release 2014.01.22.3

[space] Add support for mobile URLs

README.md

@@ -44,6 +44,10 @@ which means you can modify it, redistribute it or use it however you like.
     --bidi-workaround          Work around terminals that lack bidirectional
                                text support. Requires bidiv or fribidi
                                executable in PATH
+    --default-search PREFIX    Use this prefix for unqualified URLs. For example
+                               "gvsearch2:" downloads two videos from google
+                               videos for  youtube-dl "large apple". By default
+                               (with value "auto") youtube-dl guesses.
 
 ## Video Selection:
     --playlist-start NUMBER    playlist video to start at (default is 1)
@@ -71,6 +75,7 @@ which means you can modify it, redistribute it or use it however you like.
     --download-archive FILE    Download only videos not listed in the archive
                                file. Record the IDs of all downloaded videos in
                                it.
+    --include-ads              Download advertisements as well (experimental)
 
 ## Download Options:
     -r, --rate-limit LIMIT     maximum download rate in bytes per second (e.g.
@@ -153,7 +158,9 @@ which means you can modify it, redistribute it or use it however you like.
 ## Video Format Options:
     -f, --format FORMAT        video format code, specify the order of
                                preference using slashes: "-f 22/17/18". "-f mp4"
-                               and "-f flv" are also supported
+                               and "-f flv" are also supported. You can also use
+                               the special names "best", "bestaudio", and
+                               "worst"
     --all-formats              download all available video formats
     --prefer-free-formats      prefer free video formats unless a specific one
                                is requested

test/test_YoutubeDL.py

@@ -1,5 +1,7 @@
 #!/usr/bin/env python
 
+from __future__ import unicode_literals
+
 # Allow direct execution
 import os
 import sys
@@ -30,125 +32,155 @@ class TestFormatSelection(unittest.TestCase):
         ydl = YDL()
         ydl.params['prefer_free_formats'] = True
         formats = [
-            {u'ext': u'webm', u'height': 460},
+            {'ext': 'webm', 'height': 460},
-            {u'ext': u'mp4',  u'height': 460},
+            {'ext': 'mp4',  'height': 460},
         ]
-        info_dict = {u'formats': formats, u'extractor': u'test'}
+        info_dict = {'formats': formats, 'extractor': 'test'}
         yie = YoutubeIE(ydl)
         yie._sort_formats(info_dict['formats'])
         ydl.process_ie_result(info_dict)
         downloaded = ydl.downloaded_info_dicts[0]
-        self.assertEqual(downloaded[u'ext'], u'webm')
+        self.assertEqual(downloaded['ext'], 'webm')
 
         # Different resolution => download best quality (mp4)
         ydl = YDL()
         ydl.params['prefer_free_formats'] = True
         formats = [
-            {u'ext': u'webm', u'height': 720},
+            {'ext': 'webm', 'height': 720},
-            {u'ext': u'mp4', u'height': 1080},
+            {'ext': 'mp4', 'height': 1080},
         ]
-        info_dict[u'formats'] = formats
+        info_dict['formats'] = formats
         yie = YoutubeIE(ydl)
         yie._sort_formats(info_dict['formats'])
         ydl.process_ie_result(info_dict)
         downloaded = ydl.downloaded_info_dicts[0]
-        self.assertEqual(downloaded[u'ext'], u'mp4')
+        self.assertEqual(downloaded['ext'], 'mp4')
 
         # No prefer_free_formats => prefer mp4 and flv for greater compatibilty
         ydl = YDL()
         ydl.params['prefer_free_formats'] = False
         formats = [
-            {u'ext': u'webm', u'height': 720},
+            {'ext': 'webm', 'height': 720},
-            {u'ext': u'mp4', u'height': 720},
+            {'ext': 'mp4', 'height': 720},
-            {u'ext': u'flv', u'height': 720},
+            {'ext': 'flv', 'height': 720},
         ]
-        info_dict[u'formats'] = formats
+        info_dict['formats'] = formats
         yie = YoutubeIE(ydl)
         yie._sort_formats(info_dict['formats'])
         ydl.process_ie_result(info_dict)
         downloaded = ydl.downloaded_info_dicts[0]
-        self.assertEqual(downloaded[u'ext'], u'mp4')
+        self.assertEqual(downloaded['ext'], 'mp4')
 
         ydl = YDL()
         ydl.params['prefer_free_formats'] = False
         formats = [
-            {u'ext': u'flv', u'height': 720},
+            {'ext': 'flv', 'height': 720},
-            {u'ext': u'webm', u'height': 720},
+            {'ext': 'webm', 'height': 720},
         ]
-        info_dict[u'formats'] = formats
+        info_dict['formats'] = formats
         yie = YoutubeIE(ydl)
         yie._sort_formats(info_dict['formats'])
         ydl.process_ie_result(info_dict)
         downloaded = ydl.downloaded_info_dicts[0]
-        self.assertEqual(downloaded[u'ext'], u'flv')
+        self.assertEqual(downloaded['ext'], 'flv')
 
     def test_format_limit(self):
         formats = [
-            {u'format_id': u'meh', u'url': u'http://example.com/meh', 'preference': 1},
+            {'format_id': 'meh', 'url': 'http://example.com/meh', 'preference': 1},
-            {u'format_id': u'good', u'url': u'http://example.com/good', 'preference': 2},
+            {'format_id': 'good', 'url': 'http://example.com/good', 'preference': 2},
-            {u'format_id': u'great', u'url': u'http://example.com/great', 'preference': 3},
+            {'format_id': 'great', 'url': 'http://example.com/great', 'preference': 3},
-            {u'format_id': u'excellent', u'url': u'http://example.com/exc', 'preference': 4},
+            {'format_id': 'excellent', 'url': 'http://example.com/exc', 'preference': 4},
         ]
         info_dict = {
-            u'formats': formats, u'extractor': u'test', 'id': 'testvid'}
+            'formats': formats, 'extractor': 'test', 'id': 'testvid'}
 
         ydl = YDL()
         ydl.process_ie_result(info_dict)
         downloaded = ydl.downloaded_info_dicts[0]
-        self.assertEqual(downloaded[u'format_id'], u'excellent')
+        self.assertEqual(downloaded['format_id'], 'excellent')
 
         ydl = YDL({'format_limit': 'good'})
         assert ydl.params['format_limit'] == 'good'
         ydl.process_ie_result(info_dict.copy())
         downloaded = ydl.downloaded_info_dicts[0]
-        self.assertEqual(downloaded[u'format_id'], u'good')
+        self.assertEqual(downloaded['format_id'], 'good')
 
         ydl = YDL({'format_limit': 'great', 'format': 'all'})
         ydl.process_ie_result(info_dict.copy())
-        self.assertEqual(ydl.downloaded_info_dicts[0][u'format_id'], u'meh')
+        self.assertEqual(ydl.downloaded_info_dicts[0]['format_id'], 'meh')
-        self.assertEqual(ydl.downloaded_info_dicts[1][u'format_id'], u'good')
+        self.assertEqual(ydl.downloaded_info_dicts[1]['format_id'], 'good')
-        self.assertEqual(ydl.downloaded_info_dicts[2][u'format_id'], u'great')
+        self.assertEqual(ydl.downloaded_info_dicts[2]['format_id'], 'great')
         self.assertTrue('3' in ydl.msgs[0])
 
         ydl = YDL()
         ydl.params['format_limit'] = 'excellent'
         ydl.process_ie_result(info_dict.copy())
         downloaded = ydl.downloaded_info_dicts[0]
-        self.assertEqual(downloaded[u'format_id'], u'excellent')
+        self.assertEqual(downloaded['format_id'], 'excellent')
 
     def test_format_selection(self):
         formats = [
-            {u'format_id': u'35', u'ext': u'mp4', 'preference': 1},
+            {'format_id': '35', 'ext': 'mp4', 'preference': 1},
-            {u'format_id': u'45', u'ext': u'webm', 'preference': 2},
+            {'format_id': '45', 'ext': 'webm', 'preference': 2},
-            {u'format_id': u'47', u'ext': u'webm', 'preference': 3},
+            {'format_id': '47', 'ext': 'webm', 'preference': 3},
-            {u'format_id': u'2', u'ext': u'flv', 'preference': 4},
+            {'format_id': '2', 'ext': 'flv', 'preference': 4},
         ]
-        info_dict = {u'formats': formats, u'extractor': u'test'}
+        info_dict = {'formats': formats, 'extractor': 'test'}
 
