Merge branch 'pornovoisines' of https://github.com/Roman2K/youtube-dl into Roman2K...

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

diff --git a/youtube_dl/extractor/pornhub.py b/youtube_dl/extractor/pornhub.py
index 634142d0d27300eb82ea2f460fd2163a20208709..0c8b731cf47267568e43ccd09ff21f1683b4d992 100644 (file)
--- a/youtube_dl/extractor/pornhub.py
+++ b/youtube_dl/extractor/pornhub.py
@@ -10,6 +10,7 @@ from ..compat import (
     compat_urllib_request,
 )
 from ..utils import (
+    ExtractorError,
     str_to_int,
 )
 from ..aes import (
@@ -32,10 +33,8 @@ class PornHubIE(InfoExtractor):
     }
 
     def _extract_count(self, pattern, webpage, name):
-        count = self._html_search_regex(pattern, webpage, '%s count' % name, fatal=False)
-        if count:
-            count = str_to_int(count)
-        return count
+        return str_to_int(self._search_regex(
+            pattern, webpage, '%s count' % name, fatal=False))
 
     def _real_extract(self, url):
         video_id = self._match_id(url)
@@ -44,19 +43,31 @@ class PornHubIE(InfoExtractor):
         req.add_header('Cookie', 'age_verified=1')
         webpage = self._download_webpage(req, video_id)
 
+        error_msg = self._html_search_regex(
+            r'(?s)<div class="userMessageSection[^"]*".*?>(.*?)</div>',
+            webpage, 'error message', default=None)
+        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_uploader = self._html_search_regex(
-            r'(?s)From:&nbsp;.+?<(?:a href="/users/|a href="/channels/|<span class="username)[^>]+>(.+?)<',
+            r'(?s)From:&nbsp;.+?<(?:a href="/users/|a href="/channels/|span class="username)[^>]+>(.+?)<',
             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')
-        like_count = self._extract_count(r'<span class="votesUp">([\d,\.]+)</span>', webpage, 'like')
-        dislike_count = self._extract_count(r'<span class="votesDown">([\d,\.]+)</span>', webpage, 'dislike')
+        view_count = self._extract_count(
+            r'<span class="count">([\d,\.]+)</span> views', webpage, 'view')
+        like_count = self._extract_count(
+            r'<span class="votesUp">([\d,\.]+)</span>', webpage, 'like')
+        dislike_count = self._extract_count(
+            r'<span class="votesDown">([\d,\.]+)</span>', webpage, 'dislike')
         comment_count = self._extract_count(
-            r'All comments \(<var class="videoCommentCount">([\d,\.]+)</var>', webpage, 'comment')
+            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:
@@ -100,3 +111,33 @@ class PornHubIE(InfoExtractor):
             'formats': formats,
             'age_limit': 18,
         }
+
+
+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,
+    }]
+
+    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))
+        ]
+
+        playlist = self._parse_json(
+            self._search_regex(
+                r'playlistObject\s*=\s*({.+?});', webpage, 'playlist'),
+            playlist_id)
+
+        return self.playlist_result(
+            entries, playlist_id, playlist.get('title'), playlist.get('description'))