youtube-dl

commit 7150858d49f05a5650a12b5f4694f91dfb9595d3
parent 91c7271aabdd74c833ef570db59018e2d9f9d803
Author: Philipp Hagemeister <phihag@phihag.de>
Date:   Sat, 16 Nov 2013 01:33:12 +0100

[spiegel] Implement format selection

Diffstat:
M
youtube_dl/YoutubeDL.py
 | 
10
++++++++--
M
youtube_dl/extractor/spiegel.py
 | 
50
++++++++++++++++++++++++++++++++++++--------------

2 files changed, 44 insertions(+), 16 deletions(-)
diff --git a/youtube_dl/YoutubeDL.py b/youtube_dl/YoutubeDL.py
@@ -786,13 +786,19 @@ class YoutubeDL(object):
                 return fdict['format_note']
             res = u''
             if fdict.get('vcodec') is not None:
-                res += fdict['vcodec']
+                res += u'%-5s' % fdict['vcodec']
+            elif fdict.get('vbr') is not None:
+                res += u'video'
             if fdict.get('vbr') is not None:
                 res += u'@%4dk' % fdict['vbr']
             if fdict.get('acodec') is not None:
                 if res:
                     res += u', '
-                res += fdict['acodec']
+                res += u'%-5s' % fdict['acodec']
+            elif fdict.get('abr') is not None:
+                if res:
+                    res += u', '
+                res += 'audio'
             if fdict.get('abr') is not None:
                 res += u'@%3dk' % fdict['abr']
             return res
diff --git a/youtube_dl/extractor/spiegel.py b/youtube_dl/extractor/spiegel.py
@@ -2,18 +2,27 @@ import re
 import xml.etree.ElementTree
 
 from .common import InfoExtractor
+from ..utils import determine_ext
 
 
 class SpiegelIE(InfoExtractor):
     _VALID_URL = r'https?://(?:www\.)?spiegel\.de/video/[^/]*-(?P<videoID>[0-9]+)(?:\.html)?(?:#.*)?$'
-    _TEST = {
+    _TESTS = [{
         u'url': u'http://www.spiegel.de/video/vulkan-tungurahua-in-ecuador-ist-wieder-aktiv-video-1259285.html',
         u'file': u'1259285.mp4',
         u'md5': u'2c2754212136f35fb4b19767d242f66e',
         u'info_dict': {
             u"title": u"Vulkanausbruch in Ecuador: Der \"Feuerschlund\" ist wieder aktiv"
         }
-    }
+    },
+    {
+        u'url': u'http://www.spiegel.de/video/schach-wm-videoanalyse-des-fuenften-spiels-video-1309159.html',
+        u'file': u'1309159.mp4',
+        u'md5': u'f2cdf638d7aa47654e251e1aee360af1',
+        u'info_dict': {
+            u'title': u'Schach-WM in der Videoanalyse: Carlsen nutzt die Fehlgriffe des Titelverteidigers'
+        }
+    }]
 
     def _real_extract(self, url):
         m = re.match(self._VALID_URL, url)
@@ -21,25 +30,38 @@ class SpiegelIE(InfoExtractor):
 
         webpage = self._download_webpage(url, video_id)
 
-        video_title = self._html_search_regex(r'<div class="module-title">(.*?)</div>',
-            webpage, u'title')
+        video_title = self._html_search_regex(
+            r'<div class="module-title">(.*?)</div>', webpage, u'title')
 
         xml_url = u'http://video2.spiegel.de/flash/' + video_id + u'.xml'
-        xml_code = self._download_webpage(xml_url, video_id,
-                    note=u'Downloading XML', errnote=u'Failed to download XML')
+        xml_code = self._download_webpage(
+            xml_url, video_id,
+            note=u'Downloading XML', errnote=u'Failed to download XML')
 
         idoc = xml.etree.ElementTree.fromstring(xml_code)
-        last_type = idoc[-1]
-        filename = last_type.findall('./filename')[0].text
-        duration = float(last_type.findall('./duration')[0].text)
 
-        video_url = 'http://video2.spiegel.de/flash/' + filename
-        video_ext = filename.rpartition('.')[2]
+        formats = [
+            {
+                'format_id': n.tag.rpartition('type')[2],
+                'url': u'http://video2.spiegel.de/flash/' + n.find('./filename').text,
+                'width': int(n.find('./width').text),
+                'height': int(n.find('./height').text),
+                'abr': int(n.find('./audiobitrate').text),
+                'vbr': int(n.find('./videobitrate').text),
+                'vcodec': n.find('./codec').text,
+                'acodec': 'MP4A',
+            }
+            for n in list(idoc)
+            # Blacklist type 6, it's extremely LQ and not available on the same server
+            if n.tag.startswith('type') and n.tag != 'type6'
+        ]
+        formats.sort(key=lambda f: f['vbr'])
+        duration = float(idoc[0].findall('./duration')[0].text)
+
         info = {
             'id': video_id,
-            'url': video_url,
-            'ext': video_ext,
             'title': video_title,
             'duration': duration,
+            'formats': formats,
         }
-        return [info]
+        return info