release 2013.07.17

Added an IE and test for Freesound.org .

devscripts/youtube_genalgo.py

@@ -20,12 +20,15 @@ tests = [
     # 84
     ("qwertyuioplkjhgfdsazxcvbnm1234567890QWERTYUIOPLKJHGFDSAZXCVBNM!@#$%^&*()_-+={[};?>.<",
      "<.>?;}[{=+-_)(*&^%$#@!MNBVCXZASDFGHJKLPOIUYTREWe098765432rmnbvcxzasdfghjklpoiuyt1"),
-    # 83 - vfl26ng3K 2013/07/10
+    # 83 - vflcaqGO8 2013/07/11
     ("qwertyuioplkjhgfdsazxcvbnm1234567890QWERTYUIOPLKJHGFDSAZXCVBNM!#$%^&*()_+={[};?/>.<",
-     "qwertyuioplkjhgfdsazxcvbnm1234567890QWERTYUIOPLKJHGFDSAZXCVBNM!#$%^&*()_+={[};?/>"),
+     "urty8ioplkjhgfdsazxcvbqm1234567S90QWERTYUIOPLKJHGFDnAZXCVBNM!#$%^&*()_+={[};?/>.<"),
     # 82
     ("qwertyuioplkjhgfdsazxcvbnm1234567890QWERTYUIOPLKHGFDSAZXCVBNM!@#$%^&*(-+={[};?/>.<",
      "Q>/?;}[{=+-(*<^%$#@!MNBVCXZASDFGHKLPOIUY8REWT0q&7654321mnbvcxzasdfghjklpoiuytrew9"),
+    # 81
+    ("qwertyuioplkjhgfdsazxcvbnm1234567890QWERTYUIOPLKHGFDSAZXCVBNM!@#$%^&*(-+={[};?/>.",
+     "urty8ioplkjhgfdsazxcvbqm1234567e90QWERTYUIOPLKHGFDSnZXCVBNM!@#$%^&*(-+={[};?/>."),
 ]
 
 def find_matching(wrong, right):

test/test_utils.py

@@ -4,6 +4,7 @@
 
 import sys
 import unittest
+import xml.etree.ElementTree
 
 # Allow direct execution
 import os
@@ -16,6 +17,7 @@ from youtube_dl.utils import unescapeHTML
 from youtube_dl.utils import orderedSet
 from youtube_dl.utils import DateRange
 from youtube_dl.utils import unified_strdate
+from youtube_dl.utils import find_xpath_attr
 
 if sys.version_info < (3, 0):
     _compat_str = lambda b: b.decode('unicode-escape')
@@ -112,5 +114,18 @@ class TestUtil(unittest.TestCase):
         self.assertEqual(unified_strdate('Dec 14, 2012'), '20121214')
         self.assertEqual(unified_strdate('2012/10/11 01:56:38 +0000'), '20121011')
 
+    def test_find_xpath_attr(self):
+        testxml = u'''<root>
+            <node/>
+            <node x="a"/>
+            <node x="a" y="c" />
+            <node x="b" y="d" />
+        </root>'''
+        doc = xml.etree.ElementTree.fromstring(testxml)
+
+        self.assertEqual(find_xpath_attr(doc, './/fourohfour', 'n', 'v'), None)
+        self.assertEqual(find_xpath_attr(doc, './/node', 'x', 'a'), doc[1])
+        self.assertEqual(find_xpath_attr(doc, './/node', 'y', 'c'), doc[2])
+
 if __name__ == '__main__':
     unittest.main()

test/test_youtube_sig.py

@@ -45,7 +45,7 @@ class TestYoutubeSig(unittest.TestCase):
 
     def test_83(self):
         wrong = "qwertyuioplkjhgfdsazxcvbnm1234567890QWERTYUIOPLKJHGFDSAZXCVBNM!#$%^&*()_+={[};?/>.<"
-        right = "qwertyuioplkjhgfdsazxcvbnm1234567890QWERTYUIOPLKJHGFDSAZXCVBNM!#$%^&*()_+={[};?/>"
+        right = "urty8ioplkjhgfdsazxcvbqm1234567S90QWERTYUIOPLKJHGFDnAZXCVBNM!#$%^&*()_+={[};?/>.<"
         self.assertEqual(sig(wrong), right)
 
     def test_82(self):
@@ -53,5 +53,10 @@ class TestYoutubeSig(unittest.TestCase):
         right = "Q>/?;}[{=+-(*<^%$#@!MNBVCXZASDFGHKLPOIUY8REWT0q&7654321mnbvcxzasdfghjklpoiuytrew9"
         self.assertEqual(sig(wrong), right)
 
+    def test_81(self):
+        wrong = "qwertyuioplkjhgfdsazxcvbnm1234567890QWERTYUIOPLKHGFDSAZXCVBNM!@#$%^&*(-+={[};?/>."
+        right = "urty8ioplkjhgfdsazxcvbqm1234567e90QWERTYUIOPLKHGFDSnZXCVBNM!@#$%^&*(-+={[};?/>."
+        self.assertEqual(sig(wrong), right)
+
 if __name__ == '__main__':
     unittest.main()

youtube_dl/YoutubeDL.py

@@ -348,6 +348,7 @@ class YoutubeDL(object):
 
         result_type = ie_result.get('_type', 'video') # If not given we suppose it's a video, support the default old system
         if result_type == 'video':
+            ie_result.update(extra_info)
             if 'playlist' not in ie_result:
                 # It isn't part of a playlist
                 ie_result['playlist'] = None
@@ -528,10 +529,8 @@ class YoutubeDL(object):
                 return
 
         if self.params.get('writethumbnail', False):
-            if 'thumbnail' in info_dict:
-                thumb_format = info_dict['thumbnail'].rpartition(u'/')[2].rpartition(u'.')[2]
-                if not thumb_format:
-                    thumb_format = 'jpg'
+            if info_dict.get('thumbnail') is not None:
+                thumb_format = determine_ext(info_dict['thumbnail'], u'jpg')
                 thumb_filename = filename.rpartition('.')[0] + u'.' + thumb_format
                 self.to_screen(u'[%s] %s: Downloading thumbnail ...' %
                                (info_dict['extractor'], info_dict['id']))

youtube_dl/extractor/__init__.py

@@ -6,17 +6,21 @@ from .bandcamp import BandcampIE
 from .bliptv import BlipTVIE, BlipTVUserIE
 from .breakcom import BreakIE
 from .brightcove import BrightcoveIE
+from .canalplus import CanalplusIE
 from .collegehumor import CollegeHumorIE
 from .comedycentral import ComedyCentralIE
+from .criterion import CriterionIE
 from .cspan import CSpanIE
 from .dailymotion import DailymotionIE
 from .depositfiles import DepositFilesIE
 from .dotsub import DotsubIE
 from .dreisat import DreiSatIE
+from .ehow import EHowIE
 from .eighttracks import EightTracksIE
 from .escapist import EscapistIE
 from .facebook import FacebookIE
 from .flickr import FlickrIE
+from .freesound import FreesoundIE
 from .funnyordie import FunnyOrDieIE
 from .gamespot import GameSpotIE
 from .gametrailers import GametrailersIE
@@ -26,6 +30,7 @@ from .googlesearch import GoogleSearchIE
 from .hotnewhiphop import HotNewHipHopIE
 from .howcast import HowcastIE
 from .hypem import HypemIE
+from .ign import IGNIE, OneUPIE
 from .ina import InaIE
 from .infoq import InfoQIE
 from .instagram import InstagramIE
@@ -33,6 +38,7 @@ from .jukebox import JukeboxIE
 from .justintv import JustinTVIE
 from .keek import KeekIE
 from .liveleak import LiveLeakIE
+from .livestream import LivestreamIE
 from .metacafe import MetacafeIE
 from .mixcloud import MixcloudIE
 from .mtv import MTVIE

youtube_dl/extractor/archiveorg.py

@@ -48,6 +48,7 @@ class ArchiveOrgIE(InfoExtractor):
         formats.sort(key=lambda fdata: fdata['file_size'])
 
         info = {
+            '_type': 'video',
             'id': video_id,
             'title': title,
             'formats': formats,
@@ -63,4 +64,4 @@ class ArchiveOrgIE(InfoExtractor):
         info['url'] = formats[-1]['url']
         info['ext'] = determine_ext(formats[-1]['url'])
 
-        return self.video_result(info)
+        return info

youtube_dl/extractor/arte.py

@@ -5,6 +5,7 @@ import xml.etree.ElementTree
 from .common import InfoExtractor
 from ..utils import (
     ExtractorError,
+    find_xpath_attr,
     unified_strdate,
 )
 