-        ydl = YDL({'format': u'20/47'})
+        ydl = YDL({'format': '20/47'})
         ydl.process_ie_result(info_dict.copy())
         downloaded = ydl.downloaded_info_dicts[0]
-        self.assertEqual(downloaded['format_id'], u'47')
+        self.assertEqual(downloaded['format_id'], '47')
 
-        ydl = YDL({'format': u'20/71/worst'})
+        ydl = YDL({'format': '20/71/worst'})
         ydl.process_ie_result(info_dict.copy())
         downloaded = ydl.downloaded_info_dicts[0]
-        self.assertEqual(downloaded['format_id'], u'35')
+        self.assertEqual(downloaded['format_id'], '35')
 
         ydl = YDL()
         ydl.process_ie_result(info_dict.copy())
         downloaded = ydl.downloaded_info_dicts[0]
-        self.assertEqual(downloaded['format_id'], u'2')
+        self.assertEqual(downloaded['format_id'], '2')
 
-        ydl = YDL({'format': u'webm/mp4'})
+        ydl = YDL({'format': 'webm/mp4'})
         ydl.process_ie_result(info_dict.copy())
         downloaded = ydl.downloaded_info_dicts[0]
-        self.assertEqual(downloaded['format_id'], u'47')
+        self.assertEqual(downloaded['format_id'], '47')
 
-        ydl = YDL({'format': u'3gp/40/mp4'})
+        ydl = YDL({'format': '3gp/40/mp4'})
         ydl.process_ie_result(info_dict.copy())
         downloaded = ydl.downloaded_info_dicts[0]
-        self.assertEqual(downloaded['format_id'], u'35')
+        self.assertEqual(downloaded['format_id'], '35')
+
+    def test_format_selection_audio(self):
+        formats = [
+            {'format_id': 'audio-low', 'ext': 'webm', 'preference': 1, 'vcodec': 'none'},
+            {'format_id': 'audio-mid', 'ext': 'webm', 'preference': 2, 'vcodec': 'none'},
+            {'format_id': 'audio-high', 'ext': 'flv', 'preference': 3, 'vcodec': 'none'},
+            {'format_id': 'vid', 'ext': 'mp4', 'preference': 4},
+        ]
+        info_dict = {'formats': formats, 'extractor': 'test'}
+
+        ydl = YDL({'format': 'bestaudio'})
+        ydl.process_ie_result(info_dict.copy())
+        downloaded = ydl.downloaded_info_dicts[0]
+        self.assertEqual(downloaded['format_id'], 'audio-high')
+
+        ydl = YDL({'format': 'worstaudio'})
+        ydl.process_ie_result(info_dict.copy())
+        downloaded = ydl.downloaded_info_dicts[0]
+        self.assertEqual(downloaded['format_id'], 'audio-low')
+
+        formats = [
+            {'format_id': 'vid-low', 'ext': 'mp4', 'preference': 1},
+            {'format_id': 'vid-high', 'ext': 'mp4', 'preference': 2},
+        ]
+        info_dict = {'formats': formats, 'extractor': 'test'}
+
+        ydl = YDL({'format': 'bestaudio/worstaudio/best'})
+        ydl.process_ie_result(info_dict.copy())
+        downloaded = ydl.downloaded_info_dicts[0]
+        self.assertEqual(downloaded['format_id'], 'vid-high')
 
     def test_youtube_format_selection(self):
         order = [
@@ -200,17 +232,17 @@ class TestFormatSelection(unittest.TestCase):
 
     def test_prepare_filename(self):
         info = {
-            u'id': u'1234',
+            'id': '1234',
-            u'ext': u'mp4',
+            'ext': 'mp4',
-            u'width': None,
+            'width': None,
         }
         def fname(templ):
             ydl = YoutubeDL({'outtmpl': templ})
             return ydl.prepare_filename(info)
-        self.assertEqual(fname(u'%(id)s.%(ext)s'), u'1234.mp4')
+        self.assertEqual(fname('%(id)s.%(ext)s'), '1234.mp4')
-        self.assertEqual(fname(u'%(id)s-%(width)s.%(ext)s'), u'1234-NA.mp4')
+        self.assertEqual(fname('%(id)s-%(width)s.%(ext)s'), '1234-NA.mp4')
         # Replace missing fields with 'NA'
-        self.assertEqual(fname(u'%(uploader_date)s-%(id)s.%(ext)s'), u'NA-1234.mp4')
+        self.assertEqual(fname('%(uploader_date)s-%(id)s.%(ext)s'), 'NA-1234.mp4')
 
 
 if __name__ == '__main__':

test/test_subtitles.py

@@ -167,13 +167,13 @@ class TestTedSubtitles(BaseTestSubtitles):
     def test_subtitles(self):
         self.DL.params['writesubtitles'] = True
         subtitles = self.getSubtitles()
-        self.assertEqual(md5(subtitles['en']), '2154f31ff9b9f89a0aa671537559c21d')
+        self.assertEqual(md5(subtitles['en']), '4262c1665ff928a2dada178f62cb8d14')
 
     def test_subtitles_lang(self):
         self.DL.params['writesubtitles'] = True
         self.DL.params['subtitleslangs'] = ['fr']
         subtitles = self.getSubtitles()
-        self.assertEqual(md5(subtitles['fr']), '7616cbc6df20ec2c1204083c83871cf6')
+        self.assertEqual(md5(subtitles['fr']), '66a63f7f42c97a50f8c0e90bc7797bb5')
 
     def test_allsubtitles(self):
         self.DL.params['writesubtitles'] = True

test/test_utils.py

@@ -16,6 +16,7 @@ from youtube_dl.utils import (
     DateRange,
     encodeFilename,
     find_xpath_attr,
+    fix_xml_ampersands,
     get_meta_content,
     orderedSet,
     parse_duration,
@@ -200,5 +201,18 @@ class TestUtil(unittest.TestCase):
         self.assertEqual(parse_duration('9:12:43'), 33163)
         self.assertEqual(parse_duration('x:y'), None)
 
+    def test_fix_xml_ampersands(self):
+        self.assertEqual(
+            fix_xml_ampersands('"&x=y&z=a'), '"&x=y&z=a')
+        self.assertEqual(
+            fix_xml_ampersands('"&x=y&wrong;&z=a'),
+            '"&x=y&wrong;&z=a')
+        self.assertEqual(
+            fix_xml_ampersands('&'><"'),
+            '&'><"')
+        self.assertEqual(
+            fix_xml_ampersands('Ӓ᪼'), 'Ӓ᪼')
+        self.assertEqual(fix_xml_ampersands('&#&#'), '&#&#')
+
 if __name__ == '__main__':
     unittest.main()

test/test_youtube_signature.py

@@ -27,12 +27,6 @@ _TESTS = [
         85,
         u'3456789a0cdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRS[UVWXYZ!"#$%&\'()*+,-./:;<=>?@',
     ),
-    (
-        u'https://s.ytimg.com/yts/swfbin/watch_as3-vflg5GhxU.swf',
-        u'swf',
-        82,
-        u':/.-,+*)=\'&%$#"!ZYX0VUTSRQPONMLKJIHGFEDCBAzyxw>utsrqponmlkjihgfedcba987654321'
-    ),
 ]
 
 

youtube_dl/YoutubeDL.py

@@ -151,6 +151,9 @@ class YoutubeDL(object):
     bidi_workaround:   Work around buggy terminals without bidirectional text
                        support, using fridibi
     debug_printtraffic:Print out sent and received HTTP traffic
+    include_ads:       Download ads as well
+    default_search:    Prepend this string if an input url is not valid.
+                       'auto' for elaborate guessing
 
