Merge c555e5e478 into 668332b973

* YouPornCategoryIE
* YouPornChannelIE
* YouPornCollectionIE
* YouPornStarIE
* YouPornTagIE
* YouPornVideosIE,

test/test_download.py

@@ -10,6 +10,7 @@ sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
 
 from test.helper import (
     assertGreaterEqual,
+    assertLessEqual,
     expect_warnings,
     get_params,
     gettestcases,
@@ -122,7 +123,10 @@ def generator(test_case, tname):
         params['outtmpl'] = tname + '_' + params['outtmpl']
         if is_playlist and 'playlist' not in test_case:
             params.setdefault('extract_flat', 'in_playlist')
-            params.setdefault('playlistend', test_case.get('playlist_mincount'))
+            params.setdefault('playlistend',
+                              test_case['playlist_maxcount'] + 1
+                              if test_case.get('playlist_maxcount')
+                              else test_case.get('playlist_mincount'))
             params.setdefault('skip_download', True)
 
         ydl = YoutubeDL(params, auto_init=False)
@@ -190,6 +194,14 @@ def generator(test_case, tname):
                     'Expected at least %d in playlist %s, but got only %d' % (
                         test_case['playlist_mincount'], test_case['url'],
                         len(res_dict['entries'])))
+            if 'playlist_maxcount' in test_case:
+                assertLessEqual(
+                    self,
+                    len(res_dict['entries']),
+                    test_case['playlist_maxcount'],
+                    'Expected at most %d in playlist %s, but got %d' % (
+                        test_case['playlist_maxcount'], test_case['url'],
+                        len(res_dict['entries'])))
             if 'playlist_count' in test_case:
                 self.assertEqual(
                     len(res_dict['entries']),

youtube_dl/extractor/common.py

@@ -1415,9 +1415,21 @@ class InfoExtractor(object):
                     continue
                 info[count_key] = interaction_count
 
+        def extract_author(e):
+            author = e.get('author')
+            if not author:
+                return
+            info.update({
+                # author can be an instance of 'Organization' or 'Person' types.
+                # both types can have 'name' property(inherited from 'Thing' type). [1]
+                # however some websites are using 'Text' type instead.
+                # 1. https://schema.org/VideoObject
+                'uploader': author.get('name') if isinstance(author, dict) else author if isinstance(author, compat_str) else None,
+            })
+
         def extract_video_object(e):
             assert e['@type'] == 'VideoObject'
-            author = e.get('author')
+            extract_author(e)
             info.update({
                 'url': url_or_none(e.get('contentUrl')),
                 'title': unescapeHTML(e.get('name')),
@@ -1425,11 +1437,6 @@ class InfoExtractor(object):
                 'thumbnail': url_or_none(e.get('thumbnailUrl') or e.get('thumbnailURL')),
                 'duration': parse_duration(e.get('duration')),
                 'timestamp': unified_timestamp(e.get('uploadDate')),
-                # author can be an instance of 'Organization' or 'Person' types.
-                # both types can have 'name' property(inherited from 'Thing' type). [1]
-                # however some websites are using 'Text' type instead.
-                # 1. https://schema.org/VideoObject
-                'uploader': author.get('name') if isinstance(author, dict) else author if isinstance(author, compat_str) else None,
                 'filesize': float_or_none(e.get('contentSize')),
                 'tbr': int_or_none(e.get('bitrate')),
                 'width': int_or_none(e.get('width')),
@@ -1474,6 +1481,8 @@ class InfoExtractor(object):
                         'title': unescapeHTML(e.get('headline')),
                         'description': unescapeHTML(e.get('articleBody')),
                     })
+                elif item_type == 'SocialMediaPosting':
+                    extract_author(e)
                 elif item_type == 'VideoObject':
                     extract_video_object(e)
                     if expected_type is None:

youtube_dl/extractor/extractors.py

@@ -1653,7 +1653,15 @@ from .younow import (
     YouNowChannelIE,
     YouNowMomentIE,
 )
-from .youporn import YouPornIE
+from .youporn import (
+    YouPornIE,
+    YouPornCategoryIE,
+    YouPornChannelIE,
+    YouPornCollectionIE,
+    YouPornStarIE,
+    YouPornTagIE,
+    YouPornVideosIE,
+)
 from .yourporn import YourPornIE
 from .yourupload import YourUploadIE
 from .youtube import (

