]> jfr.im git - yt-dlp.git/blobdiff - yt_dlp/extractor/vevo.py
[ie/mlbtv] Fix extraction (#10296)
[yt-dlp.git] / yt_dlp / extractor / vevo.py
index bc018751102700a66fba956539db94bcc5105ba1..8552a609c9fd76f6ce7f100e313ba5d0e47701b3 100644 (file)
@@ -1,11 +1,8 @@
-import re
 import json
+import re
 
 from .common import InfoExtractor
-from ..compat import (
-    compat_str,
-    compat_HTTPError,
-)
+from ..networking.exceptions import HTTPError
 from ..utils import (
     ExtractorError,
     int_or_none,
@@ -24,19 +21,134 @@ def _extract_json(self, webpage, video_id):
 
 
 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/(?!playlist|genre)(?:[^/]+/(?:[^/]+/)?)?|
            https?://cache\.vevo\.com/m/html/embed\.html\?video=|
            https?://videoplayer\.vevo\.com/embed/embedded\?videoId=|
            https?://embed\.vevo\.com/.*?[?&]isrc=|
+           https?://tv\.vevo\.com/watch/artist/(?:[^/]+)/|
            vevo:)
         (?P<id>[^&?#]+)'''
+    _EMBED_REGEX = [r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//(?:cache\.)?vevo\.com/.+?)\1']
 
-    _TESTS = []
+    _TESTS = [{
+        'url': 'http://www.vevo.com/watch/hurts/somebody-to-die-for/GB1101300280',
+        'md5': '95ee28ee45e70130e3ab02b0f579ae23',
+        'info_dict': {
+            'id': 'GB1101300280',
+            'ext': 'mp4',
+            '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', 'Unable to download info'],
+    }, {
+        'note': 'v3 SMIL format',
+        'url': 'http://www.vevo.com/watch/cassadee-pope/i-wish-i-could-break-your-heart/USUV71302923',
+        'md5': 'f6ab09b034f8c22969020b042e5ac7fc',
+        'info_dict': {
+            'id': 'USUV71302923',
+            'ext': 'mp4',
+            'title': 'Cassadee Pope - I Wish I Could Break Your Heart',
+            'timestamp': 1392796919,
+            'upload_date': '20140219',
+            'uploader': 'Cassadee Pope',
+            'track': 'I Wish I Could Break Your Heart',
+            'artist': 'Cassadee Pope',
+            'genre': 'Country',
+        },
+        'expected_warnings': ['Unable to download SMIL file', 'Unable to download info'],
+    }, {
+        '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,
+            'timestamp': 1372888800,
+            'upload_date': '20130703',
+            'uploader': 'Justin Timberlake',
+            'track': 'Tunnel Vision (Explicit)',
+            'artist': 'Justin Timberlake',
+            'genre': 'Pop',
+        },
+        'expected_warnings': ['Unable to download SMIL file', 'Unable to download info'],
+    }, {
+        '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 ft. T.I. - Till I Die',
+            'age_limit': 18,
+            'timestamp': 1449468000,
+            'upload_date': '20151207',
+            'uploader': 'K Camp',
+            'track': 'Till I Die',
+            'artist': 'K Camp',
+            'genre': 'Hip-Hop',
+        },
+        'expected_warnings': ['Unable to download SMIL file', 'Unable to download info'],
+    }, {
+        'note': 'Featured test',
+        'url': 'https://www.vevo.com/watch/lemaitre/Wait/USUV71402190',
+        'md5': 'd28675e5e8805035d949dc5cf161071d',
+        'info_dict': {
+            'id': 'USUV71402190',
+            'ext': 'mp4',
+            'title': 'Lemaitre ft. LoLo - Wait',
+            'age_limit': 0,
+            'timestamp': 1413432000,
+            'upload_date': '20141016',
+            'uploader': 'Lemaitre',
+            'track': 'Wait',
+            'artist': 'Lemaitre',
+            'genre': 'Electronic',
+        },
+        'expected_warnings': ['Unable to download SMIL file', 'Unable to download info'],
+    }, {
+        '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,
+    }, {
+        # Another case available only via the webpage; using streams/streamsV3 formats
+        # Geo-restricted to Netherlands/Germany
+        'url': 'http://www.vevo.com/watch/boostee/pop-corn-clip-officiel/FR1A91600909',
+        'only_matching': True,
+    }, {
+        'url': 'https://embed.vevo.com/?isrc=USH5V1923499&partnerId=4d61b777-8023-4191-9ede-497ed6c24647&partnerAdCode=',
+        'only_matching': True,
+    }, {
+        'url': 'https://tv.vevo.com/watch/artist/janet-jackson/US0450100550',
+        'only_matching': True,
+    }]
     _VERSIONS = {
         0: 'youtube',  # only in AuthenticateVideo videoVersions
         1: 'level3',
@@ -53,14 +165,14 @@ def _initialize_api(self, video_id):
             data=json.dumps({
                 'client_id': 'SPupX1tvqFEopQ1YS6SS',
                 'grant_type': 'urn:vevo:params:oauth:grant-type:anonymous',
-            }).encode('utf-8'),
+            }).encode(),
             headers={
                 'Content-Type': 'application/json',
             })
 
