Start new release and annotate with bugs being fixed.

[youtubedl] / youtube_dl / extractor / pornhub.py

diff --git a/youtube_dl/extractor/pornhub.py b/youtube_dl/extractor/pornhub.py
index daa284ea28be63f9677471f16a0b8102f32b560e..6782848d90c289d02aeef76bb2d0925351d64121 100644 (file)
--- a/youtube_dl/extractor/pornhub.py
+++ b/youtube_dl/extractor/pornhub.py
@@ -1,36 +1,131 @@
+# coding: utf-8

 from __future__ import unicode_literals
 
 from __future__ import unicode_literals
 

-import os
+import functools
+import itertools
+import operator

 import re
 
 from .common import InfoExtractor
 from ..compat import (
 import re
 
 from .common import InfoExtractor
 from ..compat import (

-    compat_urllib_parse,
-    compat_urllib_parse_urlparse,
-    compat_urllib_request,
+    compat_HTTPError,
+    compat_str,

 )
 from ..utils import (
     ExtractorError,
 )
 from ..utils import (
     ExtractorError,

+    int_or_none,
+    js_to_json,
+    orderedSet,
+    remove_quotes,

     str_to_int,
     str_to_int,

-)
-from ..aes import (
-    aes_decrypt_text
+    url_or_none,

 )
 
 
 class PornHubIE(InfoExtractor):
 )
 
 
 class PornHubIE(InfoExtractor):

-    _VALID_URL = r'https?://(?:www\.)?pornhub\.com/view_video\.php\?viewkey=(?P<id>[0-9a-f]+)'
-    _TEST = {
+    IE_DESC = 'PornHub and Thumbzilla'
+    _VALID_URL = r'''(?x)
+                    https?://
+                        (?:
+                            (?:[^/]+\.)?pornhub\.com/(?:(?:view_video\.php|video/show)\?viewkey=|embed/)|
+                            (?:www\.)?thumbzilla\.com/video/
+                        )
+                        (?P<id>[\da-z]+)
+                    '''
+    _TESTS = [{

         'url': 'http://www.pornhub.com/view_video.php?viewkey=648719015',
         'url': 'http://www.pornhub.com/view_video.php?viewkey=648719015',

-        'md5': '882f488fa1f0026f023f33576004a2ed',
+        'md5': '1e19b41231a02eba417839222ac9d58e',

         'info_dict': {
             'id': '648719015',
             'ext': 'mp4',
         'info_dict': {
             'id': '648719015',
             'ext': 'mp4',

-            "uploader": "Babes",
-            "title": "Seductive Indian beauty strips down and fingers her pink pussy",
-            "age_limit": 18
-        }
-    }
+            'title': 'Seductive Indian beauty strips down and fingers her pink pussy',
+            'uploader': 'Babes',
+            'duration': 361,
+            'view_count': int,
+            'like_count': int,
+            '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': '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,
+        },
+    }, {
+        # subtitles
+        'url': 'https://www.pornhub.com/view_video.php?viewkey=ph5af5fef7c2aa7',
+        'info_dict': {
+            'id': 'ph5af5fef7c2aa7',
+            'ext': 'mp4',
+            'title': 'BFFS - Cute Teen Girls Share Cock On the Floor',
+            'uploader': 'BFFs',
+            'duration': 622,
+            'view_count': int,
+            'like_count': int,
+            'dislike_count': int,
+            'comment_count': int,
+            'age_limit': 18,
+            'tags': list,
+            'categories': list,
+            'subtitles': {
+                'en': [{
+                    "ext": 'srt'
+                }]
+            },
+        },
+        '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,
+    }, {
+        'url': 'http://www.pornhub.com/video/show?viewkey=648719015',
+        'only_matching': True,
+    }]
+
+    @staticmethod
+    def _extract_urls(webpage):
+        return re.findall(
+            r'<iframe[^>]+?src=["\'](?P<url>(?:https?:)?//(?:www\.)?pornhub\.com/embed/[\da-z]+)',
+            webpage)

 
     def _extract_count(self, pattern, webpage, name):
         return str_to_int(self._search_regex(
 
     def _extract_count(self, pattern, webpage, name):
         return str_to_int(self._search_regex(


@@ -39,26 +134,128 @@ class PornHubIE(InfoExtractor):
     def _real_extract(self, url):
         video_id = self._match_id(url)
 
     def _real_extract(self, url):
         video_id = self._match_id(url)
 

-        req = compat_urllib_request.Request(url)
-        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(
 
         error_msg = self._html_search_regex(

-            r'(?s)<div class="userMessageSection[^"]*".*?>(.*?)</div>',
-            webpage, 'error message', default=None)
+            r'(?s)<div[^>]+class=(["\'])(?:(?!\1).)*\b(?:removed|userMessageSection)\b(?:(?!\1).)*\1[^>]*>(?P<error>.+?)</div>',
+            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)
 
