release 2014.04.07

[UstreamIE] [generic] Added support for Ustream embed URLs (Fixes #2694)

test/test_all_urls.py

@@ -157,5 +157,11 @@ class TestAllURLsMatching(unittest.TestCase):
             'http://thedailyshow.cc.com/guests/michael-lewis/3efna8/exclusive---michael-lewis-extended-interview-pt--3',
             ['ComedyCentralShows'])
 
+    def test_yahoo_https(self):
+        # https://github.com/rg3/youtube-dl/issues/2701
+        self.assertMatch(
+            'https://screen.yahoo.com/smartwatches-latest-wearable-gadgets-163745379-cbs.html',
+            ['Yahoo'])
+
 if __name__ == '__main__':
     unittest.main()

test/test_playlists.py

@@ -324,7 +324,6 @@ class TestPlaylists(unittest.TestCase):
         self.assertEqual(result['id'], '342759')
         self.assertEqual(
             result['title'], 'General Motors Ignition Switch Recall')
-        self.assertEqual(len(result['entries']), 9)
         whole_duration = sum(e['duration'] for e in result['entries'])
         self.assertEqual(whole_duration, 14855)
 

test/test_utils.py

@@ -38,6 +38,7 @@ from youtube_dl.utils import (
     xpath_with_ns,
     parse_iso8601,
     strip_jsonp,
+    uppercase_escape,
 )
 
 if sys.version_info < (3, 0):
@@ -279,6 +280,9 @@ class TestUtil(unittest.TestCase):
         d = json.loads(stripped)
         self.assertEqual(d, [{"id": "532cb", "x": 3}])
 
+    def test_uppercase_escpae(self):
+        self.assertEqual(uppercase_escape(u'aä'), u'aä')
+        self.assertEqual(uppercase_escape(u'\\U0001d550'), u'𝕐')
 
 if __name__ == '__main__':
     unittest.main()

youtube_dl/__init__.py

@@ -52,6 +52,7 @@ __authors__  = (
     'Juan C. Olivares',
     'Mattias Harrysson',
     'phaer',
+    'Sainyam Kapoor',
 )
 
 __license__ = 'Public Domain'

youtube_dl/extractor/__init__.py

@@ -32,6 +32,7 @@ from .canal13cl import Canal13clIE
 from .canalplus import CanalplusIE
 from .canalc2 import Canalc2IE
 from .cbs import CBSIE
+from .cbsnews import CBSNewsIE
 from .ceskatelevize import CeskaTelevizeIE
 from .channel9 import Channel9IE
 from .chilloutzone import ChilloutzoneIE
@@ -62,6 +63,7 @@ from .dotsub import DotsubIE
 from .dreisat import DreiSatIE
 from .defense import DefenseGouvFrIE
 from .discovery import DiscoveryIE
+from .divxstage import DivxStageIE
 from .dropbox import DropboxIE
 from .ebaumsworld import EbaumsWorldIE
 from .ehow import EHowIE
@@ -156,6 +158,7 @@ from .mofosex import MofosexIE
 from .mooshare import MooshareIE
 from .morningstar import MorningstarIE
 from .motorsport import MotorsportIE
+from .movshare import MovShareIE
 from .mtv import (
     MTVIE,
     MTVIggyIE,
@@ -276,6 +279,7 @@ from .videodetective import VideoDetectiveIE
 from .videolecturesnet import VideoLecturesNetIE
 from .videofyme import VideofyMeIE
 from .videopremium import VideoPremiumIE
+from .videoweed import VideoWeedIE
 from .vimeo import (
     VimeoIE,
     VimeoChannelIE,

