[pornhub:playlistbase] Improve extract entries

[youtube-dl] / youtube_dl / extractor / pornhub.py

diff --git a/youtube_dl/extractor/pornhub.py b/youtube_dl/extractor/pornhub.py
index 0c8b731cf47267568e43ccd09ff21f1683b4d992..55af1332489dc39bf756fdbd5eb6ea91a9f98c78 100644 (file)
--- a/youtube_dl/extractor/pornhub.py
+++ b/youtube_dl/extractor/pornhub.py
@@ -5,12 +5,13 @@ import re
 
 from .common import InfoExtractor
 from ..compat import (
-    compat_urllib_parse,
+    compat_urllib_parse_unquote,
+    compat_urllib_parse_unquote_plus,
     compat_urllib_parse_urlparse,
-    compat_urllib_request,
 )
 from ..utils import (
     ExtractorError,
+    sanitized_Request,
     str_to_int,
 )
 from ..aes import (
@@ -19,18 +20,31 @@ from ..aes import (
 
 
 class PornHubIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:www\.)?pornhub\.com/view_video\.php\?viewkey=(?P<id>[0-9a-f]+)'
-    _TEST = {
+    _VALID_URL = r'https?://(?:[a-z]+\.)?pornhub\.com/(?:view_video\.php\?viewkey=|embed/)(?P<id>[0-9a-z]+)'
+    _TESTS = [{
         'url': 'http://www.pornhub.com/view_video.php?viewkey=648719015',
         'md5': '882f488fa1f0026f023f33576004a2ed',
         'info_dict': {
             'id': '648719015',
             'ext': 'mp4',
-            "uploader": "Babes",
-            "title": "Seductive Indian beauty strips down and fingers her pink pussy",
-            "age_limit": 18
+            'uploader': 'Babes',
+            'title': 'Seductive Indian beauty strips down and fingers her pink pussy',
+            'age_limit': 18
         }
-    }
+    }, {
+        'url': 'http://www.pornhub.com/view_video.php?viewkey=ph557bbb6676d2d',
+        'only_matching': True,
+    }, {
+        'url': 'http://fr.pornhub.com/view_video.php?viewkey=ph55ca2f9760862',
+        'only_matching': True,
+    }]
+
+    @classmethod
+    def _extract_url(cls, webpage):
+        mobj = re.search(
+            r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//(?:www\.)?pornhub\.com/embed/\d+)\1', webpage)
+        if mobj:
+            return mobj.group('url')
 
     def _extract_count(self, pattern, webpage, name):
         return str_to_int(self._search_regex(
@@ -39,7 +53,8 @@ class PornHubIE(InfoExtractor):
     def _real_extract(self, url):
         video_id = self._match_id(url)
 
-        req = compat_urllib_request.Request(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)
 
@@ -58,7 +73,7 @@ class PornHubIE(InfoExtractor):
             webpage, 'uploader', fatal=False)
         thumbnail = self._html_search_regex(r'"image_url":"([^"]+)', webpage, 'thumbnail', fatal=False)
         if thumbnail:
-            thumbnail = compat_urllib_parse.unquote(thumbnail)
+            thumbnail = compat_urllib_parse_unquote(thumbnail)
 
         view_count = self._extract_count(
             r'<span class="count">([\d,\.]+)</span> views', webpage, 'view')
@@ -69,9 +84,10 @@ 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'"quality_[0-9]{3}p":"([^"]+)', webpage)))
+        video_urls = list(map(compat_urllib_parse_unquote, re.findall(r"player_quality_[0-9]{3}p\s*=\s*'([^']+)'", webpage)))
         if webpage.find('"encrypted":true') != -1:
-            password = compat_urllib_parse.unquote_plus(self._html_search_regex(r'"video_title":"([^"]+)', webpage, 'password'))
+            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 = []
@@ -79,9 +95,9 @@ class PornHubIE(InfoExtractor):
             path = compat_urllib_parse_urlparse(video_url).path
             extension = os.path.splitext(path)[1][1:]
             format = path.split('/')[5].split('_')[:2]
-            format = "-".join(format)
+            format = '-'.join(format)
 
-            m = re.match(r'^(?P<height>[0-9]+)P-(?P<tbr>[0-9]+)K$', format)
+            m = re.match(r'^(?P<height>[0-9]+)[pP]-(?P<tbr>[0-9]+)[kK]$', format)
             if m is None:
                 height = None
                 tbr = None
@@ -113,26 +129,20 @@ class PornHubIE(InfoExtractor):
         }
 
 
-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):
+        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))
+        ]
 
     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(
@@ -141,3 +151,33 @@ class PornHubPlaylistIE(InfoExtractor):
 
         return self.playlist_result(
             entries, playlist_id, playlist.get('title'), playlist.get('description'))
+
+
+class PornHubPlaylistIE(PornHubPlaylistBaseIE):
+    _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 PornHubUserVideosIE(PornHubPlaylistBaseIE):
+    _VALID_URL = r'https?://(?:www\.)?pornhub\.com/users/(?P<id>[^/]+)/videos'
+    _TESTS = [{
+        'url': 'http://www.pornhub.com/users/rushandlia/videos',
+        'info_dict': {
+            'id': 'rushandlia',
+        },
+        'playlist_mincount': 13,
+    }]
+
+    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)