release 2013.07.23

Also detect videos that come from Youtube

.travis.yml

@@ -9,6 +9,7 @@ notifications:
     - filippo.valsorda@gmail.com
     - phihag@phihag.de
     - jaime.marquinez.ferrandiz+travis@gmail.com
+    - yasoob.khld@gmail.com
 #  irc:
 #    channels:
 #      - "irc.freenode.org#youtube-dl"

README.md

@@ -16,7 +16,9 @@ which means you can modify it, redistribute it or use it however you like.
 # OPTIONS
     -h, --help                 print this help text and exit
     --version                  print program version and exit
-    -U, --update               update this program to latest version
+    -U, --update               update this program to latest version. Make sure
+                               that you have sufficient permissions (run with
+                               sudo if needed)
     -i, --ignore-errors        continue on download errors
     --dump-user-agent          display the current browser identification
     --user-agent UA            specify a custom user agent

devscripts/youtube_genalgo.py

@@ -5,6 +5,12 @@
 import sys
 
 tests = [
+    # 92 - vflQw-fB4 2013/07/17
+    ("qwertyuioplkjhgfdsazxcvbnm1234567890QWERTYUIOPLKJHGFDSAZXCVBNM!@#$%^&*()_-+={[]}|:;?/>.<'`~\"",
+     "mrtyuioplkjhgfdsazxcvbnq1234567890QWERTY}IOPLKJHGFDSAZXCVBNM!@#$%^&*()_-+={[]\"|:;"),
+    # 90
+    ("qwertyuioplkjhgfdsazxcvbnm1234567890QWERTYUIOPLKJHGFDSAZXCVBNM!@#$%^&*()_-+={[]}|:;?/>.<'`",
+     "mrtyuioplkjhgfdsazxcvbne1234567890QWER[YUIOPLKJHGFDSAZXCVBNM!@#$%^&*()_-+={`]}|"),
     # 88
     ("qwertyuioplkjhgfdsazxcvbnm1234567890QWERTYUIOPLKJHGFDSAZXCVBNM!@#$%^&*()_-+={[]}|:;?/>.<",
      "J:|}][{=+-_)(*&;%$#@>MNBVCXZASDFGH^KLPOIUYTREWQ0987654321mnbvcxzasdfghrklpoiuytej"),
@@ -14,9 +20,9 @@ tests = [
     # 86 - vfl_ymO4Z 2013/06/27
     ("qwertyuioplkjhgfdsazxcvbnm1234567890QWERTYUIOPLKJHGFDSAZXCVBNM!@#$%^&*()_-+={[|};?/>.<",
      "ertyuioplkjhgfdsazxcvbnm1234567890QWERTYUIOPLKJHGFDSAZXCVBNM!/#$%^&*()_-+={[|};?@"),
-    # 85
+    # 85 - vflSAFCP9 2013/07/19
     ("qwertyuioplkjhgfdsazxcvbnm1234567890QWERTYUIOPLKJHGFDSAZXCVBNM!@#$%^&*()_-+={[};?/>.<",
-     "{>/?;}[.=+-_)(*&^%$#@!MqBVCXZASDFwHJKLPOIUYTREWQ0987654321mnbvcxzasdfghjklpoiuytr"),
+     "ertyuiqplkjhgfdsazx$vbnm1234567890QWERTYUIOPLKJHGFDSAZXCVBNM!@#<%^&*()_-+={[};?/c"),
     # 84
     ("qwertyuioplkjhgfdsazxcvbnm1234567890QWERTYUIOPLKJHGFDSAZXCVBNM!@#$%^&*()_-+={[};?>.<",
      "<.>?;}[{=+-_)(*&^%$#@!MNBVCXZASDFGHJKLPOIUYTREWe098765432rmnbvcxzasdfghjklpoiuyt1"),
@@ -26,6 +32,9 @@ tests = [
     # 82
     ("qwertyuioplkjhgfdsazxcvbnm1234567890QWERTYUIOPLKHGFDSAZXCVBNM!@#$%^&*(-+={[};?/>.<",
      "Q>/?;}[{=+-(*<^%$#@!MNBVCXZASDFGHKLPOIUY8REWT0q&7654321mnbvcxzasdfghjklpoiuytrew9"),
+    # 81
+    ("qwertyuioplkjhgfdsazxcvbnm1234567890QWERTYUIOPLKHGFDSAZXCVBNM!@#$%^&*(-+={[};?/>.",
+     "urty8ioplkjhgfdsazxcvbqm1234567e90QWERTYUIOPLKHGFDSnZXCVBNM!@#$%^&*(-+={[};?/>."),
 ]
 
 def find_matching(wrong, right):

test/test_youtube_sig.py

@@ -13,9 +13,14 @@ from helper import FakeYDL
 sig = YoutubeIE(FakeYDL())._decrypt_signature
 
 class TestYoutubeSig(unittest.TestCase):
-    def test_43_43(self):
-        wrong = '5AEEAE0EC39677BC65FD9021CCD115F1F2DBD5A59E4.C0B243A3E2DED6769199AF3461781E75122AE135135'
-        right = '931EA22157E1871643FA9519676DED253A342B0C.4E95A5DBD2F1F511DCC1209DF56CB77693CE0EAE'
+    def test_92(self):
+        wrong = "F9F9B6E6FD47029957AB911A964CC20D95A181A5D37A2DBEFD67D403DB0E8BE4F4910053E4E8A79.0B70B.0B80B8"
+        right = "69B6E6FD47029957AB911A9F4CC20D95A181A5D3.A2DBEFD67D403DB0E8BE4F4910053E4E8A7980B7"
+        self.assertEqual(sig(wrong), right)
+
+    def test_90(self):
+        wrong = "qwertyuioplkjhgfdsazxcvbnm1234567890QWERTYUIOPLKJHGFDSAZXCVBNM!@#$%^&*()_-+={[]}|:;?/>.<'`"
+        right = "mrtyuioplkjhgfdsazxcvbne1234567890QWER[YUIOPLKJHGFDSAZXCVBNM!@#$%^&*()_-+={`]}|"
         self.assertEqual(sig(wrong), right)
 