     The following parameters are not used by YoutubeDL itself, they are used by
     the FileDownloader:
@@ -634,6 +637,18 @@ class YoutubeDL(object):
             return available_formats[-1]
         elif format_spec == 'worst':
             return available_formats[0]
+        elif format_spec == 'bestaudio':
+            audio_formats = [
+                f for f in available_formats
+                if f.get('vcodec') == 'none']
+            if audio_formats:
+                return audio_formats[-1]
+        elif format_spec == 'worstaudio':
+            audio_formats = [
+                f for f in available_formats
+                if f.get('vcodec') == 'none']
+            if audio_formats:
+                return audio_formats[0]
         else:
             extensions = ['mp4', 'flv', 'webm', '3gp']
             if format_spec in extensions:
@@ -698,7 +713,7 @@ class YoutubeDL(object):
             self.list_formats(info_dict)
             return
 
-        req_format = self.params.get('format', 'best')
+        req_format = self.params.get('format')
         if req_format is None:
             req_format = 'best'
         formats_to_download = []
@@ -908,6 +923,14 @@ class YoutubeDL(object):
                     if info_dict.get('requested_formats') is not None:
                         downloaded = []
                         success = True
+                        merger = FFmpegMergerPP(self)
+                        if not merger._get_executable():
+                            postprocessors = []
+                            self.report_warning('You have requested multiple '
+                                'formats but ffmpeg or avconv are not installed.'
+                                ' The formats won\'t be merged')
+                        else:
+                            postprocessors = [merger]
                         for f in info_dict['requested_formats']:
                             new_info = dict(info_dict)
                             new_info.update(f)
@@ -916,7 +939,7 @@ class YoutubeDL(object):
                             downloaded.append(fname)
                             partial_success = dl(fname, new_info)
                             success = success and partial_success
-                        info_dict['__postprocessors'] = [FFmpegMergerPP(self)]
+                        info_dict['__postprocessors'] = postprocessors
                         info_dict['__files_to_merge'] = downloaded
                     else:
                         # Just a single file

youtube_dl/__init__.py

@@ -39,6 +39,7 @@ __authors__  = (
     'Sergey M.',
     'Michael Orlitzky',
     'Chris Gahan',
+    'Saimadhav Heblikar',
 )
 
 __license__ = 'Public Domain'
@@ -198,7 +199,9 @@ def parseOpts(overrideArguments=None):
     general.add_option(
         '--bidi-workaround', dest='bidi_workaround', action='store_true',
         help=u'Work around terminals that lack bidirectional text support. Requires bidiv or fribidi executable in PATH')
-
+    general.add_option('--default-search',
+            dest='default_search', metavar='PREFIX',
+            help='Use this prefix for unqualified URLs. For example "gvsearch2:" downloads two videos from google videos for  youtube-dl "large apple". By default (with value "auto") youtube-dl guesses.')
 
     selection.add_option(
         '--playlist-start',
@@ -237,7 +240,10 @@ def parseOpts(overrideArguments=None):
     selection.add_option('--download-archive', metavar='FILE',
                          dest='download_archive',
                          help='Download only videos not listed in the archive file. Record the IDs of all downloaded videos in it.')
-
+    selection.add_option(
+        '--include-ads', dest='include_ads',
+        action='store_true',
+        help='Download advertisements as well (experimental)')
 
     authentication.add_option('-u', '--username',
             dest='username', metavar='USERNAME', help='account username')
@@ -250,8 +256,8 @@ def parseOpts(overrideArguments=None):
 
 
     video_format.add_option('-f', '--format',
-            action='store', dest='format', metavar='FORMAT', default='best',
+            action='store', dest='format', metavar='FORMAT', default=None,
-            help='video format code, specify the order of preference using slashes: "-f 22/17/18". "-f mp4" and "-f flv" are also supported')
+            help='video format code, specify the order of preference using slashes: "-f 22/17/18". "-f mp4" and "-f flv" are also supported. You can also use the special names "best", "bestaudio", and "worst"')
     video_format.add_option('--all-formats',
             action='store_const', dest='format', help='download all available video formats', const='all')
     video_format.add_option('--prefer-free-formats',
@@ -615,6 +621,12 @@ def _real_main(argv=None):
         date = DateRange.day(opts.date)
     else:
         date = DateRange(opts.dateafter, opts.datebefore)
+    if opts.default_search not in ('auto', None) and ':' not in opts.default_search:
+        parser.error(u'--default-search invalid; did you forget a colon (:) at the end?')
+
+    # Do not download videos when there are audio-only formats
+    if opts.extractaudio and not opts.keepvideo and opts.format is None:
+        opts.format = 'bestaudio/best'
 
     # --all-sub automatically sets --write-sub if --write-auto-sub is not given
     # this was the old behaviour if only --all-sub was given.
@@ -715,6 +727,8 @@ def _real_main(argv=None):
         'bidi_workaround': opts.bidi_workaround,
         'debug_printtraffic': opts.debug_printtraffic,
         'prefer_ffmpeg': opts.prefer_ffmpeg,
+        'include_ads': opts.include_ads,
+        'default_search': opts.default_search,
     }
 
     with YoutubeDL(ydl_opts) as ydl:

youtube_dl/extractor/__init__.py

@@ -47,6 +47,7 @@ from .depositfiles import DepositFilesIE
 from .dotsub import DotsubIE
 from .dreisat import DreiSatIE
 from .defense import DefenseGouvFrIE
+from .dropbox import DropboxIE
 from .ebaumsworld import EbaumsWorldIE
 from .ehow import EHowIE
 from .eighttracks import EightTracksIE
@@ -62,6 +63,7 @@ from .fktv import (
     FKTVPosteckeIE,
 )
 from .flickr import FlickrIE
+from .franceinter import FranceInterIE
 from .francetv import (
     PluzzIE,
     FranceTvInfoIE,
@@ -117,7 +119,10 @@ from .mit import TechTVMITIE, MITIE
 from .mixcloud import MixcloudIE
 from .mpora import MporaIE
 from .mofosex import MofosexIE
-from .mtv import MTVIE
+from .mtv import (
+    MTVIE,
+    MTVIggyIE,
+)
 from .muzu import MuzuTVIE
 from .myspace import MySpaceIE
 from .myspass import MySpassIE
@@ -150,6 +155,7 @@ from .rottentomatoes import RottenTomatoesIE
 from .roxwel import RoxwelIE
 from .rtlnow import RTLnowIE
 from .rutube import RutubeIE
+from .servingsys import ServingSysIE
 from .sina import SinaIE
 from .slashdot import SlashdotIE
 from .slideshare import SlideshareIE
@@ -168,6 +174,7 @@ from .southparkstudios import (
 from .space import SpaceIE
 from .spankwire import SpankwireIE
 from .spiegel import SpiegelIE
+from .spike import SpikeIE
 from .stanfordoc import StanfordOpenClassroomIE
 from .statigram import StatigramIE
 from .steam import SteamIE

youtube_dl/extractor/brightcove.py

@@ -9,9 +9,11 @@ from .common import InfoExtractor
 from ..utils import (
     compat_urllib_parse,
     find_xpath_attr,
+    fix_xml_ampersands,
     compat_urlparse,
     compat_str,
     compat_urllib_request,
+    compat_parse_qs,
 