         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 = self._html_search_regex(r'<h1 [^>]+>([^<]+)', webpage, 'title')
+        # 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'<h1[^>]+class=["\']title["\'][^>]*>(?P<title>[^<]+)',
+             r'<div[^>]+data-video-title=(["\'])(?P<title>.+?)\1',
+             r'shareTitle\s*=\s*(["\'])(?P<title>.+?)\1'),
+            webpage, 'title', group='title')
+
+        video_urls = []
+        video_urls_set = set()
+        subtitles = {}
+
+        flashvars = self._parse_json(
+            self._search_regex(
+                r'var\s+flashvars_\d+\s*=\s*({.+?});', webpage, 'flashvars', default='{}'),
+            video_id)
+        if flashvars:
+            subtitle_url = url_or_none(flashvars.get('closedCaptionsFile'))
+            if subtitle_url:
+                subtitles.setdefault('en', []).append({
+                    'url': subtitle_url,
+                    'ext': 'srt',
+                })
+            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:
+            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(
         video_uploader = self._html_search_regex(

-            r'(?s)From:&nbsp;.+?<(?:a href="/users/|a href="/channels/|span class="username)[^>]+>(.+?)<',
+            r'(?s)From:&nbsp;.+?<(?:a\b[^>]+\bhref=["\']/(?:(?:user|channel)s|model|pornstar)/|span\b[^>]+\bclass=["\']username)[^>]+>(.+?)<',

             webpage, 'uploader', fatal=False)
             webpage, 'uploader', fatal=False)

-        thumbnail = self._html_search_regex(r'"image_url":"([^"]+)', webpage, 'thumbnail', fatal=False)
-        if thumbnail:
-            thumbnail = compat_urllib_parse.unquote(thumbnail)

 
         view_count = self._extract_count(
             r'<span class="count">([\d,\.]+)</span> views', webpage, 'view')
 
         view_count = self._extract_count(
             r'<span class="count">([\d,\.]+)</span> views', webpage, 'view')


@@ -69,76 +266,142 @@ class PornHubIE(InfoExtractor):
         comment_count = self._extract_count(
             r'All Comments\s*<span>\(([\d,.]+)\)', webpage, 'comment')
 
         comment_count = self._extract_count(
             r'All Comments\s*<span>\(([\d,.]+)\)', webpage, 'comment')
 

-        video_urls = list(map(compat_urllib_parse.unquote, re.findall(r'"quality_[0-9]{3}p":"([^"]+)', 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]+)P-(?P<tbr>[0-9]+)K$', 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,
             'uploader': video_uploader,
 
         return {
             'id': video_id,
             'uploader': video_uploader,

-            'title': video_title,
+            'title': title,

             'thumbnail': thumbnail,
             'thumbnail': thumbnail,

+            'duration': duration,

             'view_count': view_count,
             'like_count': like_count,
             'dislike_count': dislike_count,
             'comment_count': comment_count,
             'formats': formats,
             'age_limit': 18,
             'view_count': view_count,
             'like_count': like_count,
             'dislike_count': dislike_count,
             'comment_count': comment_count,
             'formats': formats,
             'age_limit': 18,

+            'tags': tags,
+            'categories': categories,
+            'subtitles': subtitles,

         }
 
 
         }
 
 

-class PornHubPlaylistIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:www\.)?pornhub\.com/playlist/(?P<id>\d+)'
-    _TESTS = [{
-        'url': 'http://www.pornhub.com/playlist/6201671',
-        'info_dict': {
-            'id': '6201671',
-            'title': 'P0p4',
-        },
-        'playlist_mincount': 35,
-    }]
+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="([^"]+)"',
+                container))
+        ]

 
     def _real_extract(self, url):
         playlist_id = self._match_id(url)
 
         webpage = self._download_webpage(url, playlist_id)
 
 
     def _real_extract(self, url):
         playlist_id = self._match_id(url)
 
         webpage = self._download_webpage(url, playlist_id)
 

-        entries = [
-            self.url_result('http://www.pornhub.com/%s' % video_url, 'PornHub')
-            for video_url in set(re.findall('href="/?(view_video\.php\?viewkey=\d+[^"]*)"', webpage))
-        ]
+        entries = self._extract_entries(webpage)

 
         playlist = self._parse_json(
             self._search_regex(
 
         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(
 
         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?://(?:[^/]+\.)?pornhub\.com/playlist/(?P<id>\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 PornHubUserVideosIE(PornHubPlaylistBaseIE):
+    _VALID_URL = r'https?://(?:[^/]+\.)?pornhub\.com/(?:(?:user|channel)s|model|pornstar)/(?P<id>[^/]+)/videos'
+    _TESTS = [{
+        'url': 'http://www.pornhub.com/users/zoe_ph/videos/public',
+        'info_dict': {
+            'id': 'zoe_ph',
+        },
+        'playlist_mincount': 171,
+    }, {
+        '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,
+    }, {
+        'url': 'https://www.pornhub.com/model/jayndrea/videos/upload',
+        'only_matching': True,
+    }, {
+        'url': 'https://www.pornhub.com/pornstar/jenny-blighe/videos/upload',
+        'only_matching': True,
+    }]
+
+    def _real_extract(self, url):
+        user_id = self._match_id(url)
+
+        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
+                raise
+            page_entries = self._extract_entries(webpage)
+            if not page_entries:
+                break
+            entries.extend(page_entries)
+
+        return self.playlist_result(entries, user_id)