[^<]+)', - r'<div[^>]+data-video-title=(["\'])(?P<title>.+?)\1', - r'shareTitle\s*=\s*(["\'])(?P<title>.+?)\1'), + 'twitter:title', webpage, default=None) or self._html_search_regex( + (r'(?s)<h1[^>]+class=["\']title["\'][^>]*>(?P<title>.+?)</h1>', + r'<div[^>]+data-video-title=(["\'])(?P<title>(?:(?!\1).)+)\1', + r'shareTitle["\']\s*[=:]\s*(["\'])(?P<title>(?:(?!\1).)+)\1'), webpage, 'title', group='title') + video_urls = [] + video_urls_set = set() + subtitles = {} + flashvars = self._parse_json( self._search_regex( r'var\s+flashvars_\d+\s*=\s*({.+?});', webpage, 'flashvars', default='{}'), video_id) if flashvars: + subtitle_url = url_or_none(flashvars.get('closedCaptionsFile')) + if subtitle_url: + subtitles.setdefault('en', []).append({ + 'url': subtitle_url, + 'ext': 'srt', + }) thumbnail = flashvars.get('image_url') duration = int_or_none(flashvars.get('video_duration')) + media_definitions = flashvars.get('mediaDefinitions') + if isinstance(media_definitions, list): + for definition in media_definitions: + if not isinstance(definition, dict): + continue + video_url = definition.get('videoUrl') + if not video_url or not isinstance(video_url, compat_str): + continue + if video_url in video_urls_set: + continue + video_urls_set.add(video_url) + video_urls.append( + (video_url, int_or_none(definition.get('quality')))) else: - title, thumbnail, duration = [None] * 3 + thumbnail, duration = [None] * 2 + + def extract_js_vars(webpage, pattern, default=NO_DEFAULT): + assignments = self._search_regex( + pattern, webpage, 'encoded url', default=default) + if not assignments: + return {} + + assignments = assignments.split(';') + + js_vars = {} + + def parse_js_value(inp): + inp = re.sub(r'/\*(?:(?!\*/).)*?\*/', '', inp) + if '+' in inp: + inps = inp.split('+') + return functools.reduce( + operator.concat, map(parse_js_value, inps)) + inp = inp.strip() + if inp in js_vars: + return js_vars[inp] + return remove_quotes(inp) + + for assn in assignments: + assn = assn.strip() + if not assn: + continue + assn = re.sub(r'var\s+', '', assn) + vname, value = assn.split('=', 1) + js_vars[vname] = parse_js_value(value) + return js_vars + + def add_video_url(video_url): + v_url = url_or_none(video_url) + if not v_url: + return + if v_url in video_urls_set: + return + video_urls.append((v_url, None)) + video_urls_set.add(v_url) + + if not video_urls: + FORMAT_PREFIXES = ('media', 'quality') + js_vars = extract_js_vars( + webpage, r'(var\s+(?:%s)_.+)' % '|'.join(FORMAT_PREFIXES), + default=None) + if js_vars: + for key, format_url in js_vars.items(): + if any(key.startswith(p) for p in FORMAT_PREFIXES): + add_video_url(format_url) + if not video_urls and re.search( + r'<[^>]+\bid=["\']lockedPlayer', webpage): + raise ExtractorError( + 'Video %s is locked' % video_id, expected=True) + + if not video_urls: + js_vars = extract_js_vars( + dl_webpage('tv'), r'(var.+?mediastring.+?)</script>') + add_video_url(js_vars['mediastring']) + + for mobj in re.finditer( + r'<a[^>]+\bclass=["\']downloadBtn\b[^>]+\bhref=(["\'])(?P<url>(?:(?!\1).)+)\1', + webpage): + video_url = mobj.group('url') + if video_url not in video_urls_set: + 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('/', '') + ext = determine_ext(video_url) + if ext == 'mpd': + formats.extend(self._extract_mpd_formats( + video_url, video_id, mpd_id='dash', fatal=False)) + continue + elif ext == 'm3u8': + formats.extend(self._extract_m3u8_formats( + video_url, video_id, 'mp4', entry_protocol='m3u8_native', + m3u8_id='hls', fatal=False)) + continue + tbr = None + mobj = re.search(r'(?P<height>\d+)[pP]?_(?P<tbr>\d+)[kK]', video_url) + if mobj: + if not height: + height = int(mobj.group('height')) + tbr = int(mobj.group('tbr')) + formats.append({ + 'url': video_url, + 'format_id': '%dp' % height if height else None, + 'height': height, + 'tbr': tbr, + }) + self._sort_formats(formats) video_uploader = self._html_search_regex( - r'(?s)From: .+?<(?:a href="/users/|a href="/channels/|span class="username)[^>]+>(.+?)<', + r'(?s)From: .+?<(?:a\b[^>]+\bhref=["\']/(?:(?:user|channel)s|model|pornstar)/|span\b[^>]+\bclass=["\']username)[^>]+>(.+?)<', webpage, 'uploader', fatal=False) view_count = self._extract_count( - r'<span class="count">([\d,\.]+)</span> views', webpage, 'view') + r'<span class="count">([\d,\.]+)</span> [Vv]iews', webpage, 'view') like_count = self._extract_count( r'<span class="votesUp">([\d,\.]