youtube_dl/extractor/facebook.py

@@ -23,6 +23,7 @@ from ..utils import (
     int_or_none,
     js_to_json,
     limit_length,
+    merge_dicts,
     parse_count,
     qualities,
     sanitized_Request,
@@ -82,8 +83,9 @@ class FacebookIE(InfoExtractor):
         'info_dict': {
             'id': '274175099429670',
             'ext': 'mp4',
-            'title': 're:^Asif Nawab Butt posted a video',
+            'title': 'Asif',
             'uploader': 'Asif Nawab Butt',
+            'uploader_id': '100005115464446',
             'upload_date': '20140506',
             'timestamp': 1399398998,
             'thumbnail': r're:^https?://.*',
@@ -137,18 +139,20 @@ class FacebookIE(InfoExtractor):
             'upload_date': '20160223',
             'uploader': 'Barack Obama',
         },
+        'skip': 'Gif on giphy.com gone',
     }, {
         # have 1080P, but only up to 720p in swf params
         # data.video.story.attachments[].media
         'url': 'https://www.facebook.com/cnn/videos/10155529876156509/',
-        'md5': '9571fae53d4165bbbadb17a94651dcdc',
         'info_dict': {
             'id': '10155529876156509',
             'ext': 'mp4',
-            'title': 'She survived the holocaust — and years later, she’s getting her citizenship s...',
+            'title': 'Holocaust survivor becomes US citizen',
+            'description': 'She survived the holocaust — and years later, she’s getting her citizenship so she can vote for Hillary Clinton http://cnn.it/2eERh5f',
             'timestamp': 1477818095,
             'upload_date': '20161030',
             'uploader': 'CNN',
+            'uploader_id': '5550296508',
             'thumbnail': r're:^https?://.*',
             'view_count': int,
         },
@@ -159,14 +163,17 @@ class FacebookIE(InfoExtractor):
         'info_dict': {
             'id': '1417995061575415',
             'ext': 'mp4',
-            'title': 'md5:1db063d6a8c13faa8da727817339c857',
+            'title': 'Ukrainian Scientists Worldwide | Довгоочікуване відео',
+            'description': 'Довгоочікуване відео',
             'timestamp': 1486648217,
             'upload_date': '20170209',
             'uploader': 'Yaroslav Korpan',
+            'uploader_id': '100000948048708',
         },
         'params': {
             'skip_download': True,
         },
+        'skip': 'This video is only available for registered users',
     }, {
         'url': 'https://www.facebook.com/LaGuiaDelVaron/posts/1072691702860471',
         'info_dict': {
@@ -181,16 +188,21 @@ class FacebookIE(InfoExtractor):
         'params': {
             'skip_download': True,
         },
+        # TODO: parse this webpage
+        'skip': 'This video is only available for registered users (and cannot parse webpage well)',
     }, {
         # data.node.comet_sections.content.story.attachments[].style_type_renderer.attachment.media
         'url': 'https://www.facebook.com/groups/1024490957622648/permalink/1396382447100162/',
         'info_dict': {
-            'id': '1396382447100162',
+            'id': '202882990186699',
             'ext': 'mp4',
-            'title': 'md5:19a428bbde91364e3de815383b54a235',
+
+            'title': 'birb (O v O\") | Hello? Yes your uber ride is here',
+            'description': 'md5:f122681cf504d04be12bb2bdc66c81fd',
             'timestamp': 1486035494,
             'upload_date': '20170202',
             'uploader': 'Elisabeth Ahtn',
+            'uploader_id': '100013949973717',
         },
         'params': {
             'skip_download': True,
@@ -231,8 +243,12 @@ class FacebookIE(InfoExtractor):
         'info_dict': {
             'id': '359649331226507',
             'ext': 'mp4',
-            'title': '#ESLOne VoD - Birmingham Finals Day#1 Fnatic vs. @Evil Geniuses',
+            'title': 'Fnatic vs. EG - Group A - Opening Match - ESL One Birmingham Day 1',
+            'description': '#ESLOne VoD - Birmingham Finals Day#1 Fnatic vs. @Evil Geniuses',
+            'timestamp': 1527084179,
+            'upload_date': '20180523',
             'uploader': 'ESL One Dota 2',
+            'uploader_id': '234218833769558',
         },
         'params': {
             'skip_download': True,
@@ -255,6 +271,7 @@ class FacebookIE(InfoExtractor):
             'id': '10157667649866271',
         },
         'playlist_count': 3,
+        'skip': 'Requires logging in',
     }, {
         # data.nodes[].comet_sections.content.story.attachments[].style_type_renderer.attachment.media
         'url': 'https://m.facebook.com/Alliance.Police.Department/posts/4048563708499330',
@@ -279,6 +296,7 @@ class FacebookIE(InfoExtractor):
             'upload_date': '20161122',
             'timestamp': 1479793574,
         },
+        'skip': 'No video',
     }, {
         # data.video.creation_story.attachments[].media
         'url': 'https://www.facebook.com/watch/live/?v=1823658634322275',
@@ -290,6 +308,20 @@ class FacebookIE(InfoExtractor):
         },
         'playlist_count': 1,
         'skip': 'Requires logging in',
+    }, {
+        # data.node.comet_sections.content.story.attachments[].styles.attachment.media
+        # gone : 'https://www.facebook.com/groups/352666925484718/permalink/1112505706167499/',
+        'url': 'https://www.facebook.com/groups/471738070722101/permalink/471799617382613/',
+        'info_dict': {
+            'id': '101173872166725',
+            'ext': 'mp4',
+            'title': 'md5:0ac174e60ed9d9a5eec588343f6a889c',
+            'description': 'md5:80efdff1371fa5d92758df2fb1b97bfd',
+            'upload_date': '20210517',
+            'timestamp': 1621247381,
+            'uploader': 'md5:34165f897c692f92b13c2cc1eddd8245',
+            'uploader_id': '100957655521680',
+        },
     }]
     _SUPPORTED_PAGLETS_REGEX = r'(?:pagelet_group_mall|permalink_video_pagelet|hyperfeed_story_id_[0-9a-f]+)'
     _api_config = {
@@ -378,8 +410,80 @@ class FacebookIE(InfoExtractor):
         self._login()
 
