release 2014.03.10

And don't report a warning, just log a message, it allows to pass the test from Europe.

README.md

@@ -124,8 +124,12 @@ which means you can modify it, redistribute it or use it however you like.
                                      video id, %(playlist)s for the playlist the
                                      video is in, %(playlist_index)s for the
                                      position in the playlist and %% for a
-                                     literal percent. Use - to output to stdout.
-                                     Can also be used to download to a different
+                                     literal percent. %(height)s and %(width)s
+                                     for the width and height of the video
+                                     format. %(resolution)s for a textual
+                                     description of the resolution of the video
+                                     format. Use - to output to stdout. Can also
+                                     be used to download to a different
                                      directory, for example with -o '/my/downloa
                                      ds/%(uploader)s/%(title)s-%(id)s.%(ext)s' .
     --autonumber-size NUMBER         Specifies the number of digits in

test/test_InfoExtractor.py

@@ -0,0 +1,44 @@
+#!/usr/bin/env python
+
+from __future__ import unicode_literals
+
+# Allow direct execution
+import os
+import sys
+import unittest
+sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
+
+from test.helper import FakeYDL
+from youtube_dl.extractor.common import InfoExtractor
+from youtube_dl.extractor import YoutubeIE, get_info_extractor
+
+
+class TestIE(InfoExtractor):
+    pass
+
+
+class TestInfoExtractor(unittest.TestCase):
+    def setUp(self):
+        self.ie = TestIE(FakeYDL())
+
+    def test_ie_key(self):
+        self.assertEqual(get_info_extractor(YoutubeIE.ie_key()), YoutubeIE)
+
+    def test_html_search_regex(self):
+        html = '<p id="foo">Watch this <a href="http://www.youtube.com/watch?v=BaW_jenozKc">video</a></p>'
+        search = lambda re, *args: self.ie._html_search_regex(re, html, *args)
+        self.assertEqual(search(r'<p id="foo">(.+?)</p>', 'foo'), 'Watch this video')
+
+    def test_opengraph(self):
+        ie = self.ie
+        html = '''
+            <meta name="og:title" content='Foo'/>
+            <meta content="Some video's description " name="og:description"/>
+            <meta property='og:image' content='http://domain.com/pic.jpg?key1=val1&amp;key2=val2'/>
+            '''
+        self.assertEqual(ie._og_search_title(html), 'Foo')
+        self.assertEqual(ie._og_search_description(html), 'Some video\'s description ')
+        self.assertEqual(ie._og_search_thumbnail(html), 'http://domain.com/pic.jpg?key1=val1&key2=val2')
+
+if __name__ == '__main__':
+    unittest.main()

test/test_all_urls.py

@@ -71,6 +71,10 @@ class TestAllURLsMatching(unittest.TestCase):
     def test_youtube_truncated(self):
         self.assertMatch('http://www.youtube.com/watch?', ['youtube:truncated_url'])
 
+    def test_youtube_search_matching(self):
+        self.assertMatch('http://www.youtube.com/results?search_query=making+mustard', ['youtube:search_url'])
+        self.assertMatch('https://www.youtube.com/results?baz=bar&search_query=youtube-dl+test+video&filters=video&lclk=video', ['youtube:search_url'])
+
     def test_justin_tv_channelid_matching(self):
         self.assertTrue(JustinTVIE.suitable(u"justin.tv/vanillatv"))
         self.assertTrue(JustinTVIE.suitable(u"twitch.tv/vanillatv"))

test/test_playlists.py

@@ -36,6 +36,7 @@ from youtube_dl.extractor import (
     RutubeChannelIE,
     GoogleSearchIE,
     GenericIE,
+    TEDIE,
 )
 
 
@@ -98,7 +99,7 @@ class TestPlaylists(unittest.TestCase):
         result = ie.extract('http://www.ustream.tv/channel/young-americans-for-liberty')
         self.assertIsPlaylist(result)
         self.assertEqual(result['id'], '5124905')
-        self.assertTrue(len(result['entries']) >= 11)
+        self.assertTrue(len(result['entries']) >= 6)
 
     def test_soundcloud_set(self):
         dl = FakeYDL()
@@ -253,11 +254,20 @@ class TestPlaylists(unittest.TestCase):
     def test_generic_rss_feed(self):
         dl = FakeYDL()
         ie = GenericIE(dl)
-        result = ie.extract('http://www.escapistmagazine.com/rss/videos/list/1.xml')
+        result = ie.extract('http://phihag.de/2014/youtube-dl/rss.xml')
         self.assertIsPlaylist(result)
-        self.assertEqual(result['id'], 'http://www.escapistmagazine.com/rss/videos/list/1.xml')
+        self.assertEqual(result['id'], 'http://phihag.de/2014/youtube-dl/rss.xml')
         self.assertEqual(result['title'], 'Zero Punctuation')
         self.assertTrue(len(result['entries']) > 10)
 
+    def test_ted_playlist(self):
+        dl = FakeYDL()
+        ie = TEDIE(dl)
+        result = ie.extract('http://www.ted.com/playlists/who_are_the_hackers')
+        self.assertIsPlaylist(result)
+        self.assertEqual(result['id'], '10')
+        self.assertEqual(result['title'], 'Who are the hackers?')
+        self.assertTrue(len(result['entries']) >= 6)
+
 if __name__ == '__main__':
     unittest.main()

test/test_utils.py

@@ -33,6 +33,7 @@ from youtube_dl.utils import (
     unified_strdate,
     unsmuggle_url,
     url_basename,
+    urlencode_postdata,
     xpath_with_ns,
 )
 
@@ -261,5 +262,9 @@ class TestUtil(unittest.TestCase):
             bam''')
         self.assertEqual(read_batch_urls(f), [u'foo', u'bar', u'baz', u'bam'])
 
+    def test_urlencode_postdata(self):
+        data = urlencode_postdata({'username': 'foo@bar.com', 'password': '1234'})
+        self.assertTrue(isinstance(data, bytes))
+
 if __name__ == '__main__':
     unittest.main()

test/test_youtube_lists.py

@@ -16,6 +16,7 @@ from youtube_dl.extractor import (
     YoutubeChannelIE,
     YoutubeShowIE,
     YoutubeTopListIE,
+    YoutubeSearchURLIE,
 )
 
 
@@ -133,5 +134,14 @@ class TestYoutubeLists(unittest.TestCase):
         entries = result['entries']
         self.assertTrue(len(entries) >= 5)
 
+    def test_youtube_search_url(self):
+        dl = FakeYDL()
+        ie = YoutubeSearchURLIE(dl)
+        result = ie.extract('https://www.youtube.com/results?baz=bar&search_query=youtube-dl+test+video&filters=video&lclk=video')
+        entries = result['entries']
+        self.assertIsPlaylist(result)
+        self.assertEqual(result['title'], 'youtube-dl test video')
+        self.assertTrue(len(entries) >= 5)
+
 if __name__ == '__main__':
     unittest.main()

youtube_dl/YoutubeDL.py

@@ -370,12 +370,15 @@ class YoutubeDL(object):
         Print the message to stderr, it will be prefixed with 'WARNING:'
         If stderr is a tty file the 'WARNING:' will be colored
         '''
-        if self._err_file.isatty() and os.name != 'nt':
-            _msg_header = '\033[0;33mWARNING:\033[0m'
+        if self.params.get('logger') is not None:
+            self.params['logger'].warning(message)
         else:
-            _msg_header = 'WARNING:'
-        warning_message = '%s %s' % (_msg_header, message)
-        self.to_stderr(warning_message)
+            if self._err_file.isatty() and os.name != 'nt':
+                _msg_header = '\033[0;33mWARNING:\033[0m'
+            else:
+                _msg_header = 'WARNING:'
+            warning_message = '%s %s' % (_msg_header, message)
+            self.to_stderr(warning_message)
 
     def report_error(self, message, tb=None):
         '''
@@ -409,6 +412,13 @@ class YoutubeDL(object):
             template_dict['autonumber'] = autonumber_templ % self._num_downloads
             if template_dict.get('playlist_index') is not None:
                 template_dict['playlist_index'] = '%05d' % template_dict['playlist_index']
+            if template_dict.get('resolution') is None:
+                if template_dict.get('width') and template_dict.get('height'):
+                    template_dict['resolution'] = '%dx%d' % (template_dict['width'], template_dict['height'])
+                elif template_dict.get('height'):
+                    template_dict['resolution'] = '%sp' % template_dict['height']
+                elif template_dict.get('width'):
+                    template_dict['resolution'] = '?x%d' % template_dict['width']
 
             sanitize = lambda k, v: sanitize_filename(
                 compat_str(v),
@@ -675,6 +685,9 @@ class YoutubeDL(object):
             info_dict['playlist'] = None
             info_dict['playlist_index'] = None
 
+        if 'display_id' not in info_dict and 'id' in info_dict:
+            info_dict['display_id'] = info_dict['id']
+
         # This extractors handle format selection themselves
         if info_dict['extractor'] in ['Youku']:
             if download:
@@ -908,7 +921,7 @@ class YoutubeDL(object):
                     self.to_screen('[%s] %s: Downloading thumbnail ...' %
                                    (info_dict['extractor'], info_dict['id']))
                     try:
-                        uf = compat_urllib_request.urlopen(info_dict['thumbnail'])
+                        uf = self.urlopen(info_dict['thumbnail'])
                         with open(thumb_filename, 'wb') as thumbf:
                             shutil.copyfileobj(uf, thumbf)
                         self.to_screen('[%s] %s: Writing thumbnail to: %s' %

youtube_dl/__init__.py

@@ -48,6 +48,9 @@ __authors__  = (
     'Niklas Laxström',
     'David Triendl',
     'Anthony Weems',
+    'David Wagner',
+    'Juan C. Olivares',
+    'Mattias Harrysson',
 )
 
 __license__ = 'Public Domain'
@@ -428,6 +431,8 @@ def parseOpts(overrideArguments=None):
                   '%(extractor)s for the provider (youtube, metacafe, etc), '
                   '%(id)s for the video id, %(playlist)s for the playlist the video is in, '
                   '%(playlist_index)s for the position in the playlist and %% for a literal percent. '
+                  '%(height)s and %(width)s for the width and height of the video format. '
+                  '%(resolution)s for a textual description of the resolution of the video format. '
                   'Use - to output to stdout. Can also be used to download to a different directory, '
                   'for example with -o \'/my/downloads/%(uploader)s/%(title)s-%(id)s.%(ext)s\' .'))
     filesystem.add_option('--autonumber-size',

youtube_dl/downloader/http.py

@@ -49,7 +49,7 @@ class HttpFD(FileDownloader):
         while count <= retries:
             # Establish connection
             try:
-                data = compat_urllib_request.urlopen(request)
+                data = self.ydl.urlopen(request)
                 break
             except (compat_urllib_error.HTTPError, ) as err:
                 if (err.code < 500 or err.code >= 600) and err.code != 416:
@@ -59,7 +59,7 @@ class HttpFD(FileDownloader):
                     # Unable to resume (requested range not satisfiable)
                     try:
                         # Open the connection again without the range header
-                        data = compat_urllib_request.urlopen(basic_request)
+                        data = self.ydl.urlopen(basic_request)
                         content_length = data.info()['Content-Length']
                     except (compat_urllib_error.HTTPError, ) as err:
                         if err.code < 500 or err.code >= 600:
@@ -85,6 +85,7 @@ class HttpFD(FileDownloader):
                         else:
                             # The length does not match, we start the download over
                             self.report_unable_to_resume()
+                            resume_len = 0
                             open_mode = 'wb'
                             break
             # Retry

youtube_dl/downloader/rtmp.py

@@ -1,3 +1,5 @@
+from __future__ import unicode_literals
+
 import os
 import re
 import subprocess
@@ -22,7 +24,7 @@ class RtmpFD(FileDownloader):
             proc_stderr_closed = False
             while not proc_stderr_closed:
                 # read line from stderr
-                line = u''
+                line = ''
                 while True:
                     char = proc.stderr.read(1)
                     if not char:
@@ -46,7 +48,7 @@ class RtmpFD(FileDownloader):
                     data_len = None
                     if percent > 0:
                         data_len = int(downloaded_data_len * 100 / percent)
-                    data_len_str = u'~' + format_bytes(data_len)
+                    data_len_str = '~' + format_bytes(data_len)
                     self.report_progress(percent, data_len_str, speed, eta)
                     cursor_in_new_line = False
                     self._hook_progress({
@@ -76,12 +78,12 @@ class RtmpFD(FileDownloader):
                         })
                     elif self.params.get('verbose', False):
                         if not cursor_in_new_line:
-                            self.to_screen(u'')
+                            self.to_screen('')
                         cursor_in_new_line = True
-                        self.to_screen(u'[rtmpdump] '+line)
+                        self.to_screen('[rtmpdump] '+line)
             proc.wait()
             if not cursor_in_new_line:
-                self.to_screen(u'')
+                self.to_screen('')
             return proc.returncode
 
         url = info_dict['url']
@@ -102,7 +104,7 @@ class RtmpFD(FileDownloader):
         try:
             subprocess.call(['rtmpdump', '-h'], stdout=(open(os.path.devnull, 'w')), stderr=subprocess.STDOUT)
         except (OSError, IOError):
-            self.report_error(u'RTMP download detected but "rtmpdump" could not be run')
+            self.report_error('RTMP download detected but "rtmpdump" could not be run')
             return False
 
         # Download using rtmpdump. rtmpdump returns exit code 2 when
@@ -127,7 +129,7 @@ class RtmpFD(FileDownloader):
             basic_args += ['--live']
         if conn:
             basic_args += ['--conn', conn]
-        args = basic_args + [[], ['--resume', '--skip', '1']][self.params.get('continuedl', False)]
+        args = basic_args + [[], ['--resume', '--skip', '1']][not live and self.params.get('continuedl', False)]
 
         if sys.platform == 'win32' and sys.version_info < (3, 0):
             # Windows subprocess module does not actually support Unicode
@@ -150,26 +152,35 @@ class RtmpFD(FileDownloader):
                 shell_quote = lambda args: ' '.join(map(pipes.quote, str_args))
             except ImportError:
                 shell_quote = repr
-            self.to_screen(u'[debug] rtmpdump command line: ' + shell_quote(str_args))
+            self.to_screen('[debug] rtmpdump command line: ' + shell_quote(str_args))
+
+        RD_SUCCESS = 0
+        RD_FAILED = 1
+        RD_INCOMPLETE = 2
+        RD_NO_CONNECT = 3
 
         retval = run_rtmpdump(args)
 
-        while (retval == 2 or retval == 1) and not test:
+        if retval == RD_NO_CONNECT:
+            self.report_error('[rtmpdump] Could not connect to RTMP server.')
+            return False
+
+        while (retval == RD_INCOMPLETE or retval == RD_FAILED) and not test and not live:
             prevsize = os.path.getsize(encodeFilename(tmpfilename))
-            self.to_screen(u'[rtmpdump] %s bytes' % prevsize)
+            self.to_screen('[rtmpdump] %s bytes' % prevsize)
             time.sleep(5.0) # This seems to be needed
-            retval = run_rtmpdump(basic_args + ['-e'] + [[], ['-k', '1']][retval == 1])
+            retval = run_rtmpdump(basic_args + ['-e'] + [[], ['-k', '1']][retval == RD_FAILED])
             cursize = os.path.getsize(encodeFilename(tmpfilename))
-            if prevsize == cursize and retval == 1:
+            if prevsize == cursize and retval == RD_FAILED:
                 break
              # Some rtmp streams seem abort after ~ 99.8%. Don't complain for those
-            if prevsize == cursize and retval == 2 and cursize > 1024:
-                self.to_screen(u'[rtmpdump] Could not download the whole video. This can happen for some advertisements.')
-                retval = 0
+            if prevsize == cursize and retval == RD_INCOMPLETE and cursize > 1024:
+                self.to_screen('[rtmpdump] Could not download the whole video. This can happen for some advertisements.')
+                retval = RD_SUCCESS
                 break
-        if retval == 0 or (test and retval == 2):
+        if retval == RD_SUCCESS or (test and retval == RD_INCOMPLETE):
             fsize = os.path.getsize(encodeFilename(tmpfilename))
-            self.to_screen(u'[rtmpdump] %s bytes' % fsize)
+            self.to_screen('[rtmpdump] %s bytes' % fsize)
             self.try_rename(tmpfilename, filename)
             self._hook_progress({
                 'downloaded_bytes': fsize,
@@ -179,6 +190,6 @@ class RtmpFD(FileDownloader):
             })
             return True
         else:
-            self.to_stderr(u"\n")
-            self.report_error(u'rtmpdump exited with code %d' % retval)
+            self.to_stderr('\n')
+            self.report_error('rtmpdump exited with code %d' % retval)
             return False

youtube_dl/extractor/__init__.py

@@ -1,5 +1,6 @@
 from .academicearth import AcademicEarthCourseIE
 from .addanime import AddAnimeIE
+from .aftonbladet import AftonbladetIE
 from .anitube import AnitubeIE
 from .aparat import AparatIE
 from .appletrailers import AppleTrailersIE
@@ -23,9 +24,11 @@ from .br import BRIE
 from .breakcom import BreakIE
 from .brightcove import BrightcoveIE
 from .c56 import C56IE
+from .canal13cl import Canal13clIE
 from .canalplus import CanalplusIE
 from .canalc2 import Canalc2IE
 from .cbs import CBSIE
+from .ceskatelevize import CeskaTelevizeIE
 from .channel9 import Channel9IE
 from .chilloutzone import ChilloutzoneIE
 from .cinemassacre import CinemassacreIE
@@ -89,6 +92,7 @@ from .funnyordie import FunnyOrDieIE
 from .gamekings import GamekingsIE
 from .gamespot import GameSpotIE
 from .gametrailers import GametrailersIE
+from .gdcvault import GDCVaultIE
 from .generic import GenericIE
 from .googleplus import GooglePlusIE
 from .googlesearch import GoogleSearchIE
@@ -133,6 +137,7 @@ from .lynda import (
 )
 from .m6 import M6IE
 from .macgamestore import MacGameStoreIE
+from .mailru import MailRuIE
 from .malemotion import MalemotionIE
 from .mdr import MDRIE
 from .metacafe import MetacafeIE
@@ -175,6 +180,7 @@ from .podomatic import PodomaticIE
 from .pornhd import PornHdIE
 from .pornhub import PornHubIE
 from .pornotube import PornotubeIE
+from .prosiebensat1 import ProSiebenSat1IE
 from .pyvideo import PyvideoIE
 from .radiofrance import RadioFranceIE
 from .rbmaradio import RBMARadioIE
@@ -234,6 +240,7 @@ from .tube8 import Tube8IE
 from .tudou import TudouIE
 from .tumblr import TumblrIE
 from .tutv import TutvIE
+from .tvigle import TvigleIE
 from .tvp import TvpIE
 from .unistra import UnistraIE
 from .ustream import UstreamIE, UstreamChannelIE
@@ -279,19 +286,20 @@ from .youku import YoukuIE
 from .youporn import YouPornIE
 from .youtube import (
     YoutubeIE,
-    YoutubePlaylistIE,
-    YoutubeSearchIE,
-    YoutubeSearchDateIE,
-    YoutubeUserIE,
     YoutubeChannelIE,
-    YoutubeShowIE,
-    YoutubeSubscriptionsIE,
-    YoutubeRecommendedIE,
-    YoutubeTruncatedURLIE,
-    YoutubeWatchLaterIE,
     YoutubeFavouritesIE,
     YoutubeHistoryIE,
+    YoutubePlaylistIE,
+    YoutubeRecommendedIE,
+    YoutubeSearchDateIE,
+    YoutubeSearchIE,
+    YoutubeSearchURLIE,
+    YoutubeShowIE,
+    YoutubeSubscriptionsIE,
     YoutubeTopListIE,
+    YoutubeTruncatedURLIE,
+    YoutubeUserIE,
+    YoutubeWatchLaterIE,
 )
 from .zdf import ZDFIE
 