     ExtractorError,
     unsmuggle_url,
@@ -83,17 +85,33 @@ class BrightcoveIE(InfoExtractor):
                             lambda m: m.group(1) + '/>', object_str)
         # Fix up some stupid XML, see https://github.com/rg3/youtube-dl/issues/1608
         object_str = object_str.replace('<--', '<!--')
+        object_str = fix_xml_ampersands(object_str)
 
         object_doc = xml.etree.ElementTree.fromstring(object_str)
-        assert 'BrightcoveExperience' in object_doc.attrib['class']
+
-        params = {
+        fv_el = find_xpath_attr(object_doc, './param', 'name', 'flashVars')
-            'playerID': find_xpath_attr(object_doc, './param', 'name', 'playerID').attrib['value'],
+        if fv_el is not None:
-        }
+            flashvars = dict(
+                (k, v[0])
+                for k, v in compat_parse_qs(fv_el.attrib['value']).items())
+        else:
+            flashvars = {}
+
         def find_param(name):
+            if name in flashvars:
+                return flashvars[name]
             node = find_xpath_attr(object_doc, './param', 'name', name)
             if node is not None:
                 return node.attrib['value']
             return None
+
+        params = {}
+
+        playerID = find_param('playerID')
+        if playerID is None:
+            raise ExtractorError('Cannot find player ID')
+        params['playerID'] = playerID
+
         playerKey = find_param('playerKey')
         # Not all pages define this value
         if playerKey is not None:
@@ -114,8 +132,12 @@ class BrightcoveIE(InfoExtractor):
         if it can't be found
         """
         m_brightcove = re.search(
-            r'<object[^>]+?class=([\'"])[^>]*?BrightcoveExperience.*?\1.+?</object>',
+            r'''(?sx)<object
-            webpage, re.DOTALL)
+            (?:
+                [^>]+?class=([\'"])[^>]*?BrightcoveExperience.*?\1 |
+                [^>]*?>\s*<param\s+name="movie"\s+value="https?://[^/]*brightcove\.com/
+            ).+?</object>''',
+            webpage)
         if m_brightcove is not None:
             return cls._build_brighcove_url(m_brightcove.group())
         else:
@@ -156,6 +178,7 @@ class BrightcoveIE(InfoExtractor):
         info = self._search_regex(r'var experienceJSON = ({.*?});', webpage, 'json')
         info = json.loads(info)['data']
         video_info = info['programmedContent']['videoPlayer']['mediaDTO']
+        video_info['_youtubedl_adServerURL'] = info.get('adServerURL')
 
         return self._extract_video_info(video_info)
 
@@ -193,6 +216,23 @@ class BrightcoveIE(InfoExtractor):
             info.update({
                 'url': video_info['FLVFullLengthURL'],
             })
-        else:
+
+        if self._downloader.params.get('include_ads', False):
+            adServerURL = video_info.get('_youtubedl_adServerURL')
+            if adServerURL:
+                ad_info = {
+                    '_type': 'url',
+                    'url': adServerURL,
+                }
+                if 'url' in info:
+                    return {
+                        '_type': 'playlist',
+                        'title': info['title'],
+                        'entries': [ad_info, info],
+                    }
+                else:
+                    return ad_info
+
+        if 'url' not in info and not info.get('formats'):
             raise ExtractorError('Unable to extract video url for %s' % info['id'])
         return info

youtube_dl/extractor/clipsyndicate.py

@@ -3,7 +3,7 @@ import re
 from .common import InfoExtractor
 from ..utils import (
     find_xpath_attr,
-    fix_xml_all_ampersand,
+    fix_xml_ampersands
 )
 
 
@@ -33,7 +33,7 @@ class ClipsyndicateIE(InfoExtractor):
         pdoc = self._download_xml(
             'http://eplayer.clipsyndicate.com/osmf/playlist?%s' % flvars,
             video_id, u'Downloading video info',
-            transform_source=fix_xml_all_ampersand) 
+            transform_source=fix_xml_ampersands)
 
         track_doc = pdoc.find('trackList/track')
         def find_param(name):

youtube_dl/extractor/cnn.py

@@ -25,12 +25,13 @@ class CNNIE(InfoExtractor):
         },
     },
     {
-        u"url": u"http://edition.cnn.com/video/?/video/us/2013/08/21/sot-student-gives-epic-speech.georgia-institute-of-technology&utm_source=feedburner&utm_medium=feed&utm_campaign=Feed%3A+rss%2Fcnn_topstories+%28RSS%3A+Top+Stories%29",
+        "url": "http://edition.cnn.com/video/?/video/us/2013/08/21/sot-student-gives-epic-speech.georgia-institute-of-technology&utm_source=feedburner&utm_medium=feed&utm_campaign=Feed%3A+rss%2Fcnn_topstories+%28RSS%3A+Top+Stories%29",
-        u"file": u"us_2013_08_21_sot-student-gives-epic-speech.georgia-institute-of-technology.mp4",
+        "file": "us_2013_08_21_sot-student-gives-epic-speech.georgia-institute-of-technology.mp4",
-        u"md5": u"b5cc60c60a3477d185af8f19a2a26f4e",
+        "md5": "b5cc60c60a3477d185af8f19a2a26f4e",
-        u"info_dict": {
+        "info_dict": {
-            u"title": "Student's epic speech stuns new freshmen",
+            "title": "Student's epic speech stuns new freshmen",
-            u"description": "A Georgia Tech student welcomes the incoming freshmen with an epic speech backed by music from \"2001: A Space Odyssey.\""
+            "description": "A Georgia Tech student welcomes the incoming freshmen with an epic speech backed by music from \"2001: A Space Odyssey.\"",
+            "upload_date": "20130821",
         }
     }]
 

youtube_dl/extractor/comedycentral.py

@@ -1,3 +1,5 @@
+from __future__ import unicode_literals
+
 import re
 
 from .common import InfoExtractor
@@ -15,30 +17,22 @@ class ComedyCentralIE(MTVServicesInfoExtractor):
     _VALID_URL = r'''(?x)https?://(?:www.)?comedycentral.com/
         (video-clips|episodes|cc-studios|video-collections)
         /(?P<title>.*)'''
-    _FEED_URL = u'http://comedycentral.com/feeds/mrss/'
+    _FEED_URL = 'http://comedycentral.com/feeds/mrss/'
 
     _TEST = {
-        u'url': u'http://www.comedycentral.com/video-clips/kllhuv/stand-up-greg-fitzsimmons--uncensored---too-good-of-a-mother',
+        'url': 'http://www.comedycentral.com/video-clips/kllhuv/stand-up-greg-fitzsimmons--uncensored---too-good-of-a-mother',
-        u'md5': u'4167875aae411f903b751a21f357f1ee',
+        'md5': '4167875aae411f903b751a21f357f1ee',
-        u'info_dict': {
+        'info_dict': {
-            u'id': u'cef0cbb3-e776-4bc9-b62e-8016deccb354',
+            'id': 'cef0cbb3-e776-4bc9-b62e-8016deccb354',
-            u'ext': u'mp4',
+            'ext': 'mp4',
-            u'title': u'Uncensored - Greg Fitzsimmons - Too Good of a Mother',
+            'title': 'CC:Stand-Up|Greg Fitzsimmons: Life on Stage|Uncensored - Too Good of a Mother',
-            u'description': u'After a certain point, breastfeeding becomes c**kblocking.',
+            'description': 'After a certain point, breastfeeding becomes c**kblocking.',
         },
     }
 
-    def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
-        title = mobj.group('title')
-        webpage = self._download_webpage(url, title)
-        mgid = self._search_regex(r'data-mgid="(?P<mgid>mgid:.*?)"',
-                                  webpage, u'mgid')
-        return self._get_videos_info(mgid)
-
 
 class ComedyCentralShowsIE(InfoExtractor):
-    IE_DESC = u'The Daily Show / Colbert Report'
+    IE_DESC = 'The Daily Show / Colbert Report'
     # urls can be abbreviations like :thedailyshow or :colbert
     # urls for episodes like:
     # or urls for clips like: http://www.thedailyshow.com/watch/mon-december-10-2012/any-given-gun-day
@@ -55,14 +49,14 @@ class ComedyCentralShowsIE(InfoExtractor):
                               extended-interviews/(?P<interID>[0-9]+)/playlist_tds_extended_(?P<interview_title>.*?)/.*?)))
                      $"""
     _TEST = {
-        u'url': u'http://www.thedailyshow.com/watch/thu-december-13-2012/kristen-stewart',
+        'url': 'http://www.thedailyshow.com/watch/thu-december-13-2012/kristen-stewart',
-        u'file': u'422212.mp4',
+        'file': '422212.mp4',
-        u'md5': u'4e2f5cb088a83cd8cdb7756132f9739d',
+        'md5': '4e2f5cb088a83cd8cdb7756132f9739d',
-        u'info_dict': {
+        'info_dict': {
-            u"upload_date": u"20121214", 
+            "upload_date": "20121214",
-            u"description": u"Kristen Stewart", 
+            "description": "Kristen Stewart",
-            u"uploader": u"thedailyshow", 
+            "uploader": "thedailyshow",
-            u"title": u"thedailyshow-kristen-stewart part 1"
+            "title": "thedailyshow-kristen-stewart part 1"
         }
     }
 