youtube_dl/extractor/cbsnews.py

@@ -0,0 +1,87 @@
+# encoding: utf-8
+from __future__ import unicode_literals
+
+import re
+import json
+
+from .common import InfoExtractor
+
+
+class CBSNewsIE(InfoExtractor):
+    IE_DESC = 'CBS News'
+    _VALID_URL = r'http://(?:www\.)?cbsnews\.com/(?:[^/]+/)+(?P<id>[\da-z_-]+)'
+
+    _TESTS = [
+        {
+            'url': 'http://www.cbsnews.com/news/tesla-and-spacex-elon-musks-industrial-empire/',
+            'info_dict': {
+                'id': 'tesla-and-spacex-elon-musks-industrial-empire',
+                'ext': 'flv',
+                'title': 'Tesla and SpaceX: Elon Musk\'s industrial empire',
+                'thumbnail': 'http://beta.img.cbsnews.com/i/2014/03/30/60147937-2f53-4565-ad64-1bdd6eb64679/60-0330-pelley-640x360.jpg',
+                'duration': 791,
+            },
+            'params': {
+                # rtmp download
+                'skip_download': True,
+            },
+        },
+        {
+            'url': 'http://www.cbsnews.com/videos/fort-hood-shooting-army-downplays-mental-illness-as-cause-of-attack/',
+            'info_dict': {
+                'id': 'fort-hood-shooting-army-downplays-mental-illness-as-cause-of-attack',
+                'ext': 'flv',
+                'title': 'Fort Hood shooting: Army downplays mental illness as cause of attack',
+                'thumbnail': 'http://cbsnews2.cbsistatic.com/hub/i/r/2014/04/04/0c9fbc66-576b-41ca-8069-02d122060dd2/thumbnail/140x90/6dad7a502f88875ceac38202984b6d58/en-0404-werner-replace-640x360.jpg',
+                'duration': 205,
+            },
+            'params': {
+                # rtmp download
+                'skip_download': True,
+            },
+        },
+    ]
+
+    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_info = json.loads(self._html_search_regex(
+            r'(?:<ul class="media-list items" id="media-related-items"><li data-video-info|<div id="cbsNewsVideoPlayer" data-video-player-options)=\'({.+?})\'',
+            webpage, 'video JSON info'))
+
+        item = video_info['item'] if 'item' in video_info else video_info
+        title = item.get('articleTitle') or item.get('hed')
+        duration = item.get('duration')
+        thumbnail = item.get('mediaImage') or item.get('thumbnail')
+
+        formats = []
+        for format_id in ['RtmpMobileLow', 'RtmpMobileHigh', 'Hls', 'RtmpDesktop']:
+            uri = item.get('media' + format_id + 'URI')
+            if not uri:
+                continue
+            fmt = {
+                'url': uri,
+                'format_id': format_id,
+            }
+            if uri.startswith('rtmp'):
+                fmt.update({
+                    'app': 'ondemand?auth=cbs',
+                    'play_path': 'mp4:' + uri.split('<break>')[-1],
+                    'player_url': 'http://www.cbsnews.com/[[IMPORT]]/vidtech.cbsinteractive.com/player/3_3_0/CBSI_PLAYER_HD.swf',
+                    'page_url': 'http://www.cbsnews.com',
+                    'ext': 'flv',
+                })
+            elif uri.endswith('.m3u8'):
+                fmt['ext'] = 'mp4'
+            formats.append(fmt)
+
+        return {
+            'id': video_id,
+            'title': title,
+            'thumbnail': thumbnail,
+            'duration': duration,
+            'formats': formats,
+        }

youtube_dl/extractor/dailymotion.py

@@ -8,7 +8,6 @@ from .subtitles import SubtitlesInfoExtractor
 from ..utils import (
     compat_urllib_request,
     compat_str,
-    get_element_by_attribute,
     get_element_by_id,
     orderedSet,
     str_to_int,

youtube_dl/extractor/divxstage.py

@@ -0,0 +1,27 @@
+from __future__ import unicode_literals
+
+from .novamov import NovaMovIE
+
+
+class DivxStageIE(NovaMovIE):
+    IE_NAME = 'divxstage'
+    IE_DESC = 'DivxStage'
+
+    _VALID_URL = NovaMovIE._VALID_URL_TEMPLATE % {'host': 'divxstage\.(?:eu|net|ch|co|at|ag)'}
+
+    _HOST = 'www.divxstage.eu'
+
+    _FILE_DELETED_REGEX = r'>This file no longer exists on our servers.<'
+    _TITLE_REGEX = r'<div class="video_det">\s*<strong>([^<]+)</strong>'
+    _DESCRIPTION_REGEX = r'<div class="video_det">\s*<strong>[^<]+</strong>\s*<p>([^<]+)</p>'
+
+    _TEST = {
+        'url': 'http://www.divxstage.eu/video/57f238e2e5e01',
+        'md5': '63969f6eb26533a1968c4d325be63e72',
+        'info_dict': {
+            'id': '57f238e2e5e01',
+            'ext': 'flv',
+            'title': 'youtubedl test video',
+            'description': 'This is a test video for youtubedl.',
+        }
+    }

