Merge tag 'upstream/2014.12.01'

[youtubedl] / youtube_dl / extractor / dreisat.py

diff --git a/youtube_dl/extractor/dreisat.py b/youtube_dl/extractor/dreisat.py
index 765cb1f377df132ee91deac5872877777187cd6a..69ca75423cb1d4692f1958829dc61bcf5c2bac73 100644 (file)
--- a/youtube_dl/extractor/dreisat.py
+++ b/youtube_dl/extractor/dreisat.py
@@ -1,42 +1,37 @@
-# coding: utf-8
+from __future__ import unicode_literals

 
 import re
 
 import re

-import xml.etree.ElementTree

 
 from .common import InfoExtractor
 
 from .common import InfoExtractor

-from ..utils import (
-    determine_ext,
-    unified_strdate,
-)
+from ..utils import unified_strdate

 
 
 class DreiSatIE(InfoExtractor):
     IE_NAME = '3sat'
 
 
 class DreiSatIE(InfoExtractor):
     IE_NAME = '3sat'

-    _VALID_URL = r'(?:http://)?(?:www\.)?3sat.de/mediathek/index.php\?(?:(?:mode|display)=[^&]+&)*obj=(?P<id>[0-9]+)$'
+    _VALID_URL = r'(?:http://)?(?:www\.)?3sat\.de/mediathek/(?:index\.php)?\?(?:(?:mode|display)=[^&]+&)*obj=(?P<id>[0-9]+)$'

     _TEST = {
     _TEST = {

-        u"url": u"http://www.3sat.de/mediathek/index.php?obj=36983",
-        u'file': u'36983.webm',
-        u'md5': u'57c97d0469d71cf874f6815aa2b7c944',
-        u'info_dict': {
-            u"title": u"Kaffeeland Schweiz",
-            u"description": u"Über 80 Kaffeeröstereien liefern in der Schweiz das Getränk, in das das Land so vernarrt ist: Mehr als 1000 Tassen trinkt ein Schweizer pro Jahr. SCHWEIZWEIT nimmt die Kaffeekultur unter die...", 
-            u"uploader": u"3sat",
-            u"upload_date": u"20130622"
+        'url': 'http://www.3sat.de/mediathek/index.php?obj=36983',
+        'md5': '9dcfe344732808dbfcc901537973c922',
+        'info_dict': {
+            'id': '36983',
+            'ext': 'mp4',
+            'title': 'Kaffeeland Schweiz',
+            'description': 'md5:cc4424b18b75ae9948b13929a0814033',
+            'uploader': '3sat',
+            'upload_date': '20130622'

         }
     }
 
         }
     }
 

-

     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
         video_id = mobj.group('id')
         details_url = 'http://www.3sat.de/mediathek/xmlservice/web/beitragsDetails?ak=web&id=%s' % video_id
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
         video_id = mobj.group('id')
         details_url = 'http://www.3sat.de/mediathek/xmlservice/web/beitragsDetails?ak=web&id=%s' % video_id

-        details_xml = self._download_webpage(details_url, video_id, note=u'Downloading video details')
-        details_doc = xml.etree.ElementTree.fromstring(details_xml.encode('utf-8'))
+        details_doc = self._download_xml(details_url, video_id, 'Downloading video details')

 
         thumbnail_els = details_doc.findall('.//teaserimage')
         thumbnails = [{
 
         thumbnail_els = details_doc.findall('.//teaserimage')
         thumbnails = [{

-            'width': te.attrib['key'].partition('x')[0],
-            'height': te.attrib['key'].partition('x')[2],
+            'width': int(te.attrib['key'].partition('x')[0]),
+            'height': int(te.attrib['key'].partition('x')[2]),

             'url': te.text,
         } for te in thumbnail_els]
 
             'url': te.text,
         } for te in thumbnail_els]
 


@@ -54,20 +49,14 @@ class DreiSatIE(InfoExtractor):
             'width': int(fe.find('./width').text),
             'height': int(fe.find('./height').text),
             'url': fe.find('./url').text,
             'width': int(fe.find('./width').text),
             'height': int(fe.find('./height').text),
             'url': fe.find('./url').text,

-            'ext': determine_ext(fe.find('./url').text),

             'filesize': int(fe.find('./filesize').text),
             'video_bitrate': int(fe.find('./videoBitrate').text),
             'filesize': int(fe.find('./filesize').text),
             'video_bitrate': int(fe.find('./videoBitrate').text),

-            '3sat_qualityname': fe.find('./quality').text,

         } for fe in format_els
             if not fe.find('./url').text.startswith('http://www.metafilegenerator.de/')]
 
         } for fe in format_els
             if not fe.find('./url').text.startswith('http://www.metafilegenerator.de/')]
 

-        def _sortkey(format):
-            qidx = ['low', 'med', 'high', 'veryhigh'].index(format['3sat_qualityname'])
-            prefer_http = 1 if 'rtmp' in format['url'] else 0
-            return (qidx, prefer_http, format['video_bitrate'])
-        formats.sort(key=_sortkey)
+        self._sort_formats(formats)

 
 

-        info = {
+        return {

             '_type': 'video',
             'id': video_id,
             'title': video_title,
             '_type': 'video',
             'id': video_id,
             'title': video_title,


@@ -78,8 +67,3 @@ class DreiSatIE(InfoExtractor):
             'uploader': video_uploader,
             'upload_date': upload_date,
         }
             'uploader': video_uploader,
             'upload_date': upload_date,
         }

-
-        # TODO: Remove when #980 has been merged
-        info.update(formats[-1])
-
-        return info