@@ -94,20 +88,20 @@ class ComedyCentralShowsIE(InfoExtractor):
     def _transform_rtmp_url(rtmp_video_url):
         m = re.match(r'^rtmpe?://.*?/(?P<finalid>gsp.comedystor/.*)$', rtmp_video_url)
         if not m:
-            raise ExtractorError(u'Cannot transform RTMP url')
+            raise ExtractorError('Cannot transform RTMP url')
         base = 'http://mtvnmobile.vo.llnwd.net/kip0/_pxn=1+_pxI0=Ripod-h264+_pxL0=undefined+_pxM0=+_pxK=18639+_pxE=mp4/44620/mtvnorigin/'
         return base + m.group('finalid')
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url, re.VERBOSE)
         if mobj is None:
-            raise ExtractorError(u'Invalid URL: %s' % url)
+            raise ExtractorError('Invalid URL: %s' % url)
 
         if mobj.group('shortname'):
             if mobj.group('shortname') in ('tds', 'thedailyshow'):
-                url = u'http://www.thedailyshow.com/full-episodes/'
+                url = 'http://www.thedailyshow.com/full-episodes/'
             else:
-                url = u'http://www.colbertnation.com/full-episodes/'
+                url = 'http://www.colbertnation.com/full-episodes/'
             mobj = re.match(self._VALID_URL, url, re.VERBOSE)
             assert mobj is not None
 
@@ -133,9 +127,9 @@ class ComedyCentralShowsIE(InfoExtractor):
             url = htmlHandle.geturl()
             mobj = re.match(self._VALID_URL, url, re.VERBOSE)
             if mobj is None:
-                raise ExtractorError(u'Invalid redirected URL: ' + url)
+                raise ExtractorError('Invalid redirected URL: ' + url)
             if mobj.group('episode') == '':
-                raise ExtractorError(u'Redirected URL is still not specific: ' + url)
+                raise ExtractorError('Redirected URL is still not specific: ' + url)
             epTitle = mobj.group('episode')
 
         mMovieParams = re.findall('(?:<param name="movie" value="|var url = ")(http://media.mtvnservices.com/([^"]*(?:episode|video).*?:.*?))"', webpage)
@@ -147,15 +141,15 @@ class ComedyCentralShowsIE(InfoExtractor):
 
             altMovieParams = re.findall('data-mgid="([^"]*(?:episode|video).*?:.*?)"', webpage)
             if len(altMovieParams) == 0:
-                raise ExtractorError(u'unable to find Flash URL in webpage ' + url)
+                raise ExtractorError('unable to find Flash URL in webpage ' + url)
             else:
                 mMovieParams = [("http://media.mtvnservices.com/" + altMovieParams[0], altMovieParams[0])]
 
         uri = mMovieParams[0][1]
         indexUrl = 'http://shadow.comedycentral.com/feeds/video_player/mrss/?' + compat_urllib_parse.urlencode({'uri': uri})
         idoc = self._download_xml(indexUrl, epTitle,
-                                          u'Downloading show index',
+                                          'Downloading show index',
-                                          u'unable to download episode index')
+                                          'unable to download episode index')
 
         results = []
 
@@ -170,7 +164,7 @@ class ComedyCentralShowsIE(InfoExtractor):
             configUrl = ('http://www.comedycentral.com/global/feeds/entertainment/media/mediaGenEntertainment.jhtml?' +
                         compat_urllib_parse.urlencode({'uri': mediaId}))
             cdoc = self._download_xml(configUrl, epTitle,
-                                               u'Downloading configuration for %s' % shortMediaId)
+                                               'Downloading configuration for %s' % shortMediaId)
 
             turls = []
             for rendition in cdoc.findall('.//rendition'):
@@ -178,7 +172,7 @@ class ComedyCentralShowsIE(InfoExtractor):
                 turls.append(finfo)
 
             if len(turls) == 0:
-                self._downloader.report_error(u'unable to download ' + mediaId + ': No videos found')
+                self._downloader.report_error('unable to download ' + mediaId + ': No videos found')
                 continue
 
             formats = []
@@ -192,7 +186,7 @@ class ComedyCentralShowsIE(InfoExtractor):
                     'width': w,
                 })
 
-            effTitle = showId + u'-' + epTitle + u' part ' + compat_str(partNum+1)
+            effTitle = showId + '-' + epTitle + ' part ' + compat_str(partNum+1)
             results.append({
                 'id': shortMediaId,
                 'formats': formats,

youtube_dl/extractor/common.py

@@ -1,4 +1,5 @@
 import base64
+import hashlib
 import json
 import os
 import re
@@ -219,6 +220,8 @@ class InfoExtractor(object):
                           webpage_bytes[:1024])
             if m:
                 encoding = m.group(1).decode('ascii')
+            elif webpage_bytes.startswith(b'\xff\xfe'):
+                encoding = 'utf-16'
             else:
                 encoding = 'utf-8'
         if self._downloader.params.get('dump_intermediate_pages', False):
@@ -234,6 +237,9 @@ class InfoExtractor(object):
                 url = url_or_request.get_full_url()
             except AttributeError:
                 url = url_or_request
+            if len(url) > 200:
+                h = u'___' + hashlib.md5(url).hexdigest()
+                url = url[:200 - len(h)] + h
             raw_filename = ('%s_%s.dump' % (video_id, url))
             filename = sanitize_filename(raw_filename, restricted=True)
             self.to_screen(u'Saving request to ' + filename)

youtube_dl/extractor/condenast.py

@@ -1,4 +1,5 @@
 # coding: utf-8
+from __future__ import unicode_literals
 
 import re
 import json
@@ -20,30 +21,31 @@ class CondeNastIE(InfoExtractor):
 
     # The keys are the supported sites and the values are the name to be shown
     # to the user and in the extractor description.
-    _SITES = {'wired': u'WIRED',
+    _SITES = {
-              'gq': u'GQ',
+        'wired': 'WIRED',
-              'vogue': u'Vogue',
+        'gq': 'GQ',
-              'glamour': u'Glamour',
+        'vogue': 'Vogue',
-              'wmagazine': u'W Magazine',
+        'glamour': 'Glamour',
-              'vanityfair': u'Vanity Fair',
+        'wmagazine': 'W Magazine',
-              }
+        'vanityfair': 'Vanity Fair',
+    }
 
     _VALID_URL = r'http://(video|www).(?P<site>%s).com/(?P<type>watch|series|video)/(?P<id>.+)' % '|'.join(_SITES.keys())
-    IE_DESC = u'Condé Nast media group: %s' % ', '.join(sorted(_SITES.values()))
+    IE_DESC = 'Condé Nast media group: %s' % ', '.join(sorted(_SITES.values()))
 
