Merge remote-tracking branch 'upstream/master'

test/test_all_urls.py

@@ -37,6 +37,8 @@ class TestAllURLsMatching(unittest.TestCase):
         assertPlaylist(u'https://www.youtube.com/playlist?list=PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC')
         assertPlaylist(u'https://www.youtube.com/watch?v=AV6J6_AeFEQ&playnext=1&list=PL4023E734DA416012') #668
         self.assertFalse('youtube:playlist' in self.matching_ies(u'PLtS2H6bU1M'))
+        # Top tracks
+        assertPlaylist('https://www.youtube.com/playlist?list=MCUS.20142101')
 
     def test_youtube_matching(self):
         self.assertTrue(YoutubeIE.suitable(u'PLtS2H6bU1M'))

test/test_youtube_lists.py

@@ -117,6 +117,13 @@ class TestYoutubeLists(unittest.TestCase):
         original_video = entries[0]
         self.assertEqual(original_video['id'], 'rjFaenf1T-Y')
 
+    def test_youtube_toptracks(self):
+        dl = FakeYDL()
+        ie = YoutubePlaylistIE(dl)
+        result = ie.extract('https://www.youtube.com/playlist?list=MCUS')
+        entries = result['entries']
+        self.assertEqual(len(entries), 100)
+
     def test_youtube_toplist(self):
         dl = FakeYDL()
         ie = YoutubeTopListIE(dl)

youtube_dl/__init__.py

@@ -41,6 +41,7 @@ __authors__  = (
     'Chris Gahan',
     'Saimadhav Heblikar',
     'Mike Col',
+    'Andreas Schmitz',
 )
 
 __license__ = 'Public Domain'

youtube_dl/extractor/__init__.py

@@ -143,8 +143,10 @@ from .myvideo import MyVideoIE
 from .naver import NaverIE
 from .nba import NBAIE
 from .nbc import NBCNewsIE
+from .ndr import NDRIE
 from .ndtv import NDTVIE
 from .newgrounds import NewgroundsIE
+from .nfb import NFBIE
 from .nhl import NHLIE, NHLVideocenterIE
 from .niconico import NiconicoIE
 from .ninegag import NineGagIE

youtube_dl/extractor/chilloutzone.py

@@ -1,14 +1,18 @@
+from __future__ import unicode_literals
+
 import re
 import base64
-import urllib
 import json
 
 from .common import InfoExtractor
+from ..utils import (
+    clean_html,
+    ExtractorError
+)
 
-video_container = ('.mp4', '.mkv', '.flv')
 
 class ChilloutzoneIE(InfoExtractor):
-    _VALID_URL = r'(?:https?://)?(?:www\.)?chilloutzone\.net/video/(?P<id>[\w|-]+).html'
+    _VALID_URL = r'https?://(?:www\.)?chilloutzone\.net/video/(?P<id>[\w|-]+)\.html'
     _TEST = {
         'url': 'http://www.chilloutzone.net/video/enemene-meck-alle-katzen-weg.html',
         'md5': 'a76f3457e813ea0037e5244f509e66d1',
@@ -16,6 +20,7 @@ class ChilloutzoneIE(InfoExtractor):
             'id': 'enemene-meck-alle-katzen-weg',
             'ext': 'mp4',
             'title': 'Enemene Meck - Alle Katzen weg',
+            'description': 'Ist das der Umkehrschluss des Niesenden Panda-Babys?',
         },
     }
 
@@ -23,71 +28,45 @@ class ChilloutzoneIE(InfoExtractor):
         mobj = re.match(self._VALID_URL, url)
         video_id = mobj.group('id')
 
-        webpage_url = 'http://www.chilloutzone.net/video/' + video_id + '.html'
+        webpage = self._download_webpage(url, video_id)
 
