' in version_url: + continue + elif '.mpd' in version_url: + formats.extend(self._extract_mpd_formats( + version_url, video_id, mpd_id='dash-%s' % version, + note='Downloading %s MPD information' % version, + errnote='Failed to download %s MPD information' % version, + fatal=False)) + elif '.m3u8' in version_url: formats.extend(self._extract_m3u8_formats( - renditions.find('rendition').attrib['url'], video_id, - 'mp4', 'm3u8_native', m3u8_id='hls-%s' % version, fatal=False)) - elif source_type == 'smil' and not smil_parsed: - formats.extend(self._extract_smil_formats( - renditions.find('rendition').attrib['url'], video_id, False)) - smil_parsed = True + version_url, video_id, 'mp4', 'm3u8_native', + m3u8_id='hls-%s' % version, + note='Downloading %s m3u8 information' % version, + errnote='Failed to download %s m3u8 information' % version, + fatal=False)) + else: + m = re.search(r'''(?xi) + _(?P[0-9]+)x(?P[0-9]+) + _(?P[a-z0-9]+) + _(?P[0-9]+) + _(?P[a-z0-9]+) + _(?P[0-9]+) + \.(?P[a-z0-9]+)''', version_url) + if not m: + continue + + formats.append({ + 'url': version_url, + 'format_id': 'http-%s-%s' % (version, video_version['quality']), + 'vcodec': m.group('vcodec'), + 'acodec': m.group('acodec'), + 'vbr': int(m.group('vbr')), + 'abr': int(m.group('abr')), + 'ext': m.group('ext'), + 'width': int(m.group('width')), + 'height': int(m.group('height')), + }) + else: + timestamp = int_or_none(self._search_regex( + r'/Date\((\d+)\)/', + video_info['releaseDate'], 'release date', fatal=False), + scale=1000) + artists = video_info.get('mainArtists') + if artists: + uploader = artists[0]['artistName'] + + smil_parsed = False + for video_version in video_info['videoVersions']: + version = self._VERSIONS.get(video_version['version']) + if version == 'youtube': + continue + else: + source_type = self._SOURCE_TYPES.get(video_version['sourceType']) + renditions = compat_etree_fromstring(video_version['data']) + if source_type == 'http': + for rend in renditions.findall('rendition'): + attr = rend.attrib + formats.append({ + 'url': attr['url'], + 'format_id': 'http-%s-%s' % (version, attr['name']), + 'height': int_or_none(attr.get('frameheight')), + 'width': int_or_none(attr.get('frameWidth')), + 'tbr': int_or_none(attr.get('totalBitrate')), + 'vbr': int_or_none(attr.get('videoBitrate')), + 'abr': int_or_none(attr.get('audioBitrate')), + 'vcodec': attr.get('videoCodec'), + 'acodec': attr.get('audioCodec'), + }) + elif source_type == 'hls': + formats.extend(self._extract_m3u8_formats( + renditions.find('rendition').attrib['url'], video_id, + 'mp4', 'm3u8_native', m3u8_id='hls-%s' % version, + note='Downloading %s m3u8 information' % version, + errnote='Failed to download %s m3u8 information' % version, + fatal=False)) + elif source_type == 'smil' and version == 'level3' and not smil_parsed: + formats.extend(self._extract_smil_formats( + renditions.find('rendition').attrib['url'], video_id, False)) + smil_parsed = True self._sort_formats(formats) + title = video_info['title'] + is_explicit = video_info.get('isExplicit') if is_explicit is True: age_limit = 18 @@ -207,21 +288,16 @@ class VevoIE(InfoExtractor): else: age_limit = None - timestamp = int_or_none(self._search_regex( - r'/Date\((\d+)\)/', - video_info['launchDate'], 'launch date', fatal=False), - scale=1000) if video_info else None - - duration = video_info.get('duration') or int_or_none( - self._html_search_meta('video:duration', webpage)) + duration = video_info.get('duration') return { 'id': video_id, 'title': title, 'formats': formats, - 'thumbnail': video_info.get('imageUrl'), + 'thumbnail': video_info.get('imageUrl') or video_info.get('thumbnailUrl'), 'timestamp': timestamp, - 'uploader': video_info['mainArtists'][0]['artistName'] if video_info else None, + 'uploader': uploader, 'duration': duration, + 'view_count': view_count, 'age_limit': age_limit, }