     _TEST = {
-        u'url': u'http://video.wired.com/watch/3d-printed-speakers-lit-with-led',
+        'url': 'http://video.wired.com/watch/3d-printed-speakers-lit-with-led',
-        u'file': u'5171b343c2b4c00dd0c1ccb3.mp4',
+        'file': '5171b343c2b4c00dd0c1ccb3.mp4',
-        u'md5': u'1921f713ed48aabd715691f774c451f7',
+        'md5': '1921f713ed48aabd715691f774c451f7',
-        u'info_dict': {
+        'info_dict': {
-            u'title': u'3D Printed Speakers Lit With LED',
+            'title': '3D Printed Speakers Lit With LED',
-            u'description': u'Check out these beautiful 3D printed LED speakers.  You can\'t actually buy them, but LumiGeek is working on a board that will let you make you\'re own.',
+            'description': 'Check out these beautiful 3D printed LED speakers.  You can\'t actually buy them, but LumiGeek is working on a board that will let you make you\'re own.',
         }
     }
 
     def _extract_series(self, url, webpage):
         title = self._html_search_regex(r'<div class="cne-series-info">.*?<h1>(.+?)</h1>',
-                                        webpage, u'series title', flags=re.DOTALL)
+                                        webpage, 'series title', flags=re.DOTALL)
         url_object = compat_urllib_parse_urlparse(url)
         base_url = '%s://%s' % (url_object.scheme, url_object.netloc)
         m_paths = re.finditer(r'<p class="cne-thumb-title">.*?<a href="(/watch/.+?)["\?]',
@@ -57,39 +59,41 @@ class CondeNastIE(InfoExtractor):
         description = self._html_search_regex([r'<div class="cne-video-description">(.+?)</div>',
                                                r'<div class="video-post-content">(.+?)</div>',
                                                ],
-                                              webpage, u'description',
+                                              webpage, 'description',
                                               fatal=False, flags=re.DOTALL)
         params = self._search_regex(r'var params = {(.+?)}[;,]', webpage,
-                                    u'player params', flags=re.DOTALL)
+                                    'player params', flags=re.DOTALL)
-        video_id = self._search_regex(r'videoId: [\'"](.+?)[\'"]', params, u'video id')
+        video_id = self._search_regex(r'videoId: [\'"](.+?)[\'"]', params, 'video id')
-        player_id = self._search_regex(r'playerId: [\'"](.+?)[\'"]', params, u'player id')
+        player_id = self._search_regex(r'playerId: [\'"](.+?)[\'"]', params, 'player id')
-        target = self._search_regex(r'target: [\'"](.+?)[\'"]', params, u'target')
+        target = self._search_regex(r'target: [\'"](.+?)[\'"]', params, 'target')
         data = compat_urllib_parse.urlencode({'videoId': video_id,
                                               'playerId': player_id,
                                               'target': target,
                                               })
         base_info_url = self._search_regex(r'url = [\'"](.+?)[\'"][,;]',
-                                           webpage, u'base info url',
+                                           webpage, 'base info url',
                                            default='http://player.cnevids.com/player/loader.js?')
         info_url = base_info_url + data
         info_page = self._download_webpage(info_url, video_id,
-                                           u'Downloading video info')
+                                           'Downloading video info')
-        video_info = self._search_regex(r'var video = ({.+?});', info_page, u'video info')
+        video_info = self._search_regex(r'var video = ({.+?});', info_page, 'video info')
         video_info = json.loads(video_info)
 
-        def _formats_sort_key(f):
+        formats = [{
-            type_ord = 1 if f['type'] == 'video/mp4' else 0
+            'format_id': '%s-%s' % (fdata['type'].split('/')[-1], fdata['quality']),
-            quality_ord = 1 if f['quality'] == 'high' else 0
+            'url': fdata['src'],
-            return (quality_ord, type_ord)
+            'ext': fdata['type'].split('/')[-1],
-        best_format = sorted(video_info['sources'][0], key=_formats_sort_key)[-1]
+            'quality': 1 if fdata['quality'] == 'high' else 0,
+        } for fdata in video_info['sources'][0]]
+        self._sort_formats(formats)
 
-        return {'id': video_id,
+        return {
-                'url': best_format['src'],
+            'id': video_id,
-                'ext': best_format['type'].split('/')[-1],
+            'formats': formats,
-                'title': video_info['title'],
+            'title': video_info['title'],
-                'thumbnail': video_info['poster_frame'],
+            'thumbnail': video_info['poster_frame'],
-                'description': description,
+            'description': description,
-                }
+        }
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)

youtube_dl/extractor/cspan.py

@@ -10,7 +10,7 @@ from ..utils import (
 
 
 class CSpanIE(InfoExtractor):
-    _VALID_URL = r'http://www\.c-spanvideo\.org/program/(.*)'
+    _VALID_URL = r'http://(?:www\.)?c-spanvideo\.org/program/(?P<name>.*)'
     IE_DESC = 'C-SPAN'
     _TEST = {
         'url': 'http://www.c-spanvideo.org/program/HolderonV',
@@ -24,9 +24,9 @@ class CSpanIE(InfoExtractor):
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
-        prog_name = mobj.group(1)
+        prog_name = mobj.group('name')
         webpage = self._download_webpage(url, prog_name)
-        video_id = self._search_regex(r'programid=(.*?)&', webpage, 'video id')
+        video_id = self._search_regex(r'prog(?:ram)?id=(.*?)&', webpage, 'video id')
 
         title = self._html_search_regex(
             r'<!-- title -->\n\s*<h1[^>]*>(.*?)</h1>', webpage, 'title')

youtube_dl/extractor/d8.py

@@ -1,22 +1,25 @@
 # encoding: utf-8
+from __future__ import unicode_literals
+
 from .canalplus import CanalplusIE
 
 
 class D8IE(CanalplusIE):
     _VALID_URL = r'https?://www\.d8\.tv/.*?/(?P<path>.*)'
     _VIDEO_INFO_TEMPLATE = 'http://service.canal-plus.com/video/rest/getVideosLiees/d8/%s'
-    IE_NAME = u'd8.tv'
+    IE_NAME = 'd8.tv'
 
     _TEST = {
-        u'url': u'http://www.d8.tv/d8-docs-mags/pid6589-d8-campagne-intime.html',
+        'url': 'http://www.d8.tv/d8-docs-mags/pid6589-d8-campagne-intime.html',
-        u'file': u'966289.flv',
+        'file': '966289.flv',
-        u'info_dict': {
+        'info_dict': {
-            u'title': u'Campagne intime - Documentaire exceptionnel',
+            'title': 'Campagne intime - Documentaire exceptionnel',
-            u'description': u'md5:d2643b799fb190846ae09c61e59a859f',
+            'description': 'md5:d2643b799fb190846ae09c61e59a859f',
-            u'upload_date': u'20131108',
+            'upload_date': '20131108',
         },
-        u'params': {
+        'params': {
             # rtmp
-            u'skip_download': True,
+            'skip_download': True,
         },
+        'skip': 'videos get deleted after a while',
     }

youtube_dl/extractor/dropbox.py

@@ -0,0 +1,31 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+import os.path
+import re
+
+from .common import InfoExtractor
+
+
+class DropboxIE(InfoExtractor):
+    _VALID_URL = r'https?://(?:www\.)?dropbox[.]com/s/(?P<id>[a-zA-Z0-9]{15})/(?P<title>[^?#]*)'
+    _TEST = {
+        'url': 'https://www.dropbox.com/s/mcnzehi9wo55th4/20131219_085616.mp4',
+        'file': 'mcnzehi9wo55th4.mp4',
+        'md5': 'f6d65b1b326e82fd7ab7720bea3dacae',
+        'info_dict': {
+            'title': '20131219_085616'
+        }
+    }
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        video_id = mobj.group('id')
+        title = os.path.splitext(mobj.group('title'))[0]
+        video_url = url + '?dl=1'
+
+        return {
+            'id': video_id,
+            'title': title,
+            'url': video_url,
+        }

youtube_dl/extractor/facebook.py

@@ -17,7 +17,12 @@ from ..utils import (
 class FacebookIE(InfoExtractor):
     """Information Extractor for Facebook"""
 
