]> Raphaƫl G. Git Repositories - youtubedl/blobdiff - youtube_dl/extractor/vevo.py
Update the changelog.
[youtubedl] / youtube_dl / extractor / vevo.py
index c912c3cbe7ae42b221816b50ec6a97139cb13d55..388b4debee27d7331ae7dc351338e3829e539071 100644 (file)
@@ -1,22 +1,37 @@
 from __future__ import unicode_literals
 
 import re
-import xml.etree.ElementTree
 
 from .common import InfoExtractor
+from ..compat import (
+    compat_etree_fromstring,
+    compat_str,
+    compat_urlparse,
+)
 from ..utils import (
-    compat_urllib_request,
     ExtractorError,
+    int_or_none,
+    sanitized_Request,
+    parse_iso8601,
 )
 
 
-class VevoIE(InfoExtractor):
-    """
+class VevoBaseIE(InfoExtractor):
+    def _extract_json(self, webpage, video_id, item):
+        return self._parse_json(
+            self._search_regex(
+                r'window\.__INITIAL_STORE__\s*=\s*({.+?});\s*</script>',
+                webpage, 'initial store'),
+            video_id)['default'][item]
+
+
+class VevoIE(VevoBaseIE):
+    '''
     Accepts urls from vevo.com or in the format 'vevo:{id}'
     (currently used by MTVIE and MySpaceIE)
-    """
+    '''
     _VALID_URL = r'''(?x)
-        (?:https?://www\.vevo\.com/watch/(?:[^/]+/(?:[^/]+/)?)?|
+        (?:https?://www\.vevo\.com/watch/(?!playlist|genre)(?:[^/]+/(?:[^/]+/)?)?|
            https?://cache\.vevo\.com/m/html/embed\.html\?video=|
            https?://videoplayer\.vevo\.com/embed/embedded\?videoId=|
            vevo:)
@@ -24,19 +39,19 @@ class VevoIE(InfoExtractor):
 