youtube_dl/extractor/generic.py

@@ -184,6 +184,17 @@ class GenericIE(InfoExtractor):
                 'description': 'md5:ddb2a40ecd6b6a147e400e535874947b',
             }
         },
+        # Embeded Ustream video
+        {
+            'url': 'http://www.american.edu/spa/pti/nsa-privacy-janus-2014.cfm',
+            'md5': '27b99cdb639c9b12a79bca876a073417',
+            'info_dict': {
+                'id': '45734260',
+                'ext': 'flv',
+                'uploader': 'AU SPA:  The NSA and Privacy',
+                'title': 'NSA and Privacy Forum Debate featuring General Hayden and Barton Gellman'
+            }
+        },
         # nowvideo embed hidden behind percent encoding
         {
             'url': 'http://www.waoanime.tv/the-super-dimension-fortress-macross-episode-1/',
@@ -500,17 +511,18 @@ class GenericIE(InfoExtractor):
         if mobj is not None:
             return self.url_result(mobj.group(1), 'Mpora')
 
-        # Look for embedded NovaMov player
+        # Look for embedded NovaMov-based player
         mobj = re.search(
-            r'<iframe[^>]+?src=(["\'])(?P<url>http://(?:(?:embed|www)\.)?novamov\.com/embed\.php.+?)\1', webpage)
+            r'''(?x)<iframe[^>]+?src=(["\'])
+                    (?P<url>http://(?:(?:embed|www)\.)?
+                        (?:novamov\.com|
+                           nowvideo\.(?:ch|sx|eu|at|ag|co)|
+                           videoweed\.(?:es|com)|
+                           movshare\.(?:net|sx|ag)|
+                           divxstage\.(?:eu|net|ch|co|at|ag))
+                        /embed\.php.+?)\1''', webpage)
         if mobj is not None:
-            return self.url_result(mobj.group('url'), 'NovaMov')
-
-        # Look for embedded NowVideo player
-        mobj = re.search(
-            r'<iframe[^>]+?src=(["\'])(?P<url>http://(?:(?:embed|www)\.)?nowvideo\.(?:ch|sx|eu)/embed\.php.+?)\1', webpage)
-        if mobj is not None:
-            return self.url_result(mobj.group('url'), 'NowVideo')
+            return self.url_result(mobj.group('url'))
 
         # Look for embedded Facebook player
         mobj = re.search(
@@ -556,6 +568,12 @@ class GenericIE(InfoExtractor):
         if mobj is not None:
             return self.url_result(mobj.group('url'), 'TED')
 
+        # Look for embedded Ustream videos
+        mobj = re.search(
+            r'<iframe[^>]+?src=(["\'])(?P<url>http://www\.ustream\.tv/embed/.+?)\1', webpage)
+        if mobj is not None:
+            return self.url_result(mobj.group('url'), 'Ustream')
+
         # Look for embedded arte.tv player
         mobj = re.search(
             r'<script [^>]*?src="(?P<url>http://www\.arte\.tv/playerv2/embed[^"]+)"',

youtube_dl/extractor/justintv.py

@@ -1,9 +1,12 @@
+from __future__ import unicode_literals
+
 import json
 import os
 import re
 
 from .common import InfoExtractor
 from ..utils import (
+    compat_str,
     ExtractorError,
     formatSeconds,
 )
@@ -24,34 +27,31 @@ class JustinTVIE(InfoExtractor):
         /?(?:\#.*)?$
         """
     _JUSTIN_PAGE_LIMIT = 100
-    IE_NAME = u'justin.tv'
+    IE_NAME = 'justin.tv'
+    IE_DESC = 'justin.tv and twitch.tv'
     _TEST = {
-        u'url': u'http://www.twitch.tv/thegamedevhub/b/296128360',
-        u'file': u'296128360.flv',
-        u'md5': u'ecaa8a790c22a40770901460af191c9a',
-        u'info_dict': {
-            u"upload_date": u"20110927", 
-            u"uploader_id": 25114803, 
-            u"uploader": u"thegamedevhub", 
-            u"title": u"Beginner Series - Scripting With Python Pt.1"
+        'url': 'http://www.twitch.tv/thegamedevhub/b/296128360',
+        'md5': 'ecaa8a790c22a40770901460af191c9a',
+        'info_dict': {
+            'id': '296128360',
+            'ext': 'flv',
+            'upload_date': '20110927',
+            'uploader_id': 25114803,
+            'uploader': 'thegamedevhub',
+            'title': 'Beginner Series - Scripting With Python Pt.1'
         }
     }
 
