X-Git-Url: https://git.rapsys.eu/youtubedl/blobdiff_plain/179629569ec7a2eda04957287c17fab7e55b1a7f..201856721e6959e772b5e5ddee42d4d75d966703:/youtube_dl/extractor/veoh.py?ds=sidebyside

diff --git a/youtube_dl/extractor/veoh.py b/youtube_dl/extractor/veoh.py
index 00672c9..1c44c14 100644
--- a/youtube_dl/extractor/veoh.py
+++ b/youtube_dl/extractor/veoh.py
@@ -1,47 +1,103 @@
-import re
-import json
+from __future__ import unicode_literals
 
 from .common import InfoExtractor
 from ..utils import (
-    determine_ext,
+    int_or_none,
+    parse_duration,
+    qualities,
 )
 
+
 class VeohIE(InfoExtractor):
-    _VALID_URL = r'http://www\.veoh\.com/watch/v(?P<id>\d*)'
-
-    _TEST = {
-        u'url': u'http://www.veoh.com/watch/v56314296nk7Zdmz3',
-        u'file': u'56314296.mp4',
-        u'md5': u'620e68e6a3cff80086df3348426c9ca3',
-        u'info_dict': {
-            u'title': u'Straight Backs Are Stronger',
-            u'uploader': u'LUMOback',
-            u'description': u'At LUMOback, we believe straight backs are stronger.  The LUMOback Posture & Movement Sensor:  It gently vibrates when you slouch, inspiring improved posture and mobility.  Use the app to track your data and improve your posture over time. ',
+    _VALID_URL = r'https?://(?:www\.)?veoh\.com/(?:watch|embed|iphone/#_Watch)/(?P<id>(?:v|e|yapi-)[\da-zA-Z]+)'
+
+    _TESTS = [{
+        'url': 'http://www.veoh.com/watch/v56314296nk7Zdmz3',
+        'md5': '9e7ecc0fd8bbee7a69fe38953aeebd30',
+        'info_dict': {
+            'id': 'v56314296nk7Zdmz3',
+            'ext': 'mp4',
+            'title': 'Straight Backs Are Stronger',
+            'uploader': 'LUMOback',
+            'description': 'At LUMOback, we believe straight backs are stronger.  The LUMOback Posture & Movement Sensor:  It gently vibrates when you slouch, inspiring improved posture and mobility.  Use the app to track your data and improve your posture over time. ',
+        },
+    }, {
+        'url': 'http://www.veoh.com/embed/v56314296nk7Zdmz3',
+        'only_matching': True,
+    }, {
+        'url': 'http://www.veoh.com/watch/v27701988pbTc4wzN?h1=Chile+workers+cover+up+to+avoid+skin+damage',
+        'md5': '4a6ff84b87d536a6a71e6aa6c0ad07fa',
+        'info_dict': {
+            'id': '27701988',
+            'ext': 'mp4',
+            'title': 'Chile workers cover up to avoid skin damage',
+            'description': 'md5:2bd151625a60a32822873efc246ba20d',
+            'uploader': 'afp-news',
+            'duration': 123,
+        },
+        'skip': 'This video has been deleted.',
+    }, {
+        'url': 'http://www.veoh.com/watch/v69525809F6Nc4frX',
+        'md5': '4fde7b9e33577bab2f2f8f260e30e979',
+        'note': 'Embedded ooyala video',
+        'info_dict': {
+            'id': '69525809',
+            'ext': 'mp4',
+            'title': 'Doctors Alter Plan For Preteen\'s Weight Loss Surgery',
+            'description': 'md5:f5a11c51f8fb51d2315bca0937526891',
+            'uploader': 'newsy-videos',
+        },
+        'skip': 'This video has been deleted.',
+    }, {
+        'url': 'http://www.veoh.com/watch/e152215AJxZktGS',
+        'only_matching': True,
+    }]
+
+    def _extract_video(self, source):
+        return {
+            'id': source.get('videoId'),
+            'title': source.get('title'),
+            'description': source.get('description'),
+            'thumbnail': source.get('highResImage') or source.get('medResImage'),
+            'uploader': source.get('username'),
+            'duration': int_or_none(source.get('length')),
+            'view_count': int_or_none(source.get('views')),
+            'age_limit': 18 if source.get('isMature') == 'true' or source.get('isSexy') == 'true' else 0,
+            'formats': self._extract_formats(source),
         }
-    }
 
     def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
-        video_id = mobj.group('id')
-        webpage = self._download_webpage(url, video_id)
-
-        m_youtube = re.search(r'http://www\.youtube\.com/v/(.*?)(\&|")', webpage)
-        if m_youtube is not None:
-            youtube_id = m_youtube.group(1)
-            self.to_screen(u'%s: detected Youtube video.' % video_id)
-            return self.url_result(youtube_id, 'Youtube')
-
-        self.report_extraction(video_id)
-        info = self._search_regex(r'videoDetailsJSON = \'({.*?})\';', webpage, 'info')
-        info = json.loads(info)
-        video_url =  info.get('fullPreviewHashHighPath') or info.get('fullPreviewHashLowPath')
-
-        return {'id': info['videoId'], 
-                'title': info['title'],
-                'ext': determine_ext(video_url),
-                'url': video_url,
-                'uploader': info['username'],
-                'thumbnail': info.get('highResImage') or info.get('medResImage'),
-                'description': info['description'],
-                'view_count': info['views'],
-                }
+        video_id = self._match_id(url)
+        video = self._download_json(
+            'https://www.veoh.com/watch/getVideo/' + video_id,
+            video_id)['video']
+        title = video['title']
+
+        thumbnail_url = None
+        q = qualities(['HQ', 'Regular'])
+        formats = []
+        for f_id, f_url in video.get('src', {}).items():
+            if not f_url:
+                continue
+            if f_id == 'poster':
+                thumbnail_url = f_url
+            else:
+                formats.append({
+                    'format_id': f_id,
+                    'quality': q(f_id),
+                    'url': f_url,
+                })
+        self._sort_formats(formats)
+
+        return {
+            'id': video_id,
+            'title': title,
+            'description': video.get('description'),
+            'thumbnail': thumbnail_url,
+            'uploader': video.get('author', {}).get('nickname'),
+            'duration': int_or_none(video.get('lengthBySec')) or parse_duration(video.get('length')),
+            'view_count': int_or_none(video.get('views')),
+            'formats': formats,
+            'average_rating': int_or_none(video.get('rating')),
+            'comment_count': int_or_none(video.get('numOfComments')),
+        }