     def test_88(self):
@@ -35,7 +40,7 @@ class TestYoutubeSig(unittest.TestCase):
 
     def test_85(self):
         wrong = "qwertyuioplkjhgfdsazxcvbnm1234567890QWERTYUIOPLKJHGFDSAZXCVBNM!@#$%^&*()_-+={[};?/>.<"
-        right = "{>/?;}[.=+-_)(*&^%$#@!MqBVCXZASDFwHJKLPOIUYTREWQ0987654321mnbvcxzasdfghjklpoiuytr"
+        right = "ertyuiqplkjhgfdsazx$vbnm1234567890QWERTYUIOPLKJHGFDSAZXCVBNM!@#<%^&*()_-+={[};?/c"
         self.assertEqual(sig(wrong), right)
 
     def test_84(self):
@@ -53,5 +58,10 @@ 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

@@ -529,10 +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/__init__.py

@@ -129,7 +129,7 @@ def parseOpts(overrideArguments=None):
     general.add_option('-v', '--version',
             action='version', help='print program version and exit')
     general.add_option('-U', '--update',
-            action='store_true', dest='update_self', help='update this program to latest version')
+            action='store_true', dest='update_self', help='update this program to latest version. Make sure that you have sufficient permissions (run with sudo if needed)')
     general.add_option('-i', '--ignore-errors',
             action='store_true', dest='ignoreerrors', help='continue on download errors', default=False)
     general.add_option('--dump-user-agent',
@@ -580,7 +580,7 @@ def _real_main(argv=None):
         })
 
     if opts.verbose:
-        ydl.to_screen(u'[debug] youtube-dl version ' + __version__)
+        sys.stderr.write(u'[debug] youtube-dl version ' + __version__ + u'\n')
         try:
             sp = subprocess.Popen(
                 ['git', 'rev-parse', '--short', 'HEAD'],
@@ -589,11 +589,14 @@ def _real_main(argv=None):
             out, err = sp.communicate()
             out = out.decode().strip()
             if re.match('[0-9a-f]+', out):
-                ydl.to_screen(u'[debug] Git HEAD: ' + out)
+                sys.stderr.write(u'[debug] Git HEAD: ' + out + u'\n')
         except:
-            sys.exc_clear()
-        ydl.to_screen(u'[debug] Python version %s - %s' %(platform.python_version(), platform.platform()))
-        ydl.to_screen(u'[debug] Proxy map: ' + str(proxy_handler.proxies))
+            try:
+                sys.exc_clear()
+            except:
+                pass
+        sys.stderr.write(u'[debug] Python version %s - %s' %(platform.python_version(), platform.platform()) + u'\n')
+        sys.stderr.write(u'[debug] Proxy map: ' + str(proxy_handler.proxies) + u'\n')
 
     ydl.add_default_info_extractors()
 

youtube_dl/extractor/__init__.py

@@ -6,8 +6,11 @@ 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 .condenast import CondeNastIE
+from .criterion import CriterionIE
 from .cspan import CSpanIE
 from .dailymotion import DailymotionIE
 from .depositfiles import DepositFilesIE
@@ -16,8 +19,10 @@ from .dreisat import DreiSatIE
 from .ehow import EHowIE
 from .eighttracks import EightTracksIE
 from .escapist import EscapistIE
+from .exfm import ExfmIE
 from .facebook import FacebookIE
 from .flickr import FlickrIE
+from .freesound import FreesoundIE
 from .funnyordie import FunnyOrDieIE
 from .gamespot import GameSpotIE
 from .gametrailers import GametrailersIE
@@ -27,6 +32,7 @@ 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
@@ -34,6 +40,7 @@ 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
@@ -45,6 +52,7 @@ from .pornotube import PornotubeIE
 from .rbmaradio import RBMARadioIE
 from .redtube import RedTubeIE
 from .ringtv import RingTVIE
+from .sina import SinaIE
 from .soundcloud import SoundcloudIE, SoundcloudSetIE
 from .spiegel import SpiegelIE
 from .stanfordoc import StanfordOpenClassroomIE
@@ -53,6 +61,7 @@ from .steam import SteamIE
 from .teamcoco import TeamcocoIE
 from .ted import TEDIE
 from .tf1 import TF1IE
+from .thisav import ThisAVIE
 from .traileraddict import TrailerAddictIE
 from .tudou import TudouIE
 from .tumblr import TumblrIE
@@ -63,7 +72,9 @@ from .veoh import VeohIE
 from .vevo import VevoIE
 from .vimeo import VimeoIE
 from .vine import VineIE
+from .c56 import C56IE
 from .wat import WatIE
+from .weibo import WeiboIE
 from .wimp import WimpIE
 from .worldstarhiphop import WorldStarHipHopIE
 from .xhamster import XHamsterIE
@@ -81,6 +92,7 @@ from .youtube import (
     YoutubeChannelIE,
     YoutubeShowIE,
     YoutubeSubscriptionsIE,
+    YoutubeRecommendedIE,
 )
 from .zdf import ZDFIE
 