@@ -119,7 +120,7 @@ class ArteTvIE(InfoExtractor):
         ref_xml_url = ref_xml_url.replace('.html', ',view,asPlayerXml.xml')
         ref_xml = self._download_webpage(ref_xml_url, video_id, note=u'Downloading metadata')
         ref_xml_doc = xml.etree.ElementTree.fromstring(ref_xml)
-        config_node = ref_xml_doc.find('.//video[@lang="%s"]' % lang)
+        config_node = find_xpath_attr(ref_xml_doc, './/video', 'lang', lang)
         config_xml_url = config_node.attrib['ref']
         config_xml = self._download_webpage(config_xml_url, video_id, note=u'Downloading configuration')
 

youtube_dl/extractor/brightcove.py

@@ -1,28 +1,82 @@
 import re
 import json
+import xml.etree.ElementTree
 
 from .common import InfoExtractor
+from ..utils import (
+    compat_urllib_parse,
+    find_xpath_attr,
+    compat_urlparse,
+)
 
 class BrightcoveIE(InfoExtractor):
-    _VALID_URL = r'http://.*brightcove\.com/.*\?(?P<query>.*videoPlayer=(?P<id>\d*).*)'
+    _VALID_URL = r'https?://.*brightcove\.com/(services|viewer).*\?(?P<query>.*)'
+    _FEDERATED_URL_TEMPLATE = 'http://c.brightcove.com/services/viewer/htmlFederated?%s'
+    _PLAYLIST_URL_TEMPLATE = 'http://c.brightcove.com/services/json/experience/runtime/?command=get_programming_for_experience&playerKey=%s'
+    
+    # There is a test for Brigtcove in GenericIE, that way we test both the download
+    # and the detection of videos, and we don't have to find an URL that is always valid
+
+    @classmethod
+    def _build_brighcove_url(cls, object_str):
+        """
+        Build a Brightcove url from a xml string containing
+        <object class="BrightcoveExperience">{params}</object>
+        """
+        object_doc = xml.etree.ElementTree.fromstring(object_str)
+        assert u'BrightcoveExperience' in object_doc.attrib['class']
+        params = {'flashID': object_doc.attrib['id'],
+                  'playerID': find_xpath_attr(object_doc, './param', 'name', 'playerID').attrib['value'],
+                  }
+        playerKey = find_xpath_attr(object_doc, './param', 'name', 'playerKey')
+        # Not all pages define this value
+        if playerKey is not None:
+            params['playerKey'] = playerKey.attrib['value']
+        videoPlayer = find_xpath_attr(object_doc, './param', 'name', '@videoPlayer')
+        if videoPlayer is not None:
+            params['@videoPlayer'] = videoPlayer.attrib['value']
+        data = compat_urllib_parse.urlencode(params)
+        return cls._FEDERATED_URL_TEMPLATE % data
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
-        query = mobj.group('query')
-        video_id = mobj.group('id')
+        query_str = mobj.group('query')
+        query = compat_urlparse.parse_qs(query_str)
 
-        request_url = 'http://c.brightcove.com/services/viewer/htmlFederated?%s' % query
+        videoPlayer = query.get('@videoPlayer')
+        if videoPlayer:
+            return self._get_video_info(videoPlayer[0], query_str)
+        else:
+            player_key = query['playerKey']
+            return self._get_playlist_info(player_key[0])
+
+    def _get_video_info(self, video_id, query):
+        request_url = self._FEDERATED_URL_TEMPLATE % query
         webpage = self._download_webpage(request_url, video_id)
 
         self.report_extraction(video_id)
         info = self._search_regex(r'var experienceJSON = ({.*?});', webpage, 'json')
         info = json.loads(info)['data']
         video_info = info['programmedContent']['videoPlayer']['mediaDTO']
+
+        return self._extract_video_info(video_info)
+
+    def _get_playlist_info(self, player_key):
+        playlist_info = self._download_webpage(self._PLAYLIST_URL_TEMPLATE % player_key,
+                                               player_key, u'Downloading playlist information')
+
+        playlist_info = json.loads(playlist_info)['videoList']
+        videos = [self._extract_video_info(video_info) for video_info in playlist_info['mediaCollectionDTO']['videoDTOs']]
+
+        return self.playlist_result(videos, playlist_id=playlist_info['id'],
+                                    playlist_title=playlist_info['mediaCollectionDTO']['displayName'])
+
+    def _extract_video_info(self, video_info):
         renditions = video_info['renditions']
         renditions = sorted(renditions, key=lambda r: r['size'])
         best_format = renditions[-1]
-        
-        return {'id': video_id,
+
+        return {'id': video_info['id'],
                 'title': video_info['displayName'],
                 'url': best_format['defaultURL'], 
                 'ext': 'mp4',

youtube_dl/extractor/canalplus.py

@@ -0,0 +1,46 @@
+import re
+import xml.etree.ElementTree
+
+from .common import InfoExtractor
+from ..utils import unified_strdate
+
+class CanalplusIE(InfoExtractor):
+    _VALID_URL = r'https?://www\.canalplus\.fr/.*?\?vid=(?P<id>\d+)'
+    _VIDEO_INFO_TEMPLATE = 'http://service.canal-plus.com/video/rest/getVideosLiees/cplus/%s'
+    IE_NAME = u'canalplus.fr'
+
+    _TEST = {
+        u'url': u'http://www.canalplus.fr/c-divertissement/pid3351-c-le-petit-journal.html?vid=889861',
+        u'file': u'889861.flv',
+        u'md5': u'590a888158b5f0d6832f84001fbf3e99',
+        u'info_dict': {
+            u'title': u'Le Petit Journal 20/06/13 - La guerre des drone',
+            u'upload_date': u'20130620',
+        },
+        u'skip': u'Requires rtmpdump'
+    }
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        video_id = mobj.group('id')
+        info_url = self._VIDEO_INFO_TEMPLATE % video_id
+        info_page = self._download_webpage(info_url,video_id, 
+                                           u'Downloading video info')
+
+        self.report_extraction(video_id)
+        doc = xml.etree.ElementTree.fromstring(info_page.encode('utf-8'))
+        video_info = [video for video in doc if video.find('ID').text == video_id][0]
+        infos = video_info.find('INFOS')
+        media = video_info.find('MEDIA')
+        formats = [media.find('VIDEOS/%s' % format)
+            for format in ['BAS_DEBIT', 'HAUT_DEBIT', 'HD']]
+        video_url = [format.text for format in formats if format is not None][-1]
+
+        return {'id': video_id,
+                'title': u'%s - %s' % (infos.find('TITRAGE/TITRE').text,
+                                       infos.find('TITRAGE/SOUS_TITRE').text),
+                'url': video_url,
+                'ext': 'flv',
+                'upload_date': unified_strdate(infos.find('PUBLICATION/DATE').text),
+                'thumbnail': media.find('IMAGES/GRAND').text,
+                }

youtube_dl/extractor/common.py

@@ -125,6 +125,11 @@ class InfoExtractor(object):
 
     def _download_webpage_handle(self, url_or_request, video_id, note=None, errnote=None):
         """ Returns a tuple (page content as string, URL handle) """
+
+        # Strip hashes from the URL (#1038)
+        if isinstance(url_or_request, (compat_str, str)):
+            url_or_request = url_or_request.partition('#')[0]
+
         urlh = self._request_webpage(url_or_request, video_id, note, errnote)
         content_type = urlh.headers.get('Content-Type', '')
         m = re.match(r'[a-zA-Z0-9_.-]+/[a-zA-Z0-9_.-]+\s*;\s*charset=(.+)', content_type)
@@ -169,11 +174,6 @@ class InfoExtractor(object):
         self.to_screen(u'Logging in')
 
     #Methods for following #608
-    #They set the correct value of the '_type' key
-    def video_result(self, video_info):
-        """Returns a video"""
-        video_info['_type'] = 'video'
-        return video_info
     def url_result(self, url, ie=None):
         """Returns a url that points to a page that should be processed"""
         #TODO: ie should be the class used for getting the info
@@ -262,6 +262,30 @@ class InfoExtractor(object):
         
         return (username, password)
 
+    # Helper functions for extracting OpenGraph info
+    @staticmethod
+    def _og_regex(prop):
+        return r'<meta.+?property=[\'"]og:%s[\'"].+?content=(?:"(.+?)"|\'(.+?)\')' % re.escape(prop)
+
+    def _og_search_property(self, prop, html, name=None, **kargs):
+        if name is None:
+            name = 'OpenGraph %s' % prop
+        return self._html_search_regex(self._og_regex(prop), html, name, flags=re.DOTALL, **kargs)
+
+    def _og_search_thumbnail(self, html, **kargs):
+        return self._og_search_property('image', html, u'thumbnail url', fatal=False, **kargs)
+
+    def _og_search_description(self, html, **kargs):
+        return self._og_search_property('description', html, fatal=False, **kargs)
+
+    def _og_search_title(self, html, **kargs):
+        return self._og_search_property('title', html, **kargs)
+
+    def _og_search_video_url(self, html, name='video url', **kargs):
+        return self._html_search_regex([self._og_regex('video:secure_url'),
+                                        self._og_regex('video')],
+                                       html, name, **kargs)
+
 class SearchInfoExtractor(InfoExtractor):
     """
     Base class for paged search queries extractors.