+)</span>', webpage, 'like') dislike_count = self._extract_count( @@ -156,54 +349,17 @@ class PornHubIE(InfoExtractor): comment_count = self._extract_count( r'All Comments\s*<span>$([\d,.]+)$', webpage, 'comment') - video_urls = [] - for quote, video_url in re.findall( - r'player_quality_[0-9]{3,4}p\s*=\s*(["\'])(.+?)\1;', webpage): - video_urls.append(compat_urllib_parse_unquote(re.sub( - r'{0}\s*\+\s*{0}'.format(quote), '', video_url))) - - if webpage.find('"encrypted":true') != -1: - password = compat_urllib_parse_unquote_plus( - self._search_regex(r'"video_title":"([^"]+)', webpage, 'password')) - video_urls = list(map(lambda s: aes_decrypt_text(s, password, 32).decode('utf-8'), video_urls)) - - formats = [] - for video_url in video_urls: - path = compat_urllib_parse_urlparse(video_url).path - extension = os.path.splitext(path)[1][1:] - format = path.split('/')[5].split('_')[:2] - format = '-'.join(format) - - m = re.match(r'^(?P<height>[0-9]+)[pP]-(?P<tbr>[0-9]+)[kK]$', format) - if m is None: - height = None - tbr = None - else: - height = int(m.group('height')) - tbr = int(m.group('tbr')) - - formats.append({ - 'url': video_url, - 'ext': extension, - 'format': format, - 'format_id': format, - 'tbr': tbr, - 'height': height, - }) - self._sort_formats(formats) - - page_params = self._parse_json(self._search_regex( - r'page_params\.zoneDetails\[([\'"])[^\'"]+\1\]\s*=\s*(?P<data>{[^}]+})', - webpage, 'page parameters', group='data', default='{}'), - video_id, transform_source=js_to_json, fatal=False) - tags = categories = None - if page_params: - tags = page_params.get('tags', '').split(',') - categories = page_params.get('categories', '').split(',') + def extract_list(meta_key): + div = self._search_regex( + r'(?s)<div[^>]+\bclass=["\'].*?\b%sWrapper[^>]*>(.+?)</div>' + % meta_key, webpage, meta_key, default=None) + if div: + return re.findall(r'<a[^>]+\bhref=[^>]+>([^<]+)', div) return { 'id': video_id, 'uploader': video_uploader, + 'upload_date': upload_date, 'title': title, 'thumbnail': thumbnail, 'duration': duration, @@ -213,85 +369,243 @@ class PornHubIE(InfoExtractor): 'comment_count': comment_count, 'formats': formats, 'age_limit': 18, - 'tags': tags, - 'categories': categories, + 'tags': extract_list('tags'), + 'categories': extract_list('categories'), + 'subtitles': subtitles, } -class PornHubPlaylistBaseIE(InfoExtractor): - def _extract_entries(self, webpage): +class PornHubPlaylistBaseIE(PornHubBaseIE): + def _extract_entries(self, webpage, host): + # Only process container div with main playlist content skipping + # drop-down menu that uses similar pattern for videos (see + # https://github.com/ytdl-org/youtube-dl/issues/11594). + container = self._search_regex( + r'(?s)(<div[^>]+class=["\']container.+)', webpage, + 'container', default=webpage) + return [ self.url_result( - 'http://www.pornhub.com/%s' % video_url, + 'http://www.%s/%s' % (host, video_url), PornHubIE.ie_key(), video_title=title) for video_url, title in orderedSet(re.findall( r'href="/?(view_video\.php\?.*\bviewkey=[\da-z]+[^"]*)"[^>]*\s+title="([^"]+)"', - webpage)) + container)) ] def _real_extract(self, url): - playlist_id = self._match_id(url) + mobj = re.match(self._VALID_URL, url) + host = mobj.group('host') + playlist_id = mobj.group('id') webpage = self._download_webpage(url, playlist_id) - # Only process container div with main playlist content skipping - # drop-down menu that uses similar pattern for videos (see - # https://github.com/rg3/youtube-dl/issues/11594). - container = self._search_regex( - r'(?s)(<div[^>]+class=["\']container.+)', webpage, - 'container', default=webpage) - - entries = self._extract_entries(container) + entries = self._extract_entries(webpage, host) playlist = self._parse_json( self._search_regex( - r'playlistObject\s*=\s*({.+?});', webpage, 'playlist'), - playlist_id) + r'(?:playlistObject|PLAYLIST_VIEW)\s*=\s*({.+?});', webpage, + 'playlist', default='{}'), + playlist_id, fatal=False) + title = playlist.get('title') or self._search_regex( + r'>Videos\s+in\s+(.+?)\s+[Pp]laylist<', webpage, 'title', fatal=False) return self.playlist_result( - entries, playlist_id, playlist.get('title'), playlist.get('description')) - - -class PornHubPlaylistIE(PornHubPlaylistBaseIE): - _VALID_URL = r'https?://(?:www\.)?pornhub\.com/playlist/(?P<id>\d+)' - _TESTS = [{ - 'url': 'http://www.pornhub.com/playlist/4667351', - 'info_dict': { - 'id': '4667351', - 'title': 'Nataly Hot', - }, - 'playlist_mincount': 2, - }] + entries, playlist_id, title, playlist.get('description')) -class PornHubUserVideosIE(PornHubPlaylistBaseIE): - _VALID_URL = r'https?://(?:www\.)