youtube_dl/extractor/aftonbladet.py

@@ -0,0 +1,69 @@
+# encoding: utf-8
+from __future__ import unicode_literals
+
+import datetime
+import re
+
+from .common import InfoExtractor
+
+
+class AftonbladetIE(InfoExtractor):
+    _VALID_URL = r'^http://tv\.aftonbladet\.se/webbtv.+?(?P<video_id>article[0-9]+)\.ab(?:$|[?#])'
+    _TEST = {
+        'url': 'http://tv.aftonbladet.se/webbtv/nyheter/vetenskap/rymden/article36015.ab',
+        'info_dict': {
+            'id': 'article36015',
+            'ext': 'mp4',
+            'title': 'Vulkanutbrott i rymden - nu släpper NASA bilderna',
+            'description': 'Jupiters måne mest aktiv av alla himlakroppar',
+            'upload_date': '20140306',
+        },
+    }
+
+    def _real_extract(self, url):
+        mobj = re.search(self._VALID_URL, url)
+
+        video_id = mobj.group('video_id')
+        webpage = self._download_webpage(url, video_id)
+
+        # find internal video meta data
+        META_URL = 'http://aftonbladet-play.drlib.aptoma.no/video/%s.json'
+        internal_meta_id = self._html_search_regex(
+            r'data-aptomaId="([\w\d]+)"', webpage, 'internal_meta_id')
+        internal_meta_url = META_URL % internal_meta_id
+        internal_meta_json = self._download_json(
+            internal_meta_url, video_id, 'Downloading video meta data')
+
+        # find internal video formats
+        FORMATS_URL = 'http://aftonbladet-play.videodata.drvideo.aptoma.no/actions/video/?id=%s'
+        internal_video_id = internal_meta_json['videoId']
+        internal_formats_url = FORMATS_URL % internal_video_id
+        internal_formats_json = self._download_json(
+            internal_formats_url, video_id, 'Downloading video formats')
+
+        formats = []
+        for fmt in internal_formats_json['formats']['http']['pseudostreaming']['mp4']:
+            p = fmt['paths'][0]
+            formats.append({
+                'url': 'http://%s:%d/%s/%s' % (p['address'], p['port'], p['path'], p['filename']),
+                'ext': 'mp4',
+                'width': fmt['width'],
+                'height': fmt['height'],
+                'tbr': fmt['bitrate'],
+                'protocol': 'http',
+            })
+        self._sort_formats(formats)
+
+        timestamp = datetime.datetime.fromtimestamp(internal_meta_json['timePublished'])
+        upload_date = timestamp.strftime('%Y%m%d')
+
+        return {
+            'id': video_id,
+            'title': internal_meta_json['title'],
+            'formats': formats,
+            'thumbnail': internal_meta_json['imageUrl'],
+            'description': internal_meta_json['shortPreamble'],
+            'upload_date': upload_date,
+            'duration': internal_meta_json['duration'],
+            'view_count': internal_meta_json['views'],
+        }

youtube_dl/extractor/arte.py

@@ -72,18 +72,22 @@ class ArteTvIE(InfoExtractor):
             return self._extract_liveweb(url, name, lang)
 
         if re.search(self._LIVE_URL, url) is not None:
-            raise ExtractorError(u'Arte live streams are not yet supported, sorry')
+            raise ExtractorError('Arte live streams are not yet supported, sorry')
             # self.extractLiveStream(url)
             # return
 
+        raise ExtractorError('No video found')
+
     def _extract_video(self, url, video_id, lang):
         """Extract from videos.arte.tv"""
         ref_xml_url = url.replace('/videos/', '/do_delegate/videos/')
         ref_xml_url = ref_xml_url.replace('.html', ',view,asPlayerXml.xml')
-        ref_xml_doc = self._download_xml(ref_xml_url, video_id, note=u'Downloading metadata')
+        ref_xml_doc = self._download_xml(
+            ref_xml_url, video_id, note='Downloading metadata')
         config_node = find_xpath_attr(ref_xml_doc, './/video', 'lang', lang)
         config_xml_url = config_node.attrib['ref']
-        config_xml = self._download_webpage(config_xml_url, video_id, note=u'Downloading configuration')
+        config_xml = self._download_webpage(
+            config_xml_url, video_id, note='Downloading configuration')
 
         video_urls = list(re.finditer(r'<url quality="(?P<quality>.*?)">(?P<url>.*?)</url>', config_xml))
         def _key(m):

youtube_dl/extractor/canal13cl.py

@@ -0,0 +1,48 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+import re
+
+from .common import InfoExtractor
+
+
+class Canal13clIE(InfoExtractor):
+    _VALID_URL = r'^http://(?:www\.)?13\.cl/(?:[^/?#]+/)*(?P<id>[^/?#]+)'
+    _TEST = {
+        'url': 'http://www.13.cl/t13/nacional/el-circulo-de-hierro-de-michelle-bachelet-en-su-regreso-a-la-moneda',
+        'md5': '4cb1fa38adcad8fea88487a078831755',
+        'info_dict': {
+            'id': '1403022125',
+            'display_id': 'el-circulo-de-hierro-de-michelle-bachelet-en-su-regreso-a-la-moneda',
+            'ext': 'mp4',
+            'title': 'El "círculo de hierro" de Michelle Bachelet en su regreso a La Moneda',
+            'description': '(Foto: Agencia Uno) En nueve días más, Michelle Bachelet va a asumir por segunda vez como presidenta de la República. Entre aquellos que la acompañarán hay caras que se repiten y otras que se consolidan en su entorno de colaboradores más cercanos.',
+        }
+    }
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        display_id = mobj.group('id')
+
+        webpage = self._download_webpage(url, display_id)
+
+        title = self._html_search_meta(
+            'twitter:title', webpage, 'title', fatal=True)
+        description = self._html_search_meta(
+            'twitter:description', webpage, 'description')
+        url = self._html_search_regex(
+            r'articuloVideo = \"(.*?)\"', webpage, 'url')
+        real_id = self._search_regex(
+            r'[^0-9]([0-9]{7,})[^0-9]', url, 'id', default=display_id)
+        thumbnail = self._html_search_regex(
+            r'articuloImagen = \"(.*?)\"', webpage, 'thumbnail')
+
+        return {
+            'id': real_id,
+            'display_id': display_id,
+            'url': url,
+            'title': title,
+            'description': description,
+            'ext': 'mp4',
+            'thumbnail': thumbnail,
+        }

youtube_dl/extractor/ceskatelevize.py

@@ -0,0 +1,126 @@
+# -*- coding: utf-8 -*-
+from __future__ import unicode_literals
+
+import re
+
+from .common import InfoExtractor
+from ..utils import (
+    compat_urllib_request,
+    compat_urllib_parse,
+    compat_urllib_parse_urlparse,
+    ExtractorError,
+)
+
+
+class CeskaTelevizeIE(InfoExtractor):
+    _VALID_URL = r'https?://www\.ceskatelevize\.cz/(porady|ivysilani)/(.+/)?(?P<id>[^?#]+)'
+
+    _TESTS = [
+        {
+            'url': 'http://www.ceskatelevize.cz/ivysilani/10532695142-prvni-republika/213512120230004-spanelska-chripka',
+            'info_dict': {
+                'id': '213512120230004',
+                'ext': 'flv',
+                'title': 'První republika: Španělská chřipka',
+                'duration': 3107.4,
+            },
+            'params': {
+                'skip_download': True,  # requires rtmpdump
+            },
+            'skip': 'Works only from Czech Republic.',
+        },
+        {
+            'url': 'http://www.ceskatelevize.cz/ivysilani/1030584952-tsatsiki-maminka-a-policajt',
+            'info_dict': {
+                'id': '20138143440',
+                'ext': 'flv',
+                'title': 'Tsatsiki, maminka a policajt',
+                'duration': 6754.1,
+            },
+            'params': {
+                'skip_download': True,  # requires rtmpdump
+            },
+            'skip': 'Works only from Czech Republic.',
+        },
+        {
+            'url': 'http://www.ceskatelevize.cz/ivysilani/10532695142-prvni-republika/bonus/14716-zpevacka-z-duparny-bobina',
+            'info_dict': {
+                'id': '14716',
+                'ext': 'flv',
+                'title': 'První republika: Zpěvačka z Dupárny Bobina',
+                'duration': 90,
+            },
+            'params': {
+                'skip_download': True,  # requires rtmpdump
+            },
+        },
+    ]
+
+    def _real_extract(self, url):
+        url = url.replace('/porady/', '/ivysilani/').replace('/video/', '')
+
+        mobj = re.match(self._VALID_URL, url)
+        video_id = mobj.group('id')
+
+        webpage = self._download_webpage(url, video_id)
+
+        NOT_AVAILABLE_STRING = 'This content is not available at your territory due to limited copyright.'
+        if '%s</p>' % NOT_AVAILABLE_STRING in webpage:
+            raise ExtractorError(NOT_AVAILABLE_STRING, expected=True)
+
+        typ = self._html_search_regex(r'getPlaylistUrl\(\[\{"type":"(.+?)","id":".+?"\}\],', webpage, 'type')
+        episode_id = self._html_search_regex(r'getPlaylistUrl\(\[\{"type":".+?","id":"(.+?)"\}\],', webpage, 'episode_id')
+
+        data = {
+            'playlist[0][type]': typ,
+            'playlist[0][id]': episode_id,
+            'requestUrl': compat_urllib_parse_urlparse(url).path,
+            'requestSource': 'iVysilani',
+        }
+
+        req = compat_urllib_request.Request('http://www.ceskatelevize.cz/ivysilani/ajax/get-playlist-url',
+                                            data=compat_urllib_parse.urlencode(data))
+
+        req.add_header('Content-type', 'application/x-www-form-urlencoded')
+        req.add_header('x-addr', '127.0.0.1')
+        req.add_header('X-Requested-With', 'XMLHttpRequest')
+        req.add_header('Referer', url)
+
+        playlistpage = self._download_json(req, video_id)
+
+        req = compat_urllib_request.Request(compat_urllib_parse.unquote(playlistpage['url']))
+        req.add_header('Referer', url)
+
+        playlist = self._download_xml(req, video_id)
+        
+        formats = []
+        for i in playlist.find('smilRoot/body'):
+            if 'AD' not in i.attrib['id']:
+                base_url = i.attrib['base']
+                parsedurl = compat_urllib_parse_urlparse(base_url)
+                duration = i.attrib['duration']
+
+                for video in i.findall('video'):
+                    if video.attrib['label'] != 'AD':
+                        format_id = video.attrib['label']
+                        play_path = video.attrib['src']
+                        vbr = int(video.attrib['system-bitrate'])
+
+                        formats.append({
+                            'format_id': format_id,
+                            'url': base_url,
+                            'vbr': vbr,
+                            'play_path': play_path,
+                            'app': parsedurl.path[1:] + '?' + parsedurl.query,
+                            'rtmp_live': True,
+                            'ext': 'flv',
+                        })
+
+        self._sort_formats(formats)
+
+        return {
+            'id': episode_id,
+            'title': self._html_search_regex(r'<title>(.+?) — iVysílání — Česká televize</title>', webpage, 'title'),
+            'duration': float(duration),
+            'formats': formats,
+        }

youtube_dl/extractor/collegehumor.py

@@ -35,15 +35,15 @@ class CollegeHumorIE(InfoExtractor):
     },
     # embedded youtube video
     {
-        'url': 'http://www.collegehumor.com/embed/6950457',
+        'url': 'http://www.collegehumor.com/embed/6950306',
         'info_dict': {
-            'id': 'W5gMp3ZjYg4',
+            'id': 'Z-bao9fg6Yc',
             'ext': 'mp4',
-            'title': 'Funny Dogs Protecting Babies Compilation 2014 [NEW HD]',
-            'uploader': 'Funnyplox TV',
-            'uploader_id': 'funnyploxtv',
-            'description': 'md5:7ded37421526d54afdf005e25bc2b7a3',
-            'upload_date': '20140128',
+            'title': 'Young Americans Think President John F. Kennedy Died THIS MORNING IN A CAR ACCIDENT!!!',
+            'uploader': 'Mark Dice',
+            'uploader_id': 'MarkDice',
+            'description': 'md5:62c3dab9351fac7bb44b53b69511d87f',
+            'upload_date': '20140127',
         },
         'params': {
             'skip_download': True,

youtube_dl/extractor/common.py

@@ -88,6 +88,10 @@ class InfoExtractor(object):
 
     The following fields are optional:
 
+    display_id      An alternative identifier for the video, not necessarily
+                    unique, but available before title. Typically, id is
+                    something like "4234987", title "Dancing naked mole rats",
+                    and display_id "dancing-naked-mole-rats"
     thumbnails:     A list of dictionaries (with the entries "resolution" and
                     "url") for the varying thumbnails
     thumbnail:      Full URL to a video thumbnail image.
@@ -432,14 +436,14 @@ class InfoExtractor(object):
         if secure: regexes = self._og_regexes('video:secure_url') + regexes
         return self._html_search_regex(regexes, html, name, **kargs)
 
-    def _html_search_meta(self, name, html, display_name=None):
+    def _html_search_meta(self, name, html, display_name=None, fatal=False):
         if display_name is None:
             display_name = name
         return self._html_search_regex(
             r'''(?ix)<meta
                     (?=[^>]+(?:itemprop|name|property)=["\']%s["\'])
                     [^>]+content=["\']([^"\']+)["\']''' % re.escape(name),
-            html, display_name, fatal=False)
+            html, display_name, fatal=fatal)
 
     def _dc_search_uploader(self, html):
         return self._html_search_meta('dc.creator', html, 'uploader')

youtube_dl/extractor/dailymotion.py

@@ -12,6 +12,7 @@ from ..utils import (
     get_element_by_id,
     orderedSet,
     str_to_int,
+    int_or_none,
 
     ExtractorError,
 )
@@ -124,7 +125,7 @@ class DailymotionIE(DailymotionBaseInfoExtractor, SubtitlesInfoExtractor):
             if video_url is not None:
                 m_size = re.search(r'H264-(\d+)x(\d+)', video_url)
                 if m_size is not None:
-                    width, height = m_size.group(1), m_size.group(2)
+                    width, height = map(int_or_none, (m_size.group(1), m_size.group(2)))
                 else:
                     width, height = None, None
                 formats.append({

youtube_dl/extractor/facebook.py

@@ -1,3 +1,5 @@
+from __future__ import unicode_literals
+
 import json
 import re
 import socket
@@ -9,16 +11,15 @@ from ..utils import (
     compat_urllib_error,
     compat_urllib_parse,
     compat_urllib_request,
+    urlencode_postdata,
 
     ExtractorError,
 )
 
 
 class FacebookIE(InfoExtractor):
-    """Information Extractor for Facebook"""
-
     _VALID_URL = r'''(?x)
-        (?:https?://)?(?:\w+\.)?facebook\.com/
+        https?://(?:\w+\.)?facebook\.com/
         (?:[^#?]*\#!/)?
         (?:video/video\.php|photo\.php|video/embed)\?(?:.*?)
         (?:v|video_id)=(?P<id>[0-9]+)
@@ -26,21 +27,18 @@ class FacebookIE(InfoExtractor):
     _LOGIN_URL = 'https://www.facebook.com/login.php?next=http%3A%2F%2Ffacebook.com%2Fhome.php&login_attempt=1'
     _CHECKPOINT_URL = 'https://www.facebook.com/checkpoint/?next=http%3A%2F%2Ffacebook.com%2Fhome.php&_fb_noscript=1'
     _NETRC_MACHINE = 'facebook'
-    IE_NAME = u'facebook'
+    IE_NAME = 'facebook'
     _TEST = {
-        u'url': u'https://www.facebook.com/photo.php?v=120708114770723',
-        u'file': u'120708114770723.mp4',
-        u'md5': u'48975a41ccc4b7a581abd68651c1a5a8',
-        u'info_dict': {
-            u"duration": 279,
-            u"title": u"PEOPLE ARE AWESOME 2013"
+        'url': 'https://www.facebook.com/photo.php?v=120708114770723',
+        'md5': '48975a41ccc4b7a581abd68651c1a5a8',
+        'info_dict': {
+            'id': '120708114770723',
+            'ext': 'mp4',
+            'duration': 279,
+            'title': 'PEOPLE ARE AWESOME 2013',
         }
     }
 