youtube_dl/extractor/criterion.py

@@ -0,0 +1,40 @@
+# -*- coding: utf-8 -*-
+
+import re
+
+from .common import InfoExtractor
+from ..utils import determine_ext
+
+class CriterionIE(InfoExtractor):
+    _VALID_URL = r'https?://www\.criterion\.com/films/(\d*)-.+'
+    _TEST = {
+        u'url': u'http://www.criterion.com/films/184-le-samourai',
+        u'file': u'184.mp4',
+        u'md5': u'bc51beba55685509883a9a7830919ec3',
+        u'info_dict': {
+            u"title": u"Le Samouraï",
+            u"description" : u'md5:a2b4b116326558149bef81f76dcbb93f',
+        }
+    }
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        video_id = mobj.group(1)
+        webpage = self._download_webpage(url, video_id)
+
+        final_url = self._search_regex(r'so.addVariable\("videoURL", "(.+?)"\)\;',
+                                webpage, 'video url')
+        title = self._html_search_regex(r'<meta content="(.+?)" property="og:title" />',
+                                webpage, 'video title')
+        description = self._html_search_regex(r'<meta name="description" content="(.+?)" />',
+                                webpage, 'video description')
+        thumbnail = self._search_regex(r'so.addVariable\("thumbnailURL", "(.+?)"\)\;',
+                                webpage, 'thumbnail url')
+
+        return {'id': video_id,
+                'url' : final_url,
+                'title': title,
+                'ext': determine_ext(final_url),
+                'description': description,
+                'thumbnail': thumbnail,
+                }

youtube_dl/extractor/cspan.py

@@ -34,8 +34,6 @@ class CSpanIE(InfoExtractor):
         description = self._html_search_regex(r'<meta (?:property="og:|name=")description" content="(.*?)"',
                                               webpage, 'description',
                                               flags=re.MULTILINE|re.DOTALL)
-        thumbnail = self._html_search_regex(r'<meta property="og:image" content="(.*?)"',
-                                            webpage, 'thumbnail')
 
         url = self._search_regex(r'<string name="URL">(.*?)</string>',
                                  video_info, 'video url')
@@ -49,5 +47,5 @@ class CSpanIE(InfoExtractor):
                 'url': url,
                 'play_path': path,
                 'description': description,
-                'thumbnail': thumbnail,
+                'thumbnail': self._og_search_thumbnail(webpage),
                 }

youtube_dl/extractor/dailymotion.py

@@ -39,9 +39,6 @@ class DailymotionIE(InfoExtractor):
         # Extract URL, uploader and title from webpage
         self.report_extraction(video_id)
 
-        video_title = self._html_search_regex(r'<meta property="og:title" content="(.*?)" />',
-                                              webpage, 'title')
-
         video_uploader = self._search_regex([r'(?im)<span class="owner[^\"]+?">[^<]+?<a [^>]+?>([^<]+?)</a>',
                                              # Looking for official user
                                              r'<(?:span|a) .*?rel="author".*?>([^<]+?)</'],
@@ -76,7 +73,7 @@ class DailymotionIE(InfoExtractor):
             'url':      video_url,
             'uploader': video_uploader,
             'upload_date':  video_upload_date,
-            'title':    video_title,
+            'title':    self._og_search_title(webpage),
             'ext':      video_extension,
             'thumbnail': info['thumbnail_url']
         }]

youtube_dl/extractor/dreisat.py

@@ -67,6 +67,7 @@ class DreiSatIE(InfoExtractor):
         formats.sort(key=_sortkey)
 
         info = {
+            '_type': 'video',
             'id': video_id,
             'title': video_title,
             'formats': formats,
@@ -81,4 +82,4 @@ class DreiSatIE(InfoExtractor):
         info['url'] = formats[-1]['url']
         info['ext'] = determine_ext(formats[-1]['url'])
 
-        return self.video_result(info)
+        return info

youtube_dl/extractor/ehow.py

@@ -0,0 +1,46 @@
+import re
+
+from ..utils import (
+    compat_urllib_parse,
+    determine_ext
+)
+from .common import InfoExtractor
+
+
+class EHowIE(InfoExtractor):
+    IE_NAME = u'eHow'
+    _VALID_URL = r'(?:https?://)?(?:www\.)?ehow\.com/[^/_?]*_(?P<id>[0-9]+)'
+    _TEST = {
+        u'url': u'http://www.ehow.com/video_12245069_hardwood-flooring-basics.html',
+        u'file': u'12245069.flv',
+        u'md5': u'9809b4e3f115ae2088440bcb4efbf371',
+        u'info_dict': {
+            u"title": u"Hardwood Flooring Basics",
+            u"description": u"Hardwood flooring may be time consuming, but its ultimately a pretty straightforward concept. Learn about hardwood flooring basics with help from a hardware flooring business owner in this free video...",
+   			u"uploader": u"Erick Nathan"
+        }
+    }
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        video_id = mobj.group('id')
+        webpage = self._download_webpage(url, video_id)
+        video_url = self._search_regex(r'(?:file|source)=(http[^\'"&]*)',
+            webpage, u'video URL')
+        final_url = compat_urllib_parse.unquote(video_url)        
+        uploader = self._search_regex(r'<meta name="uploader" content="(.+?)" />',
+            webpage, u'uploader')
+        title = self._og_search_title(webpage).replace(' | eHow', '')
+        ext = determine_ext(final_url)
+
+        return {
+            '_type':       'video',
+            'id':          video_id,
+            'url':         final_url,
+            'ext':         ext,
+            'title':       title,
+            'thumbnail':   self._og_search_thumbnail(webpage),
+            'description': self._og_search_description(webpage),
+            'uploader':    uploader,
+        }
+

youtube_dl/extractor/escapist.py

@@ -36,11 +36,7 @@ class EscapistIE(InfoExtractor):
         videoDesc = self._html_search_regex('<meta name="description" content="([^"]*)"',
             webpage, u'description', fatal=False)
 
-        imgUrl = self._html_search_regex('<meta property="og:image" content="([^"]*)"',
-            webpage, u'thumbnail', fatal=False)
-
-        playerUrl = self._html_search_regex('<meta property="og:video" content="([^"]*)"',
-            webpage, u'player url')
+        playerUrl = self._og_search_video_url(webpage, name='player url')
 
         title = self._html_search_regex('<meta name="title" content="([^"]*)"',
             webpage, u'player url').split(' : ')[-1]
@@ -70,7 +66,7 @@ class EscapistIE(InfoExtractor):
             'upload_date': None,
             'title': title,
             'ext': 'mp4',
-            'thumbnail': imgUrl,
+            'thumbnail': self._og_search_thumbnail(webpage),
             'description': videoDesc,
             'player_url': playerUrl,
         }

youtube_dl/extractor/flickr.py

@@ -47,21 +47,12 @@ class FlickrIE(InfoExtractor):
             raise ExtractorError(u'Unable to extract video url')
         video_url = mobj.group(1) + unescapeHTML(mobj.group(2))
 
-        video_title = self._html_search_regex(r'<meta property="og:title" content=(?:"([^"]+)"|\'([^\']+)\')',
-            webpage, u'video title')
-
-        video_description = self._html_search_regex(r'<meta property="og:description" content=(?:"([^"]+)"|\'([^\']+)\')',
-            webpage, u'description', fatal=False)
-
-        thumbnail = self._html_search_regex(r'<meta property="og:image" content=(?:"([^"]+)"|\'([^\']+)\')',
-            webpage, u'thumbnail', fatal=False)
-
         return [{
             'id':          video_id,
             'url':         video_url,
             'ext':         'mp4',
-            'title':       video_title,
-            'description': video_description,
-            'thumbnail':   thumbnail,
+            'title':       self._og_search_title(webpage),
+            'description': self._og_search_description(webpage),
+            'thumbnail':   self._og_search_thumbnail(webpage),
             'uploader_id': video_uploader_id,
         }]

