[dbtv] Simplify, modernize, extract all formats

youtube_dl/extractor/dbtv.py

@@ -4,73 +4,71 @@ from __future__ import unicode_literals
 import re
 
 from .common import InfoExtractor
-
 from ..utils import (
-  ExtractorError
+    float_or_none,
+    int_or_none,
+    clean_html,
 )
 
+
 class DBTVIE(InfoExtractor):
-  _VALID_URL = r'http://dbtv.no/(?P<id>[0-9]+)/?(?P<slug>.*)$'
-  _TEST = {
-    'url': 'http://dbtv.no/3649835190001#Skulle_teste_ut_fornøyelsespark,_men_kollegaen_var_bare_opptatt_av_bikinikroppen',
-    'md5': 'b89953ed25dacb6edb3ef6c6f430f8bc',
-    'info_dict': {
-      'id': '3649835190001',
-      'ext': 'mp4',
-      'title': 'Skulle teste ut fornøyelsespark, men kollegaen var bare opptatt av bikinikroppen',
-      'description': 'md5:d681bf2bb7dd3503892cedb9c2d0e6f2',
-      'thumbnail': 'http://gfx.dbtv.no/thumbs/still/33100.jpg',
-      'timestamp': 1404039863,
-      'upload_date': '20140629',
-      'duration': 69544,
+    _VALID_URL = r'http://dbtv\.no/(?P<id>[0-9]+)#(?P<display_id>.+)'
+    _TEST = {
+        'url': 'http://dbtv.no/3649835190001#Skulle_teste_ut_fornøyelsespark,_men_kollegaen_var_bare_opptatt_av_bikinikroppen',
+        'md5': 'b89953ed25dacb6edb3ef6c6f430f8bc',
+        'info_dict': {
+            'id': '33100',
+            'display_id': 'Skulle_teste_ut_fornøyelsespark,_men_kollegaen_var_bare_opptatt_av_bikinikroppen',
+            'ext': 'mp4',
+            'title': 'Skulle teste ut fornøyelsespark, men kollegaen var bare opptatt av bikinikroppen',
+            'description': 'md5:1504a54606c4dde3e4e61fc97aa857e0',
+            'thumbnail': 're:https?://.*\.jpg$',
+            'timestamp': 1404039863.438,
+            'upload_date': '20140629',
+            'duration': 69.544,
+            'view_count': int,
+            'categories': list,
+        }
     }
-  }
-
-  def _real_extract(self, url):
-    mobj = re.match(self._VALID_URL, url)
-    video_id = mobj.group('id')
 
-    # Download JSON file containing video info.
-    data = self._download_json('http://api.dbtv.no/discovery/%s' % video_id, video_id, 'Downloading media JSON')
-    # We only want the first video in the JSON API file.
-    video = data['playlist'][0]
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        video_id = mobj.group('id')
+        display_id = mobj.group('display_id')
 
-    # Check for full HD video, else use the standard video URL
-    for i in range(0, len(video['renditions'])):
-      if int(video['renditions'][i]['width']) == 1280:
-        video_url = video['renditions'][i]['URL']
-        break
-      else:
-        video_url = video['URL']
+        data = self._download_json(
+            'http://api.dbtv.no/discovery/%s' % video_id, display_id)
 
-    # Add access token to image or it will fail.
-    thumbnail = video['splash']
+        video = data['playlist'][0]
 
-    # Duration int.
-    duration = int(video['length'])
+        formats = [{
+            'url': f['URL'],
+            'vcodec': f.get('container'),
+            'width': int_or_none(f.get('width')),
+            'height': int_or_none(f.get('height')),
+            'vbr': float_or_none(f.get('rate'), 1000),
+            'filesize': int_or_none(f.get('size')),
+        } for f in video['renditions'] if 'URL' in f]
 
-    # Timestamp is given in milliseconds.
-    timestamp = float(str(video['publishedAt'])[0:-3])
+        if not formats:
+            for url_key, format_id in [('URL', 'mp4'), ('HLSURL', 'hls')]:
+                if url_key in video:
+                    formats.append({
+                        'url': video[url_key],
+                        'format_id': format_id,
+                    })
 
-    formats = []
+        self._sort_formats(formats)
 
-    # Video URL.
-    if video['URL'] is not None:
-      formats.append({
-        'url': video_url,
-        'format_id': 'mp4',
-        'ext': 'mp4'
-      })
-    else:
-      raise ExtractorError('No download URL found for video: %s.' % video_id, expected=True)
-
-    return {
-      'id': video_id,
-      'title': video['title'],
-      'description': video['desc'],
-      'thumbnail': thumbnail,
-      'timestamp': timestamp,
-      'duration': duration,
-      'view_count': video['views'],
-      'formats': formats,
-    }
+        return {
+            'id': video['id'],
+            'display_id': display_id,
+            'title': video['title'],
+            'description': clean_html(video['desc']),
+            'thumbnail': video.get('splash') or video.get('thumb'),
+            'timestamp': float_or_none(video.get('publishedAt'), 1000),
+            'duration': float_or_none(video.get('length'), 1000),
+            'view_count': int_or_none(video.get('views')),
+            'categories': video.get('tags'),
+            'formats': formats,
+        }