Merge branch 'master' into PornHub-issue-16078

youtube_dl/extractor/asiancrush.py

@@ -8,7 +8,6 @@ from .kaltura import KalturaIE
 from ..utils import (
     extract_attributes,
     remove_end,
-    urlencode_postdata,
 )
 
 
@@ -34,19 +33,40 @@ class AsianCrushIE(InfoExtractor):
     def _real_extract(self, url):
         video_id = self._match_id(url)
 
-        data = self._download_json(
+        webpage = self._download_webpage(url, video_id)
-            'https://www.asiancrush.com/wp-admin/admin-ajax.php', video_id,
-            data=urlencode_postdata({
-                'postid': video_id,
-                'action': 'get_channel_kaltura_vars',
-            }))
 
-        entry_id = data['entry_id']
+        entry_id, partner_id, title = [None] * 3
+
+        vars = self._parse_json(
+            self._search_regex(
+                r'iEmbedVars\s*=\s*({.+?})', webpage, 'embed vars',
+                default='{}'), video_id, fatal=False)
+        if vars:
+            entry_id = vars.get('entry_id')
+            partner_id = vars.get('partner_id')
+            title = vars.get('vid_label')
+
+        if not entry_id:
+            entry_id = self._search_regex(
+                r'\bentry_id["\']\s*:\s*["\'](\d+)', webpage, 'entry id')
+
+        player = self._download_webpage(
+            'https://api.asiancrush.com/embeddedVideoPlayer', video_id,
+            query={'id': entry_id})
+
+        kaltura_id = self._search_regex(
+            r'entry_id["\']\s*:\s*(["\'])(?P<id>(?:(?!\1).)+)\1', player,
+            'kaltura id', group='id')
+
+        if not partner_id:
+            partner_id = self._search_regex(
+                r'/p(?:artner_id)?/(\d+)', player, 'partner id',
+                default='513551')
 
         return self.url_result(
-            'kaltura:%s:%s' % (data['partner_id'], entry_id),
+            'kaltura:%s:%s' % (partner_id, kaltura_id),
-            ie=KalturaIE.ie_key(), video_id=entry_id,
+            ie=KalturaIE.ie_key(), video_id=kaltura_id,
-            video_title=data.get('vid_label'))
+            video_title=title)
 
 
 class AsianCrushPlaylistIE(InfoExtractor):

youtube_dl/extractor/common.py

@@ -211,6 +211,11 @@ class InfoExtractor(object):
                     If not explicitly set, calculated from timestamp.
     uploader_id:    Nickname or id of the video uploader.
     uploader_url:   Full URL to a personal webpage of the video uploader.
+    channel:        Full name of the channel the video is uploaded on.
+                    Note that channel fields may or may not repeat uploader
+                    fields. This depends on a particular extractor.
+    channel_id:     Id of the channel.
+    channel_url:    Full URL to a channel webpage.
     location:       Physical location where the video was filmed.
     subtitles:      The available subtitles as a dictionary in the format
                     {tag: subformats}. "tag" is usually a language code, and
@@ -1701,9 +1706,9 @@ class InfoExtractor(object):
                 # However, this is not always respected, for example, [2]
                 # contains EXT-X-STREAM-INF tag which references AUDIO
                 # rendition group but does not have CODECS and despite
-                # referencing audio group an audio group, it represents
+                # referencing an audio group it represents a complete
-                # a complete (with audio and video) format. So, for such cases
+                # (with audio and video) format. So, for such cases we will
-                # we will ignore references to rendition groups and treat them
+                # ignore references to rendition groups and treat them
                 # as complete formats.
                 if audio_group_id and codecs and f.get('vcodec') != 'none':
                     audio_group = groups.get(audio_group_id)

youtube_dl/extractor/crunchyroll.py

@@ -445,6 +445,10 @@ Format: Layer, Start, End, Style, Name, MarginL, MarginR, MarginV, Effect, Text
             webpage, 'vilos media', default='{}'), video_id)
         media_metadata = media.get('metadata') or {}
 