-    def report_login(self):
-        """Report attempt to log in."""
-        self.to_screen(u'Logging in')
-
     def _login(self):
         (useremail, password) = self._get_login_info()
         if useremail is None:
@@ -48,11 +46,13 @@ class FacebookIE(InfoExtractor):
 
         login_page_req = compat_urllib_request.Request(self._LOGIN_URL)
         login_page_req.add_header('Cookie', 'locale=en_US')
-        self.report_login()
-        login_page = self._download_webpage(login_page_req, None, note=False,
-            errnote=u'Unable to download login page')
-        lsd = self._search_regex(r'"lsd":"(\w*?)"', login_page, u'lsd')
-        lgnrnd = self._search_regex(r'name="lgnrnd" value="([^"]*?)"', login_page, u'lgnrnd')
+        login_page = self._download_webpage(login_page_req, None,
+            note='Downloading login page',
+            errnote='Unable to download login page')
+        lsd = self._search_regex(
+            r'<input type="hidden" name="lsd" value="([^"]*)"',
+            login_page, 'lsd')
+        lgnrnd = self._search_regex(r'name="lgnrnd" value="([^"]*?)"', login_page, 'lgnrnd')
 
         login_form = {
             'email': useremail,
@@ -65,27 +65,29 @@ class FacebookIE(InfoExtractor):
             'timezone': '-60',
             'trynum': '1',
             }
-        request = compat_urllib_request.Request(self._LOGIN_URL, compat_urllib_parse.urlencode(login_form))
+        request = compat_urllib_request.Request(self._LOGIN_URL, urlencode_postdata(login_form))
         request.add_header('Content-Type', 'application/x-www-form-urlencoded')
         try:
-            login_results = compat_urllib_request.urlopen(request).read()
+            login_results = self._download_webpage(request, None,
+                note='Logging in', errnote='unable to fetch login page')
             if re.search(r'<form(.*)name="login"(.*)</form>', login_results) is not None:
-                self._downloader.report_warning(u'unable to log in: bad username/password, or exceded login rate limit (~3/min). Check credentials or wait.')
+                self._downloader.report_warning('unable to log in: bad username/password, or exceded login rate limit (~3/min). Check credentials or wait.')
                 return
 
             check_form = {
-                'fb_dtsg': self._search_regex(r'"fb_dtsg":"(.*?)"', login_results, u'fb_dtsg'),
-                'nh': self._search_regex(r'name="nh" value="(\w*?)"', login_results, u'nh'),
+                'fb_dtsg': self._search_regex(r'name="fb_dtsg" value="(.+?)"', login_results, 'fb_dtsg'),
+                'nh': self._search_regex(r'name="nh" value="(\w*?)"', login_results, 'nh'),
                 'name_action_selected': 'dont_save',
-                'submit[Continue]': self._search_regex(r'<input value="(.*?)" name="submit\[Continue\]"', login_results, u'continue'),
+                'submit[Continue]': self._search_regex(r'<button[^>]+value="(.*?)"[^>]+name="submit\[Continue\]"', login_results, 'continue'),
             }
-            check_req = compat_urllib_request.Request(self._CHECKPOINT_URL, compat_urllib_parse.urlencode(check_form))
+            check_req = compat_urllib_request.Request(self._CHECKPOINT_URL, urlencode_postdata(check_form))
             check_req.add_header('Content-Type', 'application/x-www-form-urlencoded')
-            check_response = compat_urllib_request.urlopen(check_req).read()
+            check_response = self._download_webpage(check_req, None,
+                note='Confirming login')
             if re.search(r'id="checkpointSubmitButton"', check_response) is not None:
-                self._downloader.report_warning(u'Unable to confirm login, you have to login in your brower and authorize the login.')
+                self._downloader.report_warning('Unable to confirm login, you have to login in your brower and authorize the login.')
         except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
-            self._downloader.report_warning(u'unable to log in: %s' % compat_str(err))
+            self._downloader.report_warning('unable to log in: %s' % compat_str(err))
             return
 
     def _real_initialize(self):
@@ -93,8 +95,6 @@ class FacebookIE(InfoExtractor):
 
     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')
 
         url = 'https://www.facebook.com/video/video.php?v=%s' % video_id
@@ -107,10 +107,10 @@ class FacebookIE(InfoExtractor):
             m_msg = re.search(r'class="[^"]*uiInterstitialContent[^"]*"><div>(.*?)</div>', webpage)
             if m_msg is not None:
                 raise ExtractorError(
-                    u'The video is not available, Facebook said: "%s"' % m_msg.group(1),
+                    'The video is not available, Facebook said: "%s"' % m_msg.group(1),
                     expected=True)
             else:
-                raise ExtractorError(u'Cannot parse data')
+                raise ExtractorError('Cannot parse data')
         data = dict(json.loads(m.group(1)))
         params_raw = compat_urllib_parse.unquote(data['params'])
         params = json.loads(params_raw)
@@ -119,19 +119,15 @@ class FacebookIE(InfoExtractor):
         if not video_url:
             video_url = video_data['sd_src']
         if not video_url:
-            raise ExtractorError(u'Cannot find video URL')
-        video_duration = int(video_data['video_duration'])
-        thumbnail = video_data['thumbnail_src']
+            raise ExtractorError('Cannot find video URL')
 
         video_title = self._html_search_regex(
-            r'<h2 class="uiHeaderTitle">([^<]*)</h2>', webpage, u'title')
+            r'<h2 class="uiHeaderTitle">([^<]*)</h2>', webpage, 'title')
 
-        info = {
+        return {
             'id': video_id,
             'title': video_title,
             'url': video_url,
-            'ext': 'mp4',
-            'duration': video_duration,
-            'thumbnail': thumbnail,
+            'duration': int(video_data['video_duration']),
+            'thumbnail': video_data['thumbnail_src'],
         }
-        return [info]

youtube_dl/extractor/fourtube.py

@@ -8,8 +8,8 @@ from ..utils import (
     unified_strdate,
     str_to_int,
     parse_duration,
+    clean_html,
 )
-from youtube_dl.utils import clean_html
 
 
 class FourTubeIE(InfoExtractor):

youtube_dl/extractor/gamekings.py

@@ -1,3 +1,5 @@
+from __future__ import unicode_literals
+
 import re
 
 from .common import InfoExtractor
@@ -6,13 +8,14 @@ from .common import InfoExtractor
 class GamekingsIE(InfoExtractor):
     _VALID_URL = r'http://www\.gamekings\.tv/videos/(?P<name>[0-9a-z\-]+)'
     _TEST = {
-        u"url": u"http://www.gamekings.tv/videos/phoenix-wright-ace-attorney-dual-destinies-review/",
-        u'file': u'20130811.mp4',
+        'url': 'http://www.gamekings.tv/videos/phoenix-wright-ace-attorney-dual-destinies-review/',
         # MD5 is flaky, seems to change regularly
-        #u'md5': u'2f32b1f7b80fdc5cb616efb4f387f8a3',
+        # 'md5': '2f32b1f7b80fdc5cb616efb4f387f8a3',
         u'info_dict': {
-            u"title": u"Phoenix Wright: Ace Attorney \u2013 Dual Destinies Review",
-            u"description": u"Melle en Steven hebben voor de review een week in de rechtbank doorbracht met Phoenix Wright: Ace Attorney - Dual Destinies.",
+            'id': '20130811',
+            'ext': 'mp4',
+            'title': 'Phoenix Wright: Ace Attorney \u2013 Dual Destinies Review',
+            'description': 'md5:632e61a9f97d700e83f43d77ddafb6a4',
         }
     }
 

youtube_dl/extractor/gdcvault.py

@@ -0,0 +1,134 @@
+from __future__ import unicode_literals
+
+import re
+
+from .common import InfoExtractor
+from ..utils import (
+    compat_urllib_parse,
+    compat_urllib_request,
+)
+
+class GDCVaultIE(InfoExtractor):
+    _VALID_URL = r'https?://(?:www\.)?gdcvault\.com/play/(?P<id>\d+)/(?P<name>(\w|-)+)'
+    _TESTS = [
+        {
+            'url': 'http://www.gdcvault.com/play/1019721/Doki-Doki-Universe-Sweet-Simple',
+            'md5': '7ce8388f544c88b7ac11c7ab1b593704',
+            'info_dict': {
+                'id': '1019721',
+                'ext': 'mp4',
+                'title': 'Doki-Doki Universe: Sweet, Simple and Genuine (GDC Next 10)'
+            }
+        },
+        {
+            'url': 'http://www.gdcvault.com/play/1015683/Embracing-the-Dark-Art-of',
+            'info_dict': {
+                'id': '1015683',
+                'ext': 'flv',
+                'title': 'Embracing the Dark Art of Mathematical Modeling in AI'
+            },
+            'params': {
+                'skip_download': True,  # Requires rtmpdump
+            }
+        },
+    ]
+
+    def _parse_mp4(self, xml_description):
+        video_formats = []
+        mp4_video = xml_description.find('./metadata/mp4video')
+        if mp4_video is None:
+            return None
+
+        mobj = re.match(r'(?P<root>https?://.*?/).*', mp4_video.text)
+        video_root = mobj.group('root')
+        formats = xml_description.findall('./metadata/MBRVideos/MBRVideo')
+        for format in formats:
+            mobj = re.match(r'mp4\:(?P<path>.*)', format.find('streamName').text)
+            url = video_root + mobj.group('path')
+            vbr = format.find('bitrate').text
+            video_formats.append({
+                'url': url,
+                'vbr': int(vbr),
+            })
+        return video_formats
+
+    def _parse_flv(self, xml_description):
+        video_formats = []
+        akami_url = xml_description.find('./metadata/akamaiHost').text
+        slide_video_path = xml_description.find('./metadata/slideVideo').text
+        video_formats.append({
+            'url': 'rtmp://' + akami_url + '/' + slide_video_path,
+            'format_note': 'slide deck video',
+            'quality': -2,
+            'preference': -2,
+            'format_id': 'slides',
+        })
+        speaker_video_path = xml_description.find('./metadata/speakerVideo').text
+        video_formats.append({
+            'url': 'rtmp://' + akami_url + '/' + speaker_video_path,
+            'format_note': 'speaker video',
+            'quality': -1,
+            'preference': -1,
+            'format_id': 'speaker',
+        })
+        return video_formats
+
+    def _login(self, webpage_url, video_id):
+        (username, password) = self._get_login_info()
+        if username is None or password is None:
+            self.report_warning('It looks like ' + webpage_url + ' requires a login. Try specifying a username and password and try again.')
+            return None
+
+        mobj = re.match(r'(?P<root_url>https?://.*?/).*', webpage_url)
+        login_url = mobj.group('root_url') + 'api/login.php'
+        logout_url = mobj.group('root_url') + 'logout'
+
+        login_form = {
+            'email': username,
+            'password': password,
+        }
+
+        request = compat_urllib_request.Request(login_url, compat_urllib_parse.urlencode(login_form))
+        request.add_header('Content-Type', 'application/x-www-form-urlencoded')
+        self._download_webpage(request, video_id, 'Logging in')
+        start_page = self._download_webpage(webpage_url, video_id, 'Getting authenticated video page')
+        self._download_webpage(logout_url, video_id, 'Logging out')
+
+        return start_page
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+
+        video_id = mobj.group('id')
+        webpage_url = 'http://www.gdcvault.com/play/' + video_id
+        start_page = self._download_webpage(webpage_url, video_id)
+
+        xml_root = self._html_search_regex(r'<iframe src="(?P<xml_root>.*?)player.html.*?".*?</iframe>', start_page, 'xml root', None, False)
+
+        if xml_root is None:
+            # Probably need to authenticate
+            start_page = self._login(webpage_url, video_id)
+            if start_page is None:
+                self.report_warning('Could not login.')
+            else:
+                # Grab the url from the authenticated page
+                xml_root = self._html_search_regex(r'<iframe src="(?P<xml_root>.*?)player.html.*?".*?</iframe>', start_page, 'xml root')
+
+        xml_name = self._html_search_regex(r'<iframe src=".*?\?xml=(?P<xml_file>.+?\.xml).*?".*?</iframe>', start_page, 'xml filename', None, False)
+        if xml_name is None:
+            # Fallback to the older format
+            xml_name = self._html_search_regex(r'<iframe src=".*?\?xmlURL=xml/(?P<xml_file>.+?\.xml).*?".*?</iframe>', start_page, 'xml filename')
+
+        xml_decription_url = xml_root + 'xml/' + xml_name
+        xml_description = self._download_xml(xml_decription_url, video_id)
+
+        video_title = xml_description.find('./metadata/title').text
+        video_formats = self._parse_mp4(xml_description)
+        if video_formats is None:
+            video_formats = self._parse_flv(xml_description)
+
+        return {
+            'id': video_id,
+            'title': video_title,
+            'formats': video_formats,
+        }

youtube_dl/extractor/generic.py

@@ -83,10 +83,10 @@ class GenericIE(InfoExtractor):
         # Direct link to a video
         {
             'url': 'http://media.w3.org/2010/05/sintel/trailer.mp4',
-            'file': 'trailer.mp4',
             'md5': '67d406c2bcb6af27fa886f31aa934bbe',
             'info_dict': {
                 'id': 'trailer',
+                'ext': 'mp4',
                 'title': 'trailer',
                 'upload_date': '20100513',
             }
@@ -94,7 +94,6 @@ class GenericIE(InfoExtractor):
         # ooyala video
         {
             'url': 'http://www.rollingstone.com/music/videos/norwegian-dj-cashmere-cat-goes-spartan-on-with-me-premiere-20131219',
-            'file': 'BwY2RxaTrTkslxOfcan0UCf0YqyvWysJ.mp4',
             'md5': '5644c6ca5d5782c1d0d350dad9bd840c',
             'info_dict': {
                 'id': 'BwY2RxaTrTkslxOfcan0UCf0YqyvWysJ',
@@ -102,6 +101,39 @@ class GenericIE(InfoExtractor):
                 'title': '2cc213299525360.mov',  # that's what we get
             },
         },
+        # google redirect
+        {
+            'url': 'http://www.google.com/url?sa=t&rct=j&q=&esrc=s&source=web&cd=1&cad=rja&ved=0CCUQtwIwAA&url=http%3A%2F%2Fwww.youtube.com%2Fwatch%3Fv%3DcmQHVoWB5FY&ei=F-sNU-LLCaXk4QT52ICQBQ&usg=AFQjCNEw4hL29zgOohLXvpJ-Bdh2bils1Q&bvm=bv.61965928,d.bGE',
+            'info_dict': {
+                'id': 'cmQHVoWB5FY',
+                'ext': 'mp4',
+                'upload_date': '20130224',
+                'uploader_id': 'TheVerge',
+                'description': 'Chris Ziegler takes a look at the Alcatel OneTouch Fire and the ZTE Open; two of the first Firefox OS handsets to be officially announced.',
+                'uploader': 'The Verge',
+                'title': 'First Firefox OS phones side-by-side',
+            },
+            'params': {
+                'skip_download': False,
+            }
+        },
+        # embed.ly video
+        {
+            'url': 'http://www.tested.com/science/weird/460206-tested-grinding-coffee-2000-frames-second/',
+            'info_dict': {
+                'id': '9ODmcdjQcHQ',
+                'ext': 'mp4',
+                'title': 'Tested: Grinding Coffee at 2000 Frames Per Second',
+                'upload_date': '20140225',
+                'description': 'md5:06a40fbf30b220468f1e0957c0f558ff',
+                'uploader': 'Tested',
+                'uploader_id': 'testedcom',
+            },
+            # No need to test YoutubeIE here
+            'params': {
+                'skip_download': True,
+            },
+        },
     ]
 
     def report_download_webpage(self, video_id):
@@ -196,7 +228,7 @@ class GenericIE(InfoExtractor):
             else:
                 assert ':' in default_search
                 return self.url_result(default_search + url)
-        video_id = os.path.splitext(url.split('/')[-1])[0]
+        video_id = os.path.splitext(url.rstrip('/').split('/')[-1])[0]
 
         self.to_screen('%s: Requesting header' % video_id)
 
@@ -381,12 +413,25 @@ class GenericIE(InfoExtractor):
         if mobj is not None:
             return self.url_result(mobj.group('url'), 'Facebook')
 
+        # Look for embedded VK player
+        mobj = re.search(r'<iframe[^>]+?src=(["\'])(?P<url>https?://vk\.com/video_ext\.php.+?)\1', webpage)
+        if mobj is not None:
+            return self.url_result(mobj.group('url'), 'VK')
+
         # Look for embedded Huffington Post player
         mobj = re.search(
             r'<iframe[^>]+?src=(["\'])(?P<url>https?://embed\.live\.huffingtonpost\.com/.+?)\1', webpage)
         if mobj is not None:
             return self.url_result(mobj.group('url'), 'HuffPost')
 
+        # Look for embed.ly
+        mobj = re.search(r'class=["\']embedly-card["\'][^>]href=["\'](?P<url>[^"\']+)', webpage)
+        if mobj is not None:
+            return self.url_result(mobj.group('url'))
+        mobj = re.search(r'class=["\']embedly-embed["\'][^>]src=["\'][^"\']*url=(?P<url>[^&]+)', webpage)
+        if mobj is not None:
+            return self.url_result(compat_urllib_parse.unquote(mobj.group('url')))
+
         # Start with something easy: JW Player in SWFObject
         mobj = re.search(r'flashvars: [\'"](?:.*&)?file=(http[^\'"&]*)', webpage)
         if mobj is None:
@@ -411,6 +456,18 @@ class GenericIE(InfoExtractor):
         if mobj is None:
             # HTML5 video
             mobj = re.search(r'<video[^<]*(?:>.*?<source.*?)? src="([^"]+)"', webpage, flags=re.DOTALL)
+        if mobj is None:
+            mobj = re.search(
+                r'(?i)<meta\s+(?=(?:[a-z-]+="[^"]+"\s+)*http-equiv="refresh")'
+                r'(?:[a-z-]+="[^"]+"\s+)*?content="[0-9]{,2};url=\'([^\']+)\'"',
+                webpage)
+            if mobj:
+                new_url = mobj.group(1)
+                self.report_following_redirect(new_url)
+                return {
+                    '_type': 'url',
+                    'url': new_url,
+                }
         if mobj is None:
             raise ExtractorError('Unsupported URL: %s' % url)
 

youtube_dl/extractor/jukebox.py

@@ -1,56 +1,61 @@
-# coding: utf-8
+from __future__ import unicode_literals
+
 import re
 
 from .common import InfoExtractor
 from ..utils import (
     ExtractorError,
+    RegexNotFoundError,
     unescapeHTML,
 )
 