?pornhub\.com/users/(?P<id>[^/]+)/videos' +class PornHubUserIE(PornHubPlaylistBaseIE): + _VALID_URL = r'(?P<url>https?://(?:[^/]+\.)?(?P<host>pornhub(?:premium)?\.(?:com|net))/(?:(?:user|channel)s|model|pornstar)/(?P<id>[^/?#&]+))(?:[?#&]|/(?!videos)|$)' _TESTS = [{ - 'url': 'http://www.pornhub.com/users/zoe_ph/videos/public', + 'url': 'https://www.pornhub.com/model/zoe_ph', + 'playlist_mincount': 118, + }, { + 'url': 'https://www.pornhub.com/pornstar/liz-vicious', 'info_dict': { - 'id': 'zoe_ph', + 'id': 'liz-vicious', }, - 'playlist_mincount': 171, + 'playlist_mincount': 118, }, { - 'url': 'http://www.pornhub.com/users/rushandlia/videos', + 'url': 'https://www.pornhub.com/users/russianveet69', + 'only_matching': True, + }, { + 'url': 'https://www.pornhub.com/channels/povd', + 'only_matching': True, + }, { + 'url': 'https://www.pornhub.com/model/zoe_ph?abc=1', 'only_matching': True, }] def _real_extract(self, url): - user_id = self._match_id(url) + mobj = re.match(self._VALID_URL, url) + user_id = mobj.group('id') + return self.url_result( + '%s/videos' % mobj.group('url'), ie=PornHubPagedVideoListIE.ie_key(), + video_id=user_id) + + +class PornHubPagedPlaylistBaseIE(PornHubPlaylistBaseIE): + @staticmethod + def _has_more(webpage): + return re.search( + r'''(?x) + <li[^>]+\bclass=["\']page_next| + <link[^>]+\brel=["\']next| + <button[^>]+\bid=["\']moreDataBtn + ''', webpage) is not None + + def _real_extract(self, url): + mobj = re.match(self._VALID_URL, url) + host = mobj.group('host') + item_id = mobj.group('id') + + page = int_or_none(self._search_regex( + r'\bpage=(\d+)', url, 'page', default=None)) entries = [] - for page_num in itertools.count(1): + for page_num in (page, ) if page is not None else itertools.count(1): try: webpage = self._download_webpage( - url, user_id, 'Downloading page %d' % page_num, + url, item_id, 'Downloading page %d' % page_num, query={'page': page_num}) except ExtractorError as e: if isinstance(e.cause, compat_HTTPError) and e.cause.code == 404: break - page_entries = self._extract_entries(webpage) + raise + page_entries = self._extract_entries(webpage, host) if not page_entries: break entries.extend(page_entries) + if not self._has_more(webpage): + break + + return self.playlist_result(orderedSet(entries), item_id) + + +class PornHubPagedVideoListIE(PornHubPagedPlaylistBaseIE): + _VALID_URL = r'https?://(?:[^/]+\.)?(?P<host>pornhub(?:premium)?\.(?:com|net))/(?P<id>(?:[^/]+/)*[^/?#&]+)' + _TESTS = [{ + 'url': 'https://www.pornhub.com/model/zoe_ph/videos', + 'only_matching': True, + }, { + 'url': 'http://www.pornhub.com/users/rushandlia/videos', + 'only_matching': True, + }, { + 'url': 'https://www.pornhub.com/pornstar/jenny-blighe/videos', + 'info_dict': { + 'id': 'pornstar/jenny-blighe/videos', + }, + 'playlist_mincount': 149, + }, { + 'url': 'https://www.pornhub.com/pornstar/jenny-blighe/videos?page=3', + 'info_dict': { + 'id': 'pornstar/jenny-blighe/videos', + }, + 'playlist_mincount': 40, + }, { + # default sorting as Top Rated Videos + 'url': 'https://www.pornhub.com/channels/povd/videos', + 'info_dict': { + 'id': 'channels/povd/videos', + }, + 'playlist_mincount': 293, + }, { + # Top Rated Videos + 'url': 'https://www.pornhub.com/channels/povd/videos?o=ra', + 'only_matching': True, + }, { + # Most Recent Videos + 'url': 'https://www.pornhub.com/channels/povd/videos?o=da', + 'only_matching': True, + }, { + # Most Viewed Videos + 'url': 'https://www.pornhub.com/channels/povd/videos?o=vi', + 'only_matching': True, + }, { + 'url': 'http://www.pornhub.com/users/zoe_ph/videos/public', + 'only_matching': True, + }, { + # Most Viewed Videos + 'url': 'https://www.pornhub.com/pornstar/liz-vicious/videos?o=mv', + 'only_matching': True, + }, { + # Top Rated Videos + 'url': 'https://www.pornhub.com/pornstar/liz-vicious/videos?o=tr', + 'only_matching': True, + }, { + # Longest Videos + 'url': 'https://www.pornhub.com/pornstar/liz-vicious/videos?o=lg', + 'only_matching': True, + }, { + # Newest Videos + 'url': 'https://www.pornhub.com/pornstar/liz-vicious/videos?o=cm', + 'only_matching': True, + }, { + 'url': 'https://www.pornhub.com/pornstar/liz-vicious/videos/paid', + 'only_matching': True, + }, { + 'url': 'https://www.pornhub.com/pornstar/liz-vicious/videos/fanonly', + 'only_matching': True, + }, { + 'url': 'https://www.pornhub.com/video', + 'only_matching': True, + }, { + 