     def _extract_from_url(self, url, video_id):
-        webpage = self._download_webpage(
+        webpage, urlh = self._download_webpage_handle(
             url.replace('://m.facebook.com/', '://www.facebook.com/'), video_id)
+        login_or_id_check = re.search(r'(?P<login>\blogin\b)|/permalink/|[/=?]' + video_id, urlh.geturl())
+        if login_or_id_check and login_or_id_check.group('login'):
+            self.raise_login_required()
+        elif not login_or_id_check:
+            raise ExtractorError(
+                'Facebook redirected to a generic page (%s): the video may be unavailable or you may need to log in.' % urlh.geturl(),
+                expected=True)
+
+        def extract_metadata(webpage):
+            video_title = self._html_search_regex(
+                r'<h2\s+[^>]*class="uiHeaderTitle"[^>]*>([^<]*)</h2>', webpage,
+                'title', default=None)
+            if not video_title:
+                video_title = self._html_search_regex(
+                    r'(?s)<span class="fbPhotosPhotoCaption".*?id="fbPhotoPageCaption"><span class="hasCaption">(.*?)</span>',
+                    webpage, 'alternative title', default=None)
+
+            # _html_search_meta() finds '' ahead of later non-empty results
+            def html_search_meta_non_empty(patterns, html, name, **kwargs):
+                for p in patterns:
+                    result = self._html_search_meta(p, html, name, **kwargs)
+                    if result:
+                        return result
+
+            if not video_title:
+                video_title = html_search_meta_non_empty(
+                    ['og:title', 'twitter:title', 'description'],
+                    webpage, 'title', default=None)
+            if video_title:
+                video_title = limit_length(video_title, 80)
+            else:
+                video_title = 'Facebook video #%s' % video_id
+            description = html_search_meta_non_empty(
+                ['description', 'og:description', 'twitter:description'],
+                webpage, 'description', default=None)
+            uploader = clean_html(get_element_by_id(
+                'fbPhotoPageAuthorName', webpage)) or self._search_regex(
+                r'ownerName\s*:\s*"([^"]+)"', webpage, 'uploader',
+                default=None) or self._og_search_title(webpage, default=None)
+            timestamp = int_or_none(self._search_regex(
+                r'<abbr[^>]+data-utime=["\'](\d+)', webpage,
+                'timestamp', default=None))
+            thumbnail = html_search_meta_non_empty(
+                ['og:image', 'twitter:image'], webpage, 'thumbnail', default=None)
+            # some webpages contain unretrievable thumbnail urls
+            # like https://lookaside.fbsbx.com/lookaside/crawler/media/?media_id=10155168902769113&get_thumbnail=1
+            # in https://www.facebook.com/yaroslav.korpan/videos/1417995061575415/
+            if thumbnail and not re.search(r'\.(?:jpg|png)', thumbnail):
+                thumbnail = None
+            view_count = parse_count(self._search_regex(
+                [r'\bviewCount\s*:\s*["\']([\d,.]+)', r'"video_view_count"\s*:\s*([\d,.]+)'],
+                webpage, 'view count', default=None))
+            info_dict = {
+                'title': video_title,
+                'description': description,
+                'uploader': uploader,
+                'timestamp': timestamp,
+                'thumbnail': thumbnail,
+                'view_count': view_count,
+            }
+            info_json_ld = self._search_json_ld(webpage, video_id, default={})
+            if not info_json_ld:
+                # expecting title, description, uploader, comment_count and like_count
+                info_json_ld = self._search_json_ld(
+                    webpage, video_id, expected_type='SocialMediaPosting', default={})
+                # if we already have a non-generic title, honor that
+                if info_dict['title'] != 'Facebook video #%s' % video_id:
+                    info_json_ld['title'] = info_dict['title']
+            if info_json_ld.get('title'):
+                info_json_ld['title'] = limit_length(
+                    re.sub(r'\s*\|\s*Facebook$', '', info_json_ld['title']), 80)
+            return merge_dicts(info_json_ld, info_dict)
 
         video_data = None
 
@@ -477,7 +581,7 @@ class FacebookIE(InfoExtractor):
                     entries.append(info)
 
                 def parse_attachment(attachment, key='media'):
-                    media = attachment.get(key) or {}
+                    media = try_get(attachment, lambda x: x[key], dict) or {}
                     if media.get('__typename') == 'Video':
                         return parse_graphql_video(media)
 
@@ -492,7 +596,10 @@ class FacebookIE(InfoExtractor):
                         lambda x: x['attachments']
                     ], list) or []
                     for attachment in attachments:
-                        attachment = try_get(attachment, lambda x: x['style_type_renderer']['attachment'], dict)
+                        attachment = try_get(attachment, [
+                            lambda x: x['style_type_renderer']['attachment'],
+                            lambda x: x['styles']['attachment']
+                        ], dict) or {}
                         ns = try_get(attachment, lambda x: x['all_subattachments']['nodes'], list) or []
                         for n in ns:
                             parse_attachment(n)
@@ -513,7 +620,15 @@ class FacebookIE(InfoExtractor):
                     if not entries:
                         parse_graphql_video(video)
 
-                return self.playlist_result(entries, video_id)
+                if len(entries) > 1:
+                    return self.playlist_result(entries, video_id)
+
+                video_info = entries[0]
+                webpage_info = extract_metadata(webpage)
+                # honor precise duration in video info
+                if video_info.get('duration'):
+                    webpage_info['duration'] = video_info['duration']
+                return merge_dicts(webpage_info, video_info)
 
         if not video_data:
             m_msg = re.search(r'class="[^"]*uiInterstitialContent[^"]*"><div>(.*?)</div>', webpage)
@@ -633,43 +748,12 @@ class FacebookIE(InfoExtractor):
 
         process_formats(formats)
 
-        video_title = self._html_search_regex(
-            r'<h2\s+[^>]*class="uiHeaderTitle"[^>]*>([^<]*)</h2>', webpage,
-            'title', default=None)
-        if not video_title:
-            video_title = self._html_search_regex(
-                r'(?s)<span class="fbPhotosPhotoCaption".*?id="fbPhotoPageCaption"><span class="hasCaption">(.*?)</span>',
-                webpage, 'alternative title', default=None)
-        if not video_title:
-            video_title = self._html_search_meta(
-                'description', webpage, 'title', default=None)
-        if video_title:
-            video_title = limit_length(video_title, 80)
-        else:
-            video_title = 'Facebook video #%s' % video_id
-        uploader = clean_html(get_element_by_id(
-            'fbPhotoPageAuthorName', webpage)) or self._search_regex(
-            r'ownerName\s*:\s*"([^"]+)"', webpage, 'uploader',
-            default=None) or self._og_search_title(webpage, fatal=False)
-        timestamp = int_or_none(self._search_regex(
-            r'<abbr[^>]+data-utime=["\'](\d+)', webpage,
-            'timestamp', default=None))
-        thumbnail = self._html_search_meta(['og:image', 'twitter:image'], webpage)
-
-        view_count = parse_count(self._search_regex(
-            r'\bviewCount\s*:\s*["\']([\d,.]+)', webpage, 'view count',
-            default=None))
-
         info_dict = {
             'id': video_id,
-            'title': video_title,
             'formats': formats,
-            'uploader': uploader,
-            'timestamp': timestamp,
-            'thumbnail': thumbnail,
-            'view_count': view_count,
             'subtitles': subtitles,
         }
+        info_dict.update(extract_metadata(webpage))
 