-    def report_download_page(self, channel, offset):
-        """Report attempt to download a single page of videos."""
-        self.to_screen(u'%s: Downloading video information from %d to %d' %
-                (channel, offset, offset + self._JUSTIN_PAGE_LIMIT))
-
     # Return count of items, list of *valid* items
     def _parse_page(self, url, video_id):
         info_json = self._download_webpage(url, video_id,
-                                           u'Downloading video info JSON',
-                                           u'unable to download video info JSON')
+                                           'Downloading video info JSON',
+                                           'unable to download video info JSON')
 
         response = json.loads(info_json)
         if type(response) != list:
             error_text = response.get('error', 'unknown error')
-            raise ExtractorError(u'Justin.tv API: %s' % error_text)
+            raise ExtractorError('Justin.tv API: %s' % error_text)
         info = []
         for clip in response:
             video_url = clip['video_file_url']
@@ -62,7 +62,7 @@ class JustinTVIE(InfoExtractor):
                 video_id = clip['id']
                 video_title = clip.get('title', video_id)
                 info.append({
-                    'id': video_id,
+                    'id': compat_str(video_id),
                     'url': video_url,
                     'title': video_title,
                     'uploader': clip.get('channel_name', video_uploader_id),
@@ -74,8 +74,6 @@ class JustinTVIE(InfoExtractor):
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
-        if mobj is None:
-            raise ExtractorError(u'invalid URL: %s' % url)
 
         api_base = 'http://api.justin.tv'
         paged = False
@@ -89,40 +87,41 @@ class JustinTVIE(InfoExtractor):
             webpage = self._download_webpage(url, chapter_id)
             m = re.search(r'PP\.archive_id = "([0-9]+)";', webpage)
             if not m:
-                raise ExtractorError(u'Cannot find archive of a chapter')
+                raise ExtractorError('Cannot find archive of a chapter')
             archive_id = m.group(1)
 
             api = api_base + '/broadcast/by_chapter/%s.xml' % chapter_id
-            doc = self._download_xml(api, chapter_id,
-                                             note=u'Downloading chapter information',
-                                             errnote=u'Chapter information download failed')
+            doc = self._download_xml(
+                api, chapter_id,
+                note='Downloading chapter information',
+                errnote='Chapter information download failed')
             for a in doc.findall('.//archive'):
                 if archive_id == a.find('./id').text:
                     break
             else:
-                raise ExtractorError(u'Could not find chapter in chapter information')
+                raise ExtractorError('Could not find chapter in chapter information')
 
             video_url = a.find('./video_file_url').text
-            video_ext = video_url.rpartition('.')[2] or u'flv'
+            video_ext = video_url.rpartition('.')[2] or 'flv'
 
-            chapter_api_url = u'https://api.twitch.tv/kraken/videos/c' + chapter_id
-            chapter_info_json = self._download_webpage(chapter_api_url, u'c' + chapter_id,
-                                   note='Downloading chapter metadata',
-                                   errnote='Download of chapter metadata failed')
-            chapter_info = json.loads(chapter_info_json)
+            chapter_api_url = 'https://api.twitch.tv/kraken/videos/c' + chapter_id
+            chapter_info = self._download_json(
+                chapter_api_url, 'c' + chapter_id,
+                note='Downloading chapter metadata',
+                errnote='Download of chapter metadata failed')
 
             bracket_start = int(doc.find('.//bracket_start').text)
             bracket_end = int(doc.find('.//bracket_end').text)
 
