Prepare to upload

[youtubedl] / youtube_dl / extractor / pornhub.py

diff --git a/youtube_dl/extractor/pornhub.py b/youtube_dl/extractor/pornhub.py
index 017f6c55219ff3db0cf9bf745f74031882f12c54..3428458afa987fb3eb8c3be061742cefce6e2734 100644 (file)
--- a/youtube_dl/extractor/pornhub.py
+++ b/youtube_dl/extractor/pornhub.py
@@ -1,28 +1,31 @@
 # coding: utf-8
 from __future__ import unicode_literals
 
 # coding: utf-8
 from __future__ import unicode_literals
 

+import functools

 import itertools
 import itertools

-import os
+import operator
+# import os

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

-    compat_urllib_parse_unquote,
-    compat_urllib_parse_unquote_plus,
-    compat_urllib_parse_urlparse,
+    # compat_urllib_parse_unquote,
+    # compat_urllib_parse_unquote_plus,
+    # compat_urllib_parse_urlparse,

 )
 from ..utils import (
     ExtractorError,
     int_or_none,
     js_to_json,
     orderedSet,
 )
 from ..utils import (
     ExtractorError,
     int_or_none,
     js_to_json,
     orderedSet,

-    sanitized_Request,
+    # sanitized_Request,
+    remove_quotes,

     str_to_int,
 )
     str_to_int,
 )

-from ..aes import (
-    aes_decrypt_text
-)
+# from ..aes import (
+#     aes_decrypt_text
+# )

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


@@ -30,7 +33,7 @@ class PornHubIE(InfoExtractor):
     _VALID_URL = r'''(?x)
                     https?://
                         (?:
     _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<id>[\da-z]+)
                             (?:www\.)?thumbzilla\.com/video/
                         )
                         (?P<id>[\da-z]+)


@@ -94,6 +97,9 @@ class PornHubIE(InfoExtractor):
     }, {
         'url': 'https://www.thumbzilla.com/video/ph56c6114abd99a/horny-girlfriend-sex',
         '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
     }]
 
     @staticmethod


@@ -109,10 +115,14 @@ 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 = sanitized_Request(
-            'http://www.pornhub.com/view_video.php?viewkey=%s' % video_id)
-        req.add_header('Cookie', 'age_verified=1')
-        webpage = self._download_webpage(req, video_id)
+        def dl_webpage(platform):
+            return self._download_webpage(
+                'http://www.pornhub.com/view_video.php?viewkey=%s' % video_id,
+                video_id, headers={
+                    'Cookie': 'age_verified=1; platform=%s' % platform,
+                })
+
+        webpage = dl_webpage('pc')

 
         error_msg = self._html_search_regex(
             r'(?s)<div[^>]+class=(["\'])(?:(?!\1).)*\b(?:removed|userMessageSection)\b(?:(?!\1).)*\1[^>]*>(?P<error>.+?)</div>',
 
         error_msg = self._html_search_regex(
             r'(?s)<div[^>]+class=(["\'])(?:(?!\1).)*\b(?:removed|userMessageSection)\b(?:(?!\1).)*\1[^>]*>(?P<error>.+?)</div>',


@@ -123,10 +133,42 @@ class PornHubIE(InfoExtractor):
                 'PornHub said: %s' % error_msg,
                 expected=True, video_id=video_id)
 
                 'PornHub said: %s' % error_msg,
                 expected=True, video_id=video_id)
 

+        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']
+
+        title = self._search_regex(
+            r'<h1>([^>]+)</h1>', tv_webpage, 'title', default=None)
+

         # 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.
         # 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(
+        title = title or 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',
             'twitter:title', webpage, default=None) or self._search_regex(
             (r'<h1[^>]+class=["\']title["\'][^>]*>(?P<title>[^<]+)',
              r'<div[^>]+data-video-title=(["\'])(?P<title>.+?)\1',


@@ -144,7 +186,7 @@ class PornHubIE(InfoExtractor):
             title, thumbnail, duration = [None] * 3
 
         video_uploader = self._html_search_regex(
             title, thumbnail, duration = [None] * 3
 
         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/|span\b[^>]+\bclass=["\']username)[^>]+>(.+?)<',

             webpage, 'uploader', fatal=False)
 
         view_count = self._extract_count(
             webpage, 'uploader', fatal=False)
 
         view_count = self._extract_count(


@@ -156,42 +198,6 @@ 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 = []
-        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'))
-            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]+)[pP]-(?P<tbr>[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<data>{[^}]+})',
             webpage, 'page parameters', group='data', default='{}'),
         page_params = self._parse_json(self._search_regex(
             r'page_params\.zoneDetails\[([\'"])[^\'"]+\1\]\s*=\s*(?P<data>{[^}]+})',
             webpage, 'page parameters', group='data', default='{}'),


@@ -203,6 +209,7 @@ class PornHubIE(InfoExtractor):
 
         return {
             'id': video_id,
 
         return {
             'id': video_id,

+            'url': video_url,

             'uploader': video_uploader,
             'title': title,
             'thumbnail': thumbnail,
             'uploader': video_uploader,
             'title': title,
             'thumbnail': thumbnail,


@@ -211,7 +218,7 @@ class PornHubIE(InfoExtractor):
             'like_count': like_count,
             'dislike_count': dislike_count,
             'comment_count': comment_count,
             'like_count': like_count,
             'dislike_count': dislike_count,
             'comment_count': comment_count,

-            'formats': formats,
+            # 'formats': formats,

             'age_limit': 18,
             'tags': tags,
             'categories': categories,
             'age_limit': 18,
             'tags': tags,
             'categories': categories,


@@ -220,13 +227,20 @@ class PornHubIE(InfoExtractor):
 
 class PornHubPlaylistBaseIE(InfoExtractor):
     def _extract_entries(self, webpage):
 
 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="([^"]+)"',
         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):
         ]
 
     def _real_extract(self, url):


@@ -234,22 +248,18 @@ class PornHubPlaylistBaseIE(InfoExtractor):
 
         webpage = self._download_webpage(url, playlist_id)
 
 
         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)(<div[^>]+class=["\']container.+)', webpage,
-            'container', default=webpage)
-
-        entries = self._extract_entries(container)
+        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):
 
 
 class PornHubPlaylistIE(PornHubPlaylistBaseIE):


@@ -289,6 +299,7 @@ class PornHubUserVideosIE(PornHubPlaylistBaseIE):
             except ExtractorError as e:
                 if isinstance(e.cause, compat_HTTPError) and e.cause.code == 404:
                     break
             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
             page_entries = self._extract_entries(webpage)
             if not page_entries:
                 break