youtube_dl/extractor/breakcom.py

@@ -1,6 +1,8 @@
 import re
+import json
 
 from .common import InfoExtractor
+from ..utils import determine_ext
 
 
 class BreakIE(InfoExtractor):
@@ -17,17 +19,20 @@ class BreakIE(InfoExtractor):
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
         video_id = mobj.group(1).split("-")[-1]
-        webpage = self._download_webpage(url, video_id)
-        video_url = re.search(r"videoPath: '(.+?)',",webpage).group(1)
-        key = re.search(r"icon: '(.+?)',",webpage).group(1)
-        final_url = str(video_url)+"?"+str(key)
-        thumbnail_url = re.search(r"thumbnailURL: '(.+?)'",webpage).group(1)
-        title = re.search(r"sVidTitle: '(.+)',",webpage).group(1)
-        ext = video_url.split('.')[-1]
+        embed_url = 'http://www.break.com/embed/%s' % video_id
+        webpage = self._download_webpage(embed_url, video_id)
+        info_json = self._search_regex(r'var embedVars = ({.*?});', webpage,
+                                       u'info json', flags=re.DOTALL)
+        info = json.loads(info_json)
+        video_url = info['videoUri']
+        m_youtube = re.search(r'(https?://www\.youtube\.com/watch\?v=.*)', video_url)
+        if m_youtube is not None:
+            return self.url_result(m_youtube.group(1), 'Youtube')
+        final_url = video_url + '?' + info['AuthToken']
         return [{
             'id':        video_id,
             'url':       final_url,
-            'ext':       ext,
-            'title':     title,
-            'thumbnail': thumbnail_url,
+            'ext':       determine_ext(final_url),
+            'title':     info['contentName'],
+            'thumbnail': info['thumbUri'],
         }]

youtube_dl/extractor/brightcove.py

@@ -6,6 +6,7 @@ from .common import InfoExtractor
 from ..utils import (
     compat_urllib_parse,
     find_xpath_attr,
+    compat_urlparse,
 )
 
 class BrightcoveIE(InfoExtractor):
@@ -39,15 +40,15 @@ class BrightcoveIE(InfoExtractor):
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
-        query = mobj.group('query')
+        query_str = mobj.group('query')
+        query = compat_urlparse.parse_qs(query_str)
 
-        m_video_id = re.search(r'videoPlayer=(\d+)', query)
-        if m_video_id is not None:
-            video_id = m_video_id.group(1)
-            return self._get_video_info(video_id, query)
+        videoPlayer = query.get('@videoPlayer')
+        if videoPlayer:
+            return self._get_video_info(videoPlayer[0], query_str)
         else:
-            player_key = self._search_regex(r'playerKey=(.+?)(&|$)', query, 'playlist_id')
-            return self._get_playlist_info(player_key)
+            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

youtube_dl/extractor/c56.py

@@ -0,0 +1,36 @@
+# coding: utf-8
+
+import re
+import json
+
+from .common import InfoExtractor
+from ..utils import determine_ext
+
+class C56IE(InfoExtractor):
+    _VALID_URL = r'https?://((www|player)\.)?56\.com/(.+?/)?(v_|(play_album.+-))(?P<textid>.+?)\.(html|swf)'
+    IE_NAME = u'56.com'
+
+    _TEST ={
+        u'url': u'http://www.56.com/u39/v_OTM0NDA3MTY.html',
+        u'file': u'93440716.mp4',
+        u'md5': u'9dc07b5c8e978112a6441f9e75d2b59e',
+        u'info_dict': {
+            u'title': u'网事知多少 第32期：车怒',
+        },
+    }
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url, flags=re.VERBOSE)
+        text_id = mobj.group('textid')
+        info_page = self._download_webpage('http://vxml.56.com/json/%s/' % text_id,
+                                           text_id, u'Downloading video info')
+        info = json.loads(info_page)['info']
+        best_format = sorted(info['rfiles'], key=lambda f: int(f['filesize']))[-1]
+        video_url = best_format['url']
+
+        return {'id': info['vid'],
+                'title': info['Subject'],
+                'url': video_url,
+                'ext': determine_ext(video_url),
+                'thumbnail': info.get('bimg') or info.get('img'),
+                }

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/collegehumor.py

@@ -1,26 +1,26 @@
 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_parse_urlparse,
-    compat_urllib_request,
 
     ExtractorError,
 )
 
 
 class CollegeHumorIE(InfoExtractor):
-    _WORKING = False
     _VALID_URL = r'^(?:https?://)?(?:www\.)?collegehumor\.com/video/(?P<videoid>[0-9]+)/(?P<shorttitle>.*)$'
 
-    def report_manifest(self, video_id):
-        """Report information extraction."""
-        self.to_screen(u'%s: Downloading XML manifest' % video_id)
+    _TEST = {
+        u'url': u'http://www.collegehumor.com/video/6902724/comic-con-cosplay-catastrophe',
+        u'file': u'6902724.mp4',
+        u'md5': u'1264c12ad95dca142a9f0bf7968105a0',
+        u'info_dict': {
+            u'title': u'Comic-Con Cosplay Catastrophe',
+            u'description': u'Fans get creative this year at San Diego.  Too creative.  And yes, that\'s really Joss Whedon.',
+        },
+    }
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
@@ -36,10 +36,9 @@ class CollegeHumorIE(InfoExtractor):
 
         self.report_extraction(video_id)
         xmlUrl = 'http://www.collegehumor.com/moogaloop/video/' + video_id
-        try:
-            metaXml = compat_urllib_request.urlopen(xmlUrl).read()
-        except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
-            raise ExtractorError(u'Unable to download video info XML: %s' % compat_str(err))
+        metaXml = self._download_webpage(xmlUrl, video_id,
+                                         u'Downloading info XML',
+                                         u'Unable to download video info XML')
 
         mdoc = xml.etree.ElementTree.fromstring(metaXml)
         try:
@@ -52,11 +51,9 @@ class CollegeHumorIE(InfoExtractor):
             raise ExtractorError(u'Invalid metadata XML file')
 
         manifest_url += '?hdcore=2.10.3'
-        self.report_manifest(video_id)
-        try:
-            manifestXml = compat_urllib_request.urlopen(manifest_url).read()
-        except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
-            raise ExtractorError(u'Unable to download video info XML: %s' % compat_str(err))
+        manifestXml = self._download_webpage(manifest_url, video_id,
+                                             u'Downloading XML manifest',
+                                             u'Unable to download video info XML')
 
         adoc = xml.etree.ElementTree.fromstring(manifestXml)
         try:
@@ -66,9 +63,8 @@ class CollegeHumorIE(InfoExtractor):
         except IndexError as err:
             raise ExtractorError(u'Invalid manifest file')
 