             # TODO determine start (and probably fix up file)
             #  youtube-dl -v http://www.twitch.tv/firmbelief/c/1757457
-            #video_url += u'?start=' + TODO:start_timestamp
+            #video_url += '?start=' + TODO:start_timestamp
             # bracket_start is 13290, but we want 51670615
-            self._downloader.report_warning(u'Chapter detected, but we can just download the whole file. '
-                                            u'Chapter starts at %s and ends at %s' % (formatSeconds(bracket_start), formatSeconds(bracket_end)))
+            self._downloader.report_warning('Chapter detected, but we can just download the whole file. '
+                                            'Chapter starts at %s and ends at %s' % (formatSeconds(bracket_start), formatSeconds(bracket_end)))
 
             info = {
-                'id': u'c' + chapter_id,
+                'id': 'c' + chapter_id,
                 'url': video_url,
                 'ext': video_ext,
                 'title': chapter_info['title'],
@@ -131,14 +130,12 @@ class JustinTVIE(InfoExtractor):
                 'uploader': chapter_info['channel']['display_name'],
                 'uploader_id': chapter_info['channel']['name'],
             }
-            return [info]
+            return info
         else:
             video_id = mobj.group('videoid')
             api = api_base + '/broadcast/by_archive/%s.json' % video_id
 
-        self.report_extraction(video_id)
-
-        info = []
+        entries = []
         offset = 0
         limit = self._JUSTIN_PAGE_LIMIT
         while True:
@@ -146,8 +143,12 @@ class JustinTVIE(InfoExtractor):
                 self.report_download_page(video_id, offset)
             page_url = api + ('?offset=%d&limit=%d' % (offset, limit))
             page_count, page_info = self._parse_page(page_url, video_id)
-            info.extend(page_info)
+            entries.extend(page_info)
             if not paged or page_count != limit:
                 break
             offset += limit
-        return info
+        return {
+            '_type': 'playlist',
+            'id': video_id,
+            'entries': entries,
+        }

youtube_dl/extractor/morningstar.py

@@ -1,17 +1,9 @@
 # coding: utf-8
 from __future__ import unicode_literals
 
-import hashlib
-import json
 import re
-import time
 
 from .common import InfoExtractor
-from ..utils import (
-    compat_parse_qs,
-    compat_str,
-    int_or_none,
-)
 
 
 class MorningstarIE(InfoExtractor):

youtube_dl/extractor/motorsport.py

@@ -44,7 +44,7 @@ class MotorsportIE(InfoExtractor):
         e = compat_str(int(time.time()) + 24 * 60 * 60)
         base_video_url = params['location'] + '?e=' + e
         s = 'h3hg713fh32'
-        h = hashlib.md5(s + base_video_url).hexdigest()
+        h = hashlib.md5((s + base_video_url).encode('utf-8')).hexdigest()
         video_url = base_video_url + '&h=' + h
 
         uploader = self._html_search_regex(

youtube_dl/extractor/movshare.py

@@ -0,0 +1,27 @@
+from __future__ import unicode_literals
+
+from .novamov import NovaMovIE
+
+
+class MovShareIE(NovaMovIE):
+    IE_NAME = 'movshare'
+    IE_DESC = 'MovShare'
+
+    _VALID_URL = NovaMovIE._VALID_URL_TEMPLATE % {'host': 'movshare\.(?:net|sx|ag)'}
+
+    _HOST = 'www.movshare.net'
+
+    _FILE_DELETED_REGEX = r'>This file no longer exists on our servers.<'
+    _TITLE_REGEX = r'<strong>Title:</strong> ([^<]+)</p>'
+    _DESCRIPTION_REGEX = r'<strong>Description:</strong> ([^<]+)</p>'
+
+    _TEST = {
+        'url': 'http://www.movshare.net/video/559e28be54d96',
+        'md5': 'abd31a2132947262c50429e1d16c1bfd',
+        'info_dict': {
+            'id': '559e28be54d96',
+            'ext': 'flv',
+            'title': 'dissapeared image',
+            'description': 'optical illusion  dissapeared image  magic illusion',
+        }
+    }

youtube_dl/extractor/novamov.py

@@ -13,7 +13,8 @@ class NovaMovIE(InfoExtractor):
     IE_NAME = 'novamov'
     IE_DESC = 'NovaMov'
 
-    _VALID_URL = r'http://(?:(?:www\.)?%(host)s/video/|(?:(?:embed|www)\.)%(host)s/embed\.php\?(?:.*?&)?v=)(?P<videoid>[a-z\d]{13})' % {'host': 'novamov\.com'}
+    _VALID_URL_TEMPLATE = r'http://(?:(?:www\.)?%(host)s/(?:file|video)/|(?:(?:embed|www)\.)%(host)s/embed\.php\?(?:.*?&)?v=)(?P<id>[a-z\d]{13})'
+    _VALID_URL = _VALID_URL_TEMPLATE % {'host': 'novamov\.com'}
 