         return info_dict
 

youtube_dl/extractor/youporn.py

@@ -1,20 +1,38 @@
 # coding: utf-8
 from __future__ import unicode_literals
 
+import itertools
 import re
+from time import sleep
 
 from .common import InfoExtractor
 from ..utils import (
+    clean_html,
     extract_attributes,
+    ExtractorError,
+    get_element_by_class,
+    get_element_by_id,
     int_or_none,
-    str_to_int,
+    merge_dicts,
+    parse_count,
+    parse_qs,
+    T,
+    traverse_obj,
     unified_strdate,
     url_or_none,
+    urljoin,
 )
 
 
 class YouPornIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:www\.)?youporn\.com/(?:watch|embed)/(?P<id>\d+)(?:/(?P<display_id>[^/?#&]+))?'
+    _VALID_URL = (
+        r'youporn:(?P<id>\d+)',
+        r'''(?x)
+            https?://(?:www\.)?youporn\.com/(?:watch|embed)/(?P<id>\d+)
+            (?:/(?:(?P<display_id>[^/?#&]+)/?)?)?(?:[#?]|$)
+    '''
+    )
+    _EMBED_REGEX = [r'<iframe[^>]+\bsrc=["\'](?P<url>(?:https?:)?//(?:www\.)?youporn\.com/embed/\d+)']
     _TESTS = [{
         'url': 'http://www.youporn.com/watch/505835/sex-ed-is-it-safe-to-masturbate-daily/',
         'md5': '3744d24c50438cf5b6f6d59feb5055c2',
@@ -34,7 +52,7 @@ class YouPornIE(InfoExtractor):
             'tags': list,
             'age_limit': 18,
         },
-        'skip': 'This video has been disabled',
+        'skip': 'This video has been deactivated',
     }, {
         # Unknown uploader
         'url': 'http://www.youporn.com/watch/561726/big-tits-awesome-brunette-on-amazing-webcam-show/?from=related3&al=2&from_id=561726&pos=4',
@@ -66,57 +84,104 @@ class YouPornIE(InfoExtractor):
     }, {
         'url': 'https://www.youporn.com/watch/13922959/femdom-principal/',
         'only_matching': True,
+    }, {
+        'url': 'https://www.youporn.com/watch/16290308/tinderspecial-trailer1/',
+        'info_dict': {
+            'id': '16290308',
+            'age_limit': 18,
+            'categories': [],
+            'description': None,  # SEO spam using title removed
+            'display_id': 'tinderspecial-trailer1',
+            'duration': 298.0,
+            'ext': 'mp4',
+            'upload_date': '20201123',
+            'uploader': 'Ersties',
+            'tags': [],
+            'thumbnail': 'https://fi1.ypncdn.com/m=eaSaaTbWx/202011/23/16290308/original/3.jpg',
+            'timestamp': 1606147564,
+            'title': 'Tinder In Real Life',
+            'view_count': int,
+        }
     }]
 
-    @staticmethod
-    def _extract_urls(webpage):
-        return re.findall(
-            r'<iframe[^>]+\bsrc=["\']((?:https?:)?//(?:www\.)?youporn\.com/embed/\d+)',
-            webpage)
+    @classmethod
+    def _extract_urls(cls, webpage):
+        def yield_urls():
+            for p in cls._EMBED_REGEX:
+                for m in re.finditer(p, webpage):
+                    yield m.group('url')
+
+        return list(yield_urls())
 
     def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
-        video_id = mobj.group('id')
-        display_id = mobj.group('display_id') or video_id
+        # A different video ID (data-video-id) is hidden in the page but
+        # never seems to be used
+        video_id, display_id = self._match_valid_url(url).group('id', 'display_id')
+        url = 'http://www.youporn.com/watch/%s' % (video_id,)
+        webpage = self._download_webpage(
+            url, video_id, headers={'Cookie': 'age_verified=1'})
 
-        definitions = self._download_json(
-            'https://www.youporn.com/api/video/media_definitions/%s/' % video_id,
-            display_id)
+        watchable = self._search_regex(
+            r'''(<div\s[^>]*\bid\s*=\s*('|")?watch-container(?(2)\2|(?!-)\b)[^>]*>)''',
+            webpage, 'watchability', default=None)
+        if not watchable:
+            msg = re.split(r'\s{4}', clean_html(get_element_by_id(
+                'mainContent', webpage)) or '')[0]
+            raise ExtractorError(
+                ('%s says: %s' % (self.IE_NAME, msg))
+                if msg else 'Video unavailable: no reason found',
+                expected=True)
+        # internal ID ?
+        # video_id = extract_attributes(watchable).get('data-video-id')
+
+        playervars = self._search_json(
+            r'\bplayervars\s*:', webpage, 'playervars', video_id)
+
+        def get_fmt(x):
+            v_url = url_or_none(x.get('videoUrl'))
+            if v_url:
+                x['videoUrl'] = v_url
+                return (x['format'], x)
+
+        defs_by_format = dict(traverse_obj(playervars, (
+            'mediaDefinitions', lambda _, v: v.get('format'), T(get_fmt))))
+
+        def get_format_data(f):
+            if f not in defs_by_format:
+                return []
+            return self._download_json(
+                defs_by_format[f]['videoUrl'], video_id, '{0}-formats'.format(f))
 