-        url_pr = compat_urllib_parse_urlparse(manifest_url)
-        url = url_pr.scheme + '://' + url_pr.netloc + '/z' + video_id[:-2] + '/' + node_id + 'Seg1-Frag1'
+        url_pr = compat_urllib_parse_urlparse(info['thumbnail'])
 
-        info['url'] = url
-        info['ext'] = 'f4f'
+        info['url'] = url_pr.scheme + '://' + url_pr.netloc + video_id[:-2].replace('.csmil','').replace(',','')
+        info['ext'] = 'mp4'
         return [info]

youtube_dl/extractor/comedycentral.py

@@ -24,7 +24,9 @@ class ComedyCentralIE(InfoExtractor):
                          (full-episodes/(?P<episode>.*)|
                           (?P<clip>
                               (the-colbert-report-(videos|collections)/(?P<clipID>[0-9]+)/[^/]*/(?P<cntitle>.*?))
-                              |(watch/(?P<date>[^/]*)/(?P<tdstitle>.*)))))
+                              |(watch/(?P<date>[^/]*)/(?P<tdstitle>.*)))|
+                          (?P<interview>
+                              extended-interviews/(?P<interID>[0-9]+)/playlist_tds_extended_(?P<interview_title>.*?)/.*?)))
                      $"""
     _TEST = {
         u'url': u'http://www.thedailyshow.com/watch/thu-december-13-2012/kristen-stewart',
@@ -87,6 +89,9 @@ class ComedyCentralIE(InfoExtractor):
             else:
                 epTitle = mobj.group('cntitle')
             dlNewest = False
+        elif mobj.group('interview'):
+            epTitle = mobj.group('interview_title')
+            dlNewest = False
         else:
             dlNewest = not mobj.group('episode')
             if dlNewest:

youtube_dl/extractor/common.py

@@ -14,6 +14,7 @@ from ..utils import (
     clean_html,
     compiled_regex_type,
     ExtractorError,
+    unescapeHTML,
 )
 
 class InfoExtractor(object):
@@ -125,6 +126,11 @@ 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)
@@ -257,6 +263,31 @@ 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
+        escaped = self._search_regex(self._og_regex(prop), html, name, flags=re.DOTALL, **kargs)
+        return unescapeHTML(escaped)
+
+    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/condenast.py

@@ -0,0 +1,106 @@
+# coding: utf-8
+
+import re
+import json
+
+from .common import InfoExtractor
+from ..utils import (
+    compat_urllib_parse,
+    orderedSet,
+    compat_urllib_parse_urlparse,
+    compat_urlparse,
+)
+
+
+class CondeNastIE(InfoExtractor):
+    """
+    Condé Nast is a media group, some of its sites use a custom HTML5 player
+    that works the same in all of them.
+    """
+
+    # The keys are the supported sites and the values are the name to be shown
+    # to the user and in the extractor description.
+    _SITES = {'wired': u'WIRED',
+              'gq': u'GQ',
+              'vogue': u'Vogue',
+              'glamour': u'Glamour',
+              'wmagazine': u'W Magazine',
+              'vanityfair': u'Vanity Fair',
+              }
+
+    _VALID_URL = r'http://(video|www).(?P<site>%s).com/(?P<type>watch|series|video)/(?P<id>.+)' % '|'.join(_SITES.keys())
+    IE_DESC = u'Condé Nast media group: %s' % ', '.join(sorted(_SITES.values()))
+
+    _TEST = {
+        u'url': u'http://video.wired.com/watch/3d-printed-speakers-lit-with-led',
+        u'file': u'5171b343c2b4c00dd0c1ccb3.mp4',
+        u'md5': u'1921f713ed48aabd715691f774c451f7',
+        u'info_dict': {
+            u'title': u'3D Printed Speakers Lit With LED',
+            u'description': u'Check out these beautiful 3D printed LED speakers.  You can\'t actually buy them, but LumiGeek is working on a board that will let you make you\'re own.',
+        }
+    }
+
+    def _extract_series(self, url, webpage):
+        title = self._html_search_regex(r'<div class="cne-series-info">.*?<h1>(.+?)</h1>',
+                                        webpage, u'series title', flags=re.DOTALL)
+        url_object = compat_urllib_parse_urlparse(url)
+        base_url = '%s://%s' % (url_object.scheme, url_object.netloc)
+        m_paths = re.finditer(r'<p class="cne-thumb-title">.*?<a href="(/watch/.+?)["\?]',
+                              webpage, flags=re.DOTALL)
+        paths = orderedSet(m.group(1) for m in m_paths)
+        build_url = lambda path: compat_urlparse.urljoin(base_url, path)
+        entries = [self.url_result(build_url(path), 'CondeNast') for path in paths]
+        return self.playlist_result(entries, playlist_title=title)
+
+    def _extract_video(self, webpage):
+        description = self._html_search_regex([r'<div class="cne-video-description">(.+?)</div>',
+                                               r'<div class="video-post-content">(.+?)</div>',
+                                               ],
+                                              webpage, u'description',
+                                              fatal=False, flags=re.DOTALL)
+        params = self._search_regex(r'var params = {(.+?)}[;,]', webpage,
+                                    u'player params', flags=re.DOTALL)
+        video_id = self._search_regex(r'videoId: [\'"](.+?)[\'"]', params, u'video id')
+        player_id = self._search_regex(r'playerId: [\'"](.+?)[\'"]', params, u'player id')
+        target = self._search_regex(r'target: [\'"](.+?)[\'"]', params, u'target')
+        data = compat_urllib_parse.urlencode({'videoId': video_id,
+                                              'playerId': player_id,
+                                              'target': target,
+                                              })
+        base_info_url = self._search_regex(r'url = [\'"](.+?)[\'"][,;]',
+                                           webpage, u'base info url',
+                                           default='http://player.cnevids.com/player/loader.js?')
+        info_url = base_info_url + data
+        info_page = self._download_webpage(info_url, video_id,
+                                           u'Downloading video info')
+        video_info = self._search_regex(r'var video = ({.+?});', info_page, u'video info')
+        video_info = json.loads(video_info)
+
+        def _formats_sort_key(f):
+            type_ord = 1 if f['type'] == 'video/mp4' else 0
+            quality_ord = 1 if f['quality'] == 'high' else 0
+            return (quality_ord, type_ord)
+        best_format = sorted(video_info['sources'][0], key=_formats_sort_key)[-1]
+
+        return {'id': video_id,
+                'url': best_format['src'],
+                'ext': best_format['type'].split('/')[-1],
+                'title': video_info['title'],
+                'thumbnail': video_info['poster_frame'],
+                'description': description,
+                }
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        site = mobj.group('site')
+        url_type = mobj.group('type')
+        id = mobj.group('id')
+
+        self.to_screen(u'Extracting from %s with the Condé Nast extractor' % self._SITES[site])
+        webpage = self._download_webpage(url, id)
+
+        if url_type == 'series':
+            return self._extract_series(url, webpage)
+        else:
+            return self._extract_video(webpage)