     _HOST = 'www.novamov.com'
 
@@ -36,18 +37,17 @@ class NovaMovIE(InfoExtractor):
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
-        video_id = mobj.group('videoid')
+        video_id = mobj.group('id')
 
         page = self._download_webpage(
             'http://%s/video/%s' % (self._HOST, video_id), video_id, 'Downloading video page')
 
         if re.search(self._FILE_DELETED_REGEX, page) is not None:
-            raise ExtractorError(u'Video %s does not exist' % video_id, expected=True)
+            raise ExtractorError('Video %s does not exist' % video_id, expected=True)
 
         filekey = self._search_regex(self._FILEKEY_REGEX, page, 'filekey')
 
         title = self._html_search_regex(self._TITLE_REGEX, page, 'title', fatal=False)
-
         description = self._html_search_regex(self._DESCRIPTION_REGEX, page, 'description', default='', fatal=False)
 
         api_response = self._download_webpage(

youtube_dl/extractor/nowvideo.py

@@ -7,7 +7,7 @@ class NowVideoIE(NovaMovIE):
     IE_NAME = 'nowvideo'
     IE_DESC = 'NowVideo'
 
-    _VALID_URL = r'http://(?:(?:www\.)?%(host)s/video/|(?:(?:embed|www)\.)%(host)s/embed\.php\?(?:.*?&)?v=)(?P<videoid>[a-z\d]{13})' % {'host': 'nowvideo\.(?:ch|sx|eu)'}
+    _VALID_URL = NovaMovIE._VALID_URL_TEMPLATE % {'host': 'nowvideo\.(?:ch|sx|eu|at|ag|co)'}
 
     _HOST = 'www.nowvideo.ch'
 

youtube_dl/extractor/pornhd.py

@@ -39,11 +39,11 @@ class PornHdIE(InfoExtractor):
 
         formats = [
             {
-                'url': url,
+                'url': format_url,
                 'ext': format.lower(),
                 'format_id': '%s-%s' % (format.lower(), quality.lower()),
                 'quality': 1 if quality.lower() == 'high' else 0,
-            } for format, quality, url in re.findall(
+            } for format, quality, format_url in re.findall(
                 r'var __video([\da-zA-Z]+?)(Low|High)StreamUrl = \'(http://.+?)\?noProxy=1\'', webpage)
         ]
 

youtube_dl/extractor/ro220.py

@@ -18,7 +18,7 @@ class Ro220IE(InfoExtractor):
         'md5': '03af18b73a07b4088753930db7a34add',
         'info_dict': {
             "title": "Luati-le Banii sez 4 ep 1",
-            "description": "Iata-ne reveniti dupa o binemeritata vacanta. Va astept si pe Facebook cu pareri si comentarii.",
+            "description": "re:^Iata-ne reveniti dupa o binemeritata vacanta\. +Va astept si pe Facebook cu pareri si comentarii.$",
         }
     }
 

youtube_dl/extractor/rts.py

@@ -35,13 +35,13 @@ class RTSIE(InfoExtractor):
         },
         {
             'url': 'http://www.rts.ch/emissions/passe-moi-les-jumelles/5624067-entre-ciel-et-mer.html',
-            'md5': 'c197f0b2421995c63a64cc73d800f42e',
+            'md5': 'c148457a27bdc9e5b1ffe081a7a8337b',
             'info_dict': {
-                'id': '5738317',
+                'id': '5624067',
                 'ext': 'mp4',
-                'duration': 55,
-                'title': 'Bande de lancement de Passe-moi les jumelles',
-                'description': '',
+                'duration': 3720,
+                'title': 'Les yeux dans les cieux - Mon homard au Canada',
+                'description': 'md5:d22ee46f5cc5bac0912e5a0c6d44a9f7',
                 'uploader': 'Passe-moi les jumelles',
                 'upload_date': '20140404',
                 'timestamp': 1396635300,
@@ -98,17 +98,20 @@ class RTSIE(InfoExtractor):
         m = re.match(self._VALID_URL, url)
         video_id = m.group('id')
 
