From 9634de178d35c5cd767b183c2be82b14bef84209 Mon Sep 17 00:00:00 2001 From: =?utf8?q?Sergey=20M=E2=80=A4?= Date: Sat, 22 Jun 2019 08:37:07 +0700 Subject: [PATCH] [pornhub] Add support for more paged video sources --- youtube_dl/extractor/extractors.py | 5 +- youtube_dl/extractor/pornhub.py | 101 +++++++++++++++++------------ 2 files changed, 62 insertions(+), 44 deletions(-) diff --git a/youtube_dl/extractor/extractors.py b/youtube_dl/extractor/extractors.py index b1ed8a4b2..9cd7d3ac4 100644 --- a/youtube_dl/extractor/extractors.py +++ b/youtube_dl/extractor/extractors.py @@ -892,8 +892,9 @@ from .porncom import PornComIE from .pornhd import PornHdIE from .pornhub import ( PornHubIE, - PornHubPlaylistIE, - PornHubUserVideosIE, + PornHubUserIE, + PornHubPagedVideoListIE, + PornHubUserVideosUploadIE, ) from .pornotube import PornotubeIE from .pornovoisines import PornoVoisinesIE diff --git a/youtube_dl/extractor/pornhub.py b/youtube_dl/extractor/pornhub.py index 7de585604..11b8cfcf7 100644 --- a/youtube_dl/extractor/pornhub.py +++ b/youtube_dl/extractor/pornhub.py @@ -372,23 +372,8 @@ class PornHubPlaylistBaseIE(PornHubBaseIE): entries, playlist_id, title, playlist.get('description')) -class PornHubPlaylistIE(PornHubPlaylistBaseIE): - _VALID_URL = r'https?://(?:[^/]+\.)?(?Ppornhub\.(?:com|net))/playlist/(?P\d+)' - _TESTS = [{ - 'url': 'http://www.pornhub.com/playlist/4667351', - 'info_dict': { - 'id': '4667351', - 'title': 'Nataly Hot', - }, - 'playlist_mincount': 2, - }, { - 'url': 'https://de.pornhub.com/playlist/4667351', - 'only_matching': True, - }] - - class PornHubUserIE(PornHubPlaylistBaseIE): - _VALID_URL = r'(?Phttps?://(?:[^/]+\.)?pornhub\.(?:com|net)/(?:(?:user|channel)s|model|pornstar)/(?P[^/?#&]+))' + _VALID_URL = r'(?Phttps?://(?:[^/]+\.)?pornhub\.(?:com|net)/(?:(?:user|channel)s|model|pornstar)/(?P[^/?#&]+))(?:[?#&]|/(?!videos)|$)' _TESTS = [{ 'url': 'https://www.pornhub.com/model/zoe_ph', 'playlist_mincount': 118, @@ -400,23 +385,20 @@ class PornHubUserIE(PornHubPlaylistBaseIE): 'playlist_mincount': 118, }, { 'url': 'https://www.pornhub.com/users/russianveet69', - 'playlist_mincount': 0, + 'only_matching': True, }, { 'url': 'https://www.pornhub.com/channels/povd', - 'playlist_mincount': 0, + 'only_matching': True, + }, { + 'url': 'https://www.pornhub.com/model/zoe_ph?abc=1', + 'only_matching': True, }] - @classmethod - def suitable(cls, url): - return (False - if PornHubPagedVideosIE.suitable(url) or PornHubUserVideosUploadIE.suitable(url) - else super(PornHubUserIE, cls).suitable(url)) - def _real_extract(self, url): mobj = re.match(self._VALID_URL, url) user_id = mobj.group('id') return self.url_result( - '%s/videos' % mobj.group('url'), ie=PornHubPagedVideosIE.ie_key(), + '%s/videos' % mobj.group('url'), ie=PornHubPagedVideoListIE.ie_key(), video_id=user_id) @@ -424,7 +406,7 @@ class PornHubPagedPlaylistBaseIE(PornHubPlaylistBaseIE): def _real_extract(self, url): mobj = re.match(self._VALID_URL, url) host = mobj.group('host') - user_id = mobj.group('id') + item_id = mobj.group('id') page = int_or_none(self._search_regex( r'\bpage=(\d+)', url, 'page', default=None)) @@ -435,7 +417,7 @@ class PornHubPagedPlaylistBaseIE(PornHubPlaylistBaseIE): for page_num in (page, ) if page is not None else itertools.count(1): try: webpage = self._download_webpage( - page_url, user_id, 'Downloading page %d' % page_num, + page_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: @@ -448,18 +430,11 @@ class PornHubPagedPlaylistBaseIE(PornHubPlaylistBaseIE): if not self._has_more(webpage): break - return self.playlist_result(orderedSet(entries), user_id) + return self.playlist_result(orderedSet(entries), item_id) -class PornHubPagedVideosIE(PornHubPagedPlaylistBaseIE): - _VALID_URL = r'''(?x) - https?:// - (?:[^/]+\.)?(?Ppornhub\.(?:com|net))/ - (?: - (?:(?:user|channel)s|model|pornstar)/(?P[^/]+)/videos| - video/search - ) - ''' +class PornHubPagedVideoListIE(PornHubPagedPlaylistBaseIE): + _VALID_URL = r'https?://(?:[^/]+\.)?(?Ppornhub\.(?:com|net))/(?P(?:[^/]+/)*[^/?#&]+)' _TESTS = [{ 'url': 'https://www.pornhub.com/model/zoe_ph/videos', 'only_matching': True, @@ -469,20 +444,20 @@ class PornHubPagedVideosIE(PornHubPagedPlaylistBaseIE): }, { 'url': 'https://www.pornhub.com/pornstar/jenny-blighe/videos', 'info_dict': { - 'id': 'jenny-blighe', + 'id': 'pornstar/jenny-blighe/videos', }, 'playlist_mincount': 149, }, { 'url': 'https://www.pornhub.com/pornstar/jenny-blighe/videos?page=3', 'info_dict': { - 'id': 'jenny-blighe', + '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': 'povd', + 'id': 'channels/povd/videos', }, 'playlist_mincount': 293, }, { @@ -522,13 +497,55 @@ class PornHubPagedVideosIE(PornHubPagedPlaylistBaseIE): }, { '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 PornHubUserVideosUploadIE.suitable(url) - else super(PornHubPagedVideosIE, cls).suitable(url)) + if PornHubIE.suitable(url) or PornHubUserIE.suitable(url) or PornHubUserVideosUploadIE.suitable(url) + else super(PornHubPagedVideoListIE, cls).suitable(url)) def _make_page_url(self, url): return url -- 2.39.5