-        # Log that we are starting to download the page
-        self.report_download_webpage(webpage_url)
-        webpage = self._download_webpage(webpage_url, video_id)
-    
-        # Log that we are starting to parse the page
-        self.report_extraction(video_id)        
-        # Find base64 decoded file info
-        base64_video_info = self._html_search_regex(r'var cozVidData = "(.+?)";', webpage, u'video Data')
-        # decode string and find video file
+        base64_video_info = self._html_search_regex(
+            r'var cozVidData = "(.+?)";', webpage, 'video data')
         decoded_video_info = base64.b64decode(base64_video_info).decode("utf-8")
         video_info_dict = json.loads(decoded_video_info)
+
         # get video information from dict
-        media_url = video_info_dict['mediaUrl']
-        description = video_info_dict['description']
+        video_url = video_info_dict['mediaUrl']
+        description = clean_html(video_info_dict.get('description'))
         title = video_info_dict['title']
         native_platform = video_info_dict['nativePlatform']
         native_video_id = video_info_dict['nativeVideoId']
         source_priority = video_info_dict['sourcePriority']
 
-
-        # Start video extraction
-        video_url = ''
         # If nativePlatform is None a fallback mechanism is used (i.e. youtube embed)
-        if native_platform == None:
-            # Look for other video urls
-            video_url = self._html_search_regex(r'<iframe.* src="(.+?)".*', webpage, u'fallback Video URL')
-            if 'youtube' in video_url:
-                self.to_screen(u'Youtube video detected:')
-                return self.url_result(video_url, ie='Youtube')
-
-        # For debugging purposes
-        #print video_info_dict
-        #print native_platform
-        #print native_video_id
-        #print source_priority
-        #print media_url
+        if native_platform is None:
+            youtube_url = self._html_search_regex(
+                r'<iframe.* src="((?:https?:)?//(?:[^.]+\.)?youtube\.com/.+?)"',
+                webpage, 'fallback video URL', default=None)
+            if youtube_url is not None:
+                return self.url_result(youtube_url, ie='Youtube')
 
         # Non Fallback: Decide to use native source (e.g. youtube or vimeo) or
         # the own CDN
         if source_priority == 'native':
             if native_platform == 'youtube':
-                self.to_screen(u'Youtube video detected:')
-                video_url = 'https://www.youtube.com/watch?v=' + native_video_id
-                return self.url_result(video_url, ie='Youtube') 
+                return self.url_result(video_id, ie='Youtube')
             if native_platform == 'vimeo':
-                self.to_screen(u'Vimeo video detected:')
-                video_url = 'http://vimeo.com/' + native_video_id
-                return self.url_result(video_url, ie='Vimeo')
-
-        # No redirect, use coz media url
-        video_url = media_url
-        if video_url.endswith('.mp4') == False:
-            self.report_warning(u'Url does not contain a video container')
-            return []
+                return self.url_result(
+                    'http://vimeo.com/' + native_video_id, ie='Vimeo')
 
+        if not video_url:
+            raise ExtractorError('No video found')
 
-        return [{
-            'id':        video_id,
-            'url':       video_url,
-            'ext':       'mp4',
-            'title':     title,
+        return {
+            'id': video_id,
+            'url': video_url,
+            'ext': 'mp4',
+            'title': title,
             'description': description,
-        }]
-
-
-
+        }

youtube_dl/extractor/elpais.py

@@ -9,7 +9,7 @@ from ..utils import unified_strdate
 
 class ElPaisIE(InfoExtractor):
     _VALID_URL = r'https?://(?:[^.]+\.)?elpais\.com/.*/(?P<id>[^/#?]+)\.html(?:$|[?#])'
-    IE_DESCR = 'El País'
+    IE_DESC = 'El País'
 
     _TEST = {
         'url': 'http://blogs.elpais.com/la-voz-de-inaki/2014/02/tiempo-nuevo-recetas-viejas.html',

youtube_dl/extractor/mooshare.py

@@ -61,7 +61,7 @@ class MooshareIE(InfoExtractor):
         }
 
         request = compat_urllib_request.Request(
-            'http://mooshare.biz/8dqtk4bjbp8g', compat_urllib_parse.urlencode(download_form))
+            'http://mooshare.biz/%s' % video_id, compat_urllib_parse.urlencode(download_form))
         request.add_header('Content-Type', 'application/x-www-form-urlencoded')
 
         self.to_screen('%s: Waiting for timeout' % video_id)
@@ -111,4 +111,4 @@ class MooshareIE(InfoExtractor):
             'thumbnail': thumbnail,
             'duration': duration,
             'formats': formats,
-        }
+        }