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/ehow.py

@@ -28,14 +28,9 @@ class EHowIE(InfoExtractor):
         video_url = self._search_regex(r'(?:file|source)=(http[^\'"&]*)',
             webpage, u'video URL')
         final_url = compat_urllib_parse.unquote(video_url)        
-        thumbnail_url = self._search_regex(r'<meta property="og:image" content="(.+?)" />',
-            webpage, u'thumbnail URL')
         uploader = self._search_regex(r'<meta name="uploader" content="(.+?)" />',
             webpage, u'uploader')
-        title = self._search_regex(r'<meta property="og:title" content="(.+?)" />',
-            webpage, u'Video title').replace(' | eHow', '')
-        description = self._search_regex(r'<meta property="og:description" content="(.+?)" />',
-            webpage, u'video description')
+        title = self._og_search_title(webpage).replace(' | eHow', '')
         ext = determine_ext(final_url)
 
         return {
@@ -44,8 +39,8 @@ class EHowIE(InfoExtractor):
             'url':         final_url,
             'ext':         ext,
             'title':       title,
-            'thumbnail':   thumbnail_url,
-            'description': description,
+            '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/exfm.py

@@ -0,0 +1,42 @@
+import re
+import json
+
+from .common import InfoExtractor
+
+
+class ExfmIE(InfoExtractor):
+    IE_NAME = u'exfm'
+    IE_DESC = u'ex.fm'
+    _VALID_URL = r'(?:http://)?(?:www\.)?ex\.fm/song/([^/]+)'
+    _SOUNDCLOUD_URL_ = r'(?:http://)?(?:www\.)?api\.soundcloud.com/tracks/([^/]+)/stream'
+    _TEST = {
+        u'url': u'http://ex.fm/song/1bgtzg',
+        u'file': u'1bgtzg.mp3',
+        u'md5': u'8a7967a3fef10e59a1d6f86240fd41cf',
+        u'info_dict': {
+            u"title": u"We Can't Stop",
+            u"uploader": u"Miley Cyrus",
+            u'thumbnail': u'http://i1.sndcdn.com/artworks-000049666230-w9i7ef-t500x500.jpg?9d68d37'
+        }
+    }
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        song_id = mobj.group(1)
+        info_url = "http://ex.fm/api/v3/song/%s" %(song_id)
+        webpage = self._download_webpage(info_url, song_id)
+        info = json.loads(webpage)
+        song_url = re.match(self._SOUNDCLOUD_URL_,info['song']['url'])
+        if song_url is not None:
+        	song_url = song_url.group() + "?client_id=b45b1aa10f1ac2941910a7f0d10f8e28"
+        else:
+        	song_url = info['song']['url']
+        return [{
+            'id':          song_id,
+            'url':         song_url,
+            'ext':         'mp3',
+            'title':       info['song']['title'],
+            'thumbnail':   info['song']['image']['large'],
+            'uploader':    info['song']['artist'],
+            'view_count':  info['song']['loved_count'],
+        }]

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

@@ -1,63 +1,36 @@
 import re
-import xml.etree.ElementTree
 
-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'70e9a5d7-cf25-4a10-9104-6f3e7342ae0d.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')
         webpage = self._download_webpage(url, video_id)
         mgid = self._search_regex([r'data-video="(?P<mgid>mgid:.*?)"',
                                    r'data-contentId=\'(?P<mgid>mgid:.*?)\''],
                                   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')
-        doc = xml.etree.ElementTree.fromstring(info_page.encode('utf-8'))
-        default_thumb = doc.find('./channel/image/url').text
-
-        media_namespace = {'media': 'http://search.yahoo.com/mrss/'}
-        parts = [{
-            'title': video_doc.find('title').text,
-            'ext': 'flv',
-            'id': video_doc.find('guid').text.rpartition(':')[2],
-            # Videos are actually flv not mp4
-            'url': self._get_video_url(video_doc.find('media:group/media:content', media_namespace).attrib['url'], video_id),
-            # The thumbnail may not be defined, it would be ''
-            'thumbnail': video_doc.find('media:group/media:thumbnail', media_namespace).attrib['url'] or default_thumb,
-            'description': video_doc.find('description').text,
-        } for video_doc in doc.findall('./channel/item')]
-        return parts
-
-    def _get_video_url(self, mediagen_url, video_id):
-        if 'acceptMethods' not in mediagen_url:
-            mediagen_url += '&acceptMethods=fms'
-        links_webpage = self._download_webpage(mediagen_url,
-                                               video_id, u'Downloading video urls info')
-        doc = xml.etree.ElementTree.fromstring(links_webpage)
-        urls = list(doc.iter('src'))
-        if len(urls) == 0:
-            raise ExtractorError(u'Unable to extract video url')
-        # They are sorted from worst to best quality
-        return urls[-1].text
-
+        return self._get_videos_info(mgid)