'url': 'https://www.pornhub.com/video?page=3', + 'only_matching': True, + }, { + 'url': 'https://www.pornhub.com/video/search?search=123', + 'only_matching': True, + }, { + 'url': 'https://www.pornhub.com/categories/teen', + 'only_matching': True, + }, { + 'url': 'https://www.pornhub.com/categories/teen?page=3', + 'only_matching': True, + }, { + 'url': 'https://www.pornhub.com/hd', + 'only_matching': True, + }, { + 'url': 'https://www.pornhub.com/hd?page=3', + 'only_matching': True, + }, { + 'url': 'https://www.pornhub.com/described-video', + 'only_matching': True, + }, { + 'url': 'https://www.pornhub.com/described-video?page=2', + 'only_matching': True, + }, { + 'url': 'https://www.pornhub.com/video/incategories/60fps-1/hd-porn', + 'only_matching': True, + }, { + 'url': 'https://www.pornhub.com/playlist/44121572', + 'info_dict': { + 'id': 'playlist/44121572', + }, + 'playlist_mincount': 132, + }, { + 'url': 'https://www.pornhub.com/playlist/4667351', + 'only_matching': True, + }, { + 'url': 'https://de.pornhub.com/playlist/4667351', + 'only_matching': True, + }] + + @classmethod + def suitable(cls, url): + return (False + if PornHubIE.suitable(url) or PornHubUserIE.suitable(url) or PornHubUserVideosUploadIE.suitable(url) + else super(PornHubPagedVideoListIE, cls).suitable(url)) + - return self.playlist_result(entries, user_id) +class PornHubUserVideosUploadIE(PornHubPagedPlaylistBaseIE): + _VALID_URL = r'(?P<url>https?://(?:[^/]+\.)?(?P<host>pornhub(?:premium)?\.(?:com|net))/(?:(?:user|channel)s|model|pornstar)/(?P<id>[^/]+)/videos/upload)' + _TESTS = [{ + 'url': 'https://www.pornhub.com/pornstar/jenny-blighe/videos/upload', + 'info_dict': { + 'id': 'jenny-blighe', + }, + 'playlist_mincount': 129, + }, { + 'url': 'https://www.pornhub.com/model/zoe_ph/videos/upload', + 'only

X-Git-Url: http://git.bitcoin.ninja/index.cgi?p=youtube-dl;a=blobdiff_plain;f=youtube_dl%2Fextractor%2Fpornhub.py;h=c64c870dc8edbf33121b592d7e424399132503eb;hp=017f6c55219ff3db0cf9bf745f74031882f12c54;hb=540b9f5164d50eb99d9c988ece6eb6775ccaf94a;hpb=f28aeff26416515fbaf46ce087e8332a15958c23 diff --git a/youtube_dl/extractor/pornhub.py b/youtube_dl/extractor/pornhub.py index 017f6c552..c64c870dc 100644 --- a/youtube_dl/extractor/pornhub.py +++ b/youtube_dl/extractor/pornhub.py @@ -1,36 +1,58 @@ # coding: utf-8 from __future__ import unicode_literals +import functools import itertools -import os +import operator import re from .common import InfoExtractor from ..compat import ( compat_HTTPError, - compat_urllib_parse_unquote, - compat_urllib_parse_unquote_plus, - compat_urllib_parse_urlparse, + compat_str, + compat_urllib_request, ) +from .openload import PhantomJSwrapper from ..utils import ( + determine_ext, ExtractorError, int_or_none, - js_to_json, + NO_DEFAULT, orderedSet, - sanitized_Request, + remove_quotes, str_to_int, + url_or_none, ) -from ..aes import ( - aes_decrypt_text -) -class PornHubIE(InfoExtractor): +class PornHubBaseIE(InfoExtractor): + def _download_webpage_handle(self, *args, **kwargs): + def dl(*args, **kwargs): + return super(PornHubBaseIE, self)._download_webpage_handle(*args, **kwargs) + + webpage, urlh = dl(*args, **kwargs) + + if any(re.search(p, webpage) for p in ( + r']+\bonload=["\']go', + r'document\.cookie\s*=\s*["\']RNKEY=', + r'document\.location\.reload$true$')): + url_or_request = args[0] + url = (url_or_request.get_full_url() + if isinstance(url_or_request, compat_urllib_request.Request) + else url_or_request) + phantom = PhantomJSwrapper(self, required_version='2.0') + phantom.get(url, html=webpage) + webpage, urlh = dl(*args, **kwargs) + + return webpage, urlh + + +class PornHubIE(PornHubBaseIE): IE_DESC = 'PornHub and Thumbzilla' _VALID_URL = r'''(?x) https?:// (?: - (?:[a-z]+\.)?pornhub\.com/(?:view_video\.php\?viewkey=|embed/)| + (?:[^/]+\.)?(?Ppornhub(?:premium)?\.(?:com|net))/(?:(?:view_video\.php|video/show)\?viewkey=|embed/)| (?:www\.)