youtube_dl/extractor/ndr.py

@@ -0,0 +1,89 @@
+# encoding: utf-8
+from __future__ import unicode_literals
+
+import re
+
+from .common import InfoExtractor
+from ..utils import ExtractorError
+
+
+class NDRIE(InfoExtractor):
+    IE_NAME = 'ndr'
+    IE_DESC = 'NDR.de - Mediathek'
+    _VALID_URL = r'https?://www\.ndr\.de/.+?(?P<id>\d+)\.html'
+
+    _TESTS = [
+        # video
+        {
+            'url': 'http://www.ndr.de/fernsehen/sendungen/hallo_niedersachsen/media/hallonds19925.html',
+            'md5': '20eba151ff165f386643dad9c1da08f7',
+            'info_dict': {
+                'id': '19925',
+                'ext': 'mp4',
+                'title': 'Hallo Niedersachsen  ',
+                'description': 'Bei Hallo Niedersachsen um 19:30 Uhr erfahren Sie alles, was am Tag in Niedersachsen los war.',
+                'duration': 1722,
+            },
+        },
+        # audio
+        {
+            'url': 'http://www.ndr.de/903/audio191719.html',
+            'md5': '41ed601768534dd18a9ae34d84798129',
+            'info_dict': {
+                'id': '191719',
+                'ext': 'mp3',
+                'title': '"Es war schockierend"',
+                'description': 'md5:ed7ff8364793545021a6355b97e95f10',
+                'duration': 112,
+            }
+        }
+    ]
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        video_id = mobj.group('id')
+
+        page = self._download_webpage(url, video_id, 'Downloading page')
+
+        title = self._og_search_title(page)
+        description = self._og_search_description(page)
+
+        mobj = re.search(
+            r'<div class="duration"><span class="min">(?P<minutes>\d+)</span>:<span class="sec">(?P<seconds>\d+)</span></div>',
+            page)
+        duration = int(mobj.group('minutes')) * 60 + int(mobj.group('seconds')) if mobj else None
+
+        formats = []
+
+        mp3_url = re.search(r'''{src:'(?P<audio>[^']+)', type:"audio/mp3"},''', page)
+        if mp3_url:
+            formats.append({
+                'url': mp3_url.group('audio'),
+                'format_id': 'mp3',
+            })
+
+        thumbnail = None
+
+        video_url = re.search(r'''3: {src:'(?P<video>.+?)\.hi\.mp4', type:"video/mp4"},''', page)
+        if video_url:
+            thumbnail = self._html_search_regex(r'(?m)title: "NDR PLAYER",\s*poster: "([^"]+)",',
+                page, 'thumbnail', fatal=False)
+            if thumbnail:
+                thumbnail = 'http://www.ndr.de' + thumbnail
+            for format_id in ['lo', 'hi', 'hq']:
+                formats.append({
+                    'url': '%s.%s.mp4' % (video_url.group('video'), format_id),
+                    'format_id': format_id,
+                })
+
+        if not formats:
+            raise ExtractorError('No media links available for %s' % video_id)
+
+        return {
+            'id': video_id,
+            'title': title,
+            'description': description,
+            'thumbnail': thumbnail,
+            'duration': duration,
+            'formats': formats,
+        }

