Merge pull request #8130 from dyn888/master

[youtube-dl] / youtube_dl / extractor / tube8.py

diff --git a/youtube_dl/extractor/tube8.py b/youtube_dl/extractor/tube8.py
index 39f20c54688e9371f5a299e04b84919de2db4f4e..1d9271d1e70c3f2b5ad30a6c69f6602bd58ac89a 100644 (file)
--- a/youtube_dl/extractor/tube8.py
+++ b/youtube_dl/extractor/tube8.py
@@ -1,84 +1,103 @@
 from __future__ import unicode_literals
 
-import json
 import re
 
 from .common import InfoExtractor
+from ..compat import compat_str
 from ..utils import (
-    compat_urllib_parse_urlparse,
-    compat_urllib_request,
     int_or_none,
+    sanitized_Request,
     str_to_int,
 )
 from ..aes import aes_decrypt_text
 
 
 class Tube8IE(InfoExtractor):
-    _VALID_URL = r'https?://(?:www\.)?tube8\.com/(?:gay/|shemale/)?(?:[^/]+/){2}(?P<id>\d+)'
-    _TEST = {
+    _VALID_URL = r'https?://(?:www\.)?tube8\.com/(?:[^/]+/)+(?P<display_id>[^/]+)/(?P<id>\d+)'
+    _TESTS = [{
         'url': 'http://www.tube8.com/teen/kasia-music-video/229795/',
-        'md5': '44bf12b98313827dd52d35b8706a4ea0',
+        'md5': '65e20c48e6abff62ed0c3965fff13a39',
         'info_dict': {
             'id': '229795',
+            'display_id': 'kasia-music-video',
             'ext': 'mp4',
             'description': 'hot teen Kasia grinding',
             'uploader': 'unknown',
             'title': 'Kasia music video',
             'age_limit': 18,
+            'duration': 230,
         }
-    }
+    }, {
+        'url': 'http://www.tube8.com/shemale/teen/blonde-cd-gets-kidnapped-by-two-blacks-and-punished-for-being-a-slutty-girl/19569151/',
+        'only_matching': True,
+    }]
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
         video_id = mobj.group('id')
+        display_id = mobj.group('display_id')
 
-        req = compat_urllib_request.Request(url)
+        req = sanitized_Request(url)
         req.add_header('Cookie', 'age_verified=1')
-        webpage = self._download_webpage(req, video_id)
+        webpage = self._download_webpage(req, display_id)
 
-        flashvars = json.loads(self._html_search_regex(
-            r'var flashvars\s*=\s*({.+?})', webpage, 'flashvars'))
+        flashvars = self._parse_json(
+            self._search_regex(
+                r'flashvars\s*=\s*({.+?});\r?\n', webpage, 'flashvars'),
+            video_id)
 
-        video_url = flashvars['video_url']
-        if flashvars.get('encrypted') is True:
-            video_url = aes_decrypt_text(video_url, flashvars['video_title'], 32).decode('utf-8')
-        path = compat_urllib_parse_urlparse(video_url).path
-        format_id = '-'.join(path.split('/')[4].split('_')[:2])
+        formats = []
+        for key, video_url in flashvars.items():
+            if not isinstance(video_url, compat_str) or not video_url.startswith('http'):
+                continue
+            height = self._search_regex(
+                r'quality_(\d+)[pP]', key, 'height', default=None)
+            if not height:
+                continue
+            if flashvars.get('encrypted') is True:
+                video_url = aes_decrypt_text(
+                    video_url, flashvars['video_title'], 32).decode('utf-8')
+            formats.append({
+                'url': video_url,
+                'format_id': '%sp' % height,
+                'height': int(height),
+            })
+        self._sort_formats(formats)
 
         thumbnail = flashvars.get('image_url')
 
         title = self._html_search_regex(
-            r'videotitle\s*=\s*"([^"]+)', webpage, 'title')
+            r'videoTitle\s*=\s*"([^"]+)', webpage, 'title')
         description = self._html_search_regex(
-            r'>Description:</strong>(.+?)<', webpage, 'description', fatal=False)
+            r'>Description:</strong>\s*(.+?)\s*<', webpage, 'description', fatal=False)
         uploader = self._html_search_regex(
-            r'<strong class="video-username">(?:<a href="[^"]+">)?([^<]+)(?:</a>)?</strong>',
+            r'<span class="username">\s*(.+?)\s*<',
             webpage, 'uploader', fatal=False)
+        duration = int_or_none(flashvars.get('video_duration'))
 
-        like_count = int_or_none(self._html_search_regex(
-            r"rupVar\s*=\s*'(\d+)'", webpage, 'like count', fatal=False))
-        dislike_count = int_or_none(self._html_search_regex(
-            r"rdownVar\s*=\s*'(\d+)'", webpage, 'dislike count', fatal=False))
-        view_count = self._html_search_regex(
-            r'<strong>Views: </strong>([\d,\.]+)</li>', webpage, 'view count', fatal=False)
-        if view_count:
-            view_count = str_to_int(view_count)
-        comment_count = self._html_search_regex(
-            r'<span id="allCommentsCount">(\d+)</span>', webpage, 'comment count', fatal=False)
-        if comment_count:
-            comment_count = str_to_int(comment_count)
+        like_count = int_or_none(self._search_regex(
+            r'rupVar\s*=\s*"(\d+)"', webpage, 'like count', fatal=False))
+        dislike_count = int_or_none(self._search_regex(
+            r'rdownVar\s*=\s*"(\d+)"', webpage, 'dislike count', fatal=False))
+        view_count = str_to_int(self._search_regex(
+            r'<strong>Views: </strong>([\d,\.]+)\s*</li>',
+            webpage, 'view count', fatal=False))
+        comment_count = str_to_int(self._search_regex(
+            r'<span id="allCommentsCount">(\d+)</span>',
+            webpage, 'comment count', fatal=False))
 
         return {
             'id': video_id,
-            'url': video_url,
+            'display_id': display_id,
             'title': title,
             'description': description,
             'thumbnail': thumbnail,
             'uploader': uploader,
-            'format_id': format_id,
+            'duration': duration,
             'view_count': view_count,
             'like_count': like_count,
             'dislike_count': dislike_count,
             'comment_count': comment_count,
             'age_limit': 18,
+            'formats': formats,
         }