+        language = self._search_regex(
+            r'(?:vilos\.config\.player\.language|LOCALE)\s*=\s*(["\'])(?P<lang>(?:(?!\1).)+)\1',
+            webpage, 'language', default=None, group='lang')
+
         video_title = self._html_search_regex(
             r'(?s)<h1[^>]*>((?:(?!<h1).)*?<span[^>]+itemprop=["\']title["\'][^>]*>(?:(?!<h1).)+?)</h1>',
             webpage, 'video_title')
@@ -466,9 +470,22 @@ Format: Layer, Start, End, Style, Name, MarginL, MarginR, MarginV, Effect, Text
 
         formats = []
         for stream in media.get('streams', []):
-            formats.extend(self._extract_vrv_formats(
+            audio_lang = stream.get('audio_lang')
+            hardsub_lang = stream.get('hardsub_lang')
+            vrv_formats = self._extract_vrv_formats(
                 stream.get('url'), video_id, stream.get('format'),
-                stream.get('audio_lang'), stream.get('hardsub_lang')))
+                audio_lang, hardsub_lang)
+            for f in vrv_formats:
+                if not hardsub_lang:
+                    f['preference'] = 1
+                language_preference = 0
+                if audio_lang == language:
+                    language_preference += 1
+                if hardsub_lang == language:
+                    language_preference += 1
+                if language_preference:
+                    f['language_preference'] = language_preference
+            formats.extend(vrv_formats)
         if not formats:
             available_fmts = []
             for a, fmt in re.findall(r'(<a[^>]+token=["\']showmedia\.([0-9]{3,4})p["\'][^>]+>)', webpage):
@@ -557,7 +574,7 @@ Format: Layer, Start, End, Style, Name, MarginL, MarginR, MarginV, Effect, Text
                         'ext': 'flv',
                     })
                     formats.append(format_info)
-        self._sort_formats(formats, ('height', 'width', 'tbr', 'fps'))
+        self._sort_formats(formats, ('preference', 'language_preference', 'height', 'width', 'tbr', 'fps'))
 
         metadata = self._call_rpc_api(
             'VideoPlayer_GetMediaMetadata', video_id,

youtube_dl/extractor/fourtube.py

@@ -3,15 +3,45 @@ from __future__ import unicode_literals
 import re
 
 from .common import InfoExtractor
-from ..compat import compat_urlparse
+from ..compat import (
+    compat_b64decode,
+    compat_str,
+    compat_urllib_parse_unquote,
+    compat_urlparse,
+)
 from ..utils import (
+    int_or_none,
     parse_duration,
     parse_iso8601,
+    str_or_none,
     str_to_int,
+    try_get,
+    unified_timestamp,
+    url_or_none,
 )
 
 
 class FourTubeBaseIE(InfoExtractor):
+    _TKN_HOST = 'tkn.kodicdn.com'
+
+    def _extract_formats(self, url, video_id, media_id, sources):
+        token_url = 'https://%s/%s/desktop/%s' % (
+            self._TKN_HOST, media_id, '+'.join(sources))
+
+        parsed_url = compat_urlparse.urlparse(url)
+        tokens = self._download_json(token_url, video_id, data=b'', headers={
+            'Origin': '%s://%s' % (parsed_url.scheme, parsed_url.hostname),
+            'Referer': url,
+        })
+        formats = [{
+            'url': tokens[format]['token'],
+            'format_id': format + 'p',
+            'resolution': format + 'p',
+            'quality': int(format),
+        } for format in sources]
+        self._sort_formats(formats)
+        return formats
+
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
         kind, video_id, display_id = mobj.group('kind', 'id', 'display_id')
@@ -68,21 +98,7 @@ class FourTubeBaseIE(InfoExtractor):
             media_id = params[0]
             sources = ['%s' % p for p in params[2]]
 
-        token_url = 'https://tkn.kodicdn.com/{0}/desktop/{1}'.format(
+        formats = self._extract_formats(url, video_id, media_id, sources)
-            media_id, '+'.join(sources))
-
-        parsed_url = compat_urlparse.urlparse(url)
-        tokens = self._download_json(token_url, video_id, data=b'', headers={
-            'Origin': '%s://%s' % (parsed_url.scheme, parsed_url.hostname),
-            'Referer': url,
-        })
-        formats = [{
-            'url': tokens[format]['token'],
-            'format_id': format + 'p',
-            'resolution': format + 'p',
-            'quality': int(format),
-        } for format in sources]
-        self._sort_formats(formats)
 