         formats = []
-        for definition in definitions:
-            if not isinstance(definition, dict):
-                continue
-            video_url = url_or_none(definition.get('videoUrl'))
-            if not video_url:
-                continue
-            f = {
-                'url': video_url,
-                'filesize': int_or_none(definition.get('videoSize')),
-            }
-            height = int_or_none(definition.get('quality'))
+        # Try to extract only the actual master m3u8 first, avoiding the duplicate single resolution "master" m3u8s
+        for hls_url in traverse_obj(
+                get_format_data('hls'),
+                (lambda _, v: not isinstance(v['defaultQuality'], bool), 'videoUrl'),
+                (Ellipsis, 'videoUrl')):
+            formats.extend(self._extract_m3u8_formats(
+                hls_url, video_id, 'mp4', fatal=False, m3u8_id='hls',
+                entry_protocol='m3u8_native'))
+
+        for f in traverse_obj(get_format_data('mp4'), (
+                lambda _, v: v.get('videoUrl'), {
+                    'url': ('videoUrl', T(url_or_none)),
+                    'filesize': ('videoSize', T(int_or_none)),
+                    'height': ('quality', T(int_or_none)),
+                }, T(lambda x: x.get('videoUrl') and x))):
             # Video URL's path looks like this:
             #  /201012/17/505835/720p_1500k_505835/YouPorn%20-%20Sex%20Ed%20Is%20It%20Safe%20To%20Masturbate%20Daily.mp4
             #  /201012/17/505835/vl_240p_240k_505835/YouPorn%20-%20Sex%20Ed%20Is%20It%20Safe%20To%20Masturbate%20Daily.mp4
             #  /videos/201703/11/109285532/1080P_4000K_109285532.mp4
             # We will benefit from it by extracting some metadata
-            mobj = re.search(r'(?P<height>\d{3,4})[pP]_(?P<bitrate>\d+)[kK]_\d+', video_url)
+            mobj = re.search(r'(?P<height>\d{3,4})[pP]_(?P<bitrate>\d+)[kK]_\d+', f['videoUrl'])
             if mobj:
-                if not height:
-                    height = int(mobj.group('height'))
-                bitrate = int(mobj.group('bitrate'))
-                f.update({
-                    'format_id': '%dp-%dk' % (height, bitrate),
-                    'tbr': bitrate,
-                })
-            f['height'] = height
+                if not f.get('height'):
+                    f['height'] = int(mobj.group('height'))
+                f['tbr'] = int(mobj.group('bitrate'))
+                f['format_id'] = '%dp-%dk' % (f['height'], f['tbr'])
             formats.append(f)
         self._sort_formats(formats)
 
-        webpage = self._download_webpage(
-            'http://www.youporn.com/watch/%s' % video_id, display_id,
-            headers={'Cookie': 'age_verified=1'})
-
         title = self._html_search_regex(
             r'(?s)<div[^>]+class=["\']watchVideoTitle[^>]+>(.+?)</div>',
             webpage, 'title', default=None) or self._og_search_title(
@@ -131,8 +196,10 @@ class YouPornIE(InfoExtractor):
         thumbnail = self._search_regex(
             r'(?:imageurl\s*=|poster\s*:)\s*(["\'])(?P<thumbnail>.+?)\1',
             webpage, 'thumbnail', fatal=False, group='thumbnail')
-        duration = int_or_none(self._html_search_meta(
-            'video:duration', webpage, 'duration', fatal=False))
+        duration = traverse_obj(playervars, ('duration', T(int_or_none)))
+        if duration is None:
+            duration = int_or_none(self._html_search_meta(
+                'video:duration', webpage, 'duration', fatal=False))
 
         uploader = self._html_search_regex(
             r'(?s)<div[^>]+class=["\']submitByLink["\'][^>]*>(.+?)</div>',
@@ -148,11 +215,11 @@ class YouPornIE(InfoExtractor):
 
         view_count = None
         views = self._search_regex(
-            r'(<div[^>]+\bclass=["\']js_videoInfoViews["\']>)', webpage,
-            'views', default=None)
+            r'(<div\s[^>]*\bdata-value\s*=[^>]+>)\s*<label>Views:</label>',
+            webpage, 'views', default=None)
         if views:
-            view_count = str_to_int(extract_attributes(views).get('data-value'))
-        comment_count = str_to_int(self._search_regex(
+            view_count = parse_count(extract_attributes(views).get('data-value'))
+        comment_count = parse_count(self._search_regex(
             r'>All [Cc]omments? \(([\d,.]+)\)',
             webpage, 'comment count', default=None))
 
@@ -168,7 +235,10 @@ class YouPornIE(InfoExtractor):
             r'(?s)Tags:.*?</div>\s*<div[^>]+class=["\']tagBoxContent["\'][^>]*>(.+?)</div>',
             'tags')
 
