[vidio] Improve (Closes #9562)

youtube_dl/extractor/vidio.py

@@ -2,28 +2,30 @@
 from __future__ import unicode_literals
 
 import re
-from .common import InfoExtractor
 
+from .common import InfoExtractor
 from ..utils import int_or_none
 
 
 class VidioIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:www\.)?vidio\.com/watch/(?P<id>\d{6})-(?P<display_id>[^/?]+)'
+    _VALID_URL = r'https?://(?:www\.)?vidio\.com/watch/(?P<id>\d+)-(?P<display_id>[^/?#&]+)'
-    _TEST = {
+    _TESTS = [{
         'url': 'http://www.vidio.com/watch/165683-dj_ambred-booyah-live-2015',
+        'md5': 'cd2801394afc164e9775db6a140b91fe',
         'info_dict': {
             'id': '165683',
-            'title': 'DJ_AMBRED - Booyah (Live 2015)',
+            'display_id': 'dj_ambred-booyah-live-2015',
             'ext': 'mp4',
-            'thumbnail': 'https://cdn0-a.production.vidio.static6.com/uploads/video/image/165683/dj_ambred-booyah-live-2015-bfb2ba.jpg',
+            'title': 'DJ_AMBRED - Booyah (Live 2015)',
             'description': 'md5:27dc15f819b6a78a626490881adbadf8',
-            'duration': 149, 
+            'thumbnail': 're:^https?://.*\.jpg$',
+            'duration': 149,
+            'like_count': int,
         },
-        'params': {
+    }, {
-            # m3u8 download
+        'url': 'https://www.vidio.com/watch/77949-south-korea-test-fires-missile-that-can-strike-all-of-the-north',
-            'skip_download': True
+        'only_matching': True,
-        }
+    }]
-    }
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
@@ -31,18 +33,41 @@ class VidioIE(InfoExtractor):
 
         webpage = self._download_webpage(url, display_id)
 
-        video_data = self._parse_json(self._html_search_regex(
+        title = self._og_search_title(webpage)
-            r'data-json-clips\s*=\s*"\[(.+)\]"', webpage, 'video data'), display_id)
 
-        formats = self._extract_m3u8_formats(
+        m3u8_url, duration, thumbnail = [None] * 3
-            video_data['sources'][0]['file'],
+
-            display_id, ext='mp4')
+        clips = self._parse_json(
+            self._html_search_regex(
+                r'data-json-clips\s*=\s*(["\'])(?P<data>\[.+?\])\1',
+                webpage, 'video data', default='[]', group='data'),
+            display_id, fatal=False)
+        if clips:
+            clip = clips[0]
+            m3u8_url = clip.get('sources', [{}])[0].get('file')
+            duration = clip.get('clip_duration')
+            thumbnail = clip.get('image')
+
+        m3u8_url = m3u8_url or self._search_regex(
+            r'data(?:-vjs)?-clip-hls-url=(["\'])(?P<url>.+?)\1', webpage, 'hls url')
+        formats = self._extract_m3u8_formats(m3u8_url, display_id, 'mp4', entry_protocol='m3u8_native')
+
+        duration = int_or_none(duration or self._search_regex(
+            r'data-video-duration=(["\'])(?P<duartion>\d+)\1', webpage, 'duration'))
+        thumbnail = thumbnail or self._og_search_thumbnail(webpage)
+
+        like_count = int_or_none(self._search_regex(
+            (r'<span[^>]+data-comment-vote-count=["\'](\d+)',
+             r'<span[^>]+class=["\'].*?\blike(?:__|-)count\b.*?["\'][^>]*>\s*(\d+)'),
+            webpage, 'like count', fatal=False))
 
         return {
             'id': video_id,
-            'title': self._og_search_title(webpage),
+            'display_id': display_id,
-            'formats': formats,
+            'title': title,
-            'thumbnail': video_data.get('image'),
             'description': self._og_search_description(webpage),
-            'duration': int_or_none(video_data.get('clip_duration')),
+            'thumbnail': thumbnail,
+            'duration': duration,
+            'like_count': like_count,
+            'formats': formats,
         }