+
 class JukeboxIE(InfoExtractor):
     _VALID_URL = r'^http://www\.jukebox?\..+?\/.+[,](?P<video_id>[a-z0-9\-]+)\.html'
-    _IFRAME = r'<iframe .*src="(?P<iframe>[^"]*)".*>'
-    _VIDEO_URL = r'"config":{"file":"(?P<video_url>http:[^"]+[.](?P<video_ext>[^.?]+)[?]mdtk=[0-9]+)"'
-    _TITLE = r'<h1 class="inline">(?P<title>[^<]+)</h1>.*<span id="infos_article_artist">(?P<artist>[^<]+)</span>'
-    _IS_YOUTUBE = r'config":{"file":"(?P<youtube_url>http:[\\][/][\\][/]www[.]youtube[.]com[\\][/]watch[?]v=[^"]+)"'
+    _TEST = {
+        'url': 'http://www.jukebox.es/kosheen/videoclip,pride,r303r.html',
+        'md5': '5dc6477e74b1e37042ac5acedd8413e5',
+        'info_dict': {
+            'id': 'r303r',
+            'ext': 'flv',
+            'title': 'Kosheen-En Vivo Pride',
+            'uploader': 'Kosheen',
+        },
+    }
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
         video_id = mobj.group('video_id')
 
         html = self._download_webpage(url, video_id)
-
-        mobj = re.search(self._IFRAME, html)
-        if mobj is None:
-            raise ExtractorError(u'Cannot extract iframe url')
-        iframe_url = unescapeHTML(mobj.group('iframe'))
+        iframe_url = unescapeHTML(self._search_regex(r'<iframe .*src="([^"]*)"', html, 'iframe url'))
 
         iframe_html = self._download_webpage(iframe_url, video_id, 'Downloading iframe')
-        mobj = re.search(r'class="jkb_waiting"', iframe_html)
-        if mobj is not None:
-            raise ExtractorError(u'Video is not available(in your country?)!')
+        if re.search(r'class="jkb_waiting"', iframe_html) is not None:
+            raise ExtractorError('Video is not available(in your country?)!')
 
         self.report_extraction(video_id)
 
-        mobj = re.search(self._VIDEO_URL, iframe_html)
-        if mobj is None:
-            mobj = re.search(self._IS_YOUTUBE, iframe_html)
-            if mobj is None:
-                raise ExtractorError(u'Cannot extract video url')
-            youtube_url = unescapeHTML(mobj.group('youtube_url')).replace('\/','/')
-            self.to_screen(u'Youtube video detected')
-            return self.url_result(youtube_url,ie='Youtube')
-        video_url = unescapeHTML(mobj.group('video_url')).replace('\/','/')
-        video_ext = unescapeHTML(mobj.group('video_ext'))
+        try:
+            video_url = self._search_regex(r'"config":{"file":"(?P<video_url>http:[^"]+\?mdtk=[0-9]+)"',
+                iframe_html, 'video url')
+            video_url = unescapeHTML(video_url).replace('\/', '/')
+        except RegexNotFoundError:
+            youtube_url = self._search_regex(
+                r'config":{"file":"(http:\\/\\/www\.youtube\.com\\/watch\?v=[^"]+)"',
+                iframe_html, 'youtube url')
+            youtube_url = unescapeHTML(youtube_url).replace('\/', '/')
+            self.to_screen('Youtube video detected')
+            return self.url_result(youtube_url, ie='Youtube')
 
-        mobj = re.search(self._TITLE, html)
-        if mobj is None:
-            raise ExtractorError(u'Cannot extract title')
-        title = unescapeHTML(mobj.group('title'))
-        artist = unescapeHTML(mobj.group('artist'))
+        title = self._html_search_regex(r'<h1 class="inline">([^<]+)</h1>',
+            html, 'title')
+        artist = self._html_search_regex(r'<span id="infos_article_artist">([^<]+)</span>',
+            html, 'artist')
 
-        return [{'id': video_id,
-                 'url': video_url,
-                 'title': artist + '-' + title,
-                 'ext': video_ext
-                 }]
+        return {
+            'id': video_id,
+            'url': video_url,
+            'title': artist + '-' + title,
+            'uploader': artist,
+        }

youtube_dl/extractor/lifenews.py

@@ -6,7 +6,8 @@ import re
 from .common import InfoExtractor
 from ..utils import (
     int_or_none,
-    unified_strdate
+    unified_strdate,
+    ExtractorError,
 )
 
 
@@ -32,13 +33,11 @@ class LifeNewsIE(InfoExtractor):
         mobj = re.match(self._VALID_URL, url)
         video_id = mobj.group('id')
 
-        webpage = self._download_webpage('http://lifenews.ru/mobile/news/%s' % video_id, video_id, 'Downloading page')
+        webpage = self._download_webpage('http://lifenews.ru/news/%s' % video_id, video_id, 'Downloading page')
 
-        video_url = self._html_search_regex(
-            r'<video.*?src="([^"]+)".*?></video>', webpage, 'video URL')
-
-        thumbnail = self._html_search_regex(
-            r'<video.*?poster="([^"]+)".*?"></video>', webpage, 'video thumbnail')
+        videos = re.findall(r'<video.*?poster="(?P<poster>[^"]+)".*?src="(?P<video>[^"]+)".*?></video>', webpage)
+        if not videos:
+            raise ExtractorError('No media links available for %s' % video_id)
 
         title = self._og_search_title(webpage)
         TITLE_SUFFIX = ' - Первый по срочным новостям — LIFE | NEWS'
@@ -50,20 +49,26 @@ class LifeNewsIE(InfoExtractor):
         view_count = self._html_search_regex(
             r'<div class=\'views\'>(\d+)</div>', webpage, 'view count', fatal=False)
         comment_count = self._html_search_regex(
-            r'<div class=\'comments\'>(\d+)</div>', webpage, 'comment count', fatal=False)
+            r'<div class=\'comments\'>\s*<span class=\'counter\'>(\d+)</span>', webpage, 'comment count', fatal=False)
 
         upload_date = self._html_search_regex(
             r'<time datetime=\'([^\']+)\'>', webpage, 'upload date',fatal=False)
         if upload_date is not None:
             upload_date = unified_strdate(upload_date)
 
-        return {
-            'id': video_id,
-            'url': video_url,
-            'thumbnail': thumbnail,
-            'title': title,
-            'description': description,
-            'view_count': int_or_none(view_count),
-            'comment_count': int_or_none(comment_count),
-            'upload_date': upload_date,
-        }
+        def make_entry(video_id, media, video_number=None):
+            return {
+                'id': video_id,
+                'url': media[1],
+                'thumbnail': media[0],
+                'title': title if video_number is None else '%s-video%s' % (title, video_number),
+                'description': description,
+                'view_count': int_or_none(view_count),
+                'comment_count': int_or_none(comment_count),
+                'upload_date': upload_date,
+            }
+
+        if len(videos) == 1:
+            return make_entry(video_id, videos[0])
+        else:
+            return [make_entry(video_id, media, video_number+1) for video_number, media in enumerate(videos)]

youtube_dl/extractor/lynda.py

@@ -8,7 +8,9 @@ from .common import InfoExtractor
 from ..utils import (
     compat_urllib_parse,
     compat_urllib_request,
-    ExtractorError
+    ExtractorError,
+    int_or_none,
+    compat_str,
 )
 
 
@@ -19,16 +21,17 @@ class LyndaIE(SubtitlesInfoExtractor):
     _LOGIN_URL = 'https://www.lynda.com/login/login.aspx'
     _NETRC_MACHINE = 'lynda'
 
-    _SUCCESSFUL_LOGIN_REGEX = r'<a href="https://www.lynda.com/home/userAccount/ChangeContactInfo.aspx" data-qa="eyebrow_account_menu">My account'
+    _SUCCESSFUL_LOGIN_REGEX = r'isLoggedIn: true'
     _TIMECODE_REGEX = r'\[(?P<timecode>\d+:\d+:\d+[\.,]\d+)\]'
 
     ACCOUNT_CREDENTIALS_HINT = 'Use --username and --password options to provide lynda.com account credentials.'
 
     _TEST = {
         'url': 'http://www.lynda.com/Bootstrap-tutorials/Using-exercise-files/110885/114408-4.html',
-        'file': '114408.mp4',
         'md5': 'ecfc6862da89489161fb9cd5f5a6fac1',
         'info_dict': {
+            'id': '114408',
+            'ext': 'mp4',
             'title': 'Using the exercise files',
             'duration': 68
         }
@@ -41,27 +44,44 @@ class LyndaIE(SubtitlesInfoExtractor):
         mobj = re.match(self._VALID_URL, url)
         video_id = mobj.group(1)
 
-        page = self._download_webpage('http://www.lynda.com/ajax/player?videoId=%s&type=video' % video_id,
-                                      video_id, 'Downloading video JSON')
+        page = self._download_webpage('http://www.lynda.com/ajax/player?videoId=%s&type=video' % video_id, video_id,
+            'Downloading video JSON')
         video_json = json.loads(page)
 
         if 'Status' in video_json:
             raise ExtractorError('lynda returned error: %s' % video_json['Message'], expected=True)
 
         if video_json['HasAccess'] is False:
-            raise ExtractorError('Video %s is only available for members. ' % video_id + self.ACCOUNT_CREDENTIALS_HINT, expected=True)
+            raise ExtractorError(
+                'Video %s is only available for members. ' % video_id + self.ACCOUNT_CREDENTIALS_HINT, expected=True)
 
-        video_id = video_json['ID']
+        video_id = compat_str(video_json['ID'])
         duration = video_json['DurationInSeconds']
         title = video_json['Title']
 
-        formats = [{'url': fmt['Url'],
+        formats = []
+
+        fmts = video_json.get('Formats')
+        if fmts:
+            formats.extend([
+                {
+                    'url': fmt['Url'],
                     'ext': fmt['Extension'],
                     'width': fmt['Width'],
                     'height': fmt['Height'],
                     'filesize': fmt['FileSize'],
                     'format_id': str(fmt['Resolution'])
-                    } for fmt in video_json['Formats']]
+                } for fmt in fmts])
+
+        prioritized_streams = video_json.get('PrioritizedStreams')
+        if prioritized_streams:
+            formats.extend([
+                {
+                    'url': video_url,
+                    'width': int_or_none(format_id),
+                    'format_id': format_id,
+                } for format_id, video_url in prioritized_streams['0'].items()
+            ])
 
         self._sort_formats(formats)
 
@@ -91,7 +111,7 @@ class LyndaIE(SubtitlesInfoExtractor):
             'stayPut': 'false'
         }        
         request = compat_urllib_request.Request(self._LOGIN_URL, compat_urllib_parse.urlencode(login_form))
-        login_page = self._download_webpage(request, None, note='Logging in as %s' % username)
+        login_page = self._download_webpage(request, None, 'Logging in as %s' % username)
 
         # Not (yet) logged in
         m = re.search(r'loginResultJson = \'(?P<json>[^\']+)\';', login_page)
@@ -116,7 +136,7 @@ class LyndaIE(SubtitlesInfoExtractor):
                     'stayPut': 'false',
                 }
                 request = compat_urllib_request.Request(self._LOGIN_URL, compat_urllib_parse.urlencode(confirm_form))
-                login_page = self._download_webpage(request, None, note='Confirming log in and log out from another device')
+                login_page = self._download_webpage(request, None, 'Confirming log in and log out from another device')
 
         if re.search(self._SUCCESSFUL_LOGIN_REGEX, login_page) is None:
             raise ExtractorError('Unable to log in')
@@ -150,7 +170,7 @@ class LyndaIE(SubtitlesInfoExtractor):
 
     def _get_available_subtitles(self, video_id, webpage):
         url = 'http://www.lynda.com/ajax/player?videoId=%s&type=transcript' % video_id
-        sub = self._download_webpage(url, None, note=False)
+        sub = self._download_webpage(url, None, False)
         sub_json = json.loads(sub)
         return {'en': url} if len(sub_json) > 0 else {}
 
@@ -179,6 +199,9 @@ class LyndaCourseIE(InfoExtractor):
         videos = []
         (username, _) = self._get_login_info()
 
+        # Might want to extract videos right here from video['Formats'] as it seems 'Formats' is not provided
+        # by single video API anymore
+
         for chapter in course_json['Chapters']:
             for video in chapter['Videos']:
                 if username is None and video['HasAccess'] is False:

youtube_dl/extractor/mailru.py

@@ -0,0 +1,66 @@
+# encoding: utf-8
+from __future__ import unicode_literals
+
+import re
+import datetime
+
+from .common import InfoExtractor
+
+
+class MailRuIE(InfoExtractor):
+    IE_NAME = 'mailru'
+    IE_DESC = 'Видео@Mail.Ru'
+    _VALID_URL = r'http://(?:www\.)?my\.mail\.ru/video/.*#video=/?(?P<id>[^/]+/[^/]+/[^/]+/\d+)'
+
+    _TEST = {
+        'url': 'http://my.mail.ru/video/top#video=/mail/sonypicturesrus/75/76',
+        'md5': 'dea205f03120046894db4ebb6159879a',
+        'info_dict': {
+            'id': '46301138',
+            'ext': 'mp4',
+            'title': 'Новый Человек-Паук. Высокое напряжение. Восстание Электро',
+            'upload_date': '20140224',
+            'uploader': 'sonypicturesrus',
+            'uploader_id': 'sonypicturesrus@mail.ru',
+            'duration': 184,
+        }
+    }
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        video_id = mobj.group('id')
+
+        video_data = self._download_json(
+            'http://videoapi.my.mail.ru/videos/%s.json?new=1' % video_id, video_id, 'Downloading video JSON')
+
+        author = video_data['author']
+        uploader = author['name']
+        uploader_id = author['id']
+
+        movie = video_data['movie']
+        content_id = str(movie['contentId'])
+        title = movie['title']
+        thumbnail = movie['poster']
+        duration = movie['duration']
+
+        upload_date = datetime.datetime.fromtimestamp(video_data['timestamp']).strftime('%Y%m%d')
+        view_count = video_data['views_count']
+
+        formats = [
+            {
+                'url': video['url'],
+                'format_id': video['name'],
+            } for video in video_data['videos']
+        ]
+
+        return {
+            'id': content_id,
+            'title': title,
+            'thumbnail': thumbnail,
+            'upload_date': upload_date,
+            'uploader': uploader,
+            'uploader_id': uploader_id,
+            'duration': duration,
+            'view_count': view_count,
+            'formats': formats,
+        }

youtube_dl/extractor/mit.py

@@ -8,6 +8,7 @@ from .youtube import YoutubeIE
 from ..utils import (
     compat_urlparse,
     clean_html,
+    ExtractorError,
     get_element_by_id,
 )
 
@@ -72,7 +73,6 @@ class MITIE(TechTVMITIE):
 
     _TEST = {
         'url': 'http://video.mit.edu/watch/the-government-is-profiling-you-13222/',
-        'file': '.mp4',
         'md5': '7db01d5ccc1895fc5010e9c9e13648da',
         'info_dict': {
             'id': '21783',

youtube_dl/extractor/mixcloud.py

@@ -5,6 +5,7 @@ import re
 from .common import InfoExtractor
 from ..utils import (
     unified_strdate,
+    compat_urllib_parse,
     ExtractorError,
 )
 
@@ -15,8 +16,9 @@ class MixcloudIE(InfoExtractor):
 
     _TEST = {
         'url': 'http://www.mixcloud.com/dholbach/cryptkeeper/',
-        'file': 'dholbach-cryptkeeper.mp3',
         'info_dict': {
+            'id': 'dholbach-cryptkeeper',
+            'ext': 'mp3',
             'title': 'Cryptkeeper',
             'description': 'After quite a long silence from myself, finally another Drum\'n\'Bass mix with my favourite current dance floor bangers.',
             'uploader': 'Daniel Holbach',
@@ -45,7 +47,7 @@ class MixcloudIE(InfoExtractor):
         mobj = re.match(self._VALID_URL, url)
         uploader = mobj.group(1)
         cloudcast_name = mobj.group(2)
-        track_id = '-'.join((uploader, cloudcast_name))
+        track_id = compat_urllib_parse.unquote('-'.join((uploader, cloudcast_name)))
 
         webpage = self._download_webpage(url, track_id)
 

youtube_dl/extractor/mtv.py

@@ -5,9 +5,12 @@ import re
 from .common import InfoExtractor
 from ..utils import (
     compat_urllib_parse,
+    compat_urllib_request,
     ExtractorError,
     find_xpath_attr,
     fix_xml_ampersands,
+    HEADRequest,
+    unescapeHTML,
     url_basename,
     RegexNotFoundError,
 )
@@ -18,6 +21,7 @@ def _media_xml_tag(tag):
 
 
 class MTVServicesInfoExtractor(InfoExtractor):
+    _MOBILE_TEMPLATE = None
     @staticmethod
     def _id_from_uri(uri):
         return uri.split(':')[-1]
@@ -39,9 +43,29 @@ class MTVServicesInfoExtractor(InfoExtractor):
         else:
             return thumb_node.attrib['url']
 
-    def _extract_video_formats(self, mdoc):
-        if re.match(r'.*/error_country_block\.swf$', mdoc.find('.//src').text) is not None:
-            raise ExtractorError('This video is not available from your country.', expected=True)
+    def _extract_mobile_video_formats(self, mtvn_id):
+        webpage_url = self._MOBILE_TEMPLATE % mtvn_id
+        req = compat_urllib_request.Request(webpage_url)
+        # Otherwise we get a webpage that would execute some javascript
+        req.add_header('Youtubedl-user-agent', 'curl/7')
+        webpage = self._download_webpage(req, mtvn_id,
+            'Downloading mobile page')
+        metrics_url = unescapeHTML(self._search_regex(r'<a href="(http://metrics.+?)"', webpage, 'url'))
+        req = HEADRequest(metrics_url)
+        response = self._request_webpage(req, mtvn_id, 'Resolving url')
+        url = response.geturl()
+        # Transform the url to get the best quality:
+        url = re.sub(r'.+pxE=mp4', 'http://mtvnmobile.vo.llnwd.net/kip0/_pxn=0+_pxK=18639+_pxE=mp4', url, 1)
+        return [{'url': url,'ext': 'mp4'}]
+
+    def _extract_video_formats(self, mdoc, mtvn_id):
+        if re.match(r'.*/(error_country_block\.swf|geoblock\.mp4)$', mdoc.find('.//src').text) is not None:
+            if mtvn_id is not None and self._MOBILE_TEMPLATE is not None:
+                self.to_screen('The normal version is not available from your '
+                    'country, trying with the mobile version')
+                return self._extract_mobile_video_formats(mtvn_id)
+            raise ExtractorError('This video is not available from your country.',
+                expected=True)
 
         formats = []
         for rendition in mdoc.findall('.//rendition'):
@@ -94,9 +118,16 @@ class MTVServicesInfoExtractor(InfoExtractor):
             raise ExtractorError('Could not find video title')
         title = title.strip()
 