-        def download_json(video_id):
+        def download_json(internal_id):
             return self._download_json(
-                'http://www.rts.ch/a/%s.html?f=json/article' % video_id, video_id)
+                'http://www.rts.ch/a/%s.html?f=json/article' % internal_id,
+                video_id)
 
         all_info = download_json(video_id)
 
         # video_id extracted out of URL is not always a real id
         if 'video' not in all_info and 'audio' not in all_info:
             page = self._download_webpage(url, video_id)
-            video_id = self._html_search_regex(r'<(?:video|audio) data-id="(\d+)"', page, 'video id')
-            all_info = download_json(video_id)
+            internal_id = self._html_search_regex(
+                r'<(?:video|audio) data-id="([0-9]+)"', page,
+                'internal video id')
+            all_info = download_json(internal_id)
 
         info = all_info['video']['JSONinfo'] if 'video' in all_info else all_info['audio']
 

youtube_dl/extractor/teamcoco.py

@@ -9,8 +9,18 @@ from ..utils import (
 
 
 class TeamcocoIE(InfoExtractor):
-    _VALID_URL = r'http://teamcoco\.com/video/(?P<url_title>.*)'
-    _TEST = {
+    _VALID_URL = r'http://teamcoco\.com/video/(?P<video_id>[0-9]+)?/?(?P<url_title>.*)'
+    _TESTS = [
+    {
+        'url': 'http://teamcoco.com/video/80187/conan-becomes-a-mary-kay-beauty-consultant',
+        'file': '80187.mp4',
+        'md5': '3f7746aa0dc86de18df7539903d399ea',
+        'info_dict': {
+            'title': 'Conan Becomes A Mary Kay Beauty Consultant',
+            'description': 'Mary Kay is perhaps the most trusted name in female beauty, so of course Conan is a natural choice to sell their products.'
+        }
+    },
+    {
         'url': 'http://teamcoco.com/video/louis-ck-interview-george-w-bush',
         'file': '19705.mp4',
         'md5': 'cde9ba0fa3506f5f017ce11ead928f9a',
@@ -19,6 +29,7 @@ class TeamcocoIE(InfoExtractor):
             "title": "Louis C.K. Interview Pt. 1 11/3/11"
         }
     }
+    ]
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
@@ -26,11 +37,13 @@ class TeamcocoIE(InfoExtractor):
             raise ExtractorError('Invalid URL: %s' % url)
         url_title = mobj.group('url_title')
         webpage = self._download_webpage(url, url_title)
-
-        video_id = self._html_search_regex(
-            r'<article class="video" data-id="(\d+?)"',
-            webpage, 'video id')
-
+        
+        video_id = mobj.group("video_id")
+        if video_id == '':
+            video_id = self._html_search_regex(
+                r'<article class="video" data-id="(\d+?)"',
+                webpage, 'video id')
+        
         self.report_extraction(video_id)
 
         data_url = 'http://teamcoco.com/cvp/2.0/%s.xml' % video_id

youtube_dl/extractor/ted.py

@@ -37,6 +37,7 @@ class TEDIE(SubtitlesInfoExtractor):
                 'consciousness, but that half the time our brains are '
                 'actively fooling us.'),
             'uploader': 'Dan Dennett',
+            'width': 854,
         }
     }, {
         'url': 'http://www.ted.com/watch/ted-institute/ted-bcg/vishal-sikka-the-beauty-and-power-of-algorithms',
@@ -50,10 +51,10 @@ class TEDIE(SubtitlesInfoExtractor):
         }
     }]
 
-    _FORMATS_PREFERENCE = {
-        'low': 1,
-        'medium': 2,
-        'high': 3,
+    _NATIVE_FORMATS = {
+        'low': {'preference': 1, 'width': 320, 'height': 180},
+        'medium': {'preference': 2, 'width': 512, 'height': 288},
+        'high': {'preference': 3, 'width': 854, 'height': 480},
     }
 
     def _extract_info(self, webpage):
@@ -98,12 +99,14 @@ class TEDIE(SubtitlesInfoExtractor):
         talk_info = self._extract_info(webpage)['talks'][0]
 
