Merge branch 'douyutv' of https://github.com/bonfy/youtube-dl into bonfy-douyutv

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

diff --git a/youtube_dl/extractor/aftonbladet.py b/youtube_dl/extractor/aftonbladet.py
index 568ffc160d837c504e3748d00eec1be31a75de49..8442019eac3eaa0a373140d494ffa2ca420f4606 100644 (file)
--- a/youtube_dl/extractor/aftonbladet.py
+++ b/youtube_dl/extractor/aftonbladet.py
@@ -1,14 +1,11 @@
 # encoding: utf-8
 from __future__ import unicode_literals
 
-import datetime
-import re
-
 from .common import InfoExtractor
 
 
 class AftonbladetIE(InfoExtractor):
-    _VALID_URL = r'^http://tv\.aftonbladet\.se/webbtv.+(?P<video_id>article\d+)\.ab$'
+    _VALID_URL = r'^http://tv\.aftonbladet\.se/webbtv.+?(?P<video_id>article[0-9]+)\.ab(?:$|[?#])'
     _TEST = {
         'url': 'http://tv.aftonbladet.se/webbtv/nyheter/vetenskap/rymden/article36015.ab',
         'info_dict': {
@@ -16,31 +13,32 @@ class AftonbladetIE(InfoExtractor):
             'ext': 'mp4',
             'title': 'Vulkanutbrott i rymden - nu släpper NASA bilderna',
             'description': 'Jupiters måne mest aktiv av alla himlakroppar',
+            'timestamp': 1394142732,
             'upload_date': '20140306',
         },
     }
 
     def _real_extract(self, url):
-        mobj = re.search(self._VALID_URL, url)
-
-        video_id = mobj.group('video_id')
+        video_id = self._match_id(url)
         webpage = self._download_webpage(url, video_id)
 
         # find internal video meta data
-        META_URL = 'http://aftonbladet-play.drlib.aptoma.no/video/%s.json'
-        internal_meta_id = self._html_search_regex(r'data-aptomaId="([\w\d]+)"', webpage, 'internal_meta_id')
-        internal_meta_url = META_URL % internal_meta_id
-        internal_meta_json = self._download_json(internal_meta_url, video_id, 'Downloading video meta data')
+        meta_url = 'http://aftonbladet-play.drlib.aptoma.no/video/%s.json'
+        internal_meta_id = self._html_search_regex(
+            r'data-aptomaId="([\w\d]+)"', webpage, 'internal_meta_id')
+        internal_meta_url = meta_url % internal_meta_id
+        internal_meta_json = self._download_json(
+            internal_meta_url, video_id, 'Downloading video meta data')
 
         # find internal video formats
-        FORMATS_URL = 'http://aftonbladet-play.videodata.drvideo.aptoma.no/actions/video/?id=%s'
+        format_url = 'http://aftonbladet-play.videodata.drvideo.aptoma.no/actions/video/?id=%s'
         internal_video_id = internal_meta_json['videoId']
-        internal_formats_url = FORMATS_URL % internal_video_id
-        internal_formats_json = self._download_json(internal_formats_url, video_id, 'Downloading video formats')
+        internal_formats_url = format_url % internal_video_id
+        internal_formats_json = self._download_json(
+            internal_formats_url, video_id, 'Downloading video formats')
 
-        self.report_extraction(video_id)
         formats = []
-        for fmt in reversed(internal_formats_json['formats']['http']['pseudostreaming']['mp4']):
+        for fmt in internal_formats_json['formats']['http']['pseudostreaming']['mp4']:
             p = fmt['paths'][0]
             formats.append({
                 'url': 'http://%s:%d/%s/%s' % (p['address'], p['port'], p['path'], p['filename']),
@@ -50,17 +48,15 @@ class AftonbladetIE(InfoExtractor):
                 'tbr': fmt['bitrate'],
                 'protocol': 'http',
             })
+        self._sort_formats(formats)
 
-        timestamp = datetime.datetime.fromtimestamp(internal_meta_json['timePublished'])
-        upload_date = timestamp.strftime('%Y%m%d')
-
-        return [{
+        return {
             'id': video_id,
             'title': internal_meta_json['title'],
             'formats': formats,
             'thumbnail': internal_meta_json['imageUrl'],
             'description': internal_meta_json['shortPreamble'],
-            'upload_date': upload_date,
+            'timestamp': internal_meta_json['timePublished'],
             'duration': internal_meta_json['duration'],
             'view_count': internal_meta_json['views'],
-        }]
+        }