+        # This a short id that's used in the webpage urls
+        mtvn_id = None
+        mtvn_id_node = find_xpath_attr(itemdoc, './/{http://search.yahoo.com/mrss/}category',
+                'scheme', 'urn:mtvn:id')
+        if mtvn_id_node is not None:
+            mtvn_id = mtvn_id_node.text
+
         return {
             'title': title,
-            'formats': self._extract_video_formats(mediagen_doc),
+            'formats': self._extract_video_formats(mediagen_doc, mtvn_id),
             'id': video_id,
             'thumbnail': self._get_thumbnail_url(uri, itemdoc),
             'description': description,

youtube_dl/extractor/myvideo.py

@@ -1,3 +1,5 @@
+from __future__ import unicode_literals
+
 import binascii
 import base64
 import hashlib
@@ -14,18 +16,16 @@ from ..utils import (
 )
 
 
-
 class MyVideoIE(InfoExtractor):
-    """Information Extractor for myvideo.de."""
-
-    _VALID_URL = r'(?:http://)?(?:www\.)?myvideo\.de/(?:[^/]+/)?watch/([0-9]+)/([^?/]+).*'
-    IE_NAME = u'myvideo'
+    _VALID_URL = r'http://(?:www\.)?myvideo\.de/(?:[^/]+/)?watch/(?P<id>[0-9]+)/[^?/]+.*'
+    IE_NAME = 'myvideo'
     _TEST = {
-        u'url': u'http://www.myvideo.de/watch/8229274/bowling_fail_or_win',
-        u'file': u'8229274.flv',
-        u'md5': u'2d2753e8130479ba2cb7e0a37002053e',
-        u'info_dict': {
-            u"title": u"bowling-fail-or-win"
+        'url': 'http://www.myvideo.de/watch/8229274/bowling_fail_or_win',
+        'md5': '2d2753e8130479ba2cb7e0a37002053e',
+        'info_dict': {
+            'id': '8229274',
+            'ext': 'flv',
+            'title': 'bowling-fail-or-win',
         }
     }
 
@@ -53,10 +53,7 @@ class MyVideoIE(InfoExtractor):
 
     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(1)
+        video_id = mobj.group('id')
 
         GK = (
           b'WXpnME1EZGhNRGhpTTJNM01XVmhOREU0WldNNVpHTTJOakpt'
@@ -74,37 +71,33 @@ class MyVideoIE(InfoExtractor):
             video_url = mobj.group(1) + '.flv'
 
             video_title = self._html_search_regex('<title>([^<]+)</title>',
-                webpage, u'title')
+                webpage, 'title')
 
-            video_ext = self._search_regex('[.](.+?)$', video_url, u'extension')
-
-            return [{
-                'id':       video_id,
-                'url':      video_url,
-                'uploader': None,
-                'upload_date':  None,
-                'title':    video_title,
-                'ext':      video_ext,
-            }]
+            return {
+                'id': video_id,
+                'url': video_url,
+                'title': video_title,
+            }
 
         mobj = re.search(r'data-video-service="/service/data/video/%s/config' % video_id, webpage)
         if mobj is not None:
             request = compat_urllib_request.Request('http://www.myvideo.de/service/data/video/%s/config' % video_id, '')
             response = self._download_webpage(request, video_id,
-                                              u'Downloading video info')
+                                              'Downloading video info')
             info = json.loads(base64.b64decode(response).decode('utf-8'))
-            return {'id': video_id,
-                    'title': info['title'],
-                    'url': info['streaming_url'].replace('rtmpe', 'rtmpt'),
-                    'play_path': info['filename'],
-                    'ext': 'flv',
-                    'thumbnail': info['thumbnail'][0]['url'],
-                    }
+            return {
+                'id': video_id,
+                'title': info['title'],
+                'url': info['streaming_url'].replace('rtmpe', 'rtmpt'),
+                'play_path': info['filename'],
+                'ext': 'flv',
+                'thumbnail': info['thumbnail'][0]['url'],
+            }
 
         # try encxml
         mobj = re.search('var flashvars={(.+?)}', webpage)
         if mobj is None:
-            raise ExtractorError(u'Unable to extract video')
+            raise ExtractorError('Unable to extract video')
 
         params = {}
         encxml = ''
@@ -118,7 +111,7 @@ class MyVideoIE(InfoExtractor):
             params['domain'] = 'www.myvideo.de'
         xmldata_url = '%s?%s' % (encxml, compat_urllib_parse.urlencode(params))
         if 'flash_playertype=MTV' in xmldata_url:
-            self._downloader.report_warning(u'avoiding MTV player')
+            self._downloader.report_warning('avoiding MTV player')
             xmldata_url = (
                 'http://www.myvideo.de/dynamic/get_player_video_xml.php'
                 '?flash_playertype=D&ID=%s&_countlimit=4&autorun=yes'
@@ -144,7 +137,7 @@ class MyVideoIE(InfoExtractor):
             video_url = compat_urllib_parse.unquote(mobj.group(1))
             if 'myvideo2flash' in video_url:
                 self.report_warning(
-                    u'Rewriting URL to use unencrypted rtmp:// ...',
+                    'Rewriting URL to use unencrypted rtmp:// ...',
                     video_id)
                 video_url = video_url.replace('rtmpe://', 'rtmp://')
 
@@ -152,39 +145,31 @@ class MyVideoIE(InfoExtractor):
             # extract non rtmp videos
             mobj = re.search('path=\'(http.*?)\' source=\'(.*?)\'', dec_data)
             if mobj is None:
-                raise ExtractorError(u'unable to extract url')
+                raise ExtractorError('unable to extract url')
             video_url = compat_urllib_parse.unquote(mobj.group(1)) + compat_urllib_parse.unquote(mobj.group(2))
 
-        video_file = self._search_regex('source=\'(.*?)\'', dec_data, u'video file')
+        video_file = self._search_regex('source=\'(.*?)\'', dec_data, 'video file')
         video_file = compat_urllib_parse.unquote(video_file)
 
         if not video_file.endswith('f4m'):
             ppath, prefix = video_file.split('.')
             video_playpath = '%s:%s' % (prefix, ppath)
-            video_hls_playlist = ''
         else:
             video_playpath = ''
-            video_hls_playlist = (
-                video_file
-            ).replace('.f4m', '.m3u8')
 
-        video_swfobj = self._search_regex('swfobject.embedSWF\(\'(.+?)\'', webpage, u'swfobj')
+        video_swfobj = self._search_regex('swfobject.embedSWF\(\'(.+?)\'', webpage, 'swfobj')
         video_swfobj = compat_urllib_parse.unquote(video_swfobj)
 
         video_title = self._html_search_regex("<h1(?: class='globalHd')?>(.*?)</h1>",
-            webpage, u'title')
+            webpage, 'title')
 
-        return [{
-            'id':                 video_id,
-            'url':                video_url,
-            'tc_url':             video_url,
-            'uploader':           None,
-            'upload_date':        None,
-            'title':              video_title,
-            'ext':                u'flv',
-            'play_path':          video_playpath,
-            'video_file':         video_file,
-            'video_hls_playlist': video_hls_playlist,
-            'player_url':         video_swfobj,
-        }]
+        return {
+            'id': video_id,
+            'url': video_url,
+            'tc_url': video_url,
+            'title': video_title,
+            'ext': 'flv',
+            'play_path': video_playpath,
+            'player_url': video_swfobj,
+        }
 

youtube_dl/extractor/orf.py

@@ -8,6 +8,7 @@ from .common import InfoExtractor
 from ..utils import (
     HEADRequest,
     unified_strdate,
+    ExtractorError,
 )
 
 
@@ -35,7 +36,15 @@ class ORFIE(InfoExtractor):
         data_json = self._search_regex(
             r'initializeAdworx\((.+?)\);\n', webpage, 'video info')
         all_data = json.loads(data_json)
-        sdata = all_data[0]['values']['segments']
+
+        def get_segments(all_data):
+            for data in all_data:
+                if data['name'] == 'Tracker::EPISODE_DETAIL_PAGE_OVER_PROGRAM':
+                    return data['values']['segments']
+
+        sdata = get_segments(all_data)
+        if not sdata:
+            raise ExtractorError('Unable to extract segments')
 
         def quality_to_int(s):
             m = re.search('([0-9]+)', s)

youtube_dl/extractor/photobucket.py

@@ -1,76 +1,43 @@
+from __future__ import unicode_literals
+
 import datetime
 import json
 import re
 
 from .common import InfoExtractor
 
-from ..utils import (
-    ExtractorError,
-)
 
 class PhotobucketIE(InfoExtractor):
-    """Information extractor for photobucket.com."""
-
-    # TODO: the original _VALID_URL was:
-    # r'(?:http://)?(?:[a-z0-9]+\.)?photobucket\.com/.*[\?\&]current=(.*\.flv)'
-    # Check if it's necessary to keep the old extracion process
-    _VALID_URL = r'(?:http://)?(?:[a-z0-9]+\.)?photobucket\.com/.*(([\?\&]current=)|_)(?P<id>.*)\.(?P<ext>(flv)|(mp4))'
-    IE_NAME = u'photobucket'
+    _VALID_URL = r'http://(?:[a-z0-9]+\.)?photobucket\.com/.*(([\?\&]current=)|_)(?P<id>.*)\.(?P<ext>(flv)|(mp4))'
     _TEST = {
-        u'url': u'http://media.photobucket.com/user/rachaneronas/media/TiredofLinkBuildingTryBacklinkMyDomaincom_zpsc0c3b9fa.mp4.html?filters[term]=search&filters[primary]=videos&filters[secondary]=images&sort=1&o=0',
-        u'file': u'zpsc0c3b9fa.mp4',
-        u'md5': u'7dabfb92b0a31f6c16cebc0f8e60ff99',
-        u'info_dict': {
-            u"upload_date": u"20130504", 
-            u"uploader": u"rachaneronas", 
-            u"title": u"Tired of Link Building? Try BacklinkMyDomain.com!"
+        'url': 'http://media.photobucket.com/user/rachaneronas/media/TiredofLinkBuildingTryBacklinkMyDomaincom_zpsc0c3b9fa.mp4.html?filters[term]=search&filters[primary]=videos&filters[secondary]=images&sort=1&o=0',
+        'file': 'zpsc0c3b9fa.mp4',
+        'md5': '7dabfb92b0a31f6c16cebc0f8e60ff99',
+        'info_dict': {
+            'upload_date': '20130504',
+            'uploader': 'rachaneronas',
+            'title': 'Tired of Link Building? Try BacklinkMyDomain.com!',
         }
     }
 
     def _real_extract(self, url):
-        # Extract id from URL
         mobj = re.match(self._VALID_URL, url)
-        if mobj is None:
-            raise ExtractorError(u'Invalid URL: %s' % url)
-
         video_id = mobj.group('id')
-
         video_extension = mobj.group('ext')
 
-        # Retrieve video webpage to extract further information
         webpage = self._download_webpage(url, video_id)
 
         # Extract URL, uploader, and title from webpage
         self.report_extraction(video_id)
-        # We try first by looking the javascript code:
-        mobj = re.search(r'Pb\.Data\.Shared\.put\(Pb\.Data\.Shared\.MEDIA, (?P<json>.*?)\);', webpage)
-        if mobj is not None:
-            info = json.loads(mobj.group('json'))
-            return [{
-                'id':       video_id,
-                'url':      info[u'downloadUrl'],
-                'uploader': info[u'username'],
-                'upload_date':  datetime.date.fromtimestamp(info[u'creationDate']).strftime('%Y%m%d'),
-                'title':    info[u'title'],
-                'ext':      video_extension,
-                'thumbnail': info[u'thumbUrl'],
-            }]
-
-        # We try looking in other parts of the webpage
-        video_url = self._search_regex(r'<link rel="video_src" href=".*\?file=([^"]+)" />',
-            webpage, u'video URL')
-
-        mobj = re.search(r'<title>(.*) video by (.*) - Photobucket</title>', webpage)
-        if mobj is None:
-            raise ExtractorError(u'Unable to extract title')
-        video_title = mobj.group(1).decode('utf-8')
-        video_uploader = mobj.group(2).decode('utf-8')
-
-        return [{
-            'id':       video_id.decode('utf-8'),
-            'url':      video_url.decode('utf-8'),
-            'uploader': video_uploader,
-            'upload_date':  None,
-            'title':    video_title,
-            'ext':      video_extension.decode('utf-8'),
-        }]
+        info_json = self._search_regex(r'Pb\.Data\.Shared\.put\(Pb\.Data\.Shared\.MEDIA, (.*?)\);',
+            webpage, 'info json')
+        info = json.loads(info_json)
+        return {
+            'id': video_id,
+            'url': info['downloadUrl'],
+            'uploader': info['username'],
+            'upload_date': datetime.date.fromtimestamp(info['creationDate']).strftime('%Y%m%d'),
+            'title': info['title'],
+            'ext': video_extension,
+            'thumbnail': info['thumbUrl'],
+        }

youtube_dl/extractor/pornhub.py

@@ -44,7 +44,7 @@ class PornHubIE(InfoExtractor):
 
         video_urls = list(map(compat_urllib_parse.unquote , re.findall(r'"quality_[0-9]{3}p":"([^"]+)', webpage)))
         if webpage.find('"encrypted":true') != -1:
-            password = self._html_search_regex(r'"video_title":"([^"]+)', webpage, 'password').replace('+', ' ')
+            password = compat_urllib_parse.unquote_plus(self._html_search_regex(r'"video_title":"([^"]+)', webpage, 'password'))
             video_urls = list(map(lambda s: aes_decrypt_text(s, password, 32).decode('utf-8'), video_urls))
 
         formats = []

