X-Git-Url: https://git.rapsys.eu/youtubedl/blobdiff_plain/9f2b33881274af98a9145c533a1d295fad71521a..1bce1fcde7a8d835418c1a8b8d5ae441e055306a:/youtube_dl/extractor/pornhub.py diff --git a/youtube_dl/extractor/pornhub.py b/youtube_dl/extractor/pornhub.py index 5a55c25..017f6c5 100644 --- a/youtube_dl/extractor/pornhub.py +++ b/youtube_dl/extractor/pornhub.py @@ -1,10 +1,13 @@ +# coding: utf-8 from __future__ import unicode_literals +import itertools import os import re from .common import InfoExtractor from ..compat import ( + compat_HTTPError, compat_urllib_parse_unquote, compat_urllib_parse_unquote_plus, compat_urllib_parse_urlparse, @@ -12,6 +15,8 @@ from ..compat import ( from ..utils import ( ExtractorError, int_or_none, + js_to_json, + orderedSet, sanitized_Request, str_to_int, ) @@ -21,7 +26,15 @@ from ..aes import ( class PornHubIE(InfoExtractor): - _VALID_URL = r'https?://(?:[a-z]+\.)?pornhub\.com/(?:view_video\.php\?viewkey=|embed/)(?P[0-9a-z]+)' + IE_DESC = 'PornHub and Thumbzilla' + _VALID_URL = r'''(?x) + https?:// + (?: + (?:[a-z]+\.)?pornhub\.com/(?:view_video\.php\?viewkey=|embed/)| + (?:www\.)?thumbzilla\.com/video/ + ) + (?P[\da-z]+) + ''' _TESTS = [{ 'url': 'http://www.pornhub.com/view_video.php?viewkey=648719015', 'md5': '1e19b41231a02eba417839222ac9d58e', @@ -36,21 +49,58 @@ class PornHubIE(InfoExtractor): 'dislike_count': int, 'comment_count': int, 'age_limit': 18, - } + 'tags': list, + 'categories': list, + }, + }, { + # non-ASCII title + 'url': 'http://www.pornhub.com/view_video.php?viewkey=1331683002', + 'info_dict': { + 'id': '1331683002', + 'ext': 'mp4', + 'title': '重庆婷婷女王足交', + 'uploader': 'cj397186295', + '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, + }, }, { 'url': 'http://www.pornhub.com/view_video.php?viewkey=ph557bbb6676d2d', 'only_matching': True, }, { + # removed at the request of cam4.com 'url': 'http://fr.pornhub.com/view_video.php?viewkey=ph55ca2f9760862', 'only_matching': True, + }, { + # removed at the request of the copyright owner + 'url': 'http://www.pornhub.com/view_video.php?viewkey=788152859', + 'only_matching': True, + }, { + # removed by uploader + 'url': 'http://www.pornhub.com/view_video.php?viewkey=ph572716d15a111', + 'only_matching': True, + }, { + # private video + 'url': 'http://www.pornhub.com/view_video.php?viewkey=ph56fd731fce6b7', + 'only_matching': True, + }, { + 'url': 'https://www.thumbzilla.com/video/ph56c6114abd99a/horny-girlfriend-sex', + '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( @@ -65,27 +115,33 @@ class PornHubIE(InfoExtractor): webpage = self._download_webpage(req, video_id) error_msg = self._html_search_regex( - r'(?s)
(.*?)
', - webpage, 'error message', default=None) + 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) raise ExtractorError( 'PornHub said: %s' % error_msg, expected=True, video_id=video_id) + # video_title from flashvars contains whitespace instead of non-ASCII (see + # 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'), + webpage, 'title', group='title') + flashvars = self._parse_json( self._search_regex( - r'var\s+flashv1ars_\d+\s*=\s*({.+?});', webpage, 'flashvars', default='{}'), + r'var\s+flashvars_\d+\s*=\s*({.+?});', webpage, 'flashvars', default='{}'), video_id) if flashvars: - video_title = flashvars.get('video_title') thumbnail = flashvars.get('image_url') duration = int_or_none(flashvars.get('video_duration')) else: - video_title, thumbnail, duration = [None] * 3 - - if not video_title: - video_title = self._html_search_regex(r'<h1 [^>]+>([^<]+)', webpage, 'title') + title, thumbnail, duration = [None] * 3 video_uploader = self._html_search_regex( r'(?s)From: .+?<(?:a href="/users/|a href="/channels/|span class="username)[^>]+>(.+?)<', @@ -100,7 +156,12 @@ 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))) + 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')) @@ -131,10 +192,19 @@ class PornHubIE(InfoExtractor): }) 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, 'uploader': video_uploader, - 'title': video_title, + 'title': title, 'thumbnail': thumbnail, 'duration': duration, 'view_count': view_count, @@ -143,15 +213,20 @@ class PornHubIE(InfoExtractor): 'comment_count': comment_count, 'formats': formats, 'age_limit': 18, + 'tags': tags, + 'categories': categories, } class PornHubPlaylistBaseIE(InfoExtractor): def _extract_entries(self, webpage): return [ - self.url_result('http://www.pornhub.com/%s' % video_url, PornHubIE.ie_key()) - for video_url in set(re.findall( - r'href="/?(view_video\.php\?.*\bviewkey=[\da-z]+[^"]*)"', webpage)) + 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)) ] def _real_extract(self, url): @@ -159,7 +234,14 @@ class PornHubPlaylistBaseIE(InfoExtractor): webpage = self._download_webpage(url, playlist_id) - entries = self._extract_entries(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) + + entries = self._extract_entries(container) playlist = self._parse_json( self._search_regex( @@ -173,28 +255,43 @@ class PornHubPlaylistBaseIE(InfoExtractor): class PornHubPlaylistIE(PornHubPlaylistBaseIE): _VALID_URL = r'https?://(?:www\.)?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, }] class PornHubUserVideosIE(PornHubPlaylistBaseIE): _VALID_URL = r'https?://(?:www\.)?pornhub\.com/users/(?P<id>[^/]+)/videos' _TESTS = [{ - 'url': 'http://www.pornhub.com/users/rushandlia/videos', + 'url': 'http://www.pornhub.com/users/zoe_ph/videos/public', 'info_dict': { - 'id': 'rushandlia', + 'id': 'zoe_ph', }, - 'playlist_mincount': 13, + 'playlist_mincount': 171, + }, { + 'url': 'http://www.pornhub.com/users/rushandlia/videos', + 'only_matching': True, }] def _real_extract(self, url): user_id = self._match_id(url) - webpage = self._download_webpage(url, user_id) - - return self.playlist_result(self._extract_entries(webpage), user_id) + entries = [] + for page_num in itertools.count(1): + try: + webpage = self._download_webpage( + url, user_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) + if not page_entries: + break + entries.extend(page_entries) + + return self.playlist_result(entries, user_id)