youtube_dl/extractor/generic.py

@@ -120,7 +120,7 @@ class GenericIE(InfoExtractor):
 
         self.report_extraction(video_id)
         # Look for BrigthCove:
-        m_brightcove = re.search(r'<object.+?class=".*?BrightcoveExperience.*?".+?</object>', webpage, re.DOTALL)
+        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())

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

@@ -5,12 +5,13 @@ 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': {
             u"uploader_id": u"naomipq", 
-            u"title": u"Video by naomipq"
+            u"title": u"Video by naomipq",
+            u'description': u'md5:1f17f0ab29bd6fe2bfad705f58de3cb8',
         }
     }
 
@@ -18,25 +19,17 @@ 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)
-        ext = 'mp4'
+        uploader_id = self._search_regex(r'"owner":{"username":"(.+?)"',
+            webpage, u'uploader id', fatal=False)
+        desc = self._search_regex(r'"caption":"(.*?)"', webpage, u'description',
+            fatal=False)
 
         return [{
             'id':        video_id,
-            'url':       video_url,
-            'ext':       ext,
-            'title':     title,
-            'thumbnail': thumbnail_url,
-            'uploader_id' : uploader_id
+            'url':       self._og_search_video_url(webpage),
+            'ext':       'mp4',
+            'title':     u'Video by %s' % uploader_id,
+            'thumbnail': self._og_search_thumbnail(webpage),
+            'uploader_id' : uploader_id,
+            'description': desc,
         }]

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 +31,16 @@ 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",
+            u"description": u"md5:38c711dd98f5bb87acf973d573442e67"
+        }
+    }]
 
 
     def report_disclaimer(self):
@@ -73,14 +82,16 @@ 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 +101,37 @@ 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')
+        video_title = self._html_search_regex(r'(?im)<title>(.*) - Video</title>', webpage, u'title')
+        description = self._og_search_description(webpage)
+        video_uploader = self._html_search_regex(
+                r'submitter=(.*?);|googletag\.pubads\(\)\.setTargeting\("channel","([^"]+)"\);',
+                webpage, u'uploader nickname', fatal=False)
 
-        mobj = re.search(r'submitter=(.*?);', webpage)
-        if mobj is None:
-            raise ExtractorError(u'Unable to extract uploader nickname')
-        video_uploader = mobj.group(1)
-
-        return [{
-            'id':       video_id.decode('utf-8'),
-            'url':      video_url.decode('utf-8'),
-            'uploader': video_uploader.decode('utf-8'),
+        return {
+            '_type':    'video',
+            'id':       video_id,
+            'url':      video_url,
+            'description': description,
+            '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/sina.py

@@ -0,0 +1,67 @@
+# coding: utf-8
+
+import re
+import xml.etree.ElementTree
+
+from .common import InfoExtractor
+from ..utils import (
+    compat_urllib_request,
+    compat_urllib_parse,
+)
+
+
+class SinaIE(InfoExtractor):
+    _VALID_URL = r'''https?://(.*?\.)?video\.sina\.com\.cn/
+                        (
+                            (.+?/(((?P<pseudo_id>\d+).html)|(.*?(\#|(vid=))(?P<id>\d+?)($|&))))
+                            |
+                            # This is used by external sites like Weibo
+                            (api/sinawebApi/outplay.php/(?P<token>.+?)\.swf)
+                        )
+                  '''
+
+    _TEST = {
+        u'url': u'http://video.sina.com.cn/news/vlist/zt/chczlj2013/?opsubject_id=top12#110028898',
+        u'file': u'110028898.flv',
+        u'md5': u'd65dd22ddcf44e38ce2bf58a10c3e71f',
+        u'info_dict': {
+            u'title': u'《中国新闻》 朝鲜要求巴拿马立即释放被扣船员',
+        }
+    }
+
+    @classmethod
+    def suitable(cls, url):
+        return re.match(cls._VALID_URL, url, flags=re.VERBOSE) is not None
+
+    def _extract_video(self, video_id):
+        data = compat_urllib_parse.urlencode({'vid': video_id})
+        url_page = self._download_webpage('http://v.iask.com/v_play.php?%s' % data,
+            video_id, u'Downloading video url')
+        image_page = self._download_webpage(
+            'http://interface.video.sina.com.cn/interface/common/getVideoImage.php?%s' % data,
+            video_id, u'Downloading thumbnail info')
+        url_doc = xml.etree.ElementTree.fromstring(url_page.encode('utf-8'))
+
+        return {'id': video_id,
+                'url': url_doc.find('./durl/url').text,
+                'ext': 'flv',
+                'title': url_doc.find('./vname').text,
+                'thumbnail': image_page.split('=')[1],
+                }
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url, flags=re.VERBOSE)
+        video_id = mobj.group('id')
+        if mobj.group('token') is not None:
+            # The video id is in the redirected url
+            self.to_screen(u'Getting video id')
+            request = compat_urllib_request.Request(url)
+            request.get_method = lambda: 'HEAD'
+            (_, urlh) = self._download_webpage_handle(request, 'NA', False)
+            return self._real_extract(urlh.geturl())
+        elif video_id is None:
+            pseudo_id = mobj.group('pseudo_id')
+            webpage = self._download_webpage(url, pseudo_id)
+            video_id = self._search_regex(r'vid:\'(\d+?)\'', webpage, u'video id')
+
+        return self._extract_video(video_id)