youtube_dl/extractor/prosiebensat1.py

@@ -0,0 +1,297 @@
+# encoding: utf-8
+from __future__ import unicode_literals
+
+import re
+
+from hashlib import sha1
+from .common import InfoExtractor
+from ..utils import (
+    compat_urllib_parse,
+    unified_strdate,
+    clean_html,
+    RegexNotFoundError,
+)
+
+
+class ProSiebenSat1IE(InfoExtractor):
+    IE_NAME = 'prosiebensat1'
+    IE_DESC = 'ProSiebenSat.1 Digital'
+    _VALID_URL = r'https?://(?:www\.)?(?:(?:prosieben|prosiebenmaxx|sixx|sat1|kabeleins|ran|the-voice-of-germany)\.de|fem\.com)/(?P<id>.+)'
+
+    _TESTS = [
+        {
+            'url': 'http://www.prosieben.de/tv/circus-halligalli/videos/218-staffel-2-episode-18-jahresrueckblick-ganze-folge',
+            'info_dict': {
+                'id': '2104602',
+                'ext': 'mp4',
+                'title': 'Staffel 2, Episode 18 - Jahresrückblick',
+                'description': 'md5:8733c81b702ea472e069bc48bb658fc1',
+                'upload_date': '20131231',
+                'duration': 5845.04,
+            },
+            'params': {
+                # rtmp download
+                'skip_download': True,
+            },
+        },
+        {
+            'url': 'http://www.prosieben.de/videokatalog/Gesellschaft/Leben/Trends/video-Lady-Umstyling-f%C3%BCr-Audrina-Rebekka-Audrina-Fergen-billig-aussehen-Battal-Modica-700544.html',
+            'info_dict': {
+                'id': '2570327',
+                'ext': 'mp4',
+                'title': 'Lady-Umstyling für Audrina',
+                'description': 'md5:4c16d0c17a3461a0d43ea4084e96319d',
+                'upload_date': '20131014',
+                'duration': 606.76,
+            },
+            'params': {
+                # rtmp download
+                'skip_download': True,
+            },
+            'skip': 'Seems to be broken',
+        },
+        {
+            'url': 'http://www.prosiebenmaxx.de/tv/experience/video/144-countdown-fuer-die-autowerkstatt-ganze-folge',
+            'info_dict': {
+                'id': '2429369',
+                'ext': 'mp4',
+                'title': 'Countdown für die Autowerkstatt',
+                'description': 'md5:809fc051a457b5d8666013bc40698817',
+                'upload_date': '20140223',
+                'duration': 2595.04,
+            },
+            'params': {
+                # rtmp download
+                'skip_download': True,
+            },
+        },
+        {
+            'url': 'http://www.sixx.de/stars-style/video/sexy-laufen-in-ugg-boots-clip',
+            'info_dict': {
+                'id': '2904997',
+                'ext': 'mp4',
+                'title': 'Sexy laufen in Ugg Boots',
+                'description': 'md5:edf42b8bd5bc4e5da4db4222c5acb7d6',
+                'upload_date': '20140122',
+                'duration': 245.32,
+            },
+            'params': {
+                # rtmp download
+                'skip_download': True,
+            },
+        },
+        {
+            'url': 'http://www.sat1.de/film/der-ruecktritt/video/im-interview-kai-wiesinger-clip',
+            'info_dict': {
+                'id': '2906572',
+                'ext': 'mp4',
+                'title': 'Im Interview: Kai Wiesinger',
+                'description': 'md5:e4e5370652ec63b95023e914190b4eb9',
+                'upload_date': '20140225',
+                'duration': 522.56,
+            },
+            'params': {
+                # rtmp download
+                'skip_download': True,
+            },
+        },
+        {
+            'url': 'http://www.kabeleins.de/tv/rosins-restaurants/videos/jagd-auf-fertigkost-im-elsthal-teil-2-ganze-folge',
+            'info_dict': {
+                'id': '2992323',
+                'ext': 'mp4',
+                'title': 'Jagd auf Fertigkost im Elsthal - Teil 2',
+                'description': 'md5:2669cde3febe9bce13904f701e774eb6',
+                'upload_date': '20140225',
+                'duration': 2410.44,
+            },
+            'params': {
+                # rtmp download
+                'skip_download': True,
+            },
+        },
+        {
+            'url': 'http://www.ran.de/fussball/bundesliga/video/schalke-toennies-moechte-raul-zurueck-ganze-folge',
+            'info_dict': {
+                'id': '3004256',
+                'ext': 'mp4',
+                'title': 'Schalke: Tönnies möchte Raul zurück',
+                'description': 'md5:4b5b271d9bcde223b54390754c8ece3f',
+                'upload_date': '20140226',
+                'duration': 228.96,
+            },
+            'params': {
+                # rtmp download
+                'skip_download': True,
+            },
+        },
+        {
+            'url': 'http://www.the-voice-of-germany.de/video/31-andreas-kuemmert-rocket-man-clip',
+            'info_dict': {
+                'id': '2572814',
+                'ext': 'mp4',
+                'title': 'Andreas Kümmert: Rocket Man',
+                'description': 'md5:6ddb02b0781c6adf778afea606652e38',
+                'upload_date': '20131017',
+                'duration': 469.88,
+            },
+            'params': {
+                # rtmp download
+                'skip_download': True,
+            },
+        },
+        {
+            'url': 'http://www.fem.com/wellness/videos/wellness-video-clip-kurztripps-zum-valentinstag.html',
+            'info_dict': {
+                'id': '2156342',
+                'ext': 'mp4',
+                'title': 'Kurztrips zum Valentinstag',
+                'description': 'md5:8ba6301e70351ae0bedf8da00f7ba528',
+                'upload_date': '20130206',
+                'duration': 307.24,
+            },
+            'params': {
+                # rtmp download
+                'skip_download': True,
+            },
+        },
+    ]
+
+    _CLIPID_REGEXES = [
+        r'"clip_id"\s*:\s+"(\d+)"',
+        r'clipid: "(\d+)"',
+    ]
+    _TITLE_REGEXES = [
+        r'<h2 class="subtitle" itemprop="name">\s*(.+?)</h2>',
+        r'<header class="clearfix">\s*<h3>(.+?)</h3>',
+        r'<!-- start video -->\s*<h1>(.+?)</h1>',
+        r'<div class="ep-femvideos-pi4-video-txt">\s*<h2>(.+?)</h2>',
+    ]
+    _DESCRIPTION_REGEXES = [
+        r'<p itemprop="description">\s*(.+?)</p>',
+        r'<div class="videoDecription">\s*<p><strong>Beschreibung</strong>: (.+?)</p>',
+        r'<div class="g-plusone" data-size="medium"></div>\s*</div>\s*</header>\s*(.+?)\s*<footer>',
+        r'<p>(.+?)</p>\s*<div class="ep-femvideos-pi4-video-footer">',
+    ]
+    _UPLOAD_DATE_REGEXES = [
+        r'<meta property="og:published_time" content="(.+?)">',
+        r'<span>\s*(\d{2}\.\d{2}\.\d{4} \d{2}:\d{2}) \|\s*<span itemprop="duration"',
+        r'<footer>\s*(\d{2}\.\d{2}\.\d{4}) \d{2}:\d{2} Uhr',
+        r'<span style="padding-left: 4px;line-height:20px; color:#404040">(\d{2}\.\d{2}\.\d{4})</span>',
+        r'(\d{2}\.\d{2}\.\d{4}) \| \d{2}:\d{2} Min<br/>',
+    ]
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        video_id = mobj.group('id')
+
+        page = self._download_webpage(url, video_id, 'Downloading page')
+
+        def extract(patterns, name, page, fatal=False):
+            for pattern in patterns:
+                mobj = re.search(pattern, page)
+                if mobj:
+                    return clean_html(mobj.group(1))
+            if fatal:
+                raise RegexNotFoundError(u'Unable to extract %s' % name)
+            return None
+
+        clip_id = extract(self._CLIPID_REGEXES, 'clip id', page, fatal=True)
+
+        access_token = 'testclient'
+        client_name = 'kolibri-1.2.5'
+        client_location = url
+
+        videos_api_url = 'http://vas.sim-technik.de/vas/live/v2/videos?%s' % compat_urllib_parse.urlencode({
+            'access_token': access_token,
+            'client_location': client_location,
+            'client_name': client_name,
+            'ids': clip_id,
+        })
+
+        videos = self._download_json(videos_api_url, clip_id, 'Downloading videos JSON')
+
+        duration = float(videos[0]['duration'])
+        source_ids = [source['id'] for source in videos[0]['sources']]
+        source_ids_str = ','.join(map(str, source_ids))
+
+        g = '01!8d8F_)r9]4s[qeuXfP%'
+
+        client_id = g[:2] + sha1(''.join([clip_id, g, access_token, client_location, g, client_name])
+                                 .encode('utf-8')).hexdigest()
+
+        sources_api_url = 'http://vas.sim-technik.de/vas/live/v2/videos/%s/sources?%s' % (clip_id, compat_urllib_parse.urlencode({
+            'access_token': access_token,
+            'client_id': client_id,
+            'client_location': client_location,
+            'client_name': client_name,
+        }))
+
+        sources = self._download_json(sources_api_url, clip_id, 'Downloading sources JSON')
+        server_id = sources['server_id']
+
+        client_id = g[:2] + sha1(''.join([g, clip_id, access_token, server_id,
+                                          client_location, source_ids_str, g, client_name])
+                                 .encode('utf-8')).hexdigest()
+
+        url_api_url = 'http://vas.sim-technik.de/vas/live/v2/videos/%s/sources/url?%s' % (clip_id, compat_urllib_parse.urlencode({
+            'access_token': access_token,
+            'client_id': client_id,
+            'client_location': client_location,
+            'client_name': client_name,
+            'server_id': server_id,
+            'source_ids': source_ids_str,
+        }))
+
+        urls = self._download_json(url_api_url, clip_id, 'Downloading urls JSON')
+
+        title = extract(self._TITLE_REGEXES, 'title', page, fatal=True)
+        description = extract(self._DESCRIPTION_REGEXES, 'description', page)
+        thumbnail = self._og_search_thumbnail(page)
+
+        upload_date = extract(self._UPLOAD_DATE_REGEXES, 'upload date', page)
+        if upload_date:
+            upload_date = unified_strdate(upload_date)
+
+        formats = []
+
+        urls_sources = urls['sources']
+        if isinstance(urls_sources, dict):
+            urls_sources = urls_sources.values()
+
+        def fix_bitrate(bitrate):
+            return bitrate / 1000 if bitrate % 1000 == 0 else bitrate
+
+        for source in urls_sources:
+            protocol = source['protocol']
+            if protocol == 'rtmp' or protocol == 'rtmpe':
+                mobj = re.search(r'^(?P<url>rtmpe?://[^/]+/(?P<app>[^/]+))/(?P<playpath>.+)$', source['url'])
+                if not mobj:
+                    continue
+                formats.append({
+                    'url': mobj.group('url'),
+                    'app': mobj.group('app'),
+                    'play_path': mobj.group('playpath'),
+                    'player_url': 'http://livepassdl.conviva.com/hf/ver/2.79.0.17083/LivePassModuleMain.swf',
+                    'page_url': 'http://www.prosieben.de',
+                    'vbr': fix_bitrate(source['bitrate']),
+                    'ext': 'mp4',
+                    'format_id': '%s_%s' % (source['cdn'], source['bitrate']),
+                })
+            else:
+                formats.append({
+                    'url': source['url'],
+                    'vbr': fix_bitrate(source['bitrate']),
+                })
+
+        self._sort_formats(formats)
+
+        return {
+            'id': clip_id,
+            'title': title,
+            'description': description,
+            'thumbnail': thumbnail,
+            'upload_date': upload_date,
+            'duration': duration,
+            'formats': formats,
+        }

youtube_dl/extractor/rtlnow.py

@@ -1,148 +1,165 @@
 # encoding: utf-8
-
 from __future__ import unicode_literals
 
 import re
 
 from .common import InfoExtractor
 from ..utils import (
-    clean_html,
     ExtractorError,
+    clean_html,
+    unified_strdate,
+    int_or_none,
 )
 
 
 class RTLnowIE(InfoExtractor):
     """Information Extractor for RTL NOW, RTL2 NOW, RTL NITRO, SUPER RTL NOW, VOX NOW and n-tv NOW"""
-    _VALID_URL = r'(?:http://)?(?P<url>(?P<domain>rtl-now\.rtl\.de|rtl2now\.rtl2\.de|(?:www\.)?voxnow\.de|(?:www\.)?rtlnitronow\.de|(?:www\.)?superrtlnow\.de|(?:www\.)?n-tvnow\.de)/+[a-zA-Z0-9-]+/[a-zA-Z0-9-]+\.php\?(?:container_id|film_id)=(?P<video_id>[0-9]+)&player=1(?:&season=[0-9]+)?(?:&.*)?)'
-    _TESTS = [{
-        'url': 'http://rtl-now.rtl.de/ahornallee/folge-1.php?film_id=90419&player=1&season=1',
-        'file': '90419.flv',
-        'info_dict': {
-            'upload_date': '20070416',
-            'title': 'Ahornallee - Folge 1 - Der Einzug',
-            'description': 'Folge 1 - Der Einzug',
+    _VALID_URL = r'''(?x)
+                        (?:https?://)?
+                        (?P<url>
+                            (?P<domain>
+                                rtl-now\.rtl\.de|
+                                rtl2now\.rtl2\.de|
+                                (?:www\.)?voxnow\.de|
+                                (?:www\.)?rtlnitronow\.de|
+                                (?:www\.)?superrtlnow\.de|
+                                (?:www\.)?n-tvnow\.de)
+                            /+[a-zA-Z0-9-]+/[a-zA-Z0-9-]+\.php\?
+                            (?:container_id|film_id)=(?P<video_id>[0-9]+)&
+                            player=1(?:&season=[0-9]+)?(?:&.*)?
+                        )'''
+
+    _TESTS = [
+        {
+            'url': 'http://rtl-now.rtl.de/ahornallee/folge-1.php?film_id=90419&player=1&season=1',
+            'info_dict': {
+                'id': '90419',
+                'ext': 'flv',
+                'title': 'Ahornallee - Folge 1 - Der Einzug',
+                'description': 'md5:ce843b6b5901d9a7f7d04d1bbcdb12de',
+                'upload_date': '20070416',
+                'duration': 1685,
+            },
+            'params': {
+                'skip_download': True,
+            },
+            'skip': 'Only works from Germany',
         },
-        'params': {
-            'skip_download': True,
+        {
+            'url': 'http://rtl2now.rtl2.de/aerger-im-revier/episode-15-teil-1.php?film_id=69756&player=1&season=2&index=5',
+            'info_dict': {
+                'id': '69756',
+                'ext': 'flv',
+                'title': 'Ärger im Revier - Ein junger Ladendieb, ein handfester Streit u.a.',
+                'description': 'md5:3fb247005ed21a935ffc82b7dfa70cf0',
+                'thumbnail': 'http://autoimg.static-fra.de/rtl2now/219850/1500x1500/image2.jpg',
+                'upload_date': '20120519',
+                'duration': 1245,
+            },
+            'params': {
+                'skip_download': True,
+            },
+            'skip': 'Only works from Germany',
         },
-        'skip': 'Only works from Germany',
-    },
-    {
-        'url': 'http://rtl2now.rtl2.de/aerger-im-revier/episode-15-teil-1.php?film_id=69756&player=1&season=2&index=5',
-        'file': '69756.flv',
-        'info_dict': {
-            'upload_date': '20120519',
-            'title': 'Ärger im Revier - Ein junger Ladendieb, ein handfester Streit...',
-            'description': 'Ärger im Revier - Ein junger Ladendieb, ein handfester Streit u.a.',
-            'thumbnail': 'http://autoimg.static-fra.de/rtl2now/219850/1500x1500/image2.jpg',
+        {
+            'url': 'http://www.voxnow.de/voxtours/suedafrika-reporter-ii.php?film_id=13883&player=1&season=17',
+            'info_dict': {
+                'id': '13883',
+                'ext': 'flv',
+                'title': 'Voxtours - Südafrika-Reporter II',
+                'description': 'md5:de7f8d56be6fd4fed10f10f57786db00',
+                'upload_date': '20090627',
+                'duration': 1800,
+            },
+            'params': {
+                'skip_download': True,
+            },
         },
-        'params': {
-            'skip_download': True,
+        {
+            'url': 'http://superrtlnow.de/medicopter-117/angst.php?film_id=99205&player=1',
+            'info_dict': {
+                'id': '99205',
+                'ext': 'flv',
+                'title': 'Medicopter 117 - Angst!',
+                'description': 'md5:895b1df01639b5f61a04fc305a5cb94d',
+                'thumbnail': 'http://autoimg.static-fra.de/superrtlnow/287529/1500x1500/image2.jpg',
+                'upload_date': '20080928',
+                'duration': 2691,
+            },
+            'params': {
+                'skip_download': True,
+            },
         },
-        'skip': 'Only works from Germany',
-    },
-    {
-        'url': 'http://www.voxnow.de/voxtours/suedafrika-reporter-ii.php?film_id=13883&player=1&season=17',
-        'file': '13883.flv',
-        'info_dict': {
-            'upload_date': '20090627',
-            'title': 'Voxtours - Südafrika-Reporter II',
-            'description': 'Südafrika-Reporter II',
+        {
+            'url': 'http://www.n-tvnow.de/deluxe-alles-was-spass-macht/thema-ua-luxushotel-fuer-vierbeiner.php?container_id=153819&player=1&season=0',
+            'info_dict': {
+                'id': '153819',
+                'ext': 'flv',
+                'title': 'Deluxe - Alles was Spaß macht - Thema u.a.: Luxushotel für Vierbeiner',
+                'description': 'md5:c3705e1bb32e1a5b2bcd634fc065c631',
+                'thumbnail': 'http://autoimg.static-fra.de/ntvnow/383157/1500x1500/image2.jpg',
+                'upload_date': '20140221',
+                'duration': 2429,
+            },
+            'skip': 'Only works from Germany',
         },
-        'params': {
-            'skip_download': True,
-        },
-    },
-    {
-        'url': 'http://superrtlnow.de/medicopter-117/angst.php?film_id=99205&player=1',
-        'file': '99205.flv',
-        'info_dict': {
-            'upload_date': '20080928', 
-            'title': 'Medicopter 117 - Angst!',
-            'description': 'Angst!',
-            'thumbnail': 'http://autoimg.static-fra.de/superrtlnow/287529/1500x1500/image2.jpg'
-        },
-        'params': {
-            'skip_download': True,
-        },
-    },
-    {
-        'url': 'http://www.n-tvnow.de/top-gear/episode-1-2013-01-01-00-00-00.php?film_id=124903&player=1&season=10',
-        'file': '124903.flv',
-        'info_dict': {
-            'upload_date': '20130101',
-            'title': 'Top Gear vom 01.01.2013',
-            'description': 'Episode 1',
-        },
-        'params': {
-            'skip_download': True,
-        },
-        'skip': 'Only works from Germany',
-    }]
+    ]
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
-
-        webpage_url = 'http://' + mobj.group('url')
-        video_page_url = 'http://' + mobj.group('domain') + '/'
+        video_page_url = 'http://%s/' % mobj.group('domain')
         video_id = mobj.group('video_id')
 
-        webpage = self._download_webpage(webpage_url, video_id)
+        webpage = self._download_webpage('http://' + mobj.group('url'), video_id)
 
-        note_m = re.search(r'''(?sx)
-            <div[ ]style="margin-left:[ ]20px;[ ]font-size:[ ]13px;">(.*?)
-            <div[ ]id="playerteaser">''', webpage)
-        if note_m:
-            msg = clean_html(note_m.group(1))
-            raise ExtractorError(msg)
+        mobj = re.search(r'(?s)<div style="margin-left: 20px; font-size: 13px;">(.*?)<div id="playerteaser">', webpage)
+        if mobj:
+            raise ExtractorError(clean_html(mobj.group(1)), expected=True)
+
+        title = self._og_search_title(webpage)
+        description = self._og_search_description(webpage)
+        thumbnail = self._og_search_thumbnail(webpage, default=None)
+
+        upload_date = unified_strdate(self._html_search_meta('uploadDate', webpage, 'upload date'))
+
+        mobj = re.search(r'<meta itemprop="duration" content="PT(?P<seconds>\d+)S" />', webpage)
+        duration = int(mobj.group('seconds')) if mobj else None
 
-        video_title = self._html_search_regex(
-            r'<title>(?P<title>[^<]+?)( \| [^<]*)?</title>',
-            webpage, 'title')
         playerdata_url = self._html_search_regex(
-            r'\'playerdata\': \'(?P<playerdata_url>[^\']+)\'',
-            webpage, 'playerdata_url')
+            r"'playerdata': '(?P<playerdata_url>[^']+)'", webpage, 'playerdata_url')
 