-        return {
+        data = self._search_json_ld(webpage, video_id, expected_type='VideoObject', fatal=False) or {}
+        data.pop('url', None)
+
+        result = merge_dicts(data, {
             'id': video_id,
             'display_id': display_id,
             'title': title,
@@ -183,4 +253,442 @@ class YouPornIE(InfoExtractor):
             'tags': tags,
             'age_limit': age_limit,
             'formats': formats,
-        }
+        })
+        # Remove promotional non-description
+        if result.get('description', '').startswith(
+                'Watch %s online' % (result['title'],)):
+            del result['description']
+        return result
+
+
+class YouPornListBase(InfoExtractor):
+    # pattern in '.title-text' element of page section containing videos
+    _PLAYLIST_TITLEBAR_RE = r'\s+[Vv]ideos\s*$'
+    _PAGE_RETRY_COUNT = 0  # ie, no retry
+    _PAGE_RETRY_DELAY = 2  # seconds
+
+    def _get_next_url(self, url, pl_id, html):
+        return urljoin(url, self._search_regex(
+            r'''<a\s[^>]*?\bhref\s*=\s*("|')(?P<url>(?:(?!\1)[^>])+)\1''',
+            get_element_by_id('next', html) or '', 'next page',
+            group='url', default=None))
+
+    @classmethod
+    def _get_title_from_slug(cls, title_slug):
+        return re.sub(r'[_-]', ' ', title_slug)
+
+    def _entries(self, url, pl_id, html=None, page_num=None):
+
+        # separates page sections
+        PLAYLIST_SECTION_RE = (
+            r'''<div\s[^>]*\bclass\s*=\s*('|")(?:[\w$-]+\s+|\s)*?title-bar(?:\s+[\w$-]+|\s)*\1[^>]*>'''
+        )
+        # contains video link
+        VIDEO_URL_RE = r'''(?x)
+            <div\s[^>]*\bdata-video-id\s*=\s*('|")\d+\1[^>]*>\s*
+            (?:<div\b[\s\S]+?</div>\s*)*
+            <a\s[^>]*\bhref\s*=\s*('|")(?P<url>(?:(?!\2)[^>])+)\2
+        '''
+
+        def yield_pages(url, html=html, page_num=page_num):
+            fatal = not html
+            for pnum in itertools.count(start=page_num or 1):
+                if not html:
+                    html = self._download_webpage(
+                        url, pl_id, note='Downloading page %d' % pnum,
+                        fatal=fatal)
+                if not html:
+                    break
+                fatal = False
+                yield (url, html, pnum)
+                # explicit page: extract just that page
+                if page_num is not None:
+                    break
+                next_url = self._get_next_url(url, pl_id, html)
+                if not next_url or next_url == url:
+                    break
+                url, html = next_url, None
+
+        def retry_page(msg, tries_left, page_data):
+            if tries_left <= 0:
+                return
+            self.report_warning(msg, pl_id)
+            sleep(self._PAGE_RETRY_DELAY)
+            return next(
+                yield_pages(page_data[0], page_num=page_data[2]), None)
+
+        def yield_entries(html):
+            for frag in re.split(PLAYLIST_SECTION_RE, html):
+                if not frag:
+                    continue
+                t_text = get_element_by_class('title-text', frag or '')
+                if not (t_text and re.search(self._PLAYLIST_TITLEBAR_RE, t_text)):
+                    continue
+                for m in re.finditer(VIDEO_URL_RE, frag):
+                    video_url = urljoin(url, m.group('url'))
+                    if video_url:
+                        yield self.url_result(video_url)
+
+        last_first_url = None
+        for page_data in yield_pages(url, html=html, page_num=page_num):
+            # page_data: url, html, page_num
+            first_url = None
+            tries_left = self._PAGE_RETRY_COUNT + 1
+            while tries_left > 0:
+                tries_left -= 1
+                for from_ in yield_entries(page_data[1]):
+                    # may get the same page twice instead of empty page
+                    # or (site bug) intead of actual next page
+                    if not first_url:
+                        first_url = from_['url']
+                        if first_url == last_first_url:
+                            # sometimes (/porntags/) the site serves the previous page
+                            # instead but may provide the correct page after a delay
+                            page_data = retry_page(
+                                'Retrying duplicate page...', tries_left, page_data)
+                            if page_data:
+                                first_url = None
+                                break
+                            continue
+                    yield from_
+                else:
+                    if not first_url and 'no-result-paragarph1' in page_data[1]:
+                        page_data = retry_page(
+                            'Retrying empty page...', tries_left, page_data)
+                        if page_data:
+                            continue
+                    else:
+                        # success/failure
+                        break
+            # may get an infinite (?) sequence of empty pages
+            if not first_url:
+                break
+            last_first_url = first_url
+
+    def _real_extract(self, url, html=None):
+        # exceptionally, id may be None
+        m_dict = self._match_valid_url(url).groupdict()
+        pl_id, page_type, sort = (m_dict.get(k) for k in ('id', 'type', 'sort'))
+
+        qs = parse_qs(url)
+        for q, v in qs.items():
+            if v:
+                qs[q] = v[-1]
+            else:
+                del qs[q]
+
+        base_id = pl_id or 'YouPorn'
+        title = self._get_title_from_slug(base_id)
+        if page_type:
+            title = '%s %s' % (page_type.capitalize(), title)
+        base_id = [base_id.lower()]
+        if sort is None:
+            title += ' videos'
+        else:
+            title = '%s videos by %s' % (title, re.sub(r'[_-]', ' ', sort))