         if re.search(r'(?i)THIS PAGE IS CURRENTLY UNAVAILABLE IN YOUR REGION', webpage):
             self.raise_geo_restricted(
-                '%s said: This page is currently unavailable in your region' % self.IE_NAME)
+                f'{self.IE_NAME} said: This page is currently unavailable in your region')
 
         auth_info = self._parse_json(webpage, video_id)
         self._api_url_template = self.http_scheme() + '//apiv2.vevo.com/%s?token=' + auth_info['legacy_token']
@@ -69,10 +181,10 @@ def _call_api(self, path, *args, **kwargs):
         try:
             data = self._download_json(self._api_url_template % path, *args, **kwargs)
         except ExtractorError as e:
-            if isinstance(e.cause, compat_HTTPError):
-                errors = self._parse_json(e.cause.read().decode(), None)['errors']
+            if isinstance(e.cause, HTTPError):
+                errors = self._parse_json(e.cause.response.read().decode(), None)['errors']
                 error_message = ', '.join([error['message'] for error in errors])
-                raise ExtractorError('%s said: %s' % (self.IE_NAME, error_message), expected=True)
+                raise ExtractorError(f'{self.IE_NAME} said: {error_message}', expected=True)
             raise
         return data
 
@@ -82,11 +194,11 @@ def _real_extract(self, url):
         self._initialize_api(video_id)
 
         video_info = self._call_api(
-            'video/%s' % video_id, video_id, 'Downloading api video info',
+            f'video/{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,
+            f'video/{video_id}/streams', video_id,
             'Downloading video versions info',
             'Failed to download video versions info',
             fatal=False)
@@ -102,7 +214,7 @@ def _real_extract(self, url):
                 video_versions = [
                     value
                     for key, value in json_data['apollo']['data'].items()
-                    if key.startswith('%s.streams' % video_id)]
+                    if key.startswith(f'{video_id}.streams')]
 
         uploader = None
         artist = None
@@ -125,19 +237,20 @@ def _real_extract(self, 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,
+                    version_url, video_id, mpd_id=f'dash-{version}',
+                    note=f'Downloading {version} MPD information',
+                    errnote=f'Failed to download {version} MPD information',
                     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,
+                    m3u8_id=f'hls-{version}',
+                    note=f'Downloading {version} m3u8 information',
+                    errnote=f'Failed to download {version} m3u8 information',
                     fatal=False))
             else:
                 m = re.search(r'''(?xi)
+                    _(?P<quality>[a-z0-9]+)
                     _(?P<width>[0-9]+)x(?P<height>[0-9]+)
                     _(?P<vcodec>[a-z0-9]+)
                     _(?P<vbr>[0-9]+)
@@ -149,7 +262,7 @@ def _real_extract(self, url):
 
                 formats.append({
                     'url': version_url,
-                    'format_id': 'http-%s-%s' % (version, video_version['quality']),
+                    'format_id': f'http-{version}-{video_version.get("quality") or m.group("quality")}',
                     'vcodec': m.group('vcodec'),
                     'acodec': m.group('acodec'),
                     'vbr': int(m.group('vbr')),
@@ -158,17 +271,16 @@ def _real_extract(self, url):
                     'width': int(m.group('width')),
                     'height': int(m.group('height')),
                 })
-        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
+            artist = f'{artist} ft. {featured_artist}'
+        title = f'{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)
+            and isinstance(genres[0], str) else None)
 
         is_explicit = video_info.get('isExplicit')
         if is_explicit is True:
@@ -224,15 +336,15 @@ def _real_extract(self, url):
                 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())
+                return self.url_result(f'vevo:{video_id}', VevoIE.ie_key())
 
-        playlists = self._extract_json(webpage, playlist_id)['default']['%ss' % playlist_kind]
+        playlists = self._extract_json(webpage, playlist_id)['default'][f'{playlist_kind}s']
 
-        playlist = (list(playlists.values())[0]
+        playlist = (next(iter(playlists.values()))
                     if playlist_kind == 'playlist' else playlists[playlist_id])
 
         entries = [
-            self.url_result('vevo:%s' % src, VevoIE.ie_key())
+            self.url_result(f'vevo:{src}', VevoIE.ie_key())
             for src in playlist['isrcs']]
 
         return self.playlist_result(