Merge branch 'ytdl-org:master' into master

youtube_dl/extractor/extractors.py

@@ -1266,6 +1266,11 @@ from .theweatherchannel import TheWeatherChannelIE
 from .thisamericanlife import ThisAmericanLifeIE
 from .thisav import ThisAVIE
 from .thisoldhouse import ThisOldHouseIE
+from .thisvid import (
+    ThisVidIE,
+    ThisVidMemberIE,
+    ThisVidPlaylistIE,
+)
 from .threeqsdn import ThreeQSDNIE
 from .tiktok import (
     TikTokIE,

youtube_dl/extractor/generic.py

@@ -28,6 +28,7 @@ from ..utils import (
     mimetype2ext,
     orderedSet,
     parse_duration,
+    parse_resolution,
     sanitized_Request,
     smuggle_url,
     unescapeHTML,
@@ -35,6 +36,7 @@ from ..utils import (
     unsmuggle_url,
     UnsupportedError,
     url_or_none,
+    urljoin,
     xpath_attr,
     xpath_text,
     xpath_with_ns,
@@ -2227,6 +2229,97 @@ class GenericIE(InfoExtractor):
             # Sibnet embed (https://help.sibnet.ru/?sibnet_video_embed)
             'url': 'https://phpbb3.x-tk.ru/bbcode-video-sibnet-t24.html',
             'only_matching': True,
+        }, {
+            # KVS Player
+            'url': 'https://www.kvs-demo.com/videos/105/kelis-4th-of-july/',
+            'info_dict': {
+                'id': '105',
+                'display_id': 'kelis-4th-of-july',
+                'ext': 'mp4',
+                'title': 'Kelis - 4th Of July',
+                'thumbnail': r're:https://(?:www\.)?kvs-demo.com/contents/videos_screenshots/0/105/preview.jpg',
+            },
+        }, {
+            # KVS Player
+            'url': 'https://www.kvs-demo.com/embed/105/',
+            'info_dict': {
+                'id': '105',
+                'display_id': 'kelis-4th-of-july',
+                'ext': 'mp4',
+                'title': 'Kelis - 4th Of July / Embed Player',
+                'thumbnail': r're:https://(?:www\.)?kvs-demo.com/contents/videos_screenshots/0/105/preview.jpg',
+            },
+            'params': {
+                'skip_download': True,
+            },
+        }, {
+            # KVS Player (tested also in thisvid.py)
+            'url': 'https://youix.com/video/leningrad-zoj/',
+            'md5': '94f96ba95706dc3880812b27b7d8a2b8',
+            'info_dict': {
+                'id': '18485',
+                'display_id': 'leningrad-zoj',
+                'ext': 'mp4',
+                'title': 'Клип: Ленинград - ЗОЖ скачать, смотреть онлайн | Youix.com',
+                'thumbnail': r're:https://youix.com/contents/videos_screenshots/18000/18485/preview(?:_480x320_youix_com.mp4)?\.jpg',
+            },
+        }, {
+            # KVS Player
+            'url': 'https://youix.com/embed/18485',
+            'md5': '94f96ba95706dc3880812b27b7d8a2b8',
+            'info_dict': {
+                'id': '18485',
+                'display_id': 'leningrad-zoj',
+                'ext': 'mp4',
+                'title': 'Ленинград - ЗОЖ',
+                'thumbnail': r're:https://youix.com/contents/videos_screenshots/18000/18485/preview(?:_480x320_youix_com.mp4)?\.jpg',
+            },
+        }, {
+            # KVS Player
+            'url': 'https://bogmedia.org/videos/21217/40-nochey-40-nights-2016/',
+            'md5': '94166bdb26b4cb1fb9214319a629fc51',
+            'info_dict': {
+                'id': '21217',
+                'display_id': '40-nochey-2016',
+                'ext': 'mp4',
+                'title': '40 ночей (2016) - BogMedia.org',
+                'description': 'md5:4e6d7d622636eb7948275432eb256dc3',
+                'thumbnail': 'https://bogmedia.org/contents/videos_screenshots/21000/21217/preview_480p.mp4.jpg',
+            },
+        }, {
+            # KVS Player (for sites that serve kt_player.js via non-https urls)
+            'url': 'http://www.camhub.world/embed/389508',
+            'md5': 'fbe89af4cfb59c8fd9f34a202bb03e32',
+            'info_dict': {
+                'id': '389508',
+                'display_id': 'syren-de-mer-onlyfans-05-07-2020have-a-happy-safe-holiday5f014e68a220979bdb8cd-source',
+                'ext': 'mp4',
+                'title': 'Syren De Mer  onlyfans_05-07-2020Have_a_happy_safe_holiday5f014e68a220979bdb8cd_source / Embed плеер',
+                'thumbnail': r're:https?://www\.camhub\.world/contents/videos_screenshots/389000/389508/preview\.mp4\.jpg',
+            },
+        }, {
+            'url': 'https://mrdeepfakes.com/video/5/selena-gomez-pov-deep-fakes',
+            'md5': 'fec4ad5ec150f655e0c74c696a4a2ff4',
+            'info_dict': {
+                'id': '5',
+                'display_id': 'selena-gomez-pov-deep-fakes',
+                'ext': 'mp4',
+                'title': 'Selena Gomez POV (Deep Fakes) DeepFake Porn - MrDeepFakes',
+                'description': 'md5:17d1f84b578c9c26875ac5ef9a932354',
+                'height': 720,
+                'age_limit': 18,
+            },
+        }, {
+            'url': 'https://shooshtime.com/videos/284002/just-out-of-the-shower-joi/',
+            'md5': 'e2f0a4c329f7986280b7328e24036d60',
+            'info_dict': {
+                'id': '284002',
+                'display_id': 'just-out-of-the-shower-joi',
+                'ext': 'mp4',
+                'title': 'Just Out Of The Shower JOI - Shooshtime',
+                'height': 720,
+                'age_limit': 18,
+            },
         },
     ]
 
@@ -2332,6 +2425,88 @@ class GenericIE(InfoExtractor):
             'title': title,
         }
 
+    def _extract_kvs(self, url, webpage, video_id):
+
+        def getlicensetoken(license):
+            modlicense = license.replace('$', '').replace('0', '1')
+            center = int(len(modlicense) / 2)
+            fronthalf = int(modlicense[:center + 1])
+            backhalf = int(modlicense[center:])
+
+            modlicense = compat_str(4 * abs(fronthalf - backhalf))
+
+            def parts():
+                for o in range(0, center + 1):
+                    for i in range(1, 5):
+                        yield compat_str((int(license[o + i]) + int(modlicense[o])) % 10)
+
+            return ''.join(parts())
+
+        def getrealurl(video_url, license_code):
+            if not video_url.startswith('function/0/'):
+                return video_url  # not obfuscated
+
+            url_path, _, url_query = video_url.partition('?')
+            urlparts = url_path.split('/')[2:]
+            license = getlicensetoken(license_code)
+            newmagic = urlparts[5][:32]
+
+            def spells(x, o):
+                l = (o + sum(int(n) for n in license[o:])) % 32
+                for i in range(0, len(x)):
+                    yield {l: x[o], o: x[l]}.get(i, x[i])
+
+            for o in range(len(newmagic) - 1, -1, -1):
+                newmagic = ''.join(spells(newmagic, o))
+
+            urlparts[5] = newmagic + urlparts[5][32:]
+            return '/'.join(urlparts) + '?' + url_query
+
+        flashvars = self._search_regex(
+            r'(?s)<script\b[^>]*>.*?var\s+flashvars\s*=\s*(\{.+?\});.*?</script>',
+            webpage, 'flashvars')
+        flashvars = self._parse_json(flashvars, video_id, transform_source=js_to_json)
+
+        # extract the part after the last / as the display_id from the
+        # canonical URL.
+        display_id = self._search_regex(
+            r'(?:<link href="https?://[^"]+/(.+?)/?" rel="canonical"\s*/?>'
+            r'|<link rel="canonical" href="https?://[^"]+/(.+?)/?"\s*/?>)',
+            webpage, 'display_id', fatal=False
+        )
+        title = self._html_search_regex(r'<(?:h1|title)>(?:Video: )?(.+?)</(?:h1|title)>', webpage, 'title')
+
+        thumbnail = flashvars['preview_url']
+        if thumbnail.startswith('//'):
+            protocol, _, _ = url.partition('/')
+            thumbnail = protocol + thumbnail
+
+        url_keys = list(filter(re.compile(r'^video_(?:url|alt_url\d*)$').match, flashvars.keys()))
+        formats = []
+        for key in url_keys:
+            if '/get_file/' not in flashvars[key]:
+                continue
+            format_id = flashvars.get(key + '_text', key)
+            formats.append(merge_dicts(
+                parse_resolution(format_id) or parse_resolution(flashvars[key]), {
+                    'url': urljoin(url, getrealurl(flashvars[key], flashvars['license_code'])),
+                    'format_id': format_id,
+                    'ext': 'mp4',
+                    'http_headers': {'Referer': url},
+                }))
+            if not formats[-1].get('height'):
+                formats[-1]['quality'] = 1
+
+        self._sort_formats(formats)
+
+        return {
+            'id': flashvars['video_id'],
+            'display_id': display_id,
+            'title': title,
+            'thumbnail': thumbnail,
+            'formats': formats,
+        }
+
     def _real_extract(self, url):
         if url.startswith('//'):
             return self.url_result(self.http_scheme() + url)
@@ -2540,9 +2715,16 @@ class GenericIE(InfoExtractor):
         # but actually don't.
         AGE_LIMIT_MARKERS = [
             r'Proudly Labeled <a href="http://www\.rtalabel\.org/" title="Restricted to Adults">RTA</a>',
+            r'>[^<]*you acknowledge you are at least (\d+) years old',
+            r'>\s*(?:18\s+U(?:\.S\.C\.|SC)\s+)?(?:§+\s*)?2257\b',
         ]
-        if any(re.search(marker, webpage) for marker in AGE_LIMIT_MARKERS):
+        for marker in AGE_LIMIT_MARKERS:
-            age_limit = 18
+            m = re.search(marker, webpage)
+            if not m:
+                continue
+            age_limit = max(
+                age_limit or 0,
+                int_or_none(m.groups() and m.group(1), default=18))
 
         # video uploader is domain name
         video_uploader = self._search_regex(
@@ -3389,6 +3571,20 @@ class GenericIE(InfoExtractor):
                 info_dict['formats'] = formats
                 return info_dict
 
+        # Look for generic KVS player (before ld+json for tests)
+        found = self._search_regex(
+            (r'<script\b[^>]+?\bsrc\s*=\s*(["\'])https?://(?:\S+?/)+kt_player\.js\?v=(?P<ver>\d+(?:\.\d+)+)\1[^>]*>',
+             # kt_player('kt_player', 'https://i.shoosh.co/player/kt_player.swf?v=5.5.1', ...
+             r'kt_player\s*\(\s*(["\'])(?:(?!\1)[\w\W])+\1\s*,\s*(["\'])https?://(?:\S+?/)+kt_player\.swf\?v=(?P<ver>\d+(?:\.\d+)+)\2\s*,',
+             ), webpage, 'KVS player', group='ver', default=False)
+        if found:
+            self.report_extraction('%s: KVS Player' % (video_id, ))
+            if found.split('.')[0] not in ('4', '5', '6'):
+                self.report_warning('Untested major version (%s) in player engine - download may fail.' % (found, ))
+            return merge_dicts(
+                self._extract_kvs(url, webpage, video_id),
+                info_dict)
+
         # Looking for http://schema.org/VideoObject
         json_ld = self._search_json_ld(
             webpage, video_id, default={}, expected_type='VideoObject')

youtube_dl/extractor/thisvid.py

@@ -0,0 +1,218 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+import re
+import itertools
+
+from .common import InfoExtractor
+from ..compat import (
+    compat_urlparse,
+)
+from ..utils import (
+    clean_html,
+    get_element_by_class,
+    int_or_none,
+    merge_dicts,
+    url_or_none,
+    urljoin,
+)
+
+
+class ThisVidIE(InfoExtractor):
+    _VALID_URL = r'https?://(?:www\.)?thisvid\.com/(?P<type>videos|embed)/(?P<id>[A-Za-z0-9-]+)'
+    _TESTS = [{
+        'url': 'https://thisvid.com/videos/sitting-on-ball-tight-jeans/',
+        'md5': '839becb572995687e11a69dc4358a386',
+        'info_dict': {
+            'id': '3533241',
+            'ext': 'mp4',
+            'title': 'Sitting on ball tight jeans',
+            'description': 'md5:372353bb995883d1b65fddf507489acd',
+            'thumbnail': r're:https?://\w+\.thisvid\.com/(?:[^/]+/)+3533241/preview\.jpg',
+            'uploader_id': '150629',
+            'uploader': 'jeanslevisjeans',
+            'age_limit': 18,
+        }
+    }, {
+        'url': 'https://thisvid.com/embed/3533241/',
+        'md5': '839becb572995687e11a69dc4358a386',
+        'info_dict': {
+            'id': '3533241',
+            'ext': 'mp4',
+            'title': 'Sitting on ball tight jeans',
+            'thumbnail': r're:https?://\w+\.thisvid\.com/(?:[^/]+/)+3533241/preview\.jpg',
+            'uploader_id': '150629',
+            'uploader': 'jeanslevisjeans',
+            'age_limit': 18,
+        }
+    }]
+
+    def _real_extract(self, url):
+        main_id, type_ = re.match(self._VALID_URL, url).group('id', 'type')
+        webpage = self._download_webpage(url, main_id)
+
+        title = self._html_search_regex(
+            r'<title\b[^>]*?>(?:Video:\s+)?(.+?)(?:\s+-\s+ThisVid(?:\.com| tube))?</title>',
+            webpage, 'title')
+
+        if type_ == 'embed':
+            # look for more metadata
+            video_alt_url = url_or_none(self._search_regex(
+                r'''video_alt_url\s*:\s+'(%s/)',''' % (self._VALID_URL, ),
+                webpage, 'video_alt_url', default=None))
+            if video_alt_url and video_alt_url != url:
+                webpage = self._download_webpage(
+                    video_alt_url, main_id,
+                    note='Redirecting embed to main page', fatal=False) or webpage
+
+        video_holder = get_element_by_class('video-holder', webpage) or ''
+        if '>This video is a private video' in video_holder:
+            self.raise_login_required(
+                (clean_html(video_holder) or 'Private video').split('\n', 1)[0])
+
+        uploader = self._html_search_regex(
+            r'''(?s)<span\b[^>]*>Added by:\s*</span><a\b[^>]+\bclass\s*=\s*["']author\b[^>]+\bhref\s*=\s*["']https://thisvid\.com/members/([0-9]+/.{3,}?)\s*</a>''',
+            webpage, 'uploader', default='')
+        uploader = re.split(r'''/["'][^>]*>\s*''', uploader)
+        if len(uploader) == 2:
+            # id must be non-empty, uploader could be ''
+            uploader_id, uploader = uploader
+            uploader = uploader or None
+        else:
+            uploader_id = uploader = None
+
+        return merge_dicts({
+            '_type': 'url_transparent',
+            'title': title,
+            'age_limit': 18,
+            'uploader': uploader,
+            'uploader_id': uploader_id,
+        }, self.url_result(url, ie='Generic'))
+
+
+class ThisVidMemberIE(InfoExtractor):
+    _VALID_URL = r'https?://thisvid\.com/members/(?P<id>\d+)'
+    _TESTS = [{
+        'url': 'https://thisvid.com/members/2140501/',
+        'info_dict': {
+            'id': '2140501',
+            'title': 'Rafflesia\'s Profile',
+        },
+        'playlist_mincount': 16,
+    }, {
+        'url': 'https://thisvid.com/members/2140501/favourite_videos/',
+        'info_dict': {
+            'id': '2140501',
+            'title': 'Rafflesia\'s Favourite Videos',
+        },
+        'playlist_mincount': 15,
+    }, {
+        'url': 'https://thisvid.com/members/636468/public_videos/',
+        'info_dict': {
+            'id': '636468',
+            'title': 'Happymouth\'s Public Videos',
+        },
+        'playlist_mincount': 196,
+    },
+    ]
+
+    def _urls(self, html):
+        for m in re.finditer(r'''<a\b[^>]+\bhref\s*=\s*["'](?P<url>%s\b)[^>]+>''' % (ThisVidIE._VALID_URL, ), html):
+            yield m.group('url')
+
+    def _real_extract(self, url):
+        pl_id = self._match_id(url)
+        webpage = self._download_webpage(url, pl_id)
+
+        title = re.split(
+            r'(?i)\s*\|\s*ThisVid\.com\s*$',
+            self._og_search_title(webpage, default=None) or self._html_search_regex(r'(?s)<title\b[^>]*>(.+?)</title', webpage, 'title', fatal=False) or '', 1)[0] or None
+
+        def entries(page_url, html=None):
+            for page in itertools.count(1):
+                if not html:
+                    html = self._download_webpage(
+                        page_url, pl_id, note='Downloading page %d' % (page, ),
+                        fatal=False) or ''
+                for u in self._urls(html):
+                    yield u
+                next_page = get_element_by_class('pagination-next', html) or ''
+                if next_page:
+                    # member list page
+                    next_page = urljoin(url, self._search_regex(
+                        r'''<a\b[^>]+\bhref\s*=\s*("|')(?P<url>(?!#)(?:(?!\1).)+)''',
+                        next_page, 'next page link', group='url', default=None))
+                # in case a member page should have pagination-next with empty link, not just `else:`
+                if next_page is None:
+                    # playlist page
+                    parsed_url = compat_urlparse.urlparse(page_url)
+                    base_path, num = parsed_url.path.rsplit('/', 1)
+                    num = int_or_none(num)
+                    if num is None:
+                        base_path, num = parsed_url.path.rstrip('/'), 1
+                    parsed_url = parsed_url._replace(path=base_path + ('/%d' % (num + 1, )))
+                    next_page = compat_urlparse.urlunparse(parsed_url)
+                    if page_url == next_page:
+                        next_page = None
+                if not next_page:
+                    break
+                page_url, html = next_page, None
+
+        return self.playlist_from_matches(
+            entries(url, webpage), playlist_id=pl_id, playlist_title=title, ie='ThisVid')
+
+
+class ThisVidPlaylistIE(ThisVidMemberIE):
+    _VALID_URL = r'https?://thisvid\.com/playlist/(?P<id>\d+)/video/(?P<video_id>[A-Za-z0-9-]+)'
+    _TESTS = [{
+        'url': 'https://thisvid.com/playlist/6615/video/big-italian-booty-28/',
+        'info_dict': {
+            'id': '6615',
+            'title': 'Underwear Stuff',
+        },
+        'playlist_mincount': 200,
+    }, {
+        'url': 'https://thisvid.com/playlist/6615/video/big-italian-booty-28/',
+        'info_dict': {
+            'id': '1072387',
+            'ext': 'mp4',
+            'title': 'Big Italian Booty 28',
+            'description': 'md5:1bccf7b13765e18fb27bf764dba7ede2',
+            'uploader_id': '367912',
+            'uploader': 'Jcmusclefun',
+            'age_limit': 18,
+        },
+        'params': {
+            'noplaylist': True,
+        },
+    }]
+
+    def _get_video_url(self, pl_url):
+        video_id = re.match(self._VALID_URL, pl_url).group('video_id')
+        return urljoin(pl_url, '/videos/%s/' % (video_id, ))
+
+    def _urls(self, html):
+        for m in re.finditer(r'''<a\b[^>]+\bhref\s*=\s*["'](?P<url>%s\b)[^>]+>''' % (self._VALID_URL, ), html):
+            yield self._get_video_url(m.group('url'))
+
+    def _real_extract(self, url):
+        pl_id = self._match_id(url)
+
+        if self._downloader.params.get('noplaylist'):
+            self.to_screen('Downloading just the featured video because of --no-playlist')
+            return self.url_result(self._get_video_url(url), 'ThisVid')
+
+        self.to_screen(
+            'Downloading playlist %s - add --no-playlist to download just the featured video' % (pl_id, ))
+        result = super(ThisVidPlaylistIE, self)._real_extract(url)
+
+        # rework title returned as `the title - the title`
+        title = result['title']
+        t_len = len(title)
+        if t_len > 5 and t_len % 2 != 0:
+            t_len = t_len // 2
+            if title[t_len] == '-':
+                title = [t.strip() for t in (title[:t_len], title[t_len + 1:])]
+                if title[0] and title[0] == title[1]:
+                    result['title'] = title[0]
+        return result