     _TESTS = [{
         'url': 'http://www.vevo.com/watch/hurts/somebody-to-die-for/GB1101300280',
-        "md5": "95ee28ee45e70130e3ab02b0f579ae23",
+        'md5': '95ee28ee45e70130e3ab02b0f579ae23',
         'info_dict': {
             'id': 'GB1101300280',
             'ext': 'mp4',
-            "upload_date": "20130624",
-            "uploader": "Hurts",
-            "title": "Somebody to Die For",
-            "duration": 230.12,
-            "width": 1920,
-            "height": 1080,
-            # timestamp and upload_date are often incorrect; seem to change randomly
-            'timestamp': int,
-        }
+            'title': 'Hurts - Somebody to Die For',
+            'timestamp': 1372057200,
+            'upload_date': '20130624',
+            'uploader': 'Hurts',
+            'track': 'Somebody to Die For',
+            'artist': 'Hurts',
+            'genre': 'Pop',
+        },
+        'expected_warnings': ['Unable to download SMIL file'],
     }, {
         'note': 'v3 SMIL format',
         'url': 'http://www.vevo.com/watch/cassadee-pope/i-wish-i-could-break-your-heart/USUV71302923',
@@ -44,82 +59,102 @@ class VevoIE(InfoExtractor):
         'info_dict': {
             'id': 'USUV71302923',
             'ext': 'mp4',
+            'title': 'Cassadee Pope - I Wish I Could Break Your Heart',
+            'timestamp': 1392796919,
             'upload_date': '20140219',
             'uploader': 'Cassadee Pope',
-            'title': 'I Wish I Could Break Your Heart',
-            'duration': 226.101,
-            'age_limit': 0,
-            'timestamp': int,
-        }
+            'track': 'I Wish I Could Break Your Heart',
+            'artist': 'Cassadee Pope',
+            'genre': 'Country',
+        },
+        'expected_warnings': ['Unable to download SMIL file'],
     }, {
         'note': 'Age-limited video',
         'url': 'https://www.vevo.com/watch/justin-timberlake/tunnel-vision-explicit/USRV81300282',
         'info_dict': {
             'id': 'USRV81300282',
             'ext': 'mp4',
+            'title': 'Justin Timberlake - Tunnel Vision (Explicit)',
             'age_limit': 18,
-            'title': 'Tunnel Vision (Explicit)',
+            'timestamp': 1372888800,
+            'upload_date': '20130703',
             'uploader': 'Justin Timberlake',
-            'upload_date': 're:2013070[34]',
-            'timestamp': int,
+            'track': 'Tunnel Vision (Explicit)',
+            'artist': 'Justin Timberlake',
+            'genre': 'Pop',
         },
-        'params': {
-            'skip_download': 'true',
-        }
+        'expected_warnings': ['Unable to download SMIL file'],
+    }, {
+        'note': 'No video_info',
+        'url': 'http://www.vevo.com/watch/k-camp-1/Till-I-Die/USUV71503000',
+        'md5': '8b83cc492d72fc9cf74a02acee7dc1b0',
+        'info_dict': {
+            'id': 'USUV71503000',
+            'ext': 'mp4',
+            'title': 'K Camp - Till I Die',
+            'age_limit': 18,
+            'timestamp': 1449468000,
+            'upload_date': '20151207',
+            'uploader': 'K Camp',
+            'track': 'Till I Die',
+            'artist': 'K Camp',
+            'genre': 'Rap/Hip-Hop',
+        },
+    }, {
+        'note': 'Only available via webpage',
+        'url': 'http://www.vevo.com/watch/GBUV71600656',
+        'md5': '67e79210613865b66a47c33baa5e37fe',
+        'info_dict': {
+            'id': 'GBUV71600656',
+            'ext': 'mp4',
+            'title': 'ABC - Viva Love',
+            'age_limit': 0,
+            'timestamp': 1461830400,
+            'upload_date': '20160428',
+            'uploader': 'ABC',
+            'track': 'Viva Love',
+            'artist': 'ABC',
+            'genre': 'Pop',
+        },
+        'expected_warnings': ['Failed to download video versions info'],
+    }, {
+        # no genres available
+        'url': 'http://www.vevo.com/watch/INS171400764',
+        'only_matching': True,
     }]
-    _SMIL_BASE_URL = 'http://smil.lvl3.vevo.com/'
+    _SMIL_BASE_URL = 'http://smil.lvl3.vevo.com'
+    _SOURCE_TYPES = {
+        0: 'youtube',
+        1: 'brightcove',
+        2: 'http',
+        3: 'hls_ios',
+        4: 'hls',
+        5: 'smil',  # http
+        7: 'f4m_cc',
+        8: 'f4m_ak',
+        9: 'f4m_l3',
+        10: 'ism',
+        13: 'smil',  # rtmp
+        18: 'dash',
+    }
+    _VERSIONS = {
+        0: 'youtube',  # only in AuthenticateVideo videoVersions
+        1: 'level3',
+        2: 'akamai',
+        3: 'level3',
+        4: 'amazon',
+    }
 
-    def _real_initialize(self):
-        req = compat_urllib_request.Request(
-            'http://www.vevo.com/auth', data=b'')
-        webpage = self._download_webpage(
-            req, None,
-            note='Retrieving oauth token',
-            errnote='Unable to retrieve oauth token',
-            fatal=False)
-        if webpage is False:
-            self._oauth_token = None
-        else:
-            self._oauth_token = self._search_regex(
-                r'access_token":\s*"([^"]+)"',
-                webpage, 'access token', fatal=False)
-
-    def _formats_from_json(self, video_info):
-        last_version = {'version': -1}
-        for version in video_info['videoVersions']:
-            # These are the HTTP downloads, other types are for different manifests
-            if version['sourceType'] == 2:
-                if version['version'] > last_version['version']:
-                    last_version = version
-        if last_version['version'] == -1:
-            raise ExtractorError('Unable to extract last version of the video')
-
-        renditions = xml.etree.ElementTree.fromstring(last_version['data'])
-        formats = []
-        # Already sorted from worst to best quality
-        for rend in renditions.findall('rendition'):
-            attr = rend.attrib
-            format_note = '%(videoCodec)s@%(videoBitrate)4sk, %(audioCodec)s@%(audioBitrate)3sk' % attr
-            formats.append({
-                'url': attr['url'],
-                'format_id': attr['name'],
-                'format_note': format_note,
-                'height': int(attr['frameheight']),
-                'width': int(attr['frameWidth']),
-            })
-        return formats
-
-    def _formats_from_smil(self, smil_xml):
+    def _parse_smil_formats(self, smil, smil_url, video_id, namespace=None, f4m_params=None, transform_rtmp_url=None):
         formats = []
-        smil_doc = xml.etree.ElementTree.fromstring(smil_xml.encode('utf-8'))
-        els = smil_doc.findall('.//{http://www.w3.org/2001/SMIL20/Language}video')
+        els = smil.findall('.//{http://www.w3.org/2001/SMIL20/Language}video')
         for el in els:
             src = el.attrib['src']
             m = re.match(r'''(?xi)
                 (?P<ext>[a-z0-9]+):
                 (?P<path>
                     [/a-z0-9]+     # The directory and main part of the URL
-                    _(?P<cbr>[0-9]+)k
+                    _(?P<tbr>[0-9]+)k
                     _(?P<width>[0-9]+)x(?P<height>[0-9]+)
                     _(?P<vcodec>[a-z0-9]+)
                     _(?P<vbr>[0-9]+)
@@ -133,9 +168,10 @@ class VevoIE(InfoExtractor):
             format_url = self._SMIL_BASE_URL + m.group('path')
             formats.append({
                 'url': format_url,
-                'format_id': 'SMIL_' + m.group('cbr'),
+                'format_id': 'smil_' + m.group('tbr'),
                 'vcodec': m.group('vcodec'),
                 'acodec': m.group('acodec'),
+                'tbr': int(m.group('tbr')),
                 'vbr': int(m.group('vbr')),
                 'abr': int(m.group('abr')),
                 'ext': m.group('ext'),
@@ -144,40 +180,172 @@ class VevoIE(InfoExtractor):
             })
         return formats
 