+            base_id.append(sort)
+        if qs:
+            ps = ['%s=%s' % item for item in sorted(qs.items())]
+            title += ' (%s)' % ','.join(ps)
+            base_id.extend(ps)
+        pl_id = '/'.join(base_id)
+
+        return self.playlist_result(
+            self._entries(url, pl_id, html=html,
+                          page_num=int_or_none(qs.get('page'))),
+            playlist_id=pl_id, playlist_title=title)
+
+
+class YouPornCategoryIE(YouPornListBase):
+    IE_DESC = 'YouPorn category, with sorting, filtering and pagination'
+    _VALID_URL = r'''(?x)
+        https?://(?:www\.)?youporn\.com/
+        (?P<type>category)/(?P<id>[^/?#&]+)
+        (?:/(?P<sort>popular|views|rating|time|duration))?/?(?:[#?]|$)
+    '''
+    _TESTS = [{
+        'note': 'Full list with pagination',
+        'url': 'https://www.youporn.com/category/lingerie/popular/',
+        'info_dict': {
+            'id': 'lingerie/popular',
+            'title': 'Category lingerie videos by popular',
+        },
+        'playlist_mincount': 39,
+    }, {
+        'note': 'Filtered paginated list with single page result',
+        'url': 'https://www.youporn.com/category/lingerie/duration/?min_minutes=10',
+        'info_dict': {
+            'id': 'lingerie/duration/min_minutes=10',
+            'title': 'Category lingerie videos by duration (min_minutes=10)',
+        },
+        'playlist_maxcount': 30,
+    }, {
+        'note': 'Single page of full list',
+        'url': 'https://www.youporn.com/category/lingerie/popular?page=1',
+        'info_dict': {
+            'id': 'lingerie/popular/page=1',
+            'title': 'Category lingerie videos by popular (page=1)',
+        },
+        'playlist_count': 30,
+    }]
+
+
+class YouPornChannelIE(YouPornListBase):
+    IE_DESC = 'YouPorn channel, with sorting and pagination'
+    _VALID_URL = r'''(?x)
+        https?://(?:www\.)?youporn\.com/
+        (?P<type>channel)/(?P<id>[^/?#&]+)
+        (?:/(?P<sort>rating|views|duration))?/?(?:[#?]|$)
+    '''
+    _TESTS = [{
+        'note': 'Full list with pagination',
+        'url': 'https://www.youporn.com/channel/x-feeds/',
+        'info_dict': {
+            'id': 'x-feeds',
+            'title': 'Channel X-Feeds videos',
+        },
+        'playlist_mincount': 37,
+    }, {
+        'note': 'Single page of full list (no filters here)',
+        'url': 'https://www.youporn.com/channel/x-feeds/duration?page=1',
+        'info_dict': {
+            'id': 'x-feeds/duration/page=1',
+            'title': 'Channel X-Feeds videos by duration (page=1)',
+        },
+        'playlist_count': 24,
+    }]
+
+    @staticmethod
+    def _get_title_from_slug(title_slug):
+        return re.sub(r'_', ' ', title_slug).title()
+
+
+class YouPornCollectionIE(YouPornListBase):
+    IE_DESC = 'YouPorn collection (user playlist), with sorting and pagination'
+    _VALID_URL = r'''(?x)
+        https?://(?:www\.)?youporn\.com/
+        (?P<type>collection)s/videos/(?P<id>\d+)
+        (?:/(?P<sort>rating|views|time|duration))?/?(?:[#?]|$)
+    '''
+    _PLAYLIST_TITLEBAR_RE = r'^\s*Videos\s+in\s'
+    _TESTS = [{
+        'note': 'Full list with pagination',
+        'url': 'https://www.youporn.com/collections/videos/33044251/',
+        'info_dict': {
+            'id': '33044251',
+            'title': 'Collection Sexy Lips videos',
+            'uploader': 'ph-littlewillyb',
+        },
+        'playlist_mincount': 50,
+    }, {
+        'note': 'Single page of full list (no filters here)',
+        'url': 'https://www.youporn.com/collections/videos/33044251/time?page=1',
+        'info_dict': {
+            'id': '33044251/time/page=1',
+            'title': 'Collection Sexy Lips videos by time (page=1)',
+            'uploader': 'ph-littlewillyb',
+        },
+        'playlist_count': 20,
+    }]
+
+    def _real_extract(self, url):
+        pl_id = self._match_id(url)
+        html = self._download_webpage(url, pl_id)
+        playlist = super(YouPornCollectionIE, self)._real_extract(url, html=html)
+        infos = re.sub(r'\s+', ' ', clean_html(get_element_by_class(
+            'collection-infos', html)) or '')
+        title, uploader = self._search_regex(
+            r'^\s*Collection: (?P<title>.+?) \d+ VIDEOS \d+ VIEWS \d+ days LAST UPDATED From: (?P<uploader>[\w_-]+)',
+            infos, 'title/uploader', group=('title', 'uploader'), default=(None, None))
+
+        return merge_dicts({
+            'title': playlist['title'].replace(playlist['id'].split('/')[0], title),
+            'uploader': uploader,
+        }, playlist) if title else playlist
+
+
+class YouPornTagIE(YouPornListBase):
+    IE_DESC = 'YouPorn tag (porntags), with sorting, filtering and pagination'
+    _VALID_URL = r'''(?x)
+        https?://(?:www\.)?youporn\.com/
+        porn(?P<type>tag)s/(?P<id>[^/?#&]+)
+        (?:/(?P<sort>views|rating|time|duration))?/?(?:[#?]|$)
+    '''
+    _PLAYLIST_TITLEBAR_RE = r'^\s*Videos\s+tagged\s'
+    _PAGE_RETRY_COUNT = 1
+    _TESTS = [{
+        'note': 'Full list with pagination',
+        'url': 'https://www.youporn.com/porntags/austrian',
+        'info_dict': {
+            'id': 'austrian',
+            'title': 'Tag austrian videos',
+        },
+        'playlist_mincount': 35,
+        'expected_warnings': ['Retrying duplicate page'],
+    }, {
+        'note': 'Filtered paginated list with single page result',
+        'url': 'https://www.youporn.com/porntags/austrian/duration/?min_minutes=10',
+        'info_dict': {
+            'id': 'austrian/duration/min_minutes=10',
+            'title': 'Tag austrian videos by duration (min_minutes=10)',
+        },
+        # number of videos per page is (row x col) 2x3 + 6x4 + 2, or + 3,
+        # or more, varying with number of ads; let's set max as 9x4
+        # NB col 1 may not be shown in non-JS page with site CSS and zoom 100%
+        'playlist_maxcount': 32,