youtube_dl/extractor/freesound.py

@@ -0,0 +1,36 @@
+import re
+
+from .common import InfoExtractor
+from ..utils import determine_ext
+
+class FreesoundIE(InfoExtractor):
+    _VALID_URL = r'(?:https?://)?(?:www\.)?freesound\.org/people/([^/]+)/sounds/(?P<id>[^/]+)'
+    _TEST = {
+        u'url': u'http://www.freesound.org/people/miklovan/sounds/194503/',
+        u'file': u'194503.mp3',
+        u'md5': u'12280ceb42c81f19a515c745eae07650',
+        u'info_dict': {
+            u"title": u"gulls in the city.wav",
+            u"uploader" : u"miklovan",
+            u'description': u'the sounds of seagulls in the city',
+        }
+    }
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        music_id = mobj.group('id')
+        webpage = self._download_webpage(url, music_id)
+        title = self._html_search_regex(r'<div id="single_sample_header">.*?<a href="#">(.+?)</a>',
+                                webpage, 'music title', flags=re.DOTALL)
+        music_url = self._og_search_property('audio', webpage, 'music url')
+        description = self._html_search_regex(r'<div id="sound_description">(.*?)</div>',
+                                webpage, 'description', fatal=False, flags=re.DOTALL)
+
+        return [{
+            'id':       music_id,
+            'title':    title,            
+            'url':      music_url,
+            'uploader': self._og_search_property('audio:artist', webpage, 'music uploader'),
+            'ext':      determine_ext(music_url),
+            'description': description,
+        }]

youtube_dl/extractor/funnyordie.py

@@ -27,14 +27,11 @@ class FunnyOrDieIE(InfoExtractor):
         title = self._html_search_regex((r"<h1 class='player_page_h1'.*?>(?P<title>.*?)</h1>",
             r'<title>(?P<title>[^<]+?)</title>'), webpage, 'title', flags=re.DOTALL)
 
-        video_description = self._html_search_regex(r'<meta property="og:description" content="(?P<desc>.*?)"',
-            webpage, u'description', fatal=False, flags=re.DOTALL)
-
         info = {
             'id': video_id,
             'url': video_url,
             'ext': 'mp4',
             'title': title,
-            'description': video_description,
+            'description': self._og_search_description(webpage),
         }
         return [info]

youtube_dl/extractor/gametrailers.py

@@ -1,68 +1,36 @@
 import re
 
-from .common import InfoExtractor
-from ..utils import (
-    compat_urllib_parse,
+from .mtv import MTVIE, _media_xml_tag
 
-    ExtractorError,
-)
-
-class GametrailersIE(InfoExtractor):
+class GametrailersIE(MTVIE):
+    """
+    Gametrailers use the same videos system as MTVIE, it just changes the feed
+    url, where the uri is and the method to get the thumbnails.
+    """
     _VALID_URL = r'http://www.gametrailers.com/(?P<type>videos|reviews|full-episodes)/(?P<id>.*?)/(?P<title>.*)'
     _TEST = {
         u'url': u'http://www.gametrailers.com/videos/zbvr8i/mirror-s-edge-2-e3-2013--debut-trailer',
-        u'file': u'zbvr8i.flv',
-        u'md5': u'c3edbc995ab4081976e16779bd96a878',
+        u'file': u'70e9a5d7-cf25-4a10-9104-6f3e7342ae0d.mp4',
+        u'md5': u'4c8e67681a0ea7ec241e8c09b3ea8cf7',
         u'info_dict': {
-            u"title": u"E3 2013: Debut Trailer"
+            u'title': u'E3 2013: Debut Trailer',
+            u'description': u'Faith is back!  Check out the World Premiere trailer for Mirror\'s Edge 2 straight from the EA Press Conference at E3 2013!',
         },
-        u'skip': u'Requires rtmpdump'
     }
+    # Overwrite MTVIE properties we don't want
+    _TESTS = []
+
+    _FEED_URL = 'http://www.gametrailers.com/feeds/mrss'
+
+    def _get_thumbnail_url(self, uri, itemdoc):
+        search_path = '%s/%s' % (_media_xml_tag('group'), _media_xml_tag('thumbnail'))
+        return itemdoc.find(search_path).attrib['url']
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
-        if mobj is None:
-            raise ExtractorError(u'Invalid URL: %s' % url)
         video_id = mobj.group('id')
-        video_type = mobj.group('type')
         webpage = self._download_webpage(url, video_id)
-        if video_type == 'full-episodes':
-            mgid_re = r'data-video="(?P<mgid>mgid:.*?)"'
-        else:
-            mgid_re = r'data-contentId=\'(?P<mgid>mgid:.*?)\''
-        mgid = self._search_regex(mgid_re, webpage, u'mgid')
-        data = compat_urllib_parse.urlencode({'uri': mgid, 'acceptMethods': 'fms'})
-
-        info_page = self._download_webpage('http://www.gametrailers.com/feeds/mrss?' + data,
-                                           video_id, u'Downloading video info')
-        links_webpage = self._download_webpage('http://www.gametrailers.com/feeds/mediagen/?' + data,
-                                               video_id, u'Downloading video urls info')
-
-        self.report_extraction(video_id)
-        info_re = r'''<title><!\[CDATA\[(?P<title>.*?)\]\]></title>.*
-                      <description><!\[CDATA\[(?P<description>.*?)\]\]></description>.*
-                      <image>.*
-                        <url>(?P<thumb>.*?)</url>.*
-                      </image>'''
-
-        m_info = re.search(info_re, info_page, re.VERBOSE|re.DOTALL)
-        if m_info is None:
-            raise ExtractorError(u'Unable to extract video info')
-        video_title = m_info.group('title')
-        video_description = m_info.group('description')
-        video_thumb = m_info.group('thumb')
-
-        m_urls = list(re.finditer(r'<src>(?P<url>.*)</src>', links_webpage))
-        if m_urls is None or len(m_urls) == 0:
-            raise ExtractorError(u'Unable to extract video url')
-        # They are sorted from worst to best quality
-        video_url = m_urls[-1].group('url')
-
-        return {'url':         video_url,
-                'id':          video_id,
-                'title':       video_title,
-                # Videos are actually flv not mp4
-                'ext':         'flv',
-                'thumbnail':   video_thumb,
-                'description': video_description,
-                }
+        mgid = self._search_regex([r'data-video="(?P<mgid>mgid:.*?)"',
+                                   r'data-contentId=\'(?P<mgid>mgid:.*?)\''],
+                                  webpage, u'mgid')
+        return self._get_videos_info(mgid)

youtube_dl/extractor/generic.py

@@ -1,3 +1,5 @@
+# encoding: utf-8
+
 import os
 import re
 
@@ -9,20 +11,34 @@ from ..utils import (
 
     ExtractorError,
 )
+from .brightcove import BrightcoveIE
 
 class GenericIE(InfoExtractor):
     IE_DESC = u'Generic downloader that works on some sites'
     _VALID_URL = r'.*'
     IE_NAME = u'generic'
-    _TEST = {
-        u'url': u'http://www.hodiho.fr/2013/02/regis-plante-sa-jeep.html',
-        u'file': u'13601338388002.mp4',
-        u'md5': u'85b90ccc9d73b4acd9138d3af4c27f89',
-        u'info_dict': {
-            u"uploader": u"www.hodiho.fr", 
-            u"title": u"R\u00e9gis plante sa Jeep"
-        }
-    }
+    _TESTS = [
+        {
+            u'url': u'http://www.hodiho.fr/2013/02/regis-plante-sa-jeep.html',
+            u'file': u'13601338388002.mp4',
+            u'md5': u'85b90ccc9d73b4acd9138d3af4c27f89',
+            u'info_dict': {
+                u"uploader": u"www.hodiho.fr", 
+                u"title": u"R\u00e9gis plante sa Jeep"
+            }
+        },
+        {
+            u'url': u'http://www.8tv.cat/8aldia/videos/xavier-sala-i-martin-aquesta-tarda-a-8-al-dia/',
+            u'file': u'2371591881001.mp4',
+            u'md5': u'9e80619e0a94663f0bdc849b4566af19',
+            u'note': u'Test Brightcove downloads and detection in GenericIE',
+            u'info_dict': {
+                u'title': u'Xavier Sala i Martín: “Un banc que no presta és un banc zombi que no serveix per a res”',
+                u'uploader': u'8TV',
+                u'description': u'md5:a950cc4285c43e44d763d036710cd9cd',
+            }
+        },
+    ]
 
     def report_download_webpage(self, video_id):
         """Report webpage download."""
@@ -103,6 +119,13 @@ class GenericIE(InfoExtractor):
             raise ExtractorError(u'Invalid URL: %s' % url)
 
         self.report_extraction(video_id)
+        # Look for BrigthCove:
+        m_brightcove = re.search(r'<object.+?class=([\'"]).*?BrightcoveExperience.*?\1.+?</object>', webpage, re.DOTALL)
+        if m_brightcove is not None:
+            self.to_screen(u'Brightcove video detected.')
+            bc_url = BrightcoveIE._build_brighcove_url(m_brightcove.group())
+            return self.url_result(bc_url, 'Brightcove')
+
         # Start with something easy: JW Player in SWFObject
         mobj = re.search(r'flashvars: [\'"](?:.*&)?file=(http[^\'"&]*)', webpage)
         if mobj is None:

youtube_dl/extractor/hotnewhiphop.py

@@ -33,16 +33,12 @@ class HotNewHipHopIE(InfoExtractor):
 
         video_title = self._html_search_regex(r"<title>(.*)</title>",
             webpage_src, u'title')
-        
-        # Getting thumbnail and if not thumbnail sets correct title for WSHH candy video.
-        thumbnail = self._html_search_regex(r'"og:image" content="(.*)"',
-            webpage_src, u'thumbnail', fatal=False)
 
         results = [{
                     'id': video_id,
                     'url' : video_url,
                     'title' : video_title,
-                    'thumbnail' : thumbnail,
+                    'thumbnail' : self._og_search_thumbnail(webpage_src),
                     'ext' : 'mp3',
                     }]
-        return results
+        return results

youtube_dl/extractor/ign.py

@@ -0,0 +1,91 @@
+import re
+import json
+
+from .common import InfoExtractor
+from ..utils import (
+    determine_ext,
+)
+
+
+class IGNIE(InfoExtractor):
+    """
+    Extractor for some of the IGN sites, like www.ign.com, es.ign.com de.ign.com.
+    Some videos of it.ign.com are also supported
+    """
+
+    _VALID_URL = r'https?://.+?\.ign\.com/(?:videos|show_videos)(/.+)?/(?P<name_or_id>.+)'
+    IE_NAME = u'ign.com'
+
+    _CONFIG_URL_TEMPLATE = 'http://www.ign.com/videos/configs/id/%s.config'
+    _DESCRIPTION_RE = [r'<span class="page-object-description">(.+?)</span>',
+                       r'id="my_show_video">.*?<p>(.*?)</p>',
+                       ]
+
+    _TEST = {
+        u'url': u'http://www.ign.com/videos/2013/06/05/the-last-of-us-review',
+        u'file': u'8f862beef863986b2785559b9e1aa599.mp4',
+        u'md5': u'eac8bdc1890980122c3b66f14bdd02e9',
+        u'info_dict': {
+            u'title': u'The Last of Us Review',
+            u'description': u'md5:c8946d4260a4d43a00d5ae8ed998870c',
+        }
+    }
+
+    def _find_video_id(self, webpage):
+        res_id = [r'data-video-id="(.+?)"',
+                  r'<object id="vid_(.+?)"',
+                  r'<meta name="og:image" content=".*/(.+?)-(.+?)/.+.jpg"',
+                  ]
+        return self._search_regex(res_id, webpage, 'video id')
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        name_or_id = mobj.group('name_or_id')
+        webpage = self._download_webpage(url, name_or_id)
+        video_id = self._find_video_id(webpage)
+        result = self._get_video_info(video_id)
+        description = self._html_search_regex(self._DESCRIPTION_RE,
+                                              webpage, 'video description',
+                                              flags=re.DOTALL)
+        result['description'] = description
+        return result
+
+    def _get_video_info(self, video_id):
+        config_url = self._CONFIG_URL_TEMPLATE % video_id
+        config = json.loads(self._download_webpage(config_url, video_id,
+                            u'Downloading video info'))
+        media = config['playlist']['media']
+        video_url = media['url']
+
+        return {'id': media['metadata']['videoId'],
+                'url': video_url,
+                'ext': determine_ext(video_url),
+                'title': media['metadata']['title'],
+                'thumbnail': media['poster'][0]['url'].replace('{size}', 'grande'),
+                }
+
+
+class OneUPIE(IGNIE):
+    """Extractor for 1up.com, it uses the ign videos system."""
+
+    _VALID_URL = r'https?://gamevideos.1up.com/video/id/(?P<name_or_id>.+)'
+    IE_NAME = '1up.com'
+
+    _DESCRIPTION_RE = r'<div id="vid_summary">(.+?)</div>'
+
+    _TEST = {
+        u'url': u'http://gamevideos.1up.com/video/id/34976',
+        u'file': u'34976.mp4',
+        u'md5': u'68a54ce4ebc772e4b71e3123d413163d',
+        u'info_dict': {
+            u'title': u'Sniper Elite V2 - Trailer',
+            u'description': u'md5:5d289b722f5a6d940ca3136e9dae89cf',
+        }
+    }
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        id = mobj.group('name_or_id')
+        result = super(OneUPIE, self)._real_extract(url)
+        result['id'] = id
+        return result

youtube_dl/extractor/instagram.py

@@ -5,7 +5,7 @@ from .common import InfoExtractor
 class InstagramIE(InfoExtractor):
     _VALID_URL = r'(?:http://)?instagram.com/p/(.*?)/'
     _TEST = {
-        u'url': u'http://instagram.com/p/aye83DjauH/#',
+        u'url': u'http://instagram.com/p/aye83DjauH/?foo=bar#abc',
         u'file': u'aye83DjauH.mp4',
         u'md5': u'0d2da106a9d2631273e192b372806516',
         u'info_dict': {
@@ -18,25 +18,20 @@ class InstagramIE(InfoExtractor):
         mobj = re.match(self._VALID_URL, url)
         video_id = mobj.group(1)
         webpage = self._download_webpage(url, video_id)
-        video_url = self._html_search_regex(
-            r'<meta property="og:video" content="(.+?)"',
-            webpage, u'video URL')
-        thumbnail_url = self._html_search_regex(
-            r'<meta property="og:image" content="(.+?)" />',
-            webpage, u'thumbnail URL', fatal=False)
         html_title = self._html_search_regex(
             r'<title>(.+?)</title>',
             webpage, u'title', flags=re.DOTALL)
         title = re.sub(u'(?: *\(Videos?\))? \u2022 Instagram$', '', html_title).strip()
-        uploader_id = self._html_search_regex(r'content="(.*?)\'s video on Instagram',
-            webpage, u'uploader name', fatal=False)
+        uploader_id = self._html_search_regex(
+            r'<div class="media-user" id="media_user">.*?<h2><a href="[^"]*">([^<]*)</a></h2>',
+            webpage, u'uploader id', fatal=False, flags=re.DOTALL)
         ext = 'mp4'
 
         return [{
             'id':        video_id,
-            'url':       video_url,
+            'url':       self._og_search_video_url(webpage),
             'ext':       ext,
             'title':     title,
-            'thumbnail': thumbnail_url,
+            'thumbnail': self._og_search_thumbnail(webpage),
             'uploader_id' : uploader_id
         }]

youtube_dl/extractor/keek.py

@@ -24,8 +24,7 @@ class KeekIE(InfoExtractor):
         thumbnail = u'http://cdn.keek.com/keek/thumbnail/%s/w100/h75' % video_id
         webpage = self._download_webpage(url, video_id)
 
-        video_title = self._html_search_regex(r'<meta property="og:title" content="(?P<title>.*?)"',
-            webpage, u'title')
+        video_title = self._og_search_title(webpage)
 
         uploader = self._html_search_regex(r'<div class="user-name-and-bio">[\S\s]+?<h2>(?P<uploader>.+?)</h2>',
             webpage, u'uploader', fatal=False)

youtube_dl/extractor/liveleak.py

@@ -33,11 +33,9 @@ class LiveLeakIE(InfoExtractor):
         video_url = self._search_regex(r'file: "(.*?)",',
             webpage, u'video URL')
 
-        video_title = self._html_search_regex(r'<meta property="og:title" content="(?P<title>.*?)"',
-            webpage, u'title').replace('LiveLeak.com -', '').strip()
+        video_title = self._og_search_title(webpage).replace('LiveLeak.com -', '').strip()
 