?thumbzilla\.com/video/ ) (?P[\da-z]+) @@ -43,6 +65,7 @@ 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, @@ -59,7 +82,8 @@ class PornHubIE(InfoExtractor): 'id': '1331683002', 'ext': 'mp4', 'title': 'éåºå©·å©·å¥³çè¶³äº¤', - 'uploader': 'cj397186295', + 'uploader': 'Unknown', + 'upload_date': '20150213', 'duration': 1753, 'view_count': int, 'like_count': int, @@ -72,6 +96,31 @@ class PornHubIE(InfoExtractor): 'params': { 'skip_download': True, }, + }, { + # subtitles + 'url': 'https://www.pornhub.com/view_video.php?viewkey=ph5af5fef7c2aa7', + 'info_dict': { + 'id': 'ph5af5fef7c2aa7', + 'ext': 'mp4', + 'title': 'BFFS - Cute Teen Girls Share Cock On the Floor', + 'uploader': 'BFFs', + 'duration': 622, + 'view_count': int, + 'like_count': int, + 'dislike_count': int, + 'comment_count': int, + 'age_limit': 18, + 'tags': list, + 'categories': list, + 'subtitles': { + 'en': [{ + "ext": 'srt' + }] + }, + }, + 'params': { + 'skip_download': True, + }, }, { 'url': 'http://www.pornhub.com/view_video.php?viewkey=ph557bbb6676d2d', 'only_matching': True, @@ -94,12 +143,21 @@ class PornHubIE(InfoExtractor): }, { 'url': 'https://www.thumbzilla.com/video/ph56c6114abd99a/horny-girlfriend-sex', 'only_matching': True, + }, { + 'url': 'http://www.pornhub.com/video/show?viewkey=648719015', + 'only_matching': True, + }, { + 'url': 'https://www.pornhub.net/view_video.php?viewkey=203640933', + 'only_matching': True, + }, { + 'url': 'https://www.pornhubpremium.com/view_video.php?viewkey=ph5e4acdae54a82', + 'only_matching': True, }] @staticmethod def _extract_urls(webpage): return re.findall( - r']+?src=["\'](?P(?:https?:)?//(?:www\.)?pornhub\.com/embed/[\da-z]+)', + r']+?src=["\'](?P(?:https?:)?//(?:www\.)?pornhub\.(?:com|net)/embed/[\da-z]+)', webpage) def _extract_count(self, pattern, webpage, name): @@ -107,12 +165,26 @@ class PornHubIE(InfoExtractor): pattern, webpage, '%s count' % name, fatal=False)) def _real_extract(self, url): - video_id = self._match_id(url) + mobj = re.match(self._VALID_URL, url) + host = mobj.group('host') or 'pornhub.com' + video_id = mobj.group('id') + + if 'premium' in host: + if not self._downloader.params.get('cookiefile'): + raise ExtractorError( + 'PornHub Premium requires authentication.' + ' You may want to use --cookies.', + expected=True) + + self._set_cookie(host, 'age_verified', '1') - req = sanitized_Request( - 'http://www.pornhub.com/view_video.php?viewkey=%s' % video_id) - req.add_header('Cookie', 'age_verified=1') - webpage = self._download_webpage(req, video_id) + def dl_webpage(platform): + self._set_cookie(host, 'platform', platform) + return self._download_webpage( + 'https://www.%s/view_video.php?viewkey=%s' % (host, video_id), + video_id, 'Downloading %s webpage' % platform) + + webpage = dl_webpage('pc') error_msg = self._html_search_regex( r'(?s)]+class=(["\'])(?:(?!\1).)*\b(?:removed|userMessageSection)\b(?:(?!\1).)*\1[^>]*>(?P.+?)', @@ -127,28 +199,149 @@ class PornHubIE(InfoExtractor): # http://www.pornhub.com/view_video.php?viewkey=1331683002), not relying # on that anymore. title = self._html_search_meta( - 'twitter:title', webpage, default=None) or self._search_regex( - (r']+class=["\']title["\'][^>]*>(?P[^<]+)', - r'<div[^>]+data-video-title=(["\'])(?P<title>.+?)\1', - r'shareTitle\s*=\s*(["\'])(?P<title>.+?)\1'), + 'twitter:title', webpage, default=None) or self._html_search_regex( + (r'(?s)<h1[^>]+class=["\']title["\'][^>]*>(?P<title>.+?)</h1>', + r'<div[^>]+data-video-title=(["\'])(?P<title>(?:(?!\1).)+)\1', + r'shareTitle["\']\s*[=:]\s*(["\'])(?P<title>(?:(?!\1).)+)\1'), webpage, 'title', group='title') + video_urls = [] + video_urls_set = set() + subtitles = {} + flashvars = self._parse_json( self._search_regex( r'var\s+flashvars_\d+\s*=\s*({.+?});', webpage, 'flashvars', default='{}'), video_id) if flashvars: + subtitle_url = url_or_none(flashvars.get('closedCaptionsFile')) + if subtitle_url: + subtitles.setdefault('en', []).append({ + 'url': subtitle_url, + 'ext': 'srt', + }) thumbnail = flashvars.get('image_url') duration = int_or_none(flashvars.get('video_duration')) + media_definitions = flashvars.get('mediaDefinitions') + if isinstance(media_definitions, list): + for definition in media_definitions: + if not isinstance(definition, dict): + continue + video_url = definition.get('videoUrl') + if not video_url or not isinstance(video_url, compat_str): + continue + if video_url in video_urls_set: + continue + video_urls_set.add(video_url) + video_urls.append( + (video_url, int_or_none(definition.get('quality')))) else: - title, thumbnail, duration = [None] * 3 + thumbnail, duration = [None] * 2 + + def extract_js_vars(webpage, pattern, default=NO_DEFAULT): + assignments = self._search_regex( + pattern, webpage, 'encoded url', default=default) + if not assignments: + return {} + + assignments = assignments.split(';') + + js_vars = {} + + def parse_js_value(inp): + inp = re.sub(r'/\*(?:(?!\*/).)*?