Merge tag 'upstream/2014.11.21'

[youtubedl] / youtube_dl / extractor / ard.py

diff --git a/youtube_dl/extractor/ard.py b/youtube_dl/extractor/ard.py
index 957bdefcbec7666473530815df9e3eb5dc88e096..630b1faa999617c79d160a636052e2acd6d5debd 100644 (file)
--- a/youtube_dl/extractor/ard.py
+++ b/youtube_dl/extractor/ard.py
@@ -4,16 +4,21 @@ from __future__ import unicode_literals
 import re
 
 from .common import InfoExtractor
+from .generic import GenericIE
 from ..utils import (
     determine_ext,
     ExtractorError,
     qualities,
-    compat_urllib_parse_urlparse,
-    compat_urllib_parse,
+    int_or_none,
+    parse_duration,
+    unified_strdate,
+    xpath_text,
+    parse_xml,
 )
 
 
-class ARDIE(InfoExtractor):
+class ARDMediathekIE(InfoExtractor):
+    IE_NAME = 'ARD:mediathek'
     _VALID_URL = r'^https?://(?:(?:www\.)?ardmediathek\.de|mediathek\.daserste\.de)/(?:.*/)(?P<video_id>[0-9]+|[^0-9][^/\?]+)[^/\?]*(?:\?.*)?'
 
     _TESTS = [{
@@ -46,11 +51,16 @@ class ARDIE(InfoExtractor):
         else:
             video_id = m.group('video_id')
 
-        urlp = compat_urllib_parse_urlparse(url)
-        url = urlp._replace(path=compat_urllib_parse.quote(urlp.path.encode('utf-8'))).geturl()
-
         webpage = self._download_webpage(url, video_id)
 
+        if '>Der gewünschte Beitrag ist nicht mehr verfügbar.<' in webpage:
+            raise ExtractorError('Video %s is no longer available' % video_id, expected=True)
+
+        if re.search(r'[\?&]rss($|[=&])', url):
+            doc = parse_xml(webpage)
+            if doc.tag == 'rss':
+                return GenericIE()._extract_rss(url, video_id, doc)
+
         title = self._html_search_regex(
             [r'<h1(?:\s+class="boxTopHeadline")?>(.*?)</h1>',
              r'<meta name="dcterms.title" content="(.*?)"/>',
@@ -125,3 +135,61 @@ class ARDIE(InfoExtractor):
             'formats': formats,
             'thumbnail': thumbnail,
         }
+
+
+class ARDIE(InfoExtractor):
+    _VALID_URL = '(?P<mainurl>https?://(www\.)?daserste\.de/[^?#]+/videos/(?P<display_id>[^/?#]+)-(?P<id>[0-9]+))\.html'
+    _TEST = {
+        'url': 'http://www.daserste.de/information/reportage-dokumentation/dokus/videos/die-story-im-ersten-mission-unter-falscher-flagge-100.html',
+        'md5': 'd216c3a86493f9322545e045ddc3eb35',
+        'info_dict': {
+            'display_id': 'die-story-im-ersten-mission-unter-falscher-flagge',
+            'id': '100',
+            'ext': 'mp4',
+            'duration': 2600,
+            'title': 'Die Story im Ersten: Mission unter falscher Flagge',
+            'upload_date': '20140804',
+            'thumbnail': 're:^https?://.*\.jpg$',
+        }
+    }
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        display_id = mobj.group('display_id')
+
+        player_url = mobj.group('mainurl') + '~playerXml.xml'
+        doc = self._download_xml(player_url, display_id)
+        video_node = doc.find('./video')
+        upload_date = unified_strdate(xpath_text(
+            video_node, './broadcastDate'))
+        thumbnail = xpath_text(video_node, './/teaserImage//variant/url')
+
+        formats = []
+        for a in video_node.findall('.//asset'):
+            f = {
+                'format_id': a.attrib['type'],
+                'width': int_or_none(a.find('./frameWidth').text),
+                'height': int_or_none(a.find('./frameHeight').text),
+                'vbr': int_or_none(a.find('./bitrateVideo').text),
+                'abr': int_or_none(a.find('./bitrateAudio').text),
+                'vcodec': a.find('./codecVideo').text,
+                'tbr': int_or_none(a.find('./totalBitrate').text),
+            }
+            if a.find('./serverPrefix').text:
+                f['url'] = a.find('./serverPrefix').text
+                f['playpath'] = a.find('./fileName').text
+            else:
+                f['url'] = a.find('./fileName').text
+            formats.append(f)
+        self._sort_formats(formats)
+
+        return {
+            'id': mobj.group('id'),
+            'formats': formats,
+            'display_id': display_id,
+            'title': video_node.find('./title').text,
+            'duration': parse_duration(video_node.find('./duration').text),
+            'upload_date': upload_date,
+            'thumbnail': thumbnail,
+        }
+