+        'expected_warnings': ['Retrying duplicate page', 'Retrying empty page'],
+    }, {
+        'note': 'Single page of full list',
+        'url': 'https://www.youporn.com/porntags/austrian/?page=1',
+        'info_dict': {
+            'id': 'austrian/page=1',
+            'title': 'Tag austrian videos (page=1)',
+        },
+        'playlist_mincount': 32,
+        'playlist_maxcount': 34,
+        'expected_warnings': ['Retrying duplicate page', 'Retrying empty page'],
+    }]
+
+    # YP tag navigation is broken, loses sort
+    def _get_next_url(self, url, pl_id, html):
+        next_url = super(YouPornTagIE, self)._get_next_url(url, pl_id, html)
+        if next_url:
+            n = self._match_valid_url(next_url)
+            if n:
+                s = n.groupdict().get('sort')
+            if s:
+                u = self._match_valid_url(url)
+                if u:
+                    u = u.groupdict().get('sort')
+                    if s and not u:
+                        n = n.end('sort')
+                        next_url = next_url[:n] + '/' + u + next_url[n:]
+        return next_url
+
+
+class YouPornStarIE(YouPornListBase):
+    IE_DESC = 'YouPorn Pornstar, with description, sorting and pagination'
+    _VALID_URL = r'''(?x)
+        https?://(?:www\.)?youporn\.com/
+        (?P<type>pornstar)/(?P<id>[^/?#&]+)
+        (?:/(?P<sort>rating|views|duration))?/?(?:[#?]|$)
+    '''
+    _PLAYLIST_TITLEBAR_RE = r'^\s*Videos\s+[fF]eaturing\s'
+    _TESTS = [{
+        'note': 'Full list with pagination',
+        'url': 'https://www.youporn.com/pornstar/daynia/',
+        'info_dict': {
+            'id': 'daynia',
+            'title': 'Pornstar Daynia videos',
+            'description': r're:Daynia Rank \d+ Videos \d+ Views [\d,.]+ .+ Subscribers \d+',
+        },
+        'playlist_mincount': 45,
+    }, {
+        'note': 'Single page of full list (no filters here)',
+        'url': 'https://www.youporn.com/pornstar/daynia/?page=1',
+        'info_dict': {
+            'id': 'daynia/page=1',
+            'title': 'Pornstar Daynia videos (page=1)',
+            'description': 're:.{180,}',
+        },
+        'playlist_count': 26,
+    }]
+
+    @staticmethod
+    def _get_title_from_slug(title_slug):
+        return re.sub(r'_', ' ', title_slug).title()
+
+    def _real_extract(self, url):
+        pl_id = self._match_id(url)
+        html = self._download_webpage(url, pl_id)
+        playlist = super(YouPornStarIE, self)._real_extract(url, html=html)
+        INFO_ELEMENT_RE = r'''(?x)
+            <div\s[^>]*\bclass\s*=\s*('|")(?:[\w$-]+\s+|\s)*?pornstar-info-wrapper(?:\s+[\w$-]+|\s)*\1[^>]*>
+            (?P<info>[\s\S]+?)(?:</div>\s*){6,}
+        '''
+
+        infos = self._search_regex(INFO_ELEMENT_RE, html, 'infos', group='info', default='')
+        if infos:
+            infos = re.sub(
+                r'(?:\s*nl=nl)+\s*', ' ',
+                re.sub(r'(?u)\s+', ' ', clean_html(
+                    re.sub('\n', 'nl=nl', infos)))).replace('ribe Subsc', '')
+
+        return merge_dicts({
+            'description': infos.strip() or None,
+        }, playlist)
+
+
+class YouPornVideosIE(YouPornListBase):
+    IE_DESC = 'YouPorn video (browse) playlists, with sorting, filtering and pagination'
+    _VALID_URL = r'''(?x)
+        https?://(?:www\.)?youporn\.com/
+            (?:(?P<id>browse)/)?
+            (?P<sort>(?(id)
+                (?:duration|rating|time|views)|
+                (?:most_(?:favou?rit|view)ed|recommended|top_rated)?))
+            (?:[/#?]|$)
+    '''
+    _PLAYLIST_TITLEBAR_RE = r'\s+(?:[Vv]ideos|VIDEOS)\s*$'
+    _TESTS = [{
+        'note': 'Full list with pagination (too long for test)',
+        'url': 'https://www.youporn.com/',
+        'info_dict': {
+            'id': 'youporn',
+            'title': 'YouPorn videos',
+        },
+        'only_matching': True,
+    }, {
+        'note': 'Full list with pagination (too long for test)',
+        'url': 'https://www.youporn.com/recommended',
+        'info_dict': {
+            'id': 'youporn/recommended',
+            'title': 'YouPorn videos by recommended',
+        },
+        'only_matching': True,
+    }, {
+        'note': 'Full list with pagination (too long for test)',
+        'url': 'https://www.youporn.com/top_rated',
+        'info_dict': {
+            'id': 'youporn/top_rated',
+            'title': 'YouPorn videos by top rated',
+        },
+        'only_matching': True,
+    }, {
+        'note': 'Full list with pagination (too long for test)',
+        'url': 'https://www.youporn.com/browse/time',
+        'info_dict': {
+            'id': 'browse/time',
+            'title': 'YouPorn videos by time',
+        },
+        'only_matching': True,
+    }, {
+        'note': 'Filtered paginated list with single page result',
+        'url': 'https://www.youporn.com/most_favorited/?res=VR&max_minutes=2',
+        'info_dict': {
+            'id': 'youporn/most_favorited/max_minutes=2/res=VR',
+            'title': 'YouPorn videos by most favorited (max_minutes=2,res=VR)',
+        },
+        'playlist_mincount': 10,
+        'playlist_maxcount': 28,
+    }, {
+        'note': 'Filtered paginated list with several pages',
+        'url': 'https://www.youporn.com/most_favorited/?res=VR&max_minutes=5',
+        'info_dict': {
+            'id': 'youporn/most_favorited/max_minutes=5/res=VR',
+            'title': 'YouPorn videos by most favorited (max_minutes=5,res=VR)',
+        },
+        'playlist_mincount': 45,
+    }, {
+        'note': 'Single page of full list',
+        'url': 'https://www.youporn.com/browse/time?page=1',
+        'info_dict': {
+            'id': 'browse/time/page=1',
+            'title': 'YouPorn videos by time (page=1)',
+        },
+        'playlist_count': 36,
+    }]
+
+    @staticmethod
+    def _get_title_from_slug(title_slug):
+        return 'YouPorn' if title_slug == 'browse' else title_slug