         return {
             'id': video_id,
@@ -164,6 +180,7 @@ class FuxIE(FourTubeBaseIE):
 class PornTubeIE(FourTubeBaseIE):
     _VALID_URL = r'https?://(?:(?P<kind>www|m)\.)?porntube\.com/(?:videos/(?P<display_id>[^/]+)_|embed/)(?P<id>\d+)'
     _URL_TEMPLATE = 'https://www.porntube.com/videos/video_%s'
+    _TKN_HOST = 'tkn.porntube.com'
     _TESTS = [{
         'url': 'https://www.porntube.com/videos/teen-couple-doing-anal_7089759',
         'info_dict': {
@@ -171,13 +188,32 @@ class PornTubeIE(FourTubeBaseIE):
             'ext': 'mp4',
             'title': 'Teen couple doing anal',
             'uploader': 'Alexy',
-            'uploader_id': 'Alexy',
+            'uploader_id': '91488',
             'upload_date': '20150606',
             'timestamp': 1433595647,
             'duration': 5052,
             'view_count': int,
             'like_count': int,
-            'categories': list,
+            'age_limit': 18,
+        },
+        'params': {
+            'skip_download': True,
+        },
+    }, {
+        'url': 'https://www.porntube.com/videos/squirting-teen-ballerina-ecg_1331406',
+        'info_dict': {
+            'id': '1331406',
+            'ext': 'mp4',
+            'title': 'Squirting Teen Ballerina on ECG',
+            'uploader': 'Exploited College Girls',
+            'uploader_id': '665',
+            'channel': 'Exploited College Girls',
+            'channel_id': '665',
+            'upload_date': '20130920',
+            'timestamp': 1379685485,
+            'duration': 851,
+            'view_count': int,
+            'like_count': int,
             'age_limit': 18,
         },
         'params': {
@@ -191,6 +227,55 @@ class PornTubeIE(FourTubeBaseIE):
         'only_matching': True,
     }]
 
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        video_id, display_id = mobj.group('id', 'display_id')
+
+        webpage = self._download_webpage(url, display_id)
+
+        video = self._parse_json(
+            self._search_regex(
+                r'INITIALSTATE\s*=\s*(["\'])(?P<value>(?:(?!\1).)+)\1',
+                webpage, 'data', group='value'), video_id,
+            transform_source=lambda x: compat_urllib_parse_unquote(
+                compat_b64decode(x).decode('utf-8')))['page']['video']
+
+        title = video['title']
+        media_id = video['mediaId']
+        sources = [compat_str(e['height'])
+                   for e in video['encodings'] if e.get('height')]
+        formats = self._extract_formats(url, video_id, media_id, sources)
+
+        thumbnail = url_or_none(video.get('masterThumb'))
+        uploader = try_get(video, lambda x: x['user']['username'], compat_str)
+        uploader_id = str_or_none(try_get(
+            video, lambda x: x['user']['id'], int))
+        channel = try_get(video, lambda x: x['channel']['name'], compat_str)
+        channel_id = str_or_none(try_get(
+            video, lambda x: x['channel']['id'], int))
+        like_count = int_or_none(video.get('likes'))
+        dislike_count = int_or_none(video.get('dislikes'))
+        view_count = int_or_none(video.get('playsQty'))
+        duration = int_or_none(video.get('durationInSeconds'))
+        timestamp = unified_timestamp(video.get('publishedAt'))
+
+        return {
+            'id': video_id,
+            'title': title,
+            'formats': formats,
+            'thumbnail': thumbnail,
+            'uploader': uploader or channel,
+            'uploader_id': uploader_id or channel_id,
+            'channel': channel,
+            'channel_id': channel_id,
+            'timestamp': timestamp,
+            'like_count': like_count,
+            'dislike_count': dislike_count,
+            'view_count': view_count,
+            'duration': duration,
+            'age_limit': 18,
+        }
+
 
 class PornerBrosIE(FourTubeBaseIE):
     _VALID_URL = r'https?://(?:(?P<kind>www|m)\.)?pornerbros\.com/(?:videos/(?P<display_id>[^/]+)_|embed/)(?P<id>\d+)'

youtube_dl/extractor/pornhub.py

@@ -40,6 +40,7 @@ class PornHubIE(InfoExtractor):
             'ext': 'mp4',
             'title': 'Seductive Indian beauty strips down and fingers her pink pussy',
             'uploader': 'Babes',
+            'upload_date': '20130628',
             'duration': 361,
             'view_count': int,
             'like_count': int,
@@ -57,6 +58,7 @@ class PornHubIE(InfoExtractor):
             'ext': 'mp4',
             'title': '重庆婷婷女王足交',
             'uploader': 'Unknown',
+            'upload_date': '20150213',
             'duration': 1753,
             'view_count': int,
             'like_count': int,
@@ -237,8 +239,14 @@ class PornHubIE(InfoExtractor):
                 video_urls.append((video_url, None))
                 video_urls_set.add(video_url)
 
