[clipfish] improve info extraction

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

diff --git a/youtube_dl/extractor/clipfish.py b/youtube_dl/extractor/clipfish.py
index 95449da3cc3b50915a2265f1baf14e4e77652584..d142e326f8c7f4266f442a3c7604c75ea866b972 100644 (file)
--- a/youtube_dl/extractor/clipfish.py
+++ b/youtube_dl/extractor/clipfish.py
@@ -1,53 +1,54 @@
+from __future__ import unicode_literals
+
 import re
-import time
-import xml.etree.ElementTree
 
 from .common import InfoExtractor
+from ..utils import (
+    int_or_none,
+    unified_strdate,
+)
 
 
 class ClipfishIE(InfoExtractor):
-    IE_NAME = u'clipfish'
-
-    _VALID_URL = r'^https?://(?:www\.)?clipfish\.de/.*?/video/(?P<id>[0-9]+)/'
+    _VALID_URL = r'https?://(?:www\.)?clipfish\.de/(?:[^/]+/)+video/(?P<id>[0-9]+)'
     _TEST = {
-        u'url': u'http://www.clipfish.de/special/supertalent/video/4028320/supertalent-2013-ivana-opacak-singt-nobodys-perfect/',
-        u'file': u'4028320.f4v',
-        u'md5': u'5e38bda8c329fbfb42be0386a3f5a382',
-        u'info_dict': {
-            u'title': u'Supertalent 2013: Ivana Opacak singt Nobody\'s Perfect',
-            u'duration': 399,
+        'url': 'http://www.clipfish.de/special/game-trailer/video/3966754/fifa-14-e3-2013-trailer/',
+        'md5': '79bc922f3e8a9097b3d68a93780fd475',
+        'info_dict': {
+            'id': '3966754',
+            'ext': 'mp4',
+            'title': 'FIFA 14 - E3 2013 Trailer',
+            'description': 'Video zu FIFA 14: E3 2013 Trailer',
+            'upload_date': '20130611',
+            'duration': 82,
+            'view_count': int,
         }
     }
 
     def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
-        video_id = mobj.group(1)
-
-        info_url = ('http://www.clipfish.de/devxml/videoinfo/%s?ts=%d' %
-                    (video_id, int(time.time())))
-        info_xml = self._download_webpage(
-            info_url, video_id, note=u'Downloading info page')
-        doc = xml.etree.ElementTree.fromstring(info_xml)
-        title = doc.find('title').text
-        video_url = doc.find('filename').text
-        thumbnail = doc.find('imageurl').text
-        duration_str = doc.find('duration').text
-        m = re.match(
-            r'^(?P<hours>[0-9]+):(?P<minutes>[0-9]{2}):(?P<seconds>[0-9]{2}):(?P<ms>[0-9]*)$',
-            duration_str)
-        if m:
-            duration = (
-                (int(m.group('hours')) * 60 * 60) +
-                (int(m.group('minutes')) * 60) +
-                (int(m.group('seconds')))
-            )
-        else:
-            duration = None
+        video_id = self._match_id(url)
+
+        video_info = self._download_json('http://www.clipfish.de/devapi/id/%s?format=json&apikey=hbbtv' % video_id, video_id)['items'][0]
+
+        formats = [{
+            'url': video_info['media_videourl_hls'].replace('de.hls.fra.clipfish.de', 'hls.fra.clipfish.de'),
+            'ext': 'mp4',
+            'format_id': 'hls',
+        },{
+            'url': video_info['media_videourl'],
+            'format_id': 'mp4',
+            'width': int_or_none(video_info.get('width')),
+            'height': int_or_none(video_info.get('height')),
+            'tbr': int_or_none(video_info.get('bitrate')),
+        }]
 
         return {
             'id': video_id,
-            'title': title,
-            'url': video_url,
-            'thumbnail': thumbnail,
-            'duration': duration,
+            'title': video_info['title'],
+            'description': video_info.get('descr'),
+            'formats': formats,
+            'thumbnail': video_info.get('media_content_thumbnail_large') or video_info.get('media_thumbnail'),
+            'duration': int_or_none(video_info.get('media_length')),
+            'upload_date': unified_strdate(video_info.get('pubDate')),
+            'view_count': int_or_none(video_info.get('media_views'))
         }