\*/', '', inp) + if '+' in inp: + inps = inp.split('+') + return functools.reduce( + operator.concat, map(parse_js_value, inps)) + inp = inp.strip() + if inp in js_vars: + return js_vars[inp] + return remove_quotes(inp) + + for assn in assignments: + assn = assn.strip() + if not assn: + continue + assn = re.sub(r'var\s+', '', assn) + vname, value = assn.split('=', 1) + js_vars[vname] = parse_js_value(value) + return js_vars + + def add_video_url(video_url): + v_url = url_or_none(video_url) + if not v_url: + return + if v_url in video_urls_set: + return + video_urls.append((v_url, None)) + video_urls_set.add(v_url) + + if not video_urls: + FORMAT_PREFIXES = ('media', 'quality') + js_vars = extract_js_vars( + webpage, r'(var\s+(?:%s)_.+)' % '|'.join(FORMAT_PREFIXES), + default=None) + if js_vars: + for key, format_url in js_vars.items(): + if any(key.startswith(p) for p in FORMAT_PREFIXES): + add_video_url(format_url) + if not video_urls and re.search( + r'<[^>]+\bid=["\']lockedPlayer', webpage): + raise ExtractorError( + 'Video %s is locked' % video_id, expected=True) + + if not video_urls: + js_vars = extract_js_vars( + dl_webpage('tv'), r'(var.+?mediastring.+?)</script>') + add_video_url(js_vars['mediastring']) + + for mobj in re.finditer( + r'<a[^>]+\bclass=["\']downloadBtn\b[^>]+\bhref=(["\'])(?P<url>(?:(?!\1).)+)\1', + webpage): + video_url = mobj.group('url') + if video_url not in video_urls_set: + 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('/', '') + ext = determine_ext(video_url) + if ext == 'mpd': + formats.extend(self._extract_mpd_formats( + video_url, video_id, mpd_id='dash', fatal=False)) + continue + elif ext == 'm3u8': + formats.extend(self._extract_m3u8_formats( + video_url, video_id, 'mp4', entry_protocol='m3u8_native', + m3u8_id='hls', fatal=False)) + continue + tbr = None + mobj = re.search(r'(?P<height>\d+)[pP]?_(?P<tbr>\d+)[kK]', video_url) + if mobj: + if not height: + height = int(mobj.group('height')) + tbr = int(mobj.group('tbr')) + formats.append({ + 'url': video_url, + 'format_id': '%dp' % height if height else None, + 'height': height, + 'tbr': tbr, + }) + self._sort_formats(formats) video_uploader = self._html_search_regex( - r'(?s)From: .+?<(?:a href="/users/|a href="/channels/|span class="username)[^>]+>(.+?)<', + r'(?s)From: .+?<(?:a\b[^>]+\bhref=["\']/(?:(?:user|channel)s|model|pornstar)/|span\b[^>]+\bclass=["\']username)[^>]+>(.+?)<', webpage, 'uploader', fatal=False) view_count = self._extract_count( - r'<span class="count">([\d,\.]+)</span> views', webpage, 'view') + r'<span class="count">([\d,\.]+)</span> [Vv]iews', webpage, 'view') like_count = self._extract_count( r'<span class="votesUp">([\d,\.]+)</span>', webpage, 'like') dislike_count = self._extract_count( @@ -156,54 +349,17 @@ class PornHubIE(InfoExtractor): comment_count = self._extract_count( r'All Comments\s*<span>$([\d,.]+)$', webpage, 'comment') - video_urls = [] - for quote, video_url in re.findall( - r'player_quality_[0-9]{3,4}p\s*=\s*(["\'])(.+?)\1;', webpage): - video_urls.append(compat_urllib_parse_unquote(re.sub( - r'{0}\s*\+\s*{0}'.format(quote), '', video_url))) - - if webpage.find('"encrypted":true') != -1: - password = compat_urllib_parse_unquote_plus( - self._search_regex(r'"video_title":"([^"]+)', webpage, 'password')) - video_urls = list(map(lambda s: aes_decrypt_text(s, password, 32).decode('utf-8'), video_urls)) - - formats = [] - for video_url in video_urls: - path = compat_urllib_parse_urlparse(video_url).path - extension = os.path.splitext(path)[1][1:] - format = path.split('/')[5].split('_')[:2] - format = '-'.join(format) - - m = re.match(r'^(?P<height>[0-9]+)[pP]-(?P<tbr>[0-9]+)[kK]$', format) - if m is None: - height = None - tbr = None - else: - height = int(m.group('height')) - tbr = int(m.group('tbr')) - - formats.append({ - 'url': video_url, - 'ext': extension, - 'format': format, - 'format_id': format, - 'tbr': tbr, - 'height': height, - }) - self._sort_formats(formats) - - page_params = self._parse_json(self._search_regex( - r'page_params\.zoneDetails\[([\'"])[^\'"]+\1\]\s*=\s*(?P<data>{[^}]+})', - webpage, 'page parameters', group='data', default='{}'), - video_id, transform_source=js_to_json, fatal=False) - tags = categories = None - if page_params: - tags = page_params.get('tags', '').split(',') - categories = page_params.get('categories', '').split(',') + def extract_list(meta_key): + div = self._search_regex( + r'(?s)<div[^>]+\bclass=["\'].*?\b%sWrapper[^>]*>(.+?)