X-Git-Url: http://git.cielonegro.org/gitweb.cgi?a=blobdiff_plain;f=youtube_dl%2Fextractor%2Fzdf.py;h=35ece354a6ecdf7ba5705184d4ceff22d57eb3d4;hb=f49d89ee04897b3ec64af74bb9fd94957c42331c;hp=a8d899883f900a524bb8740559440a9d8bc21aac;hpb=02dbf93f0e98a56ed04b4a9e6a6d62efd6d801f9;p=youtube-dl.git

diff --git a/youtube_dl/extractor/zdf.py b/youtube_dl/extractor/zdf.py
index a8d899883..35ece354a 100644
--- a/youtube_dl/extractor/zdf.py
+++ b/youtube_dl/extractor/zdf.py
@@ -1,24 +1,38 @@
+# coding: utf-8
+
 import operator
 import re
 
 from .common import InfoExtractor
 from ..utils import (
-    parse_xml_doc,
     unified_strdate,
 )
 
 
 class ZDFIE(InfoExtractor):
-    _VALID_URL = r'^http://www\.zdf\.de\/ZDFmediathek(?P<hash>#)?\/(.*beitrag\/video\/)(?P<video_id>[^/\?]+)(?:\?.*)?'
+    _VALID_URL = r'^https?://www\.zdf\.de/ZDFmediathek(?P<hash>#)?/(.*beitrag/(?:video/)?)(?P<video_id>[0-9]+)(?:/[^/?]+)?(?:\?.*)?'
+
+    _TEST = {
+        u"url": u"http://www.zdf.de/ZDFmediathek/beitrag/video/2037704/ZDFspezial---Ende-des-Machtpokers--?bc=sts;stt",
+        u"file": u"2037704.webm",
+        u"info_dict": {
+            u"upload_date": u"20131127",
+            u"description": u"Union und SPD haben sich auf einen Koalitionsvertrag geeinigt. Aber was bedeutet das für die Bürger? Sehen Sie hierzu das ZDFspezial \"Ende des Machtpokers - Große Koalition für Deutschland\".",
+            u"uploader": u"spezial",
+            u"title": u"ZDFspezial - Ende des Machtpokers"
+        },
+        u"skip": u"Videos on ZDF.de are depublicised in short order",
+    }
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
         video_id = mobj.group('video_id')
 
         xml_url = u'http://www.zdf.de/ZDFmediathek/xmlservice/web/beitragsDetails?ak=web&id=%s' % video_id
-        info_xml = self._download_webpage(
-            xml_url, video_id, note=u'Downloading video info')
-        doc = parse_xml_doc(info_xml)
+        doc = self._download_xml(
+            xml_url, video_id,
+            note=u'Downloading video info',
+            errnote=u'Failed to download video info')
 
         title = doc.find('.//information/title').text
         description = doc.find('.//information/detail').text
@@ -52,41 +66,52 @@ class ZDFIE(InfoExtractor):
                 (?P<proto>[^_]+)_(?P<index>[^_]+)_(?P<indexproto>[^_]+)
             ''', format_id)
 
+            ext = format_m.group('container')
+            is_supported = ext != 'f4f'
+
             PROTO_ORDER = ['http', 'rtmp', 'rtsp']
             try:
                 proto_pref = -PROTO_ORDER.index(format_m.group('proto'))
             except ValueError:
-                proto_pref = 999
+                proto_pref = -999
 
             quality = fnode.find('./quality').text
             QUALITY_ORDER = ['veryhigh', '300', 'high', 'med', 'low']
             try:
                 quality_pref = -QUALITY_ORDER.index(quality)
             except ValueError:
-                quality_pref = 999
+                quality_pref = -999
 
             abr = int(fnode.find('./audioBitrate').text) // 1000
             vbr = int(fnode.find('./videoBitrate').text) // 1000
-            pref = (is_available, proto_pref, quality_pref, vbr, abr)
+            pref = (is_available, is_supported,
+                    proto_pref, quality_pref, vbr, abr)
+
+            format_note = u''
+            if not is_supported:
+                format_note += u'(unsupported)'
+            if not format_note:
+                format_note = None
 
             return {
-                'format_id': format_id,
+                'format_id': format_id + u'-' + quality,
                 'url': video_url,
-                'ext': format_m.group('container'),
+                'ext': ext,
                 'acodec': format_m.group('acodec'),
                 'vcodec': format_m.group('vcodec'),
                 'abr': abr,
                 'vbr': vbr,
                 'width': int(fnode.find('./width').text),
                 'height': int(fnode.find('./height').text),
-                'quality_name': quality,
                 'filesize': int(fnode.find('./filesize').text),
-                'format_note': None if is_available else u'(unavailable)',
+                'format_note': format_note,
                 '_pref': pref,
+                '_available': is_available,
             }
 
         format_nodes = doc.findall('.//formitaeten/formitaet')
-        formats = sorted(map(xml_to_format, format_nodes),
+        formats = sorted(filter(lambda f: f['_available'],
+                                map(xml_to_format, format_nodes)),
                          key=operator.itemgetter('_pref'))
 
         return {