X-Git-Url: http://git.bitcoin.ninja/index.cgi?a=blobdiff_plain;f=youtube_dl%2Fextractor%2Fpornhub.py;h=281a4f05e4600829a42bce31de8271f588aeae46;hb=7d2b4aa04790b51a0b2b717eed618296115061a8;hp=7a2737032ff27a73825a7787feece5d52a92507d;hpb=9a372f14b422de15acf91e25a90375688b2ba3fa;p=youtube-dl diff --git a/youtube_dl/extractor/pornhub.py b/youtube_dl/extractor/pornhub.py index 7a2737032..281a4f05e 100644 --- a/youtube_dl/extractor/pornhub.py +++ b/youtube_dl/extractor/pornhub.py @@ -1,7 +1,9 @@ # coding: utf-8 from __future__ import unicode_literals +import functools import itertools +import operator # import os import re @@ -18,6 +20,7 @@ from ..utils import ( js_to_json, orderedSet, # sanitized_Request, + remove_quotes, str_to_int, ) # from ..aes import ( @@ -30,7 +33,7 @@ class PornHubIE(InfoExtractor): _VALID_URL = r'''(?x) https?:// (?: - (?:[a-z]+\.)?pornhub\.com/(?:view_video\.php\?viewkey=|embed/)| + (?:[a-z]+\.)?pornhub\.com/(?:(?:view_video\.php|video/show)\?viewkey=|embed/)| (?:www\.)?thumbzilla\.com/video/ ) (?P[\da-z]+) @@ -94,6 +97,9 @@ 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, }] @staticmethod @@ -129,9 +135,32 @@ class PornHubIE(InfoExtractor): tv_webpage = dl_webpage('tv') - video_url = self._search_regex( - r']+\bsrc=(["\'])(?P(?:https?:)?//.+?)\1', tv_webpage, - 'video url', group='url') + assignments = self._search_regex( + r'(var.+?mediastring.+?)', 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'] title = self._search_regex( r'

([^>]+)

', tv_webpage, 'title', default=None) @@ -157,7 +186,7 @@ class PornHubIE(InfoExtractor): title, thumbnail, duration = [None] * 3 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( @@ -169,50 +198,6 @@ class PornHubIE(InfoExtractor): comment_count = self._extract_count( r'All Comments\s*\(([\d,.]+)\)', webpage, 'comment') - """ - video_variables = {} - for video_variablename, quote, video_variable in re.findall( - r'(player_quality_[0-9]{3,4}p\w+)\s*=\s*(["\'])(.+?)\2;', webpage): - video_variables[video_variablename] = video_variable - - video_urls = [] - for encoded_video_url in re.findall( - r'player_quality_[0-9]{3,4}p\s*=(.+?);', webpage): - for varname, varval in video_variables.items(): - encoded_video_url = encoded_video_url.replace(varname, varval) - video_urls.append(re.sub(r'[\s+]', '', encoded_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[0-9]+)[pP]-(?P[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{[^}]+})', webpage, 'page parameters', group='data', default='{}'), @@ -242,13 +227,20 @@ class PornHubIE(InfoExtractor): 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)(]+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): @@ -256,22 +248,18 @@ class PornHubPlaylistBaseIE(InfoExtractor): 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)(]+class=["\']container.+)', webpage, - 'container', default=webpage) - - entries = self._extract_entries(container) + entries = self._extract_entries(webpage) 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): @@ -287,7 +275,7 @@ class PornHubPlaylistIE(PornHubPlaylistBaseIE): class PornHubUserVideosIE(PornHubPlaylistBaseIE): - _VALID_URL = r'https?://(?:www\.)?pornhub\.com/users/(?P[^/]+)/videos' + _VALID_URL = r'https?://(?:www\.)?pornhub\.com/(?:user|channel)s/(?P[^/]+)/videos' _TESTS = [{ 'url': 'http://www.pornhub.com/users/zoe_ph/videos/public', 'info_dict': { @@ -297,6 +285,25 @@ 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, }] def _real_extract(self, url): @@ -311,6 +318,7 @@ 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