-        video_description = self._html_search_regex(r'<meta property="og:description" content="(?P<desc>.*?)"',
-            webpage, u'description', fatal=False)
+        video_description = self._og_search_description(webpage)
 
         video_uploader = self._html_search_regex(r'By:.*?(\w+)</a>',
             webpage, u'uploader', fatal=False)

youtube_dl/extractor/livestream.py

@@ -0,0 +1,52 @@
+import re
+import json
+
+from .common import InfoExtractor
+from ..utils import compat_urllib_parse_urlparse, compat_urlparse
+
+
+class LivestreamIE(InfoExtractor):
+    _VALID_URL = r'http://new.livestream.com/.*?/(?P<event_name>.*?)(/videos/(?P<id>\d+))?/?$'
+    _TEST = {
+        u'url': u'http://new.livestream.com/CoheedandCambria/WebsterHall/videos/4719370',
+        u'file': u'4719370.mp4',
+        u'md5': u'0d2186e3187d185a04b3cdd02b828836',
+        u'info_dict': {
+            u'title': u'Live from Webster Hall NYC',
+            u'upload_date': u'20121012',
+        }
+    }
+
+    def _extract_video_info(self, video_data):
+        video_url = video_data.get('progressive_url_hd') or video_data.get('progressive_url')
+        return {'id': video_data['id'],
+                'url': video_url,
+                'ext': 'mp4',
+                'title': video_data['caption'],
+                'thumbnail': video_data['thumbnail_url'],
+                'upload_date': video_data['updated_at'].replace('-','')[:8],
+                }
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        video_id = mobj.group('id')
+        event_name = mobj.group('event_name')
+        webpage = self._download_webpage(url, video_id or event_name)
+
+        if video_id is None:
+            # This is an event page:
+            api_url = self._search_regex(r'event_design_eventId: \'(.+?)\'',
+                                         webpage, 'api url')
+            info = json.loads(self._download_webpage(api_url, event_name,
+                                                     u'Downloading event info'))
+            videos = [self._extract_video_info(video_data['data'])
+                for video_data in info['feed']['data'] if video_data['type'] == u'video']
+            return self.playlist_result(videos, info['id'], info['full_name'])
+        else:
+            og_video = self._og_search_video_url(webpage, name=u'player url')
+            query_str = compat_urllib_parse_urlparse(og_video).query
+            query = compat_urlparse.parse_qs(query_str)
+            api_url = query['play_url'][0].replace('.smil', '')
+            info = json.loads(self._download_webpage(api_url, video_id,
+                                                     u'Downloading video info'))
+            return self._extract_video_info(info)

youtube_dl/extractor/metacafe.py

@@ -9,7 +9,7 @@ from ..utils import (
     compat_urllib_parse,
     compat_urllib_request,
     compat_str,
-
+    determine_ext,
     ExtractorError,
 )
 
@@ -20,7 +20,7 @@ class MetacafeIE(InfoExtractor):
     _DISCLAIMER = 'http://www.metacafe.com/family_filter/'
     _FILTER_POST = 'http://www.metacafe.com/f/index.php?inputType=filter&controllerGroup=user'
     IE_NAME = u'metacafe'
-    _TEST = {
+    _TESTS = [{
         u"add_ie": ["Youtube"],
         u"url":  u"http://metacafe.com/watch/yt-_aUehQsCQtM/the_electric_company_short_i_pbs_kids_go/",
         u"file":  u"_aUehQsCQtM.flv",
@@ -31,7 +31,15 @@ class MetacafeIE(InfoExtractor):
             u"uploader": u"PBS",
             u"uploader_id": u"PBS"
         }
-    }
+    },
+    {
+        u"url": u"http://www.metacafe.com/watch/an-dVVXnuY7Jh77J/the_andromeda_strain_1971_stop_the_bomb_part_3/",
+        u"file": u"an-dVVXnuY7Jh77J.mp4",
+        u"info_dict": {
+            u"title": u"The Andromeda Strain (1971): Stop the Bomb Part 3",
+            u"uploader": u"AnyClip",
+        }
+    }]
 
 
     def report_disclaimer(self):
@@ -73,14 +81,16 @@ class MetacafeIE(InfoExtractor):
             return [self.url_result('http://www.youtube.com/watch?v=%s' % mobj2.group(1), 'Youtube')]
 
         # Retrieve video webpage to extract further information
-        webpage = self._download_webpage('http://www.metacafe.com/watch/%s/' % video_id, video_id)
+        req = compat_urllib_request.Request('http://www.metacafe.com/watch/%s/' % video_id)
+        req.headers['Cookie'] = 'flashVersion=0;'
+        webpage = self._download_webpage(req, video_id)
 
         # Extract URL, uploader and title from webpage
         self.report_extraction(video_id)
         mobj = re.search(r'(?m)&mediaURL=([^&]+)', webpage)
         if mobj is not None:
             mediaURL = compat_urllib_parse.unquote(mobj.group(1))
-            video_extension = mediaURL[-3:]
+            video_ext = mediaURL[-3:]
 
             # Extract gdaKey if available
             mobj = re.search(r'(?m)&gdaKey=(.*?)&', webpage)
@@ -90,34 +100,36 @@ class MetacafeIE(InfoExtractor):
                 gdaKey = mobj.group(1)
                 video_url = '%s?__gda__=%s' % (mediaURL, gdaKey)
         else:
-            mobj = re.search(r' name="flashvars" value="(.*?)"', webpage)
-            if mobj is None:
-                raise ExtractorError(u'Unable to extract media URL')
-            vardict = compat_parse_qs(mobj.group(1))
-            if 'mediaData' not in vardict:
-                raise ExtractorError(u'Unable to extract media URL')
-            mobj = re.search(r'"mediaURL":"(?P<mediaURL>http.*?)",(.*?)"key":"(?P<key>.*?)"', vardict['mediaData'][0])
-            if mobj is None:
-                raise ExtractorError(u'Unable to extract media URL')
-            mediaURL = mobj.group('mediaURL').replace('\\/', '/')
-            video_extension = mediaURL[-3:]
-            video_url = '%s?__gda__=%s' % (mediaURL, mobj.group('key'))
+            mobj = re.search(r'<video src="([^"]+)"', webpage)
+            if mobj:
+                video_url = mobj.group(1)
+                video_ext = 'mp4'
+            else:
+                mobj = re.search(r' name="flashvars" value="(.*?)"', webpage)
+                if mobj is None:
+                    raise ExtractorError(u'Unable to extract media URL')
+                vardict = compat_parse_qs(mobj.group(1))
+                if 'mediaData' not in vardict:
+                    raise ExtractorError(u'Unable to extract media URL')
+                mobj = re.search(r'"mediaURL":"(?P<mediaURL>http.*?)",(.*?)"key":"(?P<key>.*?)"', vardict['mediaData'][0])
+                if mobj is None:
+                    raise ExtractorError(u'Unable to extract media URL')
+                mediaURL = mobj.group('mediaURL').replace('\\/', '/')
+                video_url = '%s?__gda__=%s' % (mediaURL, mobj.group('key'))
+                video_ext = determine_ext(video_url)
 
         mobj = re.search(r'(?im)<title>(.*) - Video</title>', webpage)
         if mobj is None:
             raise ExtractorError(u'Unable to extract title')
         video_title = mobj.group(1).decode('utf-8')
 
-        mobj = re.search(r'submitter=(.*?);', webpage)
-        if mobj is None:
-            raise ExtractorError(u'Unable to extract uploader nickname')
-        video_uploader = mobj.group(1)
+        video_uploader = self._html_search_regex(r'submitter=(.*?);|<p class="By">\s*By\s*<a[^>]*>(.*?)</a>', webpage, u'uploader nickname', fatal=False)
 
         return [{
-            'id':       video_id.decode('utf-8'),
-            'url':      video_url.decode('utf-8'),
-            'uploader': video_uploader.decode('utf-8'),
+            'id':       video_id,
+            'url':      video_url,
+            'uploader': video_uploader,
             'upload_date':  None,
             'title':    video_title,
-            'ext':      video_extension.decode('utf-8'),
+            'ext':      video_ext,
         }]

youtube_dl/extractor/mtv.py

@@ -1,28 +1,110 @@
 import re
-import socket
 import xml.etree.ElementTree
 
 from .common import InfoExtractor
 from ..utils import (
-    compat_http_client,
-    compat_str,
-    compat_urllib_error,
-    compat_urllib_request,
-
+    compat_urllib_parse,
     ExtractorError,
 )
 
+def _media_xml_tag(tag):
+    return '{http://search.yahoo.com/mrss/}%s' % tag
 
 class MTVIE(InfoExtractor):
-    _VALID_URL = r'^(?P<proto>https?://)?(?:www\.)?mtv\.com/videos/[^/]+/(?P<videoid>[0-9]+)/[^/]+$'
-    _WORKING = False
+    _VALID_URL = r'^https?://(?:www\.)?mtv\.com/videos/.+?/(?P<videoid>[0-9]+)/[^/]+$'
+
+    _FEED_URL = 'http://www.mtv.com/player/embed/AS3/rss/'
+
+    _TESTS = [
+        {
+            u'url': u'http://www.mtv.com/videos/misc/853555/ours-vh1-storytellers.jhtml',
+            u'file': u'853555.mp4',
+            u'md5': u'850f3f143316b1e71fa56a4edfd6e0f8',
+            u'info_dict': {
+                u'title': u'Taylor Swift - "Ours (VH1 Storytellers)"',
+                u'description': u'Album: Taylor Swift performs "Ours" for VH1 Storytellers at Harvey Mudd College.',
+            },
+        },
+        {
+            u'url': u'http://www.mtv.com/videos/taylor-swift/916187/everything-has-changed-ft-ed-sheeran.jhtml',
+            u'file': u'USCJY1331283.mp4',
+            u'md5': u'73b4e7fcadd88929292fe52c3ced8caf',
+            u'info_dict': {
+                u'title': u'Everything Has Changed',
+                u'upload_date': u'20130606',
+                u'uploader': u'Taylor Swift',
+            },
+            u'skip': u'VEVO is only available in some countries',
+        },
+    ]
+
+    @staticmethod
+    def _id_from_uri(uri):
+        return uri.split(':')[-1]
+
+    # This was originally implemented for ComedyCentral, but it also works here
+    @staticmethod
+    def _transform_rtmp_url(rtmp_video_url):
+        m = re.match(r'^rtmpe?://.*?/(?P<finalid>gsp\..+?/.*)$', rtmp_video_url)
+        if not m:
+            raise ExtractorError(u'Cannot transform RTMP url')
+        base = 'http://mtvnmobile.vo.llnwd.net/kip0/_pxn=1+_pxI0=Ripod-h264+_pxL0=undefined+_pxM0=+_pxK=18639+_pxE=mp4/44620/mtvnorigin/'
+        return base + m.group('finalid')
+
+    def _get_thumbnail_url(self, uri, itemdoc):
+        return 'http://mtv.mtvnimages.com/uri/' + uri
+
+    def _extract_video_url(self, metadataXml):
+        if '/error_country_block.swf' in metadataXml:
+            raise ExtractorError(u'This video is not available from your country.', expected=True)
+        mdoc = xml.etree.ElementTree.fromstring(metadataXml.encode('utf-8'))
+        renditions = mdoc.findall('.//rendition')
+
+        # For now, always pick the highest quality.
+        rendition = renditions[-1]
+
+        try:
+            _,_,ext = rendition.attrib['type'].partition('/')
+            format = ext + '-' + rendition.attrib['width'] + 'x' + rendition.attrib['height'] + '_' + rendition.attrib['bitrate']
+            rtmp_video_url = rendition.find('./src').text
+        except KeyError:
+            raise ExtractorError('Invalid rendition field.')
+        video_url = self._transform_rtmp_url(rtmp_video_url)
+        return {'ext': ext, 'url': video_url, 'format': format}
+
+    def _get_video_info(self, itemdoc):
+        uri = itemdoc.find('guid').text
+        video_id = self._id_from_uri(uri)
+        self.report_extraction(video_id)
+        mediagen_url = itemdoc.find('%s/%s' % (_media_xml_tag('group'), _media_xml_tag('content'))).attrib['url']
+        if 'acceptMethods' not in mediagen_url:
+            mediagen_url += '&acceptMethods=fms'
+        mediagen_page = self._download_webpage(mediagen_url, video_id,
+                                               u'Downloading video urls')
+        video_info = self._extract_video_url(mediagen_page)
+
+        description_node = itemdoc.find('description')
+        if description_node is not None:
+            description = description_node.text
+        else:
+            description = None
+        video_info.update({'title': itemdoc.find('title').text,
+                           'id': video_id,
+                           'thumbnail': self._get_thumbnail_url(uri, itemdoc),
+                           'description': description,
+                           })
+        return video_info
+
+    def _get_videos_info(self, uri):
+        video_id = self._id_from_uri(uri)
+        data = compat_urllib_parse.urlencode({'uri': uri})
+        infoXml = self._download_webpage(self._FEED_URL +'?' + data, video_id,
+                                         u'Downloading info')
+        idoc = xml.etree.ElementTree.fromstring(infoXml.encode('utf-8'))
+        return [self._get_video_info(item) for item in idoc.findall('.//item')]
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
-        if mobj is None:
-            raise ExtractorError(u'Invalid URL: %s' % url)
-        if not mobj.group('proto'):
-            url = 'http://' + url
         video_id = mobj.group('videoid')
 
         webpage = self._download_webpage(url, video_id)
@@ -35,46 +117,5 @@ class MTVIE(InfoExtractor):
             self.to_screen(u'Vevo video detected: %s' % vevo_id)
             return self.url_result('vevo:%s' % vevo_id, ie='Vevo')
 
-        #song_name = self._html_search_regex(r'<meta name="mtv_vt" content="([^"]+)"/>',
-        #    webpage, u'song name', fatal=False)
-
-        video_title = self._html_search_regex(r'<meta name="mtv_an" content="([^"]+)"/>',
-            webpage, u'title')
-
-        mtvn_uri = self._html_search_regex(r'<meta name="mtvn_uri" content="([^"]+)"/>',
-            webpage, u'mtvn_uri', fatal=False)
-
-        content_id = self._search_regex(r'MTVN.Player.defaultPlaylistId = ([0-9]+);',
-            webpage, u'content id', fatal=False)
-
-        videogen_url = 'http://www.mtv.com/player/includes/mediaGen.jhtml?uri=' + mtvn_uri + '&id=' + content_id + '&vid=' + video_id + '&ref=www.mtvn.com&viewUri=' + mtvn_uri
-        self.report_extraction(video_id)
-        request = compat_urllib_request.Request(videogen_url)
-        try:
-            metadataXml = compat_urllib_request.urlopen(request).read()
-        except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
-            raise ExtractorError(u'Unable to download video metadata: %s' % compat_str(err))
-
-        mdoc = xml.etree.ElementTree.fromstring(metadataXml)
-        renditions = mdoc.findall('.//rendition')
-
-        # For now, always pick the highest quality.
-        rendition = renditions[-1]
-
-        try:
-            _,_,ext = rendition.attrib['type'].partition('/')
-            format = ext + '-' + rendition.attrib['width'] + 'x' + rendition.attrib['height'] + '_' + rendition.attrib['bitrate']
-            video_url = rendition.find('./src').text
-        except KeyError:
-            raise ExtractorError('Invalid rendition field.')
-
-        info = {
-            'id': video_id,
-            'url': video_url,
-            'upload_date': None,
-            'title': video_title,
-            'ext': ext,
-            'format': format,
-        }
-
-        return [info]
+        uri = self._html_search_regex(r'/uri/(.*?)\?', webpage, u'uri')
+        return self._get_videos_info(uri)

youtube_dl/extractor/nba.py

@@ -30,8 +30,7 @@ class NBAIE(InfoExtractor):
         video_url = u'http://ht-mobile.cdn.turner.com/nba/big' + video_id + '_nba_1280x720.mp4'
 
         shortened_video_id = video_id.rpartition('/')[2]
-        title = self._html_search_regex(r'<meta property="og:title" content="(.*?)"',
-            webpage, 'title', default=shortened_video_id).replace('NBA.com: ', '')
+        title = self._og_search_title(webpage, default=shortened_video_id).replace('NBA.com: ', '')
 
         # It isn't there in the HTML it returns to us
         # uploader_date = self._html_search_regex(r'<b>Date:</b> (.*?)</div>', webpage, 'upload_date', fatal=False)

youtube_dl/extractor/statigram.py

@@ -18,12 +18,6 @@ class StatigramIE(InfoExtractor):
         mobj = re.match(self._VALID_URL, url)
         video_id = mobj.group(1)
         webpage = self._download_webpage(url, video_id)
-        video_url = self._html_search_regex(
-            r'<meta property="og:video:secure_url" content="(.+?)">',
-            webpage, u'video URL')
-        thumbnail_url = self._html_search_regex(
-            r'<meta property="og:image" content="(.+?)" />',
-            webpage, u'thumbnail URL', fatal=False)
         html_title = self._html_search_regex(
             r'<title>(.+?)</title>',
             webpage, u'title')
@@ -34,9 +28,9 @@ class StatigramIE(InfoExtractor):
 
         return [{
             'id':        video_id,
-            'url':       video_url,
+            'url':       self._og_search_video_url(webpage),
             'ext':       ext,
             'title':     title,
-            'thumbnail': thumbnail_url,
+            'thumbnail': self._og_search_thumbnail(webpage),
             'uploader_id' : uploader_id
         }]

youtube_dl/extractor/steam.py

@@ -23,14 +23,16 @@ class SteamIE(InfoExtractor):
                 u"file": u"81300.flv",
                 u"md5": u"f870007cee7065d7c76b88f0a45ecc07",
                 u"info_dict": {
-                        u"title": u"Terraria 1.1 Trailer"
+                        u"title": u"Terraria 1.1 Trailer",
+                        u'playlist_index': 1,
                 }
             },
             {
                 u"file": u"80859.flv",
                 u"md5": u"61aaf31a5c5c3041afb58fb83cbb5751",
                 u"info_dict": {
-                    u"title": u"Terraria Trailer"
+                    u"title": u"Terraria Trailer",
+                    u'playlist_index': 2,
                 }
             }
         ]

youtube_dl/extractor/teamcoco.py

@@ -30,15 +30,6 @@ class TeamcocoIE(InfoExtractor):
 
         self.report_extraction(video_id)
 
-        video_title = self._html_search_regex(r'<meta property="og:title" content="(.+?)"',
-            webpage, u'title')
-
-        thumbnail = self._html_search_regex(r'<meta property="og:image" content="(.+?)"',
-            webpage, u'thumbnail', fatal=False)
-
-        video_description = self._html_search_regex(r'<meta property="og:description" content="(.*?)"',
-            webpage, u'description', fatal=False)
-
         data_url = 'http://teamcoco.com/cvp/2.0/%s.xml' % video_id
         data = self._download_webpage(data_url, video_id, 'Downloading data webpage')
 
@@ -49,7 +40,7 @@ class TeamcocoIE(InfoExtractor):
             'id':          video_id,
             'url':         video_url,
             'ext':         'mp4',
-            'title':       video_title,
-            'thumbnail':   thumbnail,
-            'description': video_description,
+            'title':       self._og_search_title(webpage),
+            'thumbnail':   self._og_search_thumbnail(webpage),
+            'description': self._og_search_description(webpage),
         }]

youtube_dl/extractor/traileraddict.py

@@ -24,11 +24,8 @@ class TrailerAddictIE(InfoExtractor):
                 webpage, 'video title').replace(' - Trailer Addict','')
         view_count = self._search_regex(r'Views: (.+?)<br />',
                 webpage, 'Views Count')
-        description = self._search_regex(r'<meta property="og:description" content="(.+?)" />',
-                webpage, 'video description')
-        video_id = self._search_regex(r'<meta property="og:video" content="(.+?)" />',
-                webpage, 'Video id').split('=')[1]
-        
+        video_id = self._og_search_property('video', webpage, 'Video id').split('=')[1]
+
         info_url = "http://www.traileraddict.com/fvar.php?tid=%s" %(str(video_id))
         info_webpage = self._download_webpage(info_url, video_id , "Downloading the info webpage")
         
@@ -44,6 +41,6 @@ class TrailerAddictIE(InfoExtractor):
             'ext'         : ext,
             'title'       : title,
             'thumbnail'   : thumbnail_url,
-            'description' : description,
+            'description' : self._og_search_description(webpage),
             'view_count'  : view_count,
         }]

youtube_dl/extractor/tutv.py

@@ -22,8 +22,6 @@ class TutvIE(InfoExtractor):
         video_id = mobj.group('id')
 
         webpage = self._download_webpage(url, video_id)
-        title = self._html_search_regex(
-            r'<meta property="og:title" content="(.*?)">', webpage, u'title')
         internal_id = self._search_regex(r'codVideo=([0-9]+)', webpage, u'internal video ID')
 
         data_url = u'http://tu.tv/flvurl.php?codVideo=' + str(internal_id)
@@ -36,6 +34,6 @@ class TutvIE(InfoExtractor):
             'id': internal_id,
             'url': video_url,
             'ext': ext,
-            'title': title,
+            'title': self._og_search_title(webpage),
         }
         return [info]

youtube_dl/extractor/vine.py

@@ -27,12 +27,6 @@ class VineIE(InfoExtractor):
         video_url = self._html_search_regex(r'<meta property="twitter:player:stream" content="(.+?)"',
             webpage, u'video URL')
 
-        video_title = self._html_search_regex(r'<meta property="og:title" content="(.+?)"',
-            webpage, u'title')
-
-        thumbnail = self._html_search_regex(r'<meta property="og:image" content="(.+?)(\?.*?)?"',
-            webpage, u'thumbnail', fatal=False)
-
         uploader = self._html_search_regex(r'<div class="user">.*?<h2>(.+?)</h2>',
             webpage, u'uploader', fatal=False, flags=re.DOTALL)
 
@@ -40,7 +34,7 @@ class VineIE(InfoExtractor):
             'id':        video_id,
             'url':       video_url,
             'ext':       'mp4',
-            'title':     video_title,
-            'thumbnail': thumbnail,
+            'title':     self._og_search_title(webpage),
+            'thumbnail': self._og_search_thumbnail(webpage),
             'uploader':  uploader,
         }]

youtube_dl/extractor/youjizz.py

@@ -40,8 +40,20 @@ class YouJizzIE(InfoExtractor):
         webpage = self._download_webpage(embed_page_url, video_id)
 
         # Get the video URL
-        video_url = self._search_regex(r'so.addVariable\("file",encodeURIComponent\("(?P<source>[^"]+)"\)\);',
-            webpage, u'video URL')
+        m_playlist = re.search(r'so.addVariable\("playlist", ?"(?P<playlist>.+?)"\);', webpage)
+        if m_playlist is not None:
+            playlist_url = m_playlist.group('playlist')
+            playlist_page = self._download_webpage(playlist_url, video_id,
+                                                   u'Downloading playlist page')
+            m_levels = list(re.finditer(r'<level bitrate="(\d+?)" file="(.*?)"', playlist_page))
+            if len(m_levels) == 0:
+                raise ExtractorError(u'Unable to extract video url')
+            videos = [(int(m.group(1)), m.group(2)) for m in m_levels]
+            (_, video_url) = sorted(videos)[0]
+            video_url = video_url.replace('%252F', '%2F')
+        else:
+            video_url = self._search_regex(r'so.addVariable\("file",encodeURIComponent\("(?P<source>[^"]+)"\)\);',
+                                           webpage, u'video URL')
 
         info = {'id': video_id,
                 'url': video_url,

youtube_dl/extractor/youtube.py

@@ -190,9 +190,11 @@ class YoutubeIE(InfoExtractor):
         elif len(s) == 84:
             return s[83:36:-1] + s[2] + s[35:26:-1] + s[3] + s[25:3:-1] + s[26]
         elif len(s) == 83:
-            return s[:81]
+            return s[6] + s[3:6] + s[33] + s[7:24] + s[0] + s[25:33] + s[53] + s[34:53] + s[24] + s[54:]
         elif len(s) == 82:
             return s[36] + s[79:67:-1] + s[81] + s[66:40:-1] + s[33] + s[39:36:-1] + s[40] + s[35] + s[0] + s[67] + s[32:0:-1] + s[34]
+        elif len(s) == 81:
+            return s[6] + s[3:6] + s[33] + s[7:24] + s[0] + s[25:33] + s[2] + s[34:53] + s[24] + s[54:81]
 
         else:
             raise ExtractorError(u'Unable to decrypt signature, key length %d not supported; retrying might work' % (len(s)))
@@ -567,6 +569,8 @@ class YoutubeIE(InfoExtractor):
             self.report_rtmp_download()
             video_url_list = [(None, video_info['conn'][0])]
         elif 'url_encoded_fmt_stream_map' in video_info and len(video_info['url_encoded_fmt_stream_map']) >= 1:
+            if 'rtmpe%3Dyes' in video_info['url_encoded_fmt_stream_map'][0]:
+                raise ExtractorError('rtmpe downloads are not supported, see https://github.com/rg3/youtube-dl/issues/343 for more information.', expected=True)
             url_map = {}
             for url_data_str in video_info['url_encoded_fmt_stream_map'][0].split(','):
                 url_data = compat_parse_qs(url_data_str)

youtube_dl/utils.py

@@ -35,6 +35,11 @@ try:
 except ImportError: # Python 2
     from urlparse import urlparse as compat_urllib_parse_urlparse
 
+try:
+    import urllib.parse as compat_urlparse
+except ImportError: # Python 2
+    import urlparse as compat_urlparse
+
 try:
     import http.cookiejar as compat_cookiejar
 except ImportError: # Python 2
@@ -198,6 +203,20 @@ else:
         with open(fn, 'w', encoding='utf-8') as f:
             json.dump(obj, f)
 
+if sys.version_info >= (2,7):
+    def find_xpath_attr(node, xpath, key, val):
+        """ Find the xpath xpath[@key=val] """
+        assert re.match(r'^[a-zA-Z]+$', key)
+        assert re.match(r'^[a-zA-Z@]*$', val)
+        expr = xpath + u"[@%s='%s']" % (key, val)
+        return node.find(expr)
+else:
+    def find_xpath_attr(node, xpath, key, val):
+        for f in node.findall(xpath):
+            if f.attrib.get(key) == val:
+                return f
+        return None
+
 def htmlentity_transform(matchobj):
     """Transforms an HTML entity to a character.
 
@@ -631,12 +650,12 @@ def unified_strdate(date_str):
             pass
     return upload_date
 
-def determine_ext(url):
+def determine_ext(url, default_ext=u'unknown_video'):
     guess = url.partition(u'?')[0].rpartition(u'.')[2]
     if re.match(r'^[A-Za-z0-9]+$', guess):
         return guess
     else:
-        return u'unknown_video'
+        return default_ext
 
 def date_from_str(date_str):
     """

youtube_dl/version.py

@@ -1,2 +1,2 @@
 
-__version__ = '2013.07.10'
+__version__ = '2013.07.17'