-        playerdata = self._download_webpage(playerdata_url, video_id)
-        mobj = re.search(r'<title><!\[CDATA\[(?P<description>.+?)(?:\s+- (?:Sendung )?vom (?P<upload_date_d>[0-9]{2})\.(?P<upload_date_m>[0-9]{2})\.(?:(?P<upload_date_Y>[0-9]{4})|(?P<upload_date_y>[0-9]{2})) [0-9]{2}:[0-9]{2} Uhr)?\]\]></title>', playerdata)
-        if mobj:
-            video_description = mobj.group('description')
-            if mobj.group('upload_date_Y'):
-                video_upload_date = mobj.group('upload_date_Y')
-            elif mobj.group('upload_date_y'):
-                video_upload_date = '20' + mobj.group('upload_date_y')
+        playerdata = self._download_xml(playerdata_url, video_id, 'Downloading player data XML')
+
+        videoinfo = playerdata.find('./playlist/videoinfo')
+        
+        formats = []
+        for filename in videoinfo.findall('filename'):
+            mobj = re.search(r'(?P<url>rtmpe://(?:[^/]+/){2})(?P<play_path>.+)', filename.text)
+            if mobj:
+                fmt = {
+                    'url': mobj.group('url'),
+                    'play_path': 'mp4:' + mobj.group('play_path'),
+                    'page_url': video_page_url,
+                    'player_url': video_page_url + 'includes/vodplayer.swf',
+                }
             else:
-                video_upload_date = None
-            if video_upload_date:
-                video_upload_date += mobj.group('upload_date_m') + mobj.group('upload_date_d')
-        else:
-            video_description = None
-            video_upload_date = None
-            self._downloader.report_warning('Unable to extract description and upload date')
-
-        # Thumbnail: not every video has an thumbnail
-        mobj = re.search(r'<meta property="og:image" content="(?P<thumbnail>[^"]+)">', webpage)
-        if mobj:
-            video_thumbnail = mobj.group('thumbnail')
-        else:
-            video_thumbnail = None
-
-        mobj = re.search(r'<filename [^>]+><!\[CDATA\[(?P<url>rtmpe://(?:[^/]+/){2})(?P<play_path>[^\]]+)\]\]></filename>', playerdata)
-        if mobj is None:
-            raise ExtractorError('Unable to extract media URL')
-        video_url = mobj.group('url')
-        video_play_path = 'mp4:' + mobj.group('play_path')
-        video_player_url = video_page_url + 'includes/vodplayer.swf'
+                fmt = {
+                    'url': filename.text,
+                }
+            fmt.update({
+                'width': int_or_none(filename.get('width')),
+                'height': int_or_none(filename.get('height')),
+                'vbr': int_or_none(filename.get('bitrate')),
+                'ext': 'flv',
+            })
+            formats.append(fmt)
 
         return {
             'id': video_id,
-            'url': video_url,
-            'play_path': video_play_path,
-            'page_url': video_page_url,
-            'player_url': video_player_url,
-            'ext': 'flv',
-            'title': video_title,
-            'description': video_description,
-            'upload_date': video_upload_date,
-            'thumbnail': video_thumbnail,
-        }
+            'title': title,
+            'description': description,
+            'thumbnail': thumbnail,
+            'upload_date': upload_date,
+            'duration': duration,
+            'formats': formats,
+        }

youtube_dl/extractor/soundcloud.py

@@ -54,6 +54,7 @@ class SoundcloudIE(InfoExtractor):
                 'id': '47127627',
                 'ext': 'mp3',
                 'title': 'Goldrushed',
+                'description': 'From Stockholm Sweden\r\nPovel / Magnus / Filip / David\r\nwww.theroyalconcept.com',
                 'uploader': 'The Royal Concept',
                 'upload_date': '20120521',
             },
@@ -217,7 +218,7 @@ class SoundcloudIE(InfoExtractor):
         return self._extract_info_dict(info, full_title, secret_token=token)
 
 class SoundcloudSetIE(SoundcloudIE):
-    _VALID_URL = r'^(?:https?://)?(?:www\.)?soundcloud\.com/([\w\d-]+)/sets/([\w\d-]+)(?:[?].*)?$'
+    _VALID_URL = r'https?://(?:www\.)?soundcloud\.com/([\w\d-]+)/sets/([\w\d-]+)'
     IE_NAME = 'soundcloud:set'
     # it's in tests/test_playlists.py
     _TESTS = []

youtube_dl/extractor/spike.py

@@ -1,10 +1,15 @@
 from __future__ import unicode_literals
 
+import re
+
 from .mtv import MTVServicesInfoExtractor
 
 
 class SpikeIE(MTVServicesInfoExtractor):
-    _VALID_URL = r'https?://www\.spike\.com/(video-clips|episodes)/.+'
+    _VALID_URL = r'''(?x)https?://
+        (www\.spike\.com/(video-clips|episodes)/.+|
+         m\.spike\.com/videos/video.rbml\?id=(?P<mobile_id>[^&]+))
+        '''
     _TEST = {
         'url': 'http://www.spike.com/video-clips/lhtu8m/auction-hunters-can-allen-ride-a-hundred-year-old-motorcycle',
         'md5': '1a9265f32b0c375793d6c4ce45255256',
@@ -17,3 +22,11 @@ class SpikeIE(MTVServicesInfoExtractor):
     }
 
     _FEED_URL = 'http://www.spike.com/feeds/mrss/'
+    _MOBILE_TEMPLATE = 'http://m.spike.com/videos/video.rbml?id=%s'
+
+    def _real_extract(self, url):
+        mobj = re.search(self._VALID_URL, url)
+        mobile_id = mobj.group('mobile_id')
+        if mobile_id is not None:
+            url = 'http://www.spike.com/video-clips/%s' % mobile_id
+        return super(SpikeIE, self)._real_extract(url)

youtube_dl/extractor/ted.py

@@ -6,115 +6,111 @@ import re
 from .subtitles import SubtitlesInfoExtractor
 
 from ..utils import (
-    RegexNotFoundError,
+    compat_str,
 )
 
 
 class TEDIE(SubtitlesInfoExtractor):
-    _VALID_URL=r'''http://www\.ted\.com/
-                   (
-                        ((?P<type_playlist>playlists)/(?P<playlist_id>\d+)) # We have a playlist
-                        |
-                        ((?P<type_talk>talks)) # We have a simple talk
-                   )
-                   (/lang/(.*?))? # The url may contain the language
-                   /(?P<name>\w+) # Here goes the name and then ".html"
-                   '''
+    _VALID_URL = r'''(?x)http://www\.ted\.com/
+        (
+            (?P<type_playlist>playlists(?:/\d+)?) # We have a playlist
+            |
+            ((?P<type_talk>talks)) # We have a simple talk
+        )
+        (/lang/(.*?))? # The url may contain the language
+        /(?P<name>\w+) # Here goes the name and then ".html"
+        '''
     _TEST = {
         'url': 'http://www.ted.com/talks/dan_dennett_on_our_consciousness.html',
-        'file': '102.mp4',
         'md5': '4ea1dada91e4174b53dac2bb8ace429d',
         'info_dict': {
-            "description": "md5:c6fa72e6eedbd938c9caf6b2702f5922",
-            "title": "Dan Dennett: The illusion of consciousness"
+            'id': '102',
+            'ext': 'mp4',
+            'title': 'The illusion of consciousness',
+            'description': ('Philosopher Dan Dennett makes a compelling '
+                'argument that not only don\'t we understand our own '
+                'consciousness, but that half the time our brains are '
+                'actively fooling us.'),
+            'uploader': 'Dan Dennett',
         }
     }
 
-    @classmethod
-    def suitable(cls, url):
-        """Receives a URL and returns True if suitable for this IE."""
-        return re.match(cls._VALID_URL, url, re.VERBOSE) is not None
+    _FORMATS_PREFERENCE = {
+        'low': 1,
+        'medium': 2,
+        'high': 3,
+    }
+
+    def _extract_info(self, webpage):
+        info_json = self._search_regex(r'q\("\w+.init",({.+})\)</script>',
+            webpage, 'info json')
+        return json.loads(info_json)
 
     def _real_extract(self, url):
-        m=re.match(self._VALID_URL, url, re.VERBOSE)
+        m = re.match(self._VALID_URL, url, re.VERBOSE)
+        name = m.group('name')
         if m.group('type_talk'):
-            return self._talk_info(url)
-        else :
-            playlist_id=m.group('playlist_id')
-            name=m.group('name')
-            self.to_screen(u'Getting info of playlist %s: "%s"' % (playlist_id,name))
-            return [self._playlist_videos_info(url,name,playlist_id)]
+            return self._talk_info(url, name)
+        else:
+            return self._playlist_videos_info(url, name)
 
-
-    def _playlist_videos_info(self, url, name, playlist_id):
+    def _playlist_videos_info(self, url, name):
         '''Returns the videos of the playlist'''
 
-        webpage = self._download_webpage(
-            url, playlist_id, 'Downloading playlist webpage')
-        matches = re.finditer(
-            r'<p\s+class="talk-title[^"]*"><a\s+href="(?P<talk_url>/talks/[^"]+\.html)">[^<]*</a></p>',
-            webpage)
-
-        playlist_title = self._html_search_regex(r'div class="headline">\s*?<h1>\s*?<span>(.*?)</span>',
-                                                 webpage, 'playlist title')
+        webpage = self._download_webpage(url, name,
+            'Downloading playlist webpage')
+        info = self._extract_info(webpage)
+        playlist_info = info['playlist']
 
         playlist_entries = [
-            self.url_result(u'http://www.ted.com' + m.group('talk_url'), 'TED')
-            for m in matches
+            self.url_result(u'http://www.ted.com/talks/' + talk['slug'], self.ie_key())
+            for talk in info['talks']
         ]
         return self.playlist_result(
-            playlist_entries, playlist_id=playlist_id, playlist_title=playlist_title)
+            playlist_entries,
+            playlist_id=compat_str(playlist_info['id']),
+            playlist_title=playlist_info['title'])
 
-    def _talk_info(self, url, video_id=0):
-        """Return the video for the talk in the url"""
-        m = re.match(self._VALID_URL, url,re.VERBOSE)
-        video_name = m.group('name')
-        webpage = self._download_webpage(url, video_id, 'Downloading \"%s\" page' % video_name)
+    def _talk_info(self, url, video_name):
+        webpage = self._download_webpage(url, 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>',
-                                        webpage, 'title')
-        json_data = self._search_regex(r'<script.*?>var talkDetails = ({.*?})</script>',
-                                    webpage, 'json data')
-        info = json.loads(json_data)
-        desc = self._html_search_regex(r'<div class="talk-intro">.*?<p.*?>(.*?)</p>',
-                                       webpage, 'description', flags = re.DOTALL)
-        
-        thumbnail = self._search_regex(r'</span>[\s.]*</div>[\s.]*<img src="(.*?)"',
-                                       webpage, 'thumbnail')
+
+        talk_info = self._extract_info(webpage)['talks'][0]
+
         formats = [{
             'ext': 'mp4',
-            'url': stream['file'],
-            'format': stream['id']
-        } for stream in info['htmlStreams']]
-
-        video_id = info['id']
+            '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()]
+        self._sort_formats(formats)
 
+        video_id = compat_str(talk_info['id'])
         # subtitles
-        video_subtitles = self.extract_subtitles(video_id, webpage)
+        video_subtitles = self.extract_subtitles(video_id, talk_info)
         if self._downloader.params.get('listsubtitles', False):
-            self._list_available_subtitles(video_id, webpage)
+            self._list_available_subtitles(video_id, talk_info)
             return
 
         return {
             'id': video_id,
-            'title': title,
-            'thumbnail': thumbnail,
-            'description': desc,
+            'title': talk_info['title'],
+            'uploader': talk_info['speaker'],
+            'thumbnail': talk_info['thumb'],
+            'description': self._og_search_description(webpage),
             'subtitles': video_subtitles,
             'formats': formats,
         }
 
-    def _get_available_subtitles(self, video_id, webpage):
-        try:
-            options = self._search_regex(r'(?:<select name="subtitles_language_select" id="subtitles_language_select">)(.*?)(?:</select>)', webpage, 'subtitles_language_select', flags=re.DOTALL)
-            languages = re.findall(r'(?:<option value=")(\S+)"', options)
-            if languages:
-                sub_lang_list = {}
-                for l in languages:
-                    url = 'http://www.ted.com/talks/subtitles/id/%s/lang/%s/format/srt' % (video_id, l)
-                    sub_lang_list[l] = url
-                return sub_lang_list
-        except RegexNotFoundError:
+    def _get_available_subtitles(self, video_id, talk_info):
+        languages = [lang['languageCode'] for lang in talk_info.get('languages', [])]
+        if languages:
+            sub_lang_list = {}
+            for l in languages:
+                url = 'http://www.ted.com/talks/subtitles/id/%s/lang/%s/format/srt' % (video_id, l)
+                sub_lang_list[l] = url
+            return sub_lang_list
+        else:
             self._downloader.report_warning(u'video doesn\'t have subtitles')
-        return {}
+            return {}

youtube_dl/extractor/tinypic.py

@@ -3,7 +3,7 @@ from __future__ import unicode_literals
 import re
 
 from .common import InfoExtractor
-from youtube_dl.utils import ExtractorError
+from ..utils import ExtractorError
 
 
 class TinyPicIE(InfoExtractor):

youtube_dl/extractor/tvigle.py

@@ -0,0 +1,84 @@
+# encoding: utf-8
+from __future__ import unicode_literals
+
+import re
+
+from .common import InfoExtractor
+from ..utils import (
+    unified_strdate,
+    clean_html,
+    int_or_none,
+)
+
+
+class TvigleIE(InfoExtractor):
+    IE_NAME = 'tvigle'
+    IE_DESC = 'Интернет-телевидение Tvigle.ru'
+    _VALID_URL = r'http://(?:www\.)?tvigle\.ru/category/.+?[\?&]v(?:ideo)?=(?P<id>\d+)'
+
+    _TESTS = [
+        {
+            'url': 'http://www.tvigle.ru/category/cinema/1608/?video=503081',
+            'md5': '09afba4616666249f087efc6dcf83cb3',
+            'info_dict': {
+                'id': '503081',
+                'ext': 'flv',
+                'title': 'Брат 2 ',
+                'description': 'md5:f5a42970f50648cee3d7ad740f3ae769',
+                'upload_date': '20110919',
+            },
+        },
+        {
+            'url': 'http://www.tvigle.ru/category/men/vysotskiy_vospominaniya02/?flt=196&v=676433',
+            'md5': 'e7efe5350dd5011d0de6550b53c3ba7b',
+            'info_dict': {
+                'id': '676433',
+                'ext': 'flv',
+                'title': 'Ведущий телепрограммы «60 минут» (США) о Владимире Высоцком',
+                'description': 'md5:027f7dc872948f14c96d19b4178428a4',
+                'upload_date': '20121218',
+            },
+        },
+    ]
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        video_id = mobj.group('id')
+
+        video_data = self._download_xml(
+            'http://www.tvigle.ru/xml/single.php?obj=%s' % video_id, video_id, 'Downloading video XML')
+
+        video = video_data.find('./video')
+
+        title = video.get('name')
+        description = video.get('anons')
+        if description:
+            description = clean_html(description)
+        thumbnail = video_data.get('img')
+        upload_date = unified_strdate(video.get('date'))
+        like_count = int_or_none(video.get('vtp'))
+
+        formats = []
+        for num, (format_id, format_note) in enumerate([['low_file', 'SQ'], ['file', 'HQ'], ['hd', 'HD 720']]):
+            video_url = video.get(format_id)
+            if not video_url:
+                continue
+            formats.append({
+                'url': video_url,
+                'format_id': format_id,
+                'format_note': format_note,
+                'quality': num,
+            })
+
+        self._sort_formats(formats)
+
+        return {
+            'id': video_id,
+            'title': title,
+            'description': description,
+            'thumbnail': thumbnail,
+            'upload_date': upload_date,
+            'like_count': like_count,
+            'age_limit': 18,
+            'formats': formats,
+        }

youtube_dl/extractor/vesti.py

@@ -13,7 +13,7 @@ from ..utils import (
 class VestiIE(InfoExtractor):
     IE_NAME = 'vesti'
     IE_DESC = 'Вести.Ru'
-    _VALID_URL = r'http://(?:.+?\.)?vesti\.ru/(?P<id>.+)'
+    _VALID_URL = r'http://(?:(?:.+?\.)?vesti\.ru|(?:2\.)?russia\.tv|tvkultura\.ru|rutv\.ru)/(?P<id>.+)'
 
     _TESTS = [
         {
@@ -30,6 +30,20 @@ class VestiIE(InfoExtractor):
                 'skip_download': True,
             },
         },
+        {
+            'url': 'http://www.vesti.ru/doc.html?id=1349233',
+            'info_dict': {
+                'id': '773865',
+                'ext': 'mp4',
+                'title': 'Участники митинга штурмуют Донецкую областную администрацию',
+                'description': 'md5:1a160e98b3195379b4c849f2f4958009',
+                'duration': 210,
+            },
+            'params': {
+                # m3u8 download
+                'skip_download': True,
+            },
+        },
         {
             'url': 'http://www.vesti.ru/only_video.html?vid=576180',
             'info_dict': {
@@ -44,6 +58,20 @@ class VestiIE(InfoExtractor):
                 'skip_download': True,
             },
         },
+        {
+            'url': 'http://hitech.vesti.ru/news/view/id/4000',
+            'info_dict': {
+                'id': '766888',
+                'ext': 'mp4',
+                'title': 'Вести.net: интернет-гиганты начали перетягивание программных "одеял"',
+                'description': 'md5:65ddd47f9830c4f42ed6475f8730c995',
+                'duration': 279,
+            },
+            'params': {
+                # m3u8 download
+                'skip_download': True,
+            },
+        },
         {
             'url': 'http://sochi2014.vesti.ru/video/index/video_id/766403',
             'info_dict': {
@@ -72,7 +100,77 @@ class VestiIE(InfoExtractor):
                 'skip_download': True,
             },
             'skip': 'Translation has finished'
-        }
+        },
+        {
+            'url': 'http://russia.tv/video/show/brand_id/5169/episode_id/970443/video_id/975648',
+            'info_dict': {
+                'id': '771852',
+                'ext': 'mp4',
+                'title': 'Прямой эфир. Жертвы загадочной болезни: смерть от старости в 17 лет',
+                'description': 'md5:b81c8c55247a4bd996b43ce17395b2d8',
+                'duration': 3096,
+            },
+            'params': {
+                # m3u8 download
+                'skip_download': True,
+            },
+        },
+        {
+            'url': 'http://russia.tv/brand/show/brand_id/57638',
+            'info_dict': {
+                'id': '774016',
+                'ext': 'mp4',
+                'title': 'Чужой в семье Сталина',
+                'description': '',
+                'duration': 2539,
+            },
+            'params': {
+                # m3u8 download
+                'skip_download': True,
+            },
+        },
+        {
+            'url': 'http://2.russia.tv/video/show/brand_id/48863/episode_id/972920/video_id/978667/viewtype/picture',
+            'info_dict': {
+                'id': '775081',
+                'ext': 'mp4',
+                'title': 'XXII зимние Олимпийские игры. Россияне заняли весь пьедестал в лыжных гонках',
+                'description': 'md5:15d3741dd8d04b203fbc031c6a47fb0f',
+                'duration': 101,
+            },
+            'params': {
+                # m3u8 download
+                'skip_download': True,
+            },
+        },
+        {
+            'url': 'http://tvkultura.ru/video/show/brand_id/31724/episode_id/972347/video_id/978186',
+            'info_dict': {
+                'id': '774471',
+                'ext': 'mp4',
+                'title': 'Монологи на все времена',
+                'description': 'md5:18d8b5e6a41fb1faa53819471852d5d5',
+                'duration': 2906,
+            },
+            'params': {
+                # m3u8 download
+                'skip_download': True,
+            },
+        },
+        {
+            'url': 'http://rutv.ru/brand/show/id/6792/channel/75',
+            'info_dict': {
+                'id': '125521',
+                'ext': 'mp4',
+                'title': 'Грустная дама червей. Х/ф',
+                'description': '',
+                'duration': 4882,
+            },
+            'params': {
+                # m3u8 download
+                'skip_download': True,
+            },
+        },
     ]
 
     def _real_extract(self, url):
@@ -81,16 +179,26 @@ class VestiIE(InfoExtractor):
 
         page = self._download_webpage(url, video_id, 'Downloading page')
 