+        upload_date = None
         formats = []
         for video_url, height in video_urls:
+            if not upload_date:
+                upload_date = self._search_regex(
+                    r'/(\d{6}/\d{2})/', video_url, 'upload data', default=None)
+                if upload_date:
+                    upload_date = upload_date.replace('/', '')
             tbr = None
             mobj = re.search(r'(?P<height>\d+)[pP]?_(?P<tbr>\d+)[kK]', video_url)
             if mobj:
@@ -278,6 +286,7 @@ class PornHubIE(InfoExtractor):
         return {
             'id': video_id,
             'uploader': video_uploader,
+            'upload_date': upload_date,
             'title': title,
             'thumbnail': thumbnail,
             'duration': duration,

youtube_dl/extractor/twitch.py

@@ -559,7 +559,8 @@ class TwitchStreamIE(TwitchBaseIE):
                     TwitchAllVideosIE,
                     TwitchUploadsIE,
                     TwitchPastBroadcastsIE,
-                    TwitchHighlightsIE))
+                    TwitchHighlightsIE,
+                    TwitchClipsIE))
                 else super(TwitchStreamIE, cls).suitable(url))
 
     def _real_extract(self, url):
@@ -633,7 +634,7 @@ class TwitchStreamIE(TwitchBaseIE):
 
 class TwitchClipsIE(TwitchBaseIE):
     IE_NAME = 'twitch:clips'
-    _VALID_URL = r'https?://clips\.twitch\.tv/(?:[^/]+/)*(?P<id>[^/?#&]+)'
+    _VALID_URL = r'https?://(?:clips\.twitch\.tv/(?:[^/]+/)*|(?:www\.)?twitch\.tv/[^/]+/clip/)(?P<id>[^/?#&]+)'
 
     _TESTS = [{
         'url': 'https://clips.twitch.tv/FaintLightGullWholeWheat',
@@ -653,6 +654,9 @@ class TwitchClipsIE(TwitchBaseIE):
         # multiple formats
         'url': 'https://clips.twitch.tv/rflegendary/UninterestedBeeDAESuppy',
         'only_matching': True,
+    }, {
+        'url': 'https://www.twitch.tv/sergeynixon/clip/StormyThankfulSproutFutureMan',
+        'only_matching': True,
     }]
 
     def _real_extract(self, url):

youtube_dl/extractor/vimeo.py

@@ -299,10 +299,13 @@ class VimeoIE(VimeoBaseInfoExtractor):
                 'uploader_url': r're:https?://(?:www\.)?vimeo\.com/atencio',
                 'uploader_id': 'atencio',
                 'uploader': 'Peter Atencio',
+                'channel_id': 'keypeele',
+                'channel_url': r're:https?://(?:www\.)?vimeo\.com/channels/keypeele',
                 'timestamp': 1380339469,
                 'upload_date': '20130928',
                 'duration': 187,
             },
+            'expected_warnings': ['Unable to download JSON metadata'],
         },
         {
             'url': 'http://vimeo.com/76979871',
@@ -355,11 +358,13 @@ class VimeoIE(VimeoBaseInfoExtractor):
             'url': 'https://vimeo.com/channels/tributes/6213729',
             'info_dict': {
                 'id': '6213729',
-                'ext': 'mov',
+                'ext': 'mp4',
                 'title': 'Vimeo Tribute: The Shining',
                 'uploader': 'Casey Donahue',
                 'uploader_url': r're:https?://(?:www\.)?vimeo\.com/caseydonahue',
                 'uploader_id': 'caseydonahue',
+                'channel_url': r're:https?://(?:www\.)?vimeo\.com/channels/tributes',
+                'channel_id': 'tributes',
                 'timestamp': 1250886430,
                 'upload_date': '20090821',
                 'description': 'md5:bdbf314014e58713e6e5b66eb252f4a6',
@@ -465,6 +470,9 @@ class VimeoIE(VimeoBaseInfoExtractor):
         if 'Referer' not in headers:
             headers['Referer'] = url
 