youtube_dl/extractor/nfb.py

@@ -0,0 +1,76 @@
+from __future__ import unicode_literals
+
+import re
+
+from .common import InfoExtractor
+from ..utils import (
+    compat_urllib_request,
+    compat_urllib_parse,
+)
+
+
+class NFBIE(InfoExtractor):
+    IE_NAME = 'nfb'
+    IE_DESC = 'National Film Board of Canada'
+    _VALID_URL = r'https?://(?:www\.)?(nfb|onf)\.ca/film/(?P<id>[\da-z_-]+)'
+
+    _TEST = {
+        'url': 'https://www.nfb.ca/film/qallunaat_why_white_people_are_funny',
+        'info_dict': {
+            'id': 'qallunaat_why_white_people_are_funny',
+            'ext': 'mp4',
+            'title': 'Qallunaat! Why White People Are Funny ',
+            'description': 'md5:836d8aff55e087d04d9f6df554d4e038',
+            'duration': 3128,
+            'uploader': 'Mark Sandiford',
+            'uploader_id': 'mark-sandiford',
+        },
+        'params': {
+            # rtmp download
+            'skip_download': True,
+        }
+    }
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        video_id = mobj.group('id')
+
+        page = self._download_webpage('https://www.nfb.ca/film/%s' % video_id, video_id, 'Downloading film page')
+
+        uploader_id = self._html_search_regex(r'<a class="director-link" href="/explore-all-directors/([^/]+)/"',
+            page, 'director id', fatal=False)
+        uploader = self._html_search_regex(r'<em class="director-name" itemprop="name">([^<]+)</em>',
+            page, 'director name', fatal=False)
+
+        request = compat_urllib_request.Request('https://www.nfb.ca/film/%s/player_config' % video_id,
+            compat_urllib_parse.urlencode({'getConfig': 'true'}).encode('ascii'))
+        request.add_header('Content-Type', 'application/x-www-form-urlencoded')
+        request.add_header('X-NFB-Referer', 'http://www.nfb.ca/medias/flash/NFBVideoPlayer.swf')
+
+        config = self._download_xml(request, video_id, 'Downloading player config XML')
+
+        thumbnail = config.find("./player/stream/media[@type='posterImage']/assets/asset[@quality='high']/default/url").text
+        video = config.find("./player/stream/media[@type='video']")
+        duration = int(video.get('duration'))
+        title = video.find('title').text
+        description = video.find('description').text
+
+        # It seems assets always go from lower to better quality, so no need to sort
+        formats = [{
+            'url': x.find('default/streamerURI').text + '/',
+            'play_path': x.find('default/url').text,
+            'rtmp_live': False,
+            'ext': 'mp4',
+            'format_id': x.get('quality'),
+        } for x in video.findall('assets/asset')]
+
+        return {
+            'id': video_id,
+            'title': title,
+            'description': description,
+            'thumbnail': thumbnail,
+            'duration': duration,
+            'uploader': uploader,
+            'uploader_id': uploader_id,
+            'formats': formats,
+        }

youtube_dl/extractor/youtube.py

@@ -1422,7 +1422,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor):
 
 class YoutubePlaylistIE(YoutubeBaseInfoExtractor):
     IE_DESC = u'YouTube.com playlists'
-    _VALID_URL = r"""(?:
+    _VALID_URL = r"""(?x)(?:
                         (?:https?://)?
                         (?:\w+\.)?
                         youtube\.com/
@@ -1431,7 +1431,11 @@ class YoutubePlaylistIE(YoutubeBaseInfoExtractor):
                            \? (?:.*?&)*? (?:p|a|list)=
                         |  p/
                         )
-                        ((?:PL|EC|UU|FL|RD)?[0-9A-Za-z-_]{10,})
+                        (
+                            (?:PL|EC|UU|FL|RD)?[0-9A-Za-z-_]{10,}
+                            # Top tracks, they can also include dots 
+                            |(?:MC)[\w\.]*
+                        )
                         .*
                      |
                         ((?:PL|EC|UU|FL|RD)[0-9A-Za-z-_]{10,})
@@ -1441,11 +1445,6 @@ class YoutubePlaylistIE(YoutubeBaseInfoExtractor):
     _VIDEO_RE = r'href="/watch\?v=(?P<id>[0-9A-Za-z_-]{11})&amp;[^"]*?index=(?P<index>\d+)'
     IE_NAME = u'youtube:playlist'
 
-    @classmethod
-    def suitable(cls, url):
-        """Receives a URL and returns True if suitable for this IE."""
-        return re.match(cls._VALID_URL, url, re.VERBOSE) is not None
-
     def _real_initialize(self):
         self._login()
 
@@ -1469,7 +1468,7 @@ class YoutubePlaylistIE(YoutubeBaseInfoExtractor):
 
     def _real_extract(self, url):
         # Extract playlist id
-        mobj = re.match(self._VALID_URL, url, re.VERBOSE)
+        mobj = re.match(self._VALID_URL, url)
         if mobj is None:
             raise ExtractorError(u'Invalid URL: %s' % url)
         playlist_id = mobj.group(1) or mobj.group(2)

youtube_dl/version.py

@@ -1,2 +1,2 @@
 
-__version__ = '2014.02.06.1'
+__version__ = '2014.02.06.3'