Merge tag 'upstream/2014.06.07'

[youtubedl] / youtube_dl / extractor / pornhub.py

diff --git a/youtube_dl/extractor/pornhub.py b/youtube_dl/extractor/pornhub.py
index 8b3471919565d4c7044d51eb24e8ef01cc8e77fc..4118ee9560e03d2fa1eea171766ef4893e274aa5 100644 (file)
--- a/youtube_dl/extractor/pornhub.py
+++ b/youtube_dl/extractor/pornhub.py
@@ -1,3 +1,5 @@
+from __future__ import unicode_literals
+
 import os
 import re
 
@@ -6,24 +8,32 @@ from ..utils import (
     compat_urllib_parse_urlparse,
     compat_urllib_request,
     compat_urllib_parse,
+    str_to_int,
 )
 from ..aes import (
     aes_decrypt_text
 )
 
+
 class PornHubIE(InfoExtractor):
-    _VALID_URL = r'^(?:https?://)?(?:www\.)?(?P<url>pornhub\.com/view_video\.php\?viewkey=(?P<videoid>[0-9]+))'
+    _VALID_URL = r'^(?:https?://)?(?:www\.)?(?P<url>pornhub\.com/view_video\.php\?viewkey=(?P<videoid>[0-9a-f]+))'
     _TEST = {
-        u'url': u'http://www.pornhub.com/view_video.php?viewkey=648719015',
-        u'file': u'648719015.mp4',
-        u'md5': u'882f488fa1f0026f023f33576004a2ed',
-        u'info_dict': {
-            u"uploader": u"BABES-COM", 
-            u"title": u"Seductive Indian beauty strips down and fingers her pink pussy",
-            u"age_limit": 18
+        'url': 'http://www.pornhub.com/view_video.php?viewkey=648719015',
+        'file': '648719015.mp4',
+        'md5': '882f488fa1f0026f023f33576004a2ed',
+        'info_dict': {
+            "uploader": "BABES-COM",
+            "title": "Seductive Indian beauty strips down and fingers her pink pussy",
+            "age_limit": 18
         }
     }
 
+    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
+
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
         video_id = mobj.group('videoid')
@@ -33,15 +43,23 @@ class PornHubIE(InfoExtractor):
         req.add_header('Cookie', 'age_verified=1')
         webpage = self._download_webpage(req, video_id)
 
-        video_title = self._html_search_regex(r'<h1 [^>]+>([^<]+)', webpage, u'title')
-        video_uploader = self._html_search_regex(r'<b>From: </b>(?:\s|<[^>]*>)*(.+?)<', webpage, u'uploader', fatal=False)
-        thumbnail = self._html_search_regex(r'"image_url":"([^"]+)', webpage, u'thumbnail', fatal=False)
+        video_title = self._html_search_regex(r'<h1 [^>]+>([^<]+)', webpage, 'title')
+        video_uploader = self._html_search_regex(
+            r'(?s)From:&nbsp;.+?<(?:a href="/users/|<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')
+        comment_count = self._extract_count(
+            r'All comments \(<var class="videoCommentCount">([\d,\.]+)</var>', 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 = self._html_search_regex(r'"video_title":"([^"]+)', webpage, u'password').replace('+', ' ')
+            password = compat_urllib_parse.unquote_plus(self._html_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 = []
@@ -50,19 +68,34 @@ class PornHubIE(InfoExtractor):
             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,
             })
-        formats.sort(key=lambda format: list(map(lambda s: s.zfill(6), format['format'].split('-'))))
+        self._sort_formats(formats)
 
         return {
             'id': video_id,
             'uploader': video_uploader,
             'title': video_title,
             'thumbnail': thumbnail,
+            'view_count': view_count,
+            'like_count': like_count,
+            'dislike_count': dislike_count,
+            'comment_count': comment_count,
             'formats': formats,
             'age_limit': 18,
         }