         formats = [{
-            'ext': 'mp4',
             'url': format_url,
             'format_id': format_id,
             'format': format_id,
-            'preference': self._FORMATS_PREFERENCE.get(format_id, -1),
         } for (format_id, format_url) in talk_info['nativeDownloads'].items()]
+        for f in formats:
+            finfo = self._NATIVE_FORMATS.get(f['format_id'])
+            if finfo:
+                f.update(finfo)
         self._sort_formats(formats)
 
         video_id = compat_str(talk_info['id'])

youtube_dl/extractor/ustream.py

@@ -11,7 +11,7 @@ from ..utils import (
 
 
 class UstreamIE(InfoExtractor):
-    _VALID_URL = r'https?://www\.ustream\.tv/recorded/(?P<videoID>\d+)'
+    _VALID_URL = r'https?://www\.ustream\.tv/(?P<type>recorded|embed)/(?P<videoID>\d+)'
     IE_NAME = 'ustream'
     _TEST = {
         'url': 'http://www.ustream.tv/recorded/20274954',
@@ -25,6 +25,13 @@ class UstreamIE(InfoExtractor):
 
     def _real_extract(self, url):
         m = re.match(self._VALID_URL, url)
+        if m.group('type') == 'embed':
+            video_id = m.group('videoID')
+            webpage = self._download_webpage(url, video_id)
+            desktop_video_id = self._html_search_regex(r'ContentVideoIds=\["([^"]*?)"\]', webpage, 'desktop_video_id')
+            desktop_url = 'http://www.ustream.tv/recorded/' + desktop_video_id
+            return self.url_result(desktop_url, 'Ustream')
+
         video_id = m.group('videoID')
 
         video_url = 'http://tcdn.ustream.tv/video/%s' % video_id

youtube_dl/extractor/videoweed.py

@@ -0,0 +1,26 @@
+from __future__ import unicode_literals
+
+from .novamov import NovaMovIE
+
+
+class VideoWeedIE(NovaMovIE):
+    IE_NAME = 'videoweed'
+    IE_DESC = 'VideoWeed'
+
+    _VALID_URL = NovaMovIE._VALID_URL_TEMPLATE % {'host': 'videoweed\.(?:es|com)'}
+
+    _HOST = 'www.videoweed.es'
+
+    _FILE_DELETED_REGEX = r'>This file no longer exists on our servers.<'
+    _TITLE_REGEX = r'<h1 class="text_shadow">([^<]+)</h1>'
+
+    _TEST = {
+        'url': 'http://www.videoweed.es/file/b42178afbea14',
+        'md5': 'abd31a2132947262c50429e1d16c1bfd',
+        'info_dict': {
+            'id': 'b42178afbea14',
+            'ext': 'flv',
+            'title': 'optical illusion  dissapeared image magic illusion',
+            'description': ''
+        },
+    }

youtube_dl/extractor/yahoo.py

@@ -1,6 +1,7 @@
 from __future__ import unicode_literals
 
 import itertools
+import json
 import re
 
 from .common import InfoExtractor, SearchInfoExtractor
@@ -14,7 +15,7 @@ from ..utils import (
 
 class YahooIE(InfoExtractor):
     IE_DESC = 'Yahoo screen'
-    _VALID_URL = r'http://screen\.yahoo\.com/.*?-(?P<id>\d*?)\.html'
+    _VALID_URL = r'https?://screen\.yahoo\.com/.*?-(?P<id>[0-9]+)(?:-[a-z]+)?\.html'
     _TESTS = [
         {
             'url': 'http://screen.yahoo.com/julian-smith-travis-legg-watch-214727115.html',

youtube_dl/utils.py

@@ -2,6 +2,7 @@
 # -*- coding: utf-8 -*-
 
 import calendar
+import codecs
 import contextlib
 import ctypes
 import datetime
@@ -1263,9 +1264,11 @@ class PagedList(object):
 
 
 def uppercase_escape(s):
+    unicode_escape = codecs.getdecoder('unicode_escape')
     return re.sub(
         r'\\U[0-9a-fA-F]{8}',
-        lambda m: m.group(0).decode('unicode-escape'), s)
+        lambda m: unicode_escape(m.group(0))[0],
+        s)
 
 try:
     struct.pack(u'!I', 0)

youtube_dl/version.py

@@ -1,2 +1,2 @@
 
-__version__ = '2014.04.04.4'
+__version__ = '2014.04.07'