-    def _download_api_formats(self, video_id):
-        if not self._oauth_token:
-            self._downloader.report_warning(
-                'No oauth token available, skipping API HLS download')
-            return []
-
-        api_url = 'https://apiv2.vevo.com/video/%s/streams/hls?token=%s' % (
-            video_id, self._oauth_token)
-        api_data = self._download_json(
-            api_url, video_id,
-            note='Downloading HLS formats',
-            errnote='Failed to download HLS format list', fatal=False)
-        if api_data is None:
-            return []
-
-        m3u8_url = api_data[0]['url']
-        return self._extract_m3u8_formats(
-            m3u8_url, video_id, entry_protocol='m3u8_native', ext='mp4',
-            preference=0)
+    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:
+            self.raise_geo_restricted(
+                '%s said: This page is currently unavailable in your region' % self.IE_NAME)
+
+        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, *args, **kwargs):
+        return self._download_json(self._api_url_template % path, *args, **kwargs)
 
     def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
-        video_id = mobj.group('id')
+        video_id = self._match_id(url)
 
-        json_url = 'http://videoplayer.vevo.com/VideoService/AuthenticateVideo?isrc=%s' % video_id
-        response = self._download_json(json_url, video_id)
-        video_info = response['video']
+        json_url = 'http://api.vevo.com/VideoService/AuthenticateVideo?isrc=%s' % video_id
+        response = self._download_json(
+            json_url, video_id, 'Downloading video info',
+            'Unable to download info', fatal=False) or {}
+        video_info = response.get('video') or {}
+        artist = None
+        featured_artist = None
+        uploader = None
+        view_count = None
+        formats = []
 
         if not video_info:
-            if 'statusMessage' in response:
-                raise ExtractorError('%s said: %s' % (self.IE_NAME, response['statusMessage']), expected=True)
-            raise ExtractorError('Unable to extract videos')
+            try:
+                self._initialize_api(video_id)
+            except ExtractorError:
+                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)
+
+                raise
 
-        formats = self._formats_from_json(video_info)
+            video_info = self._call_api(
+                'video/%s' % video_id, video_id, 'Downloading api video info',
+                'Failed to download video info')
+
+            video_versions = self._call_api(
+                'video/%s/streams' % video_id, video_id,
+                'Downloading video versions info',
+                'Failed to download video versions info',
+                fatal=False)
+
+            # Some videos are only available via webpage (e.g.
+            # https://github.com/rg3/youtube-dl/issues/9366)
+            if not video_versions:
+                webpage = self._download_webpage(url, video_id)
+                video_versions = self._extract_json(webpage, video_id, 'streams')[video_id][0]
+
+            timestamp = parse_iso8601(video_info.get('releaseDate'))
+            artists = video_info.get('artists')
+            if artists:
+                artist = 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(
+                        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<width>[0-9]+)x(?P<height>[0-9]+)
+                        _(?P<vcodec>[a-z0-9]+)
+                        _(?P<vbr>[0-9]+)
+                        _(?P<acodec>[a-z0-9]+)
+                        _(?P<abr>[0-9]+)
+                        \.(?P<ext>[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:
+                artist = uploader = artists[0]['artistName']
+
+            featured_artists = video_info.get('featuredArtists')
+            if featured_artists:
+                featured_artist = featured_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)
+
+        track = video_info['title']
+        if featured_artist:
+            artist = '%s ft. %s' % (artist, featured_artist)
+        title = '%s - %s' % (artist, track) if artist else track
+
+        genres = video_info.get('genres')
+        genre = (
+            genres[0] if genres and isinstance(genres, list) and
+            isinstance(genres[0], compat_str) else None)
 