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/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/ted.py

@@ -67,7 +67,7 @@ class TEDIE(InfoExtractor):
         webpage = self._download_webpage(url, video_id, 'Downloading \"%s\" page' % video_name)
         self.report_extraction(video_name)
         # If the url includes the language we get the title translated
-        title = self._html_search_regex(r'<span id="altHeadline" >(?P<title>.*)</span>',
+        title = self._html_search_regex(r'<span .*?id="altHeadline".+?>(?P<title>.*)</span>',
                                         webpage, 'title')
         json_data = self._search_regex(r'<script.*?>var talkDetails = ({.*?})</script>',
                                     webpage, 'json data')

youtube_dl/extractor/thisav.py

@@ -0,0 +1,47 @@
+#coding: utf-8
+
+import re
+
+from .common import InfoExtractor
+from ..utils import (
+    determine_ext,
+)
+
+class ThisAVIE(InfoExtractor):
+    _VALID_URL = r'https?://(?:www\.)?thisav\.com/video/(?P<id>[0-9]+)/.*'
+    _TEST = {
+        u"url": u"http://www.thisav.com/video/47734/%98%26sup1%3B%83%9E%83%82---just-fit.html",
+        u"file": u"47734.flv",
+        u"md5": u"0480f1ef3932d901f0e0e719f188f19b",
+        u"info_dict": {
+            u"title": u"高樹マリア - Just fit",
+            u"uploader": u"dj7970",
+            u"uploader_id": u"dj7970"
+        }
+    }
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+
+        video_id = mobj.group('id')
+        webpage = self._download_webpage(url, video_id)
+        title = self._html_search_regex(r'<h1>([^<]*)</h1>', webpage, u'title')
+        video_url = self._html_search_regex(
+            r"addVariable\('file','([^']+)'\);", webpage, u'video url')
+        uploader = self._html_search_regex(
+            r': <a href="http://www.thisav.com/user/[0-9]+/(?:[^"]+)">([^<]+)</a>',
+            webpage, u'uploader name', fatal=False)
+        uploader_id = self._html_search_regex(
+            r': <a href="http://www.thisav.com/user/[0-9]+/([^"]+)">(?:[^<]+)</a>',
+            webpage, u'uploader id', fatal=False)
+        ext = determine_ext(video_url)
+        
+        return {
+            '_type':       'video',
+            'id':          video_id,
+            'url':         video_url,
+            'uploader':    uploader,
+            'uploader_id': uploader_id,
+            'title':       title,
+            'ext':         ext,
+        }

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/weibo.py

@@ -0,0 +1,48 @@
+# coding: utf-8
+
+import re
+import json
+
+from .common import InfoExtractor
+
+class WeiboIE(InfoExtractor):
+    """
+    The videos in Weibo come from different sites, this IE just finds the link
+    to the external video and returns it.
+    """
+    _VALID_URL = r'https?://video\.weibo\.com/v/weishipin/t_(?P<id>.+?)\.htm'
+
+    _TEST = {
+        u'url': u'http://video.weibo.com/v/weishipin/t_zjUw2kZ.htm',
+        u'file': u'98322879.flv',
+        u'info_dict': {
+            u'title': u'魔声耳机最新广告“All Eyes On Us”',
+        },
+        u'note': u'Sina video',
+        u'params': {
+            u'skip_download': True,
+        },
+    }
+
+    # Additional example videos from different sites
+    # Youku: http://video.weibo.com/v/weishipin/t_zQGDWQ8.htm
+    # 56.com: http://video.weibo.com/v/weishipin/t_zQ44HxN.htm
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url, flags=re.VERBOSE)
+        video_id = mobj.group('id')
+        info_url = 'http://video.weibo.com/?s=v&a=play_list&format=json&mix_video_id=t_%s' % video_id
+        info_page = self._download_webpage(info_url, video_id)
+        info = json.loads(info_page)
+
+        videos_urls = map(lambda v: v['play_page_url'], info['result']['data'])
+        #Prefer sina video since they have thumbnails
+        videos_urls = sorted(videos_urls, key=lambda u: u'video.sina.com' in u)
+        player_url = videos_urls[-1]
+        m_sina = re.match(r'https?://video.sina.com.cn/v/b/(\d+)-\d+.html', player_url)
+        if m_sina is not None:
+            self.to_screen('Sina video detected')
+            sina_id = m_sina.group(1)
+            player_url = 'http://you.video.sina.com.cn/swf/quotePlayer.swf?vid=%s' % sina_id
+        return self.url_result(player_url)
+

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/youku.py

@@ -13,7 +13,7 @@ from ..utils import (
 
 
 class YoukuIE(InfoExtractor):
-    _VALID_URL =  r'(?:http://)?v\.youku\.com/v_show/id_(?P<ID>[A-Za-z0-9]+)\.html'
+    _VALID_URL =  r'(?:http://)?(v|player)\.youku\.com/(v_show/id_|player\.php/sid/)(?P<ID>[A-Za-z0-9]+)(\.html|/v.swf)'
     _TEST =   {
         u"url": u"http://v.youku.com/v_show/id_XNDgyMDQ2NTQw.html",
         u"file": u"XNDgyMDQ2NTQw_part00.flv",

youtube_dl/extractor/youtube.py

@@ -179,20 +179,26 @@ class YoutubeIE(InfoExtractor):
     def _decrypt_signature(self, s):
         """Turn the encrypted s field into a working signature"""
 