-        mobj = re.search(r'<meta property="og:video" content=".+?\.swf\?v?id=(?P<id>\d+).*?" />', page)
+        mobj = re.search(
+            r'<meta property="og:video" content="http://www\.vesti\.ru/i/flvplayer_videoHost\.swf\?vid=(?P<id>\d+)',
+            page)
+        if mobj:
+            video_id = mobj.group('id')
+            page = self._download_webpage('http://www.vesti.ru/only_video.html?vid=%s' % video_id, video_id,
+                'Downloading video page')
+
+        mobj = re.search(
+            r'<meta property="og:video" content="http://player\.rutv\.ru/flash2v/container\.swf\?id=(?P<id>\d+)', page)
         if mobj:
             video_type = 'video'
             video_id = mobj.group('id')
         else:
             mobj = re.search(
-                r'<iframe.+?src="http://player\.rutv\.ru/iframe/(?P<type>[^/]+)/id/(?P<id>\d+)[^"]*".*?></iframe>', page)
+                r'<iframe.+?src="http://player\.rutv\.ru/iframe/(?P<type>[^/]+)/id/(?P<id>\d+)[^"]*".*?></iframe>',
+                page)
 
             if not mobj:
-                raise ExtractorError('No media found')
+                raise ExtractorError('No media found', expected=True)
 
             video_type = mobj.group('type')
             video_id = mobj.group('id')
@@ -113,8 +221,8 @@ class VestiIE(InfoExtractor):
         priority_transport = playlist['priority_transport']
 
         thumbnail = media['picture']
-        width = media['width']
-        height = media['height']
+        width = int_or_none(media['width'])
+        height = int_or_none(media['height'])
         description = media['anons']
         title = media['title']
         duration = int_or_none(media.get('duration'))

youtube_dl/extractor/videodetective.py

@@ -1,22 +1,23 @@
+from __future__ import unicode_literals
+
 import re
 
 from .common import InfoExtractor
 from .internetvideoarchive import InternetVideoArchiveIE
-from ..utils import (
-    compat_urlparse,
-)
+from ..utils import compat_urlparse
 
 
 class VideoDetectiveIE(InfoExtractor):
     _VALID_URL = r'https?://www\.videodetective\.com/[^/]+/[^/]+/(?P<id>\d+)'
 
     _TEST = {
-        u'url': u'http://www.videodetective.com/movies/kick-ass-2/194487',
-        u'file': u'194487.mp4',
-        u'info_dict': {
-            u'title': u'KICK-ASS 2',
-            u'description': u'md5:65ba37ad619165afac7d432eaded6013',
-            u'duration': 135,
+        'url': 'http://www.videodetective.com/movies/kick-ass-2/194487',
+        'info_dict': {
+            'id': '194487',
+            'ext': 'mp4',
+            'title': 'KICK-ASS 2',
+            'description': 'md5:65ba37ad619165afac7d432eaded6013',
+            'duration': 135,
         },
     }
 
@@ -26,5 +27,4 @@ class VideoDetectiveIE(InfoExtractor):
         webpage = self._download_webpage(url, video_id)
         og_video = self._og_search_video_url(webpage)
         query = compat_urlparse.urlparse(og_video).query
-        return self.url_result(InternetVideoArchiveIE._build_url(query),
-            ie=InternetVideoArchiveIE.ie_key())
+        return self.url_result(InternetVideoArchiveIE._build_url(query), ie=InternetVideoArchiveIE.ie_key())

youtube_dl/extractor/vimeo.py

@@ -8,6 +8,7 @@ import itertools
 from .common import InfoExtractor
 from .subtitles import SubtitlesInfoExtractor
 from ..utils import (
+    compat_HTTPError,
     compat_urllib_parse,
     compat_urllib_request,
     clean_html,
@@ -172,7 +173,18 @@ class VimeoIE(SubtitlesInfoExtractor):
 
         # Retrieve video webpage to extract further information
         request = compat_urllib_request.Request(url, None, headers)
-        webpage = self._download_webpage(request, video_id)
+        try:
+            webpage = self._download_webpage(request, video_id)
+        except ExtractorError as ee:
+            if isinstance(ee.cause, compat_HTTPError) and ee.cause.code == 403:
+                errmsg = ee.cause.read()
+                if b'Because of its privacy settings, this video cannot be played here' in errmsg:
+                    raise ExtractorError(
+                        'Cannot download embed-only video without embedding '
+                        'URL. Please call youtube-dl with the URL of the page '
+                        'that embeds this video.',
+                        expected=True)
+            raise
 
         # Now we begin extracting as much information as we can from what we
         # retrieved. First we extract the information common to all extractors,
@@ -221,7 +233,9 @@ class VimeoIE(SubtitlesInfoExtractor):
         # Extract video thumbnail
         video_thumbnail = config["video"].get("thumbnail")
         if video_thumbnail is None:
-            _, video_thumbnail = sorted((int(width), t_url) for (width, t_url) in config["video"]["thumbs"].items())[-1]
+            video_thumbs = config["video"].get("thumbs")
+            if video_thumbs and isinstance(video_thumbs, dict):
+                _, video_thumbnail = sorted((int(width), t_url) for (width, t_url) in video_thumbs.items())[-1]
 
         # Extract video description
         video_description = None

youtube_dl/extractor/vk.py

@@ -16,7 +16,7 @@ from ..utils import (
 
 class VKIE(InfoExtractor):
     IE_NAME = 'vk.com'
-    _VALID_URL = r'https?://vk\.com/(?:videos.*?\?.*?z=)?video(?P<id>.*?)(?:\?|%2F|$)'
+    _VALID_URL = r'https?://vk\.com/(?:video_ext\.php\?.*?\boid=(?P<oid>\d+).*?\bid=(?P<id>\d+)|(?:videos.*?\?.*?z=)?video(?P<videoid>.*?)(?:\?|%2F|$))'
     _NETRC_MACHINE = 'vk'
 
     _TESTS = [
@@ -42,6 +42,18 @@ class VKIE(InfoExtractor):
                 'duration': 558,
             }
         },
+        {
+            'note': 'Embedded video',
+            'url': 'http://vk.com/video_ext.php?oid=32194266&id=162925554&hash=7d8c2e0d5e05aeaa&hd=1',
+            'md5': 'c7ce8f1f87bec05b3de07fdeafe21a0a',
+            'info_dict': {
+                'id': '162925554',
+                'ext': 'mp4',
+                'uploader': 'Vladimir Gavrin',
+                'title': 'Lin Dan',
+                'duration': 101,
+            }
+        },
         {
             'url': 'http://vk.com/video-8871596_164049491',
             'md5': 'a590bcaf3d543576c9bd162812387666',
@@ -54,7 +66,7 @@ class VKIE(InfoExtractor):
                 'duration': 8352,
             },
             'skip': 'Requires vk account credentials',
-        }
+        },
     ]
 
     def _login(self):
@@ -82,7 +94,10 @@ class VKIE(InfoExtractor):
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
-        video_id = mobj.group('id')
+        video_id = mobj.group('videoid')
+
+        if not video_id:
+            video_id = '%s_%s' % (mobj.group('oid'), mobj.group('id'))
 
         info_url = 'http://vk.com/al_video.php?act=show&al=1&video=%s' % video_id
         info_page = self._download_webpage(info_url, video_id)

youtube_dl/extractor/vube.py

@@ -13,7 +13,7 @@ class VubeIE(InfoExtractor):
 
     _TEST = {
         'url': 'http://vube.com/Chiara+Grispo+Video+Channel/YL2qNPkqon',
-        'md5': 'f81dcf6d0448e3291f54380181695821',
+        'md5': 'db7aba89d4603dadd627e9d1973946fe',
         'info_dict': {
             'id': 'YL2qNPkqon',
             'ext': 'mp4',
@@ -77,4 +77,4 @@ class VubeIE(InfoExtractor):
             'like_count': like_count,
             'dislike_count': dislike_count,
             'comment_count': comment_count,
-        }
+        }

youtube_dl/extractor/xhamster.py

@@ -103,6 +103,7 @@ class XHamsterIE(InfoExtractor):
         }]
 
         if not hd:
+            mrss_url = self._search_regex(r'<link rel="canonical" href="([^"]+)', webpage, 'mrss_url')
             webpage = self._download_webpage(mrss_url + '?hd', video_id, note='Downloading HD webpage')
             if is_hd(webpage):
                 video_url = extract_video_url(webpage)

youtube_dl/extractor/xnxx.py

@@ -1,55 +1,49 @@
+# encoding: utf-8
+from __future__ import unicode_literals
+
 import re
 
 from .common import InfoExtractor
 from ..utils import (
     compat_urllib_parse,
-
-    ExtractorError,
 )
 
 
 class XNXXIE(InfoExtractor):
-    _VALID_URL = r'^(?:https?://)?(?:video|www)\.xnxx\.com/video([0-9]+)/(.*)'
-    VIDEO_URL_RE = r'flv_url=(.*?)&'
-    VIDEO_TITLE_RE = r'<title>(.*?)\s+-\s+XNXX.COM'
-    VIDEO_THUMB_RE = r'url_bigthumb=(.*?)&amp;'
+    _VALID_URL = r'^https?://(?:video|www)\.xnxx\.com/video(?P<id>[0-9]+)/(.*)'
     _TEST = {
-        u'url': u'http://video.xnxx.com/video1135332/lida_naked_funny_actress_5_',
-        u'file': u'1135332.flv',
-        u'md5': u'0831677e2b4761795f68d417e0b7b445',
-        u'info_dict': {
-            u"title": u"lida \u00bb Naked Funny Actress  (5)",
-            u"age_limit": 18,
+        'url': 'http://video.xnxx.com/video1135332/lida_naked_funny_actress_5_',
+        'md5': '0831677e2b4761795f68d417e0b7b445',
+        'info_dict': {
+            'id': '1135332',
+            'ext': 'flv',
+            'title': 'lida » Naked Funny Actress  (5)',
+            'age_limit': 18,
         }
     }
 
     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(1)
+        video_id = mobj.group('id')
 
         # Get webpage content
         webpage = self._download_webpage(url, video_id)
 
-        video_url = self._search_regex(self.VIDEO_URL_RE,
-            webpage, u'video URL')
+        video_url = self._search_regex(r'flv_url=(.*?)&amp;',
+            webpage, 'video URL')
         video_url = compat_urllib_parse.unquote(video_url)
 
-        video_title = self._html_search_regex(self.VIDEO_TITLE_RE,
-            webpage, u'title')
+        video_title = self._html_search_regex(r'<title>(.*?)\s+-\s+XNXX.COM',
+            webpage, 'title')
 
-        video_thumbnail = self._search_regex(self.VIDEO_THUMB_RE,
-            webpage, u'thumbnail', fatal=False)
+        video_thumbnail = self._search_regex(r'url_bigthumb=(.*?)&amp;',
+            webpage, 'thumbnail', fatal=False)
 
-        return [{
+        return {
             'id': video_id,
             'url': video_url,
-            'uploader': None,
-            'upload_date': None,
             'title': video_title,
             'ext': 'flv',
             'thumbnail': video_thumbnail,
-            'description': None,
             'age_limit': 18,
-        }]
+        }

youtube_dl/extractor/xtube.py

@@ -7,19 +7,24 @@ from .common import InfoExtractor
 from ..utils import (
     compat_urllib_parse_urlparse,
     compat_urllib_request,
+    parse_duration,
+    str_to_int,
 )
 
+
 class XTubeIE(InfoExtractor):
-    _VALID_URL = r'^(?:https?://)?(?:www\.)?(?P<url>xtube\.com/watch\.php\?v=(?P<videoid>[^/?&]+))'
+    _VALID_URL = r'https?://(?:www\.)?(?P<url>xtube\.com/watch\.php\?v=(?P<videoid>[^/?&]+))'
     _TEST = {
         'url': 'http://www.xtube.com/watch.php?v=kVTUy_G222_',
-        'file': 'kVTUy_G222_.mp4',
         'md5': '092fbdd3cbe292c920ef6fc6a8a9cdab',
         'info_dict': {
-            "title": "strange erotica",
-            "description": "surreal gay themed erotica...almost an ET kind of thing",
-            "uploader": "greenshowers",
-            "age_limit": 18,
+            'id': 'kVTUy_G222_',
+            'ext': 'mp4',
+            'title': 'strange erotica',
+            'description': 'surreal gay themed erotica...almost an ET kind of thing',
+            'uploader': 'greenshowers',
+            'duration': 450,
+            'age_limit': 18,
         }
     }
 
@@ -32,10 +37,23 @@ class XTubeIE(InfoExtractor):
         req.add_header('Cookie', 'age_verified=1')
         webpage = self._download_webpage(req, video_id)
 
-        video_title = self._html_search_regex(r'<div class="p_5px[^>]*>([^<]+)', webpage, 'title')
-        video_uploader = self._html_search_regex(r'so_s\.addVariable\("owner_u", "([^"]+)', webpage, 'uploader', fatal=False)
-        video_description = self._html_search_regex(r'<p class="video_description">([^<]+)', webpage, 'description', fatal=False)
-        video_url= self._html_search_regex(r'var videoMp4 = "([^"]+)', webpage, 'video_url').replace('\\/', '/')
+        video_title = self._html_search_regex(r'<p class="title">([^<]+)', webpage, 'title')
+        video_uploader = self._html_search_regex(
+            r'so_s\.addVariable\("owner_u", "([^"]+)', webpage, 'uploader', fatal=False)
+        video_description = self._html_search_regex(
+            r'<p class="fieldsDesc">([^<]+)', webpage, 'description', fatal=False)
+        video_url = self._html_search_regex(r'var videoMp4 = "([^"]+)', webpage, 'video_url').replace('\\/', '/')
+        duration = parse_duration(self._html_search_regex(
+            r'<span class="bold">Runtime:</span> ([^<]+)</p>', webpage, 'duration', fatal=False))
+        view_count = self._html_search_regex(
+            r'<span class="bold">Views:</span> ([\d,\.]+)</p>', webpage, 'view count', fatal=False)
+        if view_count:
+            view_count = str_to_int(view_count)
+        comment_count = self._html_search_regex(
+            r'<div id="commentBar">([\d,\.]+) Comments</div>', webpage, 'comment count', fatal=False)
+        if comment_count:
+            comment_count = str_to_int(comment_count)
+
         path = compat_urllib_parse_urlparse(video_url).path
         extension = os.path.splitext(path)[1][1:]
         format = path.split('/')[5].split('_')[:2]
@@ -48,6 +66,9 @@ class XTubeIE(InfoExtractor):
             'title': video_title,
             'uploader': video_uploader,
             'description': video_description,
+            'duration': duration,
+            'view_count': view_count,
+            'comment_count': comment_count,
             'url': video_url,
             'ext': extension,
             'format': format,

youtube_dl/extractor/youtube.py

@@ -1645,7 +1645,7 @@ class YoutubeChannelIE(InfoExtractor):
 
 class YoutubeUserIE(InfoExtractor):
     IE_DESC = u'YouTube.com user videos (URL or "ytuser" keyword)'
-    _VALID_URL = r'(?:(?:(?:https?://)?(?:\w+\.)?youtube\.com/(?:user/)?(?!(?:attribution_link|watch)(?:$|[^a-z_A-Z0-9-])))|ytuser:)(?!feed/)([A-Za-z0-9_-]+)'
+    _VALID_URL = r'(?:(?:(?:https?://)?(?:\w+\.)?youtube\.com/(?:user/)?(?!(?:attribution_link|watch|results)(?:$|[^a-z_A-Z0-9-])))|ytuser:)(?!feed/)([A-Za-z0-9_-]+)'
     _TEMPLATE_URL = 'https://gdata.youtube.com/feeds/api/users/%s'
     _GDATA_PAGE_SIZE = 50
     _GDATA_URL = 'https://gdata.youtube.com/feeds/api/users/%s/uploads?max-results=%d&start-index=%d&alt=json'
@@ -1744,12 +1744,50 @@ class YoutubeSearchIE(SearchInfoExtractor):
                   for video_id in video_ids]
         return self.playlist_result(videos, query)
 
+
 class YoutubeSearchDateIE(YoutubeSearchIE):
     IE_NAME = YoutubeSearchIE.IE_NAME + ':date'
     _API_URL = 'https://gdata.youtube.com/feeds/api/videos?q=%s&start-index=%i&max-results=50&v=2&alt=jsonc&orderby=published'
     _SEARCH_KEY = 'ytsearchdate'
     IE_DESC = u'YouTube.com searches, newest videos first'
 
+
+class YoutubeSearchURLIE(InfoExtractor):
+    IE_DESC = u'YouTube.com search URLs'
+    IE_NAME = u'youtube:search_url'
+    _VALID_URL = r'https?://(?:www\.)?youtube\.com/results\?(.*?&)?search_query=(?P<query>[^&]+)(?:[&]|$)'
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        query = compat_urllib_parse.unquote_plus(mobj.group('query'))
+
+        webpage = self._download_webpage(url, query)
+        result_code = self._search_regex(
+            r'(?s)<ol id="search-results"(.*?)</ol>', webpage, u'result HTML')
+
+        part_codes = re.findall(
+            r'(?s)<h3 class="yt-lockup-title">(.*?)</h3>', result_code)
+        entries = []
+        for part_code in part_codes:
+            part_title = self._html_search_regex(
+                r'(?s)title="([^"]+)"', part_code, 'item title', fatal=False)
+            part_url_snippet = self._html_search_regex(
+                r'(?s)href="([^"]+)"', part_code, 'item URL')
+            part_url = compat_urlparse.urljoin(
+                'https://www.youtube.com/', part_url_snippet)
+            entries.append({
+                '_type': 'url',
+                'url': part_url,
+                'title': part_title,
+            })
+
+        return {
+            '_type': 'playlist',
+            'entries': entries,
+            'title': query,
+        }
+
+
 class YoutubeShowIE(InfoExtractor):
     IE_DESC = u'YouTube.com (multi-season) shows'
     _VALID_URL = r'https?://www\.youtube\.com/show/(.*)'

youtube_dl/utils.py

@@ -772,6 +772,7 @@ def unified_strdate(date_str):
         '%B %d %Y',
         '%b %d %Y',
         '%Y-%m-%d',
+        '%d.%m.%Y',
         '%d/%m/%Y',
         '%Y/%m/%d %H:%M:%S',
         '%Y-%m-%d %H:%M:%S',
@@ -1262,3 +1263,7 @@ def read_batch_urls(batch_fd):
 
     with contextlib.closing(batch_fd) as fd:
         return [url for url in map(fixup, fd) if url]
+
+
+def urlencode_postdata(*args, **kargs):
+    return compat_urllib_parse.urlencode(*args, **kargs).encode('ascii')

youtube_dl/version.py

@@ -1,2 +1,2 @@
 
-__version__ = '2014.02.26'
+__version__ = '2014.03.10'