-    _VALID_URL = r'^(?:https?://)?(?:\w+\.)?facebook\.com/(?:[^#?]*#!/)?(?:video/video|photo)\.php\?(?:.*?)v=(?P<ID>\d+)(?:.*)'
+    _VALID_URL = r'''(?x)
+        (?:https?://)?(?:\w+\.)?facebook\.com/
+        (?:[^#?]*\#!/)?
+        (?:video/video\.php|photo\.php|video/embed)\?(?:.*?)
+        (?:v|video_id)=(?P<id>[0-9]+)
+        (?:.*)'''
     _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'
@@ -90,7 +95,7 @@ class FacebookIE(InfoExtractor):
         mobj = re.match(self._VALID_URL, url)
         if mobj is None:
             raise ExtractorError(u'Invalid URL: %s' % url)
-        video_id = mobj.group('ID')
+        video_id = mobj.group('id')
 
         url = 'https://www.facebook.com/video/video.php?v=%s' % video_id
         webpage = self._download_webpage(url, video_id)

youtube_dl/extractor/flickr.py

@@ -1,3 +1,5 @@
+from __future__ import unicode_literals
+
 import re
 
 from .common import InfoExtractor
@@ -11,13 +13,13 @@ class FlickrIE(InfoExtractor):
     """Information Extractor for Flickr videos"""
     _VALID_URL = r'(?:https?://)?(?:www\.|secure\.)?flickr\.com/photos/(?P<uploader_id>[\w\-_@]+)/(?P<id>\d+).*'
     _TEST = {
-        u'url': u'http://www.flickr.com/photos/forestwander-nature-pictures/5645318632/in/photostream/',
+        'url': 'http://www.flickr.com/photos/forestwander-nature-pictures/5645318632/in/photostream/',
-        u'file': u'5645318632.mp4',
+        'file': '5645318632.mp4',
-        u'md5': u'6fdc01adbc89d72fc9c4f15b4a4ba87b',
+        'md5': '6fdc01adbc89d72fc9c4f15b4a4ba87b',
-        u'info_dict': {
+        'info_dict': {
-            u"description": u"Waterfalls in the Springtime at Dark Hollow Waterfalls. These are located just off of Skyline Drive in Virginia. They are only about 6/10 of a mile hike but it is a pretty steep hill and a good climb back up.", 
+            "description": "Waterfalls in the Springtime at Dark Hollow Waterfalls. These are located just off of Skyline Drive in Virginia. They are only about 6/10 of a mile hike but it is a pretty steep hill and a good climb back up.", 
-            u"uploader_id": u"forestwander-nature-pictures", 
+            "uploader_id": "forestwander-nature-pictures", 
-            u"title": u"Dark Hollow Waterfalls"
+            "title": "Dark Hollow Waterfalls"
         }
     }
 
@@ -29,13 +31,13 @@ class FlickrIE(InfoExtractor):
         webpage_url = 'http://www.flickr.com/photos/' + video_uploader_id + '/' + video_id
         webpage = self._download_webpage(webpage_url, video_id)
 
-        secret = self._search_regex(r"photo_secret: '(\w+)'", webpage, u'secret')
+        secret = self._search_regex(r"photo_secret: '(\w+)'", webpage, 'secret')
 
         first_url = 'https://secure.flickr.com/apps/video/video_mtl_xml.gne?v=x&photo_id=' + video_id + '&secret=' + secret + '&bitrate=700&target=_self'
         first_xml = self._download_webpage(first_url, video_id, 'Downloading first data webpage')
 
         node_id = self._html_search_regex(r'<Item id="id">(\d+-\d+)</Item>',
-            first_xml, u'node_id')
+            first_xml, 'node_id')
 
         second_url = 'https://secure.flickr.com/video_playlist.gne?node_id=' + node_id + '&tech=flash&mode=playlist&bitrate=700&secret=' + secret + '&rd=video.yahoo.com&noad=1'
         second_xml = self._download_webpage(second_url, video_id, 'Downloading second data webpage')
@@ -44,7 +46,7 @@ class FlickrIE(InfoExtractor):
 
         mobj = re.search(r'<STREAM APP="(.+?)" FULLPATH="(.+?)"', second_xml)
         if mobj is None:
-            raise ExtractorError(u'Unable to extract video url')
+            raise ExtractorError('Unable to extract video url')
         video_url = mobj.group(1) + unescapeHTML(mobj.group(2))
 
         return [{

youtube_dl/extractor/franceinter.py

@@ -0,0 +1,38 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+import re
+
+from .common import InfoExtractor
+
+
+class FranceInterIE(InfoExtractor):
+    _VALID_URL = r'http://(?:www\.)?franceinter\.fr/player/reecouter\?play=(?P<id>[0-9]{6})'
+    _TEST = {
+        'url': 'http://www.franceinter.fr/player/reecouter?play=793962',
+        'file': '793962.mp3',
+        'md5': '4764932e466e6f6c79c317d2e74f6884',
+        "info_dict": {
+            "title": "L’Histoire dans les jeux vidéo",
+        },
+    }
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        video_id = mobj.group('id')
+
+        webpage = self._download_webpage(url, video_id)
+        title = self._html_search_regex(
+            r'<span class="roll_overflow">(.*?)</span></h1>', webpage, 'title')
+        path = self._search_regex(
+            r'&urlAOD=(.*?)&startTime', webpage, 'video url')
+        video_url = 'http://www.franceinter.fr/' + path
+
+        return {
+            'id': video_id,
+            'formats': [{
+                'url': video_url,
+                'vcodec': 'none',
+            }],
+            'title': title,
+        }

youtube_dl/extractor/gamespot.py

@@ -1,3 +1,5 @@
+from __future__ import unicode_literals
+
 import re
 import json
 
@@ -13,12 +15,12 @@ from ..utils import (
 class GameSpotIE(InfoExtractor):
     _VALID_URL = r'(?:http://)?(?:www\.)?gamespot\.com/.*-(?P<page_id>\d+)/?'
     _TEST = {
-        u"url": u"http://www.gamespot.com/arma-iii/videos/arma-iii-community-guide-sitrep-i-6410818/",
+        "url": "http://www.gamespot.com/arma-iii/videos/arma-iii-community-guide-sitrep-i-6410818/",
-        u"file": u"gs-2300-6410818.mp4",
+        "file": "gs-2300-6410818.mp4",
-        u"md5": u"b2a30deaa8654fcccd43713a6b6a4825",
+        "md5": "b2a30deaa8654fcccd43713a6b6a4825",
-        u"info_dict": {
+        "info_dict": {
-            u"title": u"Arma 3 - Community Guide: SITREP I",
+            "title": "Arma 3 - Community Guide: SITREP I",
-            u'description': u'Check out this video where some of the basics of Arma 3 is explained.',
+            'description': 'Check out this video where some of the basics of Arma 3 is explained.',
         }
     }
 

youtube_dl/extractor/gametrailers.py

@@ -1,4 +1,4 @@
-import re
+from __future__ import unicode_literals
 
 from .mtv import MTVServicesInfoExtractor
 
@@ -6,22 +6,13 @@ from .mtv import MTVServicesInfoExtractor
 class GametrailersIE(MTVServicesInfoExtractor):
     _VALID_URL = r'http://www\.gametrailers\.com/(?P<type>videos|reviews|full-episodes)/(?P<id>.*?)/(?P<title>.*)'
     _TEST = {
-        u'url': u'http://www.gametrailers.com/videos/zbvr8i/mirror-s-edge-2-e3-2013--debut-trailer',
+        'url': 'http://www.gametrailers.com/videos/zbvr8i/mirror-s-edge-2-e3-2013--debut-trailer',
-        u'file': u'70e9a5d7-cf25-4a10-9104-6f3e7342ae0d.mp4',
+        'file': '70e9a5d7-cf25-4a10-9104-6f3e7342ae0d.mp4',
-        u'md5': u'4c8e67681a0ea7ec241e8c09b3ea8cf7',
+        'md5': '4c8e67681a0ea7ec241e8c09b3ea8cf7',
-        u'info_dict': {
+        'info_dict': {
-            u'title': u'E3 2013: Debut Trailer',
+            'title': 'Mirror\'s Edge 2|E3 2013: Debut Trailer',
-            u'description': u'Faith is back!  Check out the World Premiere trailer for Mirror\'s Edge 2 straight from the EA Press Conference at E3 2013!',
+            'description': 'Faith is back!  Check out the World Premiere trailer for Mirror\'s Edge 2 straight from the EA Press Conference at E3 2013!',
         },
     }
 