-        if len(s) == 88:
+        if len(s) == 92:
+            return s[25] + s[3:25] + s[0] + s[26:42] + s[79] + s[43:79] + s[91] + s[80:83]
+        elif len(s) == 90:
+            return s[25] + s[3:25] + s[2] + s[26:40] + s[77] + s[41:77] + s[89] + s[78:81]
+        elif len(s) == 88:
             return s[48] + s[81:67:-1] + s[82] + s[66:62:-1] + s[85] + s[61:48:-1] + s[67] + s[47:12:-1] + s[3] + s[11:3:-1] + s[2] + s[12]
         elif len(s) == 87:
             return s[62] + s[82:62:-1] + s[83] + s[61:52:-1] + s[0] + s[51:2:-1]
         elif len(s) == 86:
             return s[2:63] + s[82] + s[64:82] + s[63]
         elif len(s) == 85:
-            return s[76] + s[82:76:-1] + s[83] + s[75:60:-1] + s[0] + s[59:50:-1] + s[1] + s[49:2:-1]
+            return s[2:8] + s[0] + s[9:21] + s[65] + s[22:65] + s[84] + s[66:82] + s[21]
         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[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)))
@@ -587,8 +593,9 @@ class YoutubeIE(InfoExtractor):
                             else:
                                 player = u'html5 player %s' % self._search_regex(r'html5player-(.+?)\.js', video_webpage,
                                     'html5 player', fatal=False)
-                            self.to_screen('encrypted signature length %d (%d.%d), itag %s, %s' %
-                                (len(s), len(s.split('.')[0]), len(s.split('.')[1]), url_data['itag'][0], player))
+                            parts_sizes = u'.'.join(compat_str(len(part)) for part in s.split('.'))
+                            self.to_screen(u'encrypted signature length %d (%s), itag %s, %s' %
+                                (len(s), parts_sizes, url_data['itag'][0], player))
                         signature = self._decrypt_signature(url_data['s'][0])
                         url += '&signature=' + signature
                     if 'ratebypass' not in url:
@@ -891,12 +898,12 @@ class YoutubeShowIE(InfoExtractor):
         return [self.url_result('https://www.youtube.com' + season.group(1), 'YoutubePlaylist') for season in m_seasons]
 
 
-class YoutubeSubscriptionsIE(YoutubeIE):
-    """It's a subclass of YoutubeIE because we need to login"""
-    IE_DESC = u'YouTube.com subscriptions feed, "ytsubs" keyword(requires authentication)'
-    _VALID_URL = r'https?://www\.youtube\.com/feed/subscriptions|:ytsubs(?:criptions)?'
-    IE_NAME = u'youtube:subscriptions'
-    _FEED_TEMPLATE = 'http://www.youtube.com/feed_ajax?action_load_system_feed=1&feed_name=subscriptions&paging=%s'
+class YoutubeFeedsInfoExtractor(YoutubeIE):
+    """
+    Base class for extractors that fetch info from
+    http://www.youtube.com/feed_ajax
+    Subclasses must define the _FEED_NAME and _PLAYLIST_TITLE properties.
+    """
     _PAGING_STEP = 30
 
     # Overwrite YoutubeIE properties we don't want
@@ -905,18 +912,27 @@ class YoutubeSubscriptionsIE(YoutubeIE):
     def suitable(cls, url):
         return re.match(cls._VALID_URL, url) is not None
 
+    @property
+    def _FEED_TEMPLATE(self):
+        return 'http://www.youtube.com/feed_ajax?action_load_system_feed=1&feed_name=%s&paging=%%s' % self._FEED_NAME
+
+    @property
+    def IE_NAME(self):
+        return u'youtube:%s' % self._FEED_NAME
+
     def _real_initialize(self):
         (username, password) = self._get_login_info()
         if username is None:
             raise ExtractorError(u'No login info available, needed for downloading the Youtube subscriptions.', expected=True)
-        super(YoutubeSubscriptionsIE, self)._real_initialize()
+        super(YoutubeFeedsInfoExtractor, self)._real_initialize()
 
     def _real_extract(self, url):
         feed_entries = []
         # The step argument is available only in 2.7 or higher
         for i in itertools.count(0):
             paging = i*self._PAGING_STEP
-            info = self._download_webpage(self._FEED_TEMPLATE % paging, 'feed',
+            info = self._download_webpage(self._FEED_TEMPLATE % paging,
+                                          u'%s feed' % self._FEED_NAME,
                                           u'Downloading page %s' % i)
             info = json.loads(info)
             feed_html = info['feed_html']
@@ -925,4 +941,16 @@ class YoutubeSubscriptionsIE(YoutubeIE):
             feed_entries.extend(self.url_result(id, 'Youtube') for id in ids)
             if info['paging'] is None:
                 break
-        return self.playlist_result(feed_entries, playlist_title='Youtube Subscriptions')
+        return self.playlist_result(feed_entries, playlist_title=self._PLAYLIST_TITLE)
+
+class YoutubeSubscriptionsIE(YoutubeFeedsInfoExtractor):
+    IE_DESC = u'YouTube.com subscriptions feed, "ytsubs" keyword(requires authentication)'
+    _VALID_URL = r'https?://www\.youtube\.com/feed/subscriptions|:ytsubs(?:criptions)?'
+    _FEED_NAME = 'subscriptions'
+    _PLAYLIST_TITLE = u'Youtube Subscriptions'
+
+class YoutubeRecommendedIE(YoutubeFeedsInfoExtractor):
+    IE_DESC = u'YouTube.com recommended videos, "ytrec" keyword (requires authentication)'
+    _VALID_URL = r'https?://www\.youtube\.com/feed/recommended|:ytrec(?:ommended)?'
+    _FEED_NAME = 'recommended'
+    _PLAYLIST_TITLE = u'Youtube Recommended videos'

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
@@ -645,12 +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__ = '2013.07.11'
+__version__ = '2013.07.23'