</div>' + % meta_key, webpage, meta_key, default=None) + if div: + return re.findall(r'<a[^>]+\bhref=[^>]+>([^<]+)', div) return { 'id': video_id, 'uploader': video_uploader, + 'upload_date': upload_date, 'title': title, 'thumbnail': thumbnail, 'duration': duration, @@ -213,85 +369,243 @@ class PornHubIE(InfoExtractor): 'comment_count': comment_count, 'formats': formats, 'age_limit': 18, - 'tags': tags, - 'categories': categories, + 'tags': extract_list('tags'), + 'categories': extract_list('categories'), + 'subtitles': subtitles, } -class PornHubPlaylistBaseIE(InfoExtractor): - def _extract_entries(self, webpage): +class PornHubPlaylistBaseIE(PornHubBaseIE): + def _extract_entries(self, webpage, host): + # Only process container div with main playlist content skipping + # drop-down menu that uses similar pattern for videos (see + # https://github.com/ytdl-org/youtube-dl/issues/11594). + container = self._search_regex( + r'(?s)(<div[^>]+class=["\']container.+)', webpage, + 'container', default=webpage) + return [ self.url_result( - 'http://www.pornhub.com/%s' % video_url, + 'http://www.%s/%s' % (host, video_url), PornHubIE.ie_key(), video_title=title) for video_url, title in orderedSet(re.findall( r'href="/?(view_video\.php\?.*\bviewkey=[\da-z]+[^"]*)"[^>]*\s+title="([^"]+)"', - webpage)) + container)) ] def _real_extract(self, url): - playlist_id = self._match_id(url) + mobj = re.match(self._VALID_URL, url) + host = mobj.group('host') + playlist_id = mobj.group('id') webpage = self._download_webpage(url, playlist_id) - # Only process container div with main playlist content skipping - # drop-down menu that uses similar pattern for videos (see - # https://github.com/rg3/youtube-dl/issues/11594). - container = self._search_regex( - r'(?s)(<div[^>]+class=["\']container.+)', webpage, - 'container', default=webpage) - - entries = self._extract_entries(container) + entries = self._extract_entries(webpage, host) playlist = self._parse_json( self._search_regex( - r'playlistObject\s*=\s*({.+?});', webpage, 'playlist'), - playlist_id) + r'(?:playlistObject|PLAYLIST_VIEW)\s*=\s*({.+?});', webpage, + 'playlist', default='{}'), + playlist_id, fatal=False) + title = playlist.get('title') or self._search_regex( + r'>Videos\s+in\s+(.+?)\s+[Pp]laylist<', webpage, 'title', fatal=False) return self.playlist_result( - entries, playlist_id, playlist.get('title'), playlist.get('description')) - - -class PornHubPlaylistIE(PornHubPlaylistBaseIE): - _VALID_URL = r'https?://(?:www\.)?pornhub\.com/playlist/(?P<id>\d+)' - _TESTS = [{ - 'url': 'http://www.pornhub.com/playlist/4667351', - 'info_dict': { - 'id': '4667351', - 'title': 'Nataly Hot', - }, - 'playlist_mincount': 2, - }] + entries, playlist_id, title, playlist.get('description')) -class PornHubUserVideosIE(PornHubPlaylistBaseIE): - _VALID_URL = r'https?://(?:www\.)?pornhub\.com/users/(?P<id>[^/]+)/videos' +class PornHubUserIE(PornHubPlaylistBaseIE): + _VALID_URL = r'(?P<url>https?://(?:[^/]+\.)?(?P<host>pornhub(?:premium)?\.(?:com|net))/(?:(?:user|channel)s|model|pornstar)/(?P<id>[^/?#&]+))(?:[?#&]|/(?!videos)|$)' _TESTS = [{ - 'url': 'http://www.pornhub.com/users/zoe_ph/videos/public', + 'url': 'https://www.pornhub.com/model/zoe_ph', + 'playlist_mincount': 118, + }, { + 'url': 'https://www.pornhub.com/pornstar/liz-vicious', 'info_dict': { - 'id': 'zoe_ph', + 'id': 'liz-vicious', }, - 'playlist_mincount': 171, + 'playlist_mincount': 118, }, { - 'url': 'http://www.pornhub.com/users/rushandlia/videos', + 'url': 'https://www.pornhub.com/users/russianveet69', + 'only_matching': True, + }, { + 'url': 'https://www.pornhub.com/channels/povd', + 'only_matching': True, + }, { + 'url': 'https://www.pornhub.com/model/zoe_ph?abc=1', 'only_matching': True, }] def _real_extract(self, url): - user_id = self._match_id(url) + mobj = re.match(self._VALID_URL, url) + user_id = mobj.group('id') + return self.url_result( + '%s/videos' % mobj.group('url'), ie=PornHubPagedVideoListIE.ie_key(), + video_id=user_id) + + +class PornHubPagedPlaylistBaseIE(PornHubPlaylistBaseIE): + @staticmethod + def _has_more(webpage): + return re.search( + r'''(?x) + <li[^>]+\bclass=["\']page_next| + <link[^>]+\brel=["\']next| + <button[^>]+\bid=["\']moreDataBtn + ''', webpage) is not None + + def _real_extract(self, url): + mobj = re.match(self._VALID_URL, url) + host = mobj.group('host') + item_id = mobj.group('id') + + page = int_or_none(self._search_regex( + r'\bpage=(\d+)', url, 'page', default=None)) entries = [] - for page_num in itertools.count(1): + for page_num in (page, ) if page is not None else itertools.count(1): try: webpage = self._download_webpage( - url, user_id, 'Downloading page %d' % page_num, + url, item_id, 'Downloading page %d' % page_num, query={'page': page_num}) except ExtractorError as e: if isinstance(e.cause, compat_HTTPError) and e.cause.code == 404: break - page_entries = self._extract_entries(webpage) + raise + page_entries = self._extract_entries(webpage, host) if not page_entries: break entries.extend(page_entries) + if not self._has_more(webpage): + break + + return self.playlist_result(orderedSet(entries), item_id) + + +class PornHubPagedVideoListIE(PornHubPagedPlaylistBaseIE): + _VALID_URL = r'https?://(?:[^/]+\.)?(?P<host>pornhub(?:premium)?\.(?:com|net))/(?P<id>(?:[^/]+/)*[^/?#&]+)' + _TESTS = [{ + 'url': 'https://www.pornhub.com/model/zoe_ph/videos', + 'only_matching': True, + }, { + 'url': 'http://www.pornhub.com/users/rushandlia/videos', + 'only_matching': True, + }, { + 'url': 'https://www.pornhub.com/pornstar/jenny-blighe/videos', + 'info_dict': { + 'id': 'pornstar/jenny-blighe/videos', + }, + 'playlist_mincount': 149, + }, { + 'url': 'https://www.pornhub.com/pornstar/jenny-blighe/videos?page=3', + 'info_dict': { + 'id': 'pornstar/jenny-blighe/videos', + }, + 'playlist_mincount': 40, + }, { + # default sorting as Top Rated Videos + 'url': 'https://www.pornhub.com/channels/povd/videos', + 'info_dict': { + 'id': 'channels/povd/videos', + }, + 'playlist_mincount': 293, + }, { + # Top Rated Videos + 'url': 'https://www.pornhub.com/channels/povd/videos?o=ra', + 'only_matching': True, + }, { + # Most Recent Videos + 'url': 'https://www.pornhub.com/channels/povd/videos?o=da', + 'only_matching': True, + }, { + # Most Viewed Videos + 'url': 'https://www.pornhub.com/channels/povd/videos?o=vi', + 'only_matching': True, + }, { + 'url': 'http://www.pornhub.com/users/zoe_ph/videos/public', + 'only_matching': True, + }, { + # Most Viewed Videos + 'url': 'https://www.pornhub.com/pornstar/liz-vicious/videos?o=mv', + 'only_matching': True, + }, { + # Top Rated Videos + 'url': 'https://www.pornhub.com/pornstar/liz-vicious/videos?o=tr', + 'only_matching': True, + }, { + # Longest Videos + 'url': 'https://www.pornhub.com/pornstar/liz-vicious/videos?o=lg', + 'only_matching': True, + }, { + # Newest Videos + 'url': 'https://www.pornhub.com/pornstar/liz-vicious/videos?o=cm', + 'only_matching': True, + }, { + 'url': 'https://www.pornhub.com/pornstar/liz-vicious/videos/paid', + 'only_matching': True, + }, { + 'url': 'https://www.pornhub.com/pornstar/liz-vicious/videos/fanonly', + 'only_matching': True, + }, { + 'url': 'https://www.pornhub.com/video', + 'only_matching': True, + }, { + 'url': 'https://www.pornhub.com/video?page=3', + 'only_matching': True, + }, { + 'url': 'https://www.pornhub.com/video/search?search=123', + 'only_matching': True, + }, { + 'url': 'https://www.pornhub.com/categories/teen', + 'only_matching': True, + }, { + 'url': 'https://www.pornhub.com/categories/teen?page=3', + 'only_matching': True, + }, { + 'url': 'https://www.pornhub.com/hd', + 'only_matching': True, + }, { + 'url': 'https://www.pornhub.com/hd?page=3', + 'only_matching': True, + }, { + 'url': 'https://www.pornhub.com/described-video', + 'only_matching': True, + }, { + 'url': 'https://www.pornhub.com/described-video?page=2', + 'only_matching': True, + }, { + 'url': 'https://www.pornhub.com/video/incategories/60fps-1/hd-porn', + 'only_matching': True, + }, { + 'url': 'https://www.pornhub.com/playlist/44121572', + 'info_dict': { + 'id': 'playlist/44121572', + }, + 'playlist_mincount': 132, + }, { + 'url': 'https://www.pornhub.com/playlist/4667351', + 'only_matching': True, + }, { + 'url': 'https://de.pornhub.com/playlist/4667351', + 'only_matching': True, + }] + + @classmethod + def suitable(cls, url): + return (False + if PornHubIE.suitable(url) or PornHubUserIE.suitable(url) or PornHubUserVideosUploadIE.suitable(url) + else super(PornHubPagedVideoListIE, cls).suitable(url)) + - return self.playlist_result(entries, user_id) +class PornHubUserVideosUploadIE(PornHubPagedPlaylistBaseIE): + _VALID_URL = r'(?P<url>https?://(?:[^/]+\.)?(?P<host>pornhub(?:premium)?\.(?:com|net))/(?:(?:user|channel)s|model|pornstar)/(?P<id>[^/]+)/videos/upload)' + _TESTS = [{ + 'url': 'https://www.pornhub.com/pornstar/jenny-blighe/videos/upload', + 'info_dict': { + 'id': 'jenny-blighe', + }, + 'playlist_mincount': 129, + }, { + 'url': 'https://www.pornhub.com/model/zoe_ph/videos/upload', + 'only_matching': True, + }]