X-Git-Url: https://git.rapsys.eu/youtubedl/blobdiff_plain/9f2b33881274af98a9145c533a1d295fad71521a..50ac49cc58db6cb928094be89415983aca6bbd34:/youtube_dl/extractor/pornhub.py diff --git a/youtube_dl/extractor/pornhub.py b/youtube_dl/extractor/pornhub.py index 5a55c25..20976c1 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 +15,7 @@ from ..compat import ( from ..utils import ( ExtractorError, int_or_none, + orderedSet, sanitized_Request, str_to_int, ) @@ -21,7 +25,15 @@ 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[0-9a-z]+) + ''' _TESTS = [{ 'url': 'http://www.pornhub.com/view_video.php?viewkey=648719015', 'md5': '1e19b41231a02eba417839222ac9d58e', @@ -36,13 +48,47 @@ class PornHubIE(InfoExtractor): 'dislike_count': int, 'comment_count': int, 'age_limit': 18, - } + }, + }, { + # 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, + }, + '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 @@ -65,27 +111,33 @@ 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)[^>]+>(.+?)<', @@ -134,7 +186,7 @@ class PornHubIE(InfoExtractor): return { 'id': video_id, 'uploader': video_uploader, - 'title': video_title, + 'title': title, 'thumbnail': thumbnail, 'duration': duration, 'view_count': view_count, @@ -149,9 +201,12 @@ class PornHubIE(InfoExtractor): 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): @@ -185,16 +240,31 @@ class PornHubPlaylistIE(PornHubPlaylistBaseIE): 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)