     _FEED_URL = 'http://www.gametrailers.com/feeds/mrss'
-
-    def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
-        video_id = mobj.group('id')
-        webpage = self._download_webpage(url, video_id)
-        mgid = self._search_regex([r'data-video="(?P<mgid>mgid:.*?)"',
-                                   r'data-contentId=\'(?P<mgid>mgid:.*?)\''],
-                                  webpage, u'mgid')
-        return self._get_videos_info(mgid)

youtube_dl/extractor/generic.py

@@ -92,11 +92,12 @@ 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',
                 'ext': 'mp4',
-                'title': '2cc213299525360.mov', #that's what we get
+                'title': '2cc213299525360.mov',  # that's what we get
             },
         },
     ]
@@ -161,8 +162,19 @@ class GenericIE(InfoExtractor):
     def _real_extract(self, url):
         parsed_url = compat_urlparse.urlparse(url)
         if not parsed_url.scheme:
-            self._downloader.report_warning('The url doesn\'t specify the protocol, trying with http')
+            default_search = self._downloader.params.get('default_search')
-            return self.url_result('http://' + url)
+            if default_search is None:
+                default_search = 'auto'
+
+            if default_search == 'auto':
+                if '/' in url:
+                    self._downloader.report_warning('The url doesn\'t specify the protocol, trying with http')
+                    return self.url_result('http://' + url)
+                else:
+                    return self.url_result('ytsearch:' + url)
+            else:
+                assert ':' in default_search
+                return self.url_result(default_search + url)
         video_id = os.path.splitext(url.split('/')[-1])[0]
 
         self.to_screen('%s: Requesting header' % video_id)
@@ -318,6 +330,12 @@ class GenericIE(InfoExtractor):
         if mobj is not None:
             return self.url_result(mobj.group('url'), 'Novamov')
 
+        # Look for embedded Facebook player
+        mobj = re.search(
+            r'<iframe[^>]+?src=(["\'])(?P<url>https://www.facebook.com/video/embed.+?)\1', webpage)
+        if mobj is not None:
+            return self.url_result(mobj.group('url'), 'Facebook')
+
         # Start with something easy: JW Player in SWFObject
         mobj = re.search(r'flashvars: [\'"](?:.*&)?file=(http[^\'"&]*)', webpage)
         if mobj is None:

youtube_dl/extractor/hotnewhiphop.py

@@ -1,17 +1,25 @@
+from __future__ import unicode_literals
+
 import re
 import base64
 
 from .common import InfoExtractor
+from ..utils import (
+    compat_urllib_parse,
+    compat_urllib_request,
+    ExtractorError,
+    HEADRequest,
+)
 
 
 class HotNewHipHopIE(InfoExtractor):
     _VALID_URL = r'http://www\.hotnewhiphop.com/.*\.(?P<id>.*)\.html'
     _TEST = {
-        u'url': u"http://www.hotnewhiphop.com/freddie-gibbs-lay-it-down-song.1435540.html",
+        'url': 'http://www.hotnewhiphop.com/freddie-gibbs-lay-it-down-song.1435540.html',
-        u'file': u'1435540.mp3',
+        'file': '1435540.mp3',
-        u'md5': u'2c2cd2f76ef11a9b3b581e8b232f3d96',
+        'md5': '2c2cd2f76ef11a9b3b581e8b232f3d96',
-        u'info_dict': {
+        'info_dict': {
-            u"title": u'Freddie Gibbs "Lay It Down"'
+            'title': 'Freddie Gibbs - Lay It Down'
         }
     }
 
@@ -21,24 +29,41 @@ class HotNewHipHopIE(InfoExtractor):
 
         webpage_src = self._download_webpage(url, video_id)
 
-        video_url_base64 = self._search_regex(r'data-path="(.*?)"',
+        video_url_base64 = self._search_regex(
-            webpage_src, u'video URL', fatal=False)
+            r'data-path="(.*?)"', webpage_src, u'video URL', fatal=False)
 
-        if video_url_base64 == None:
+        if video_url_base64 is None:
-            video_url = self._search_regex(r'"contentUrl" content="(.*?)"', webpage_src,
+            video_url = self._search_regex(
-                u'video URL')
+                r'"contentUrl" content="(.*?)"', webpage_src, u'video URL')
             return self.url_result(video_url, ie='Youtube')
 
-        video_url = base64.b64decode(video_url_base64).decode('utf-8')
+        reqdata = compat_urllib_parse.urlencode([
+            ('mediaType', 's'),
+            ('mediaId', video_id),
+        ])
+        r = compat_urllib_request.Request(
+            'http://www.hotnewhiphop.com/ajax/media/getActions/', data=reqdata)
+        r.add_header('Content-Type', 'application/x-www-form-urlencoded')
+        mkd = self._download_json(
+            r, video_id, note='Requesting media key',
+            errnote='Could not download media key')
+        if 'mediaKey' not in mkd:
+            raise ExtractorError('Did not get a media key')
 
-        video_title = self._html_search_regex(r"<title>(.*)</title>",
+        redirect_url = base64.b64decode(video_url_base64).decode('utf-8')
-            webpage_src, u'title')
+        redirect_req = HEADRequest(redirect_url)
+        req = self._request_webpage(
+            redirect_req, video_id,
+            note='Resolving final URL', errnote='Could not resolve final URL')
+        video_url = req.geturl()
+        if video_url.endswith('.html'):
+            raise ExtractorError('Redirect failed')
 
-        results = [{
+        video_title = self._og_search_title(webpage_src).strip()
-                    'id': video_id,
+
-                    'url' : video_url,
+        return {
-                    'title' : video_title,
+            'id': video_id,
-                    'thumbnail' : self._og_search_thumbnail(webpage_src),
+            'url': video_url,
-                    'ext' : 'mp3',
+            'title': video_title,
-                    }]
+            'thumbnail': self._og_search_thumbnail(webpage_src),
-        return results
+        }

youtube_dl/extractor/imdb.py

@@ -68,22 +68,15 @@ class ImdbListIE(InfoExtractor):
         mobj = re.match(self._VALID_URL, url)
         list_id = mobj.group('id')
 
-        # RSS XML is sometimes malformed
+        webpage = self._download_webpage(url, list_id)
-        rss = self._download_webpage('http://rss.imdb.com/list/%s' % list_id, list_id, 'Downloading list RSS')
+        list_code = self._search_regex(
-        list_title = self._html_search_regex(r'<title>(.*?)</title>', rss, 'list title')
+            r'(?s)<div\s+class="list\sdetail">(.*?)class="see-more"',
+            webpage, 'list code')
+        entries = [
+            self.url_result('http://www.imdb.com' + m, 'Imdb')
+            for m in re.findall(r'href="(/video/imdb/vi[^"]+)"', webpage)]
 
-        # Export is independent of actual author_id, but returns 404 if no author_id is provided.
+        list_title = self._html_search_regex(
-        # However, passing dummy author_id seems to be enough.
+            r'<h1 class="header">(.*?)</h1>', webpage, 'list title')
-        csv = self._download_webpage('http://www.imdb.com/list/export?list_id=%s&author_id=ur00000000' % list_id,
-                                     list_id, 'Downloading list CSV')
-        
-        entries = []
-        for item in csv.split('\n')[1:]:
-            cols = item.split(',')
-            if len(cols) < 2:
-                continue
-            item_id = cols[1][1:-1]
-            if item_id.startswith('vi'):
-                entries.append(self.url_result('http://www.imdb.com/video/imdb/%s' % item_id, 'Imdb'))
 
         return self.playlist_result(entries, list_id, