X-Git-Url: http://git.cielonegro.org/gitweb.cgi?a=blobdiff_plain;f=youtube_dl%2Fextractor%2Fpornhub.py;h=97f988da4ac249874e7bf47c570a89be5eb9bc8b;hb=234a85858cd50d6c88848c2e0ceffeabe26306e4;hp=d2c92531b0745ed189da699f4392725f2d4cddd6;hpb=5b6a74856babce30e9e72701259f790322281d3a;p=youtube-dl.git diff --git a/youtube_dl/extractor/pornhub.py b/youtube_dl/extractor/pornhub.py index d2c92531b..97f988da4 100644 --- a/youtube_dl/extractor/pornhub.py +++ b/youtube_dl/extractor/pornhub.py @@ -1,27 +1,24 @@ # 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, ) from ..utils import ( ExtractorError, int_or_none, + js_to_json, orderedSet, - sanitized_Request, + remove_quotes, str_to_int, ) -from ..aes import ( - aes_decrypt_text -) class PornHubIE(InfoExtractor): @@ -29,10 +26,10 @@ class PornHubIE(InfoExtractor): _VALID_URL = r'''(?x) https?:// (?: - (?:[a-z]+\.)?pornhub\.com/(?:view_video\.php\?viewkey=|embed/)| + (?:[^/]+\.)?pornhub\.com/(?:(?:view_video\.php|video/show)\?viewkey=|embed/)| (?:www\.)?thumbzilla\.com/video/ ) - (?P[0-9a-z]+) + (?P[\da-z]+) ''' _TESTS = [{ 'url': 'http://www.pornhub.com/view_video.php?viewkey=648719015', @@ -48,6 +45,8 @@ class PornHubIE(InfoExtractor): 'dislike_count': int, 'comment_count': int, 'age_limit': 18, + 'tags': list, + 'categories': list, }, }, { # non-ASCII title @@ -56,13 +55,15 @@ class PornHubIE(InfoExtractor): 'id': '1331683002', 'ext': 'mp4', 'title': '重庆婷婷女王足交', - 'uploader': 'cj397186295', + 'uploader': 'Unknown', 'duration': 1753, 'view_count': int, 'like_count': int, 'dislike_count': int, 'comment_count': int, 'age_limit': 18, + 'tags': list, + 'categories': list, }, 'params': { 'skip_download': True, @@ -89,14 +90,16 @@ 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, }] - @classmethod - def _extract_url(cls, webpage): - mobj = re.search( - r']+?src=(["\'])(?P(?:https?:)?//(?:www\.)?pornhub\.com/embed/\d+)\1', webpage) - if mobj: - return mobj.group('url') + @staticmethod + def _extract_urls(webpage): + return re.findall( + r']+?src=["\'](?P(?:https?:)?//(?:www\.)?pornhub\.com/embed/[\da-z]+)', + webpage) def _extract_count(self, pattern, webpage, name): return str_to_int(self._search_regex( @@ -105,13 +108,18 @@ class PornHubIE(InfoExtractor): def _real_extract(self, url): video_id = self._match_id(url) - 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) + self._set_cookie('pornhub.com', 'age_verified', '1') + + def dl_webpage(platform): + self._set_cookie('pornhub.com', 'platform', platform) + return self._download_webpage( + 'http://www.pornhub.com/view_video.php?viewkey=%s' % video_id, + video_id, 'Downloading %s webpage' % platform) + + webpage = dl_webpage('pc') error_msg = self._html_search_regex( - r'(?s)]+class=(["\']).*?\b(?:removed|userMessageSection)\b.*?\1[^>]*>(?P.+?)', + r'(?s)]+class=(["\'])(?:(?!\1).)*\b(?:removed|userMessageSection)\b(?:(?!\1).)*\1[^>]*>(?P.+?)', webpage, 'error message', default=None, group='error') if error_msg: error_msg = re.sub(r'\s+', ' ', error_msg) @@ -129,6 +137,9 @@ class PornHubIE(InfoExtractor): r'shareTitle\s*=\s*(["\'])(?P.+?)\1'), webpage, 'title', group='title') + video_urls = [] + video_urls_set = set() + flashvars = self._parse_json( self._search_regex( r'var\s+flashvars_\d+\s*=\s*({.+?});', webpage, 'flashvars', default='{}'), @@ -136,11 +147,81 @@ class PornHubIE(InfoExtractor): if flashvars: 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 + + if not video_urls: + tv_webpage = dl_webpage('tv') + + assignments = self._search_regex( + r'(var.+?mediastring.+?)</script>', tv_webpage, + 'encoded url').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) + + video_url = js_vars['mediastring'] + if video_url not in video_urls_set: + video_urls.append((video_url, None)) + video_urls_set.add(video_url) + + 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) + + formats = [] + for video_url, height in video_urls: + 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/|span\b[^>]+\bclass=["\']username)[^>]+>(.+?)<', webpage, 'uploader', fatal=False) view_count = self._extract_count( @@ -152,36 +233,14 @@ class PornHubIE(InfoExtractor): comment_count = self._extract_count( r'All Comments\s*<span>\(([\d,.]+)\)', webpage, 'comment') - video_urls = list(map(compat_urllib_parse_unquote, re.findall(r"player_quality_[0-9]{3}p\s*=\s*'([^']+)'", webpage))) - 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(',') return { 'id': video_id, @@ -195,18 +254,27 @@ class PornHubIE(InfoExtractor): 'comment_count': comment_count, 'formats': formats, 'age_limit': 18, + 'tags': tags, + 'categories': categories, } class PornHubPlaylistBaseIE(InfoExtractor): def _extract_entries(self, webpage): + # 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) + return [ self.url_result( 'http://www.pornhub.com/%s' % 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): @@ -218,27 +286,33 @@ class PornHubPlaylistBaseIE(InfoExtractor): 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')) + entries, playlist_id, title, playlist.get('description')) class PornHubPlaylistIE(PornHubPlaylistBaseIE): - _VALID_URL = r'https?://(?:www\.)?pornhub\.com/playlist/(?P<id>\d+)' + _VALID_URL = r'https?://(?:[^/]+\.)?pornhub\.com/playlist/(?P<id>\d+)' _TESTS = [{ - 'url': 'http://www.pornhub.com/playlist/6201671', + 'url': 'http://www.pornhub.com/playlist/4667351', 'info_dict': { - 'id': '6201671', - 'title': 'P0p4', + 'id': '4667351', + 'title': 'Nataly Hot', }, - 'playlist_mincount': 35, + 'playlist_mincount': 2, + }, { + 'url': 'https://de.pornhub.com/playlist/4667351', + 'only_matching': True, }] class PornHubUserVideosIE(PornHubPlaylistBaseIE): - _VALID_URL = r'https?://(?:www\.)?pornhub\.com/users/(?P<id>[^/]+)/videos' + _VALID_URL = r'https?://(?:[^/]+\.)?pornhub\.com/(?:user|channel)s/(?P<id>[^/]+)/videos' _TESTS = [{ 'url': 'http://www.pornhub.com/users/zoe_ph/videos/public', 'info_dict': { @@ -248,6 +322,28 @@ class PornHubUserVideosIE(PornHubPlaylistBaseIE): }, { 'url': 'http://www.pornhub.com/users/rushandlia/videos', 'only_matching': True, + }, { + # default sorting as Top Rated Videos + 'url': 'https://www.pornhub.com/channels/povd/videos', + 'info_dict': { + 'id': 'povd', + }, + '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, }] def _real_extract(self, url): @@ -262,6 +358,7 @@ class PornHubUserVideosIE(PornHubPlaylistBaseIE): except ExtractorError as e: if isinstance(e.cause, compat_HTTPError) and e.cause.code == 404: break + raise page_entries = self._extract_entries(webpage) if not page_entries: break