         is_explicit = video_info.get('isExplicit')
         if is_explicit is True:
@@ -187,20 +355,87 @@ class VevoIE(InfoExtractor):
         else:
             age_limit = None
 
-        # Download via HLS API
-        formats.extend(self._download_api_formats(video_id))
-
-        self._sort_formats(formats)
-        timestamp_ms = int(self._search_regex(
-            r'/Date\((\d+)\)/', video_info['launchDate'], 'launch date'))
+        duration = video_info.get('duration')
 
         return {
             'id': video_id,
-            'title': video_info['title'],
+            'title': title,
             'formats': formats,
-            'thumbnail': video_info['imageUrl'],
-            'timestamp': timestamp_ms // 1000,
-            'uploader': video_info['mainArtists'][0]['artistName'],
-            'duration': video_info['duration'],
+            'thumbnail': video_info.get('imageUrl') or video_info.get('thumbnailUrl'),
+            'timestamp': timestamp,
+            'uploader': uploader,
+            'duration': duration,
+            'view_count': view_count,
             'age_limit': age_limit,
+            'track': track,
+            'artist': uploader,
+            'genre': genre,
         }
+
+
+class VevoPlaylistIE(VevoBaseIE):
+    _VALID_URL = r'https?://www\.vevo\.com/watch/(?P<kind>playlist|genre)/(?P<id>[^/?#&]+)'
+
+    _TESTS = [{
+        'url': 'http://www.vevo.com/watch/playlist/dadbf4e7-b99f-4184-9670-6f0e547b6a29',
+        'info_dict': {
+            'id': 'dadbf4e7-b99f-4184-9670-6f0e547b6a29',
+            'title': 'Best-Of: Birdman',
+        },
+        'playlist_count': 10,
+    }, {
+        'url': 'http://www.vevo.com/watch/genre/rock',
+        'info_dict': {
+            'id': 'rock',
+            'title': 'Rock',
+        },
+        'playlist_count': 20,
+    }, {
+        'url': 'http://www.vevo.com/watch/playlist/dadbf4e7-b99f-4184-9670-6f0e547b6a29?index=0',
+        'md5': '32dcdfddddf9ec6917fc88ca26d36282',
+        'info_dict': {
+            'id': 'USCMV1100073',
+            'ext': 'mp4',
+            'title': 'Birdman - Y.U. MAD',
+            'timestamp': 1323417600,
+            'upload_date': '20111209',
+            'uploader': 'Birdman',
+            'track': 'Y.U. MAD',
+            'artist': 'Birdman',
+            'genre': 'Rap/Hip-Hop',
+        },
+        'expected_warnings': ['Unable to download SMIL file'],
+    }, {
+        'url': 'http://www.vevo.com/watch/genre/rock?index=0',
+        'only_matching': True,
+    }]
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        playlist_id = mobj.group('id')
+        playlist_kind = mobj.group('kind')
+
+        webpage = self._download_webpage(url, playlist_id)
+
+        qs = compat_urlparse.parse_qs(compat_urlparse.urlparse(url).query)
+        index = qs.get('index', [None])[0]
+
+        if index:
+            video_id = self._search_regex(
+                r'<meta[^>]+content=(["\'])vevo://video/(?P<id>.+?)\1[^>]*>',
+                webpage, 'video id', default=None, group='id')
+            if video_id:
+                return self.url_result('vevo:%s' % video_id, VevoIE.ie_key())
+
+        playlists = self._extract_json(webpage, playlist_id, '%ss' % playlist_kind)
+
+        playlist = (list(playlists.values())[0]
+                    if playlist_kind == 'playlist' else playlists[playlist_id])
+
+        entries = [
+            self.url_result('vevo:%s' % src, VevoIE.ie_key())
+            for src in playlist['isrcs']]
+
+        return self.playlist_result(
+            entries, playlist.get('playlistId') or playlist_id,
+            playlist.get('name'), playlist.get('description'))