+        channel_id = self._search_regex(
+            r'vimeo\.com/channels/([^/]+)', url, 'channel id', default=None)
+
         # Extract ID from URL
         mobj = re.match(self._VALID_URL, url)
         video_id = mobj.group('id')
@@ -563,19 +571,23 @@ class VimeoIE(VimeoBaseInfoExtractor):
             if config.get('view') == 4:
                 config = self._verify_player_video_password(redirect_url, video_id)
 
+        vod = config.get('video', {}).get('vod', {})
+
         def is_rented():
             if '>You rented this title.<' in webpage:
                 return True
             if config.get('user', {}).get('purchased'):
                 return True
-            label = try_get(
+            for purchase_option in vod.get('purchase_options', []):
-                config, lambda x: x['video']['vod']['purchase_options'][0]['label_string'], compat_str)
+                if purchase_option.get('purchased'):
-            if label and label.startswith('You rented this'):
+                    return True
-                return True
+                label = purchase_option.get('label_string')
+                if label and (label.startswith('You rented this') or label.endswith(' remaining')):
+                    return True
             return False
 
-        if is_rented():
+        if is_rented() and vod.get('is_trailer'):
-            feature_id = config.get('video', {}).get('vod', {}).get('feature_id')
+            feature_id = vod.get('feature_id')
             if feature_id and not data.get('force_feature_id', False):
                 return self.url_result(smuggle_url(
                     'https://player.vimeo.com/player/%s' % feature_id,
@@ -652,6 +664,8 @@ class VimeoIE(VimeoBaseInfoExtractor):
                 r'<link[^>]+rel=["\']license["\'][^>]+href=(["\'])(?P<license>(?:(?!\1).)+)\1',
                 webpage, 'license', default=None, group='license')
 
+        channel_url = 'https://vimeo.com/channels/%s' % channel_id if channel_id else None
+
         info_dict = {
             'id': video_id,
             'formats': formats,
@@ -662,6 +676,8 @@ class VimeoIE(VimeoBaseInfoExtractor):
             'like_count': like_count,
             'comment_count': comment_count,
             'license': cc_license,
+            'channel_id': channel_id,
+            'channel_url': channel_url,
         }
 
         info_dict = merge_dicts(info_dict, info_dict_config, json_ld)

youtube_dl/extractor/vrv.py

@@ -90,7 +90,13 @@ class VRVIE(VRVBaseIE):
     def _extract_vrv_formats(self, url, video_id, stream_format, audio_lang, hardsub_lang):
         if not url or stream_format not in ('hls', 'dash'):
             return []
-        stream_id = hardsub_lang or audio_lang
+        assert audio_lang or hardsub_lang
+        stream_id_list = []
+        if audio_lang:
+            stream_id_list.append('audio-%s' % audio_lang)
+        if hardsub_lang:
+            stream_id_list.append('hardsub-%s' % hardsub_lang)
+        stream_id = '-'.join(stream_id_list)
         format_id = '%s-%s' % (stream_format, stream_id)
         if stream_format == 'hls':
             adaptive_formats = self._extract_m3u8_formats(

youtube_dl/extractor/youtube.py

@@ -490,6 +490,8 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
                 'uploader': 'Philipp Hagemeister',
                 'uploader_id': 'phihag',
                 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/phihag',
+                'channel_id': 'UCLqxVugv74EIW3VWh2NOa3Q',
+                'channel_url': r're:https?://(?:www\.)?youtube\.com/channel/UCLqxVugv74EIW3VWh2NOa3Q',
                 'upload_date': '20121002',
                 'license': 'Standard YouTube License',
                 'description': 'test chars:  "\'/\\ä↭𝕐\ntest URL: https://github.com/rg3/youtube-dl/issues/1892\n\nThis is a test video for youtube-dl.\n\nFor more information, contact phihag@phihag.de .',
@@ -1907,6 +1909,10 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
         else:
             self._downloader.report_warning('unable to extract uploader nickname')
 
+        channel_id = self._html_search_meta(
+            'channelId', video_webpage, 'channel id')
+        channel_url = 'http://www.youtube.com/channel/%s' % channel_id if channel_id else None
+
         # thumbnail image
         # We try first to get a high quality image:
         m_thumb = re.search(r'<span itemprop="thumbnail".*?href="(.*?)">',
@@ -2078,6 +2084,8 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
             'uploader': video_uploader,
             'uploader_id': video_uploader_id,
             'uploader_url': video_uploader_url,
+            'channel_id': channel_id,
+            'channel_url': channel_url,
             'upload_date': upload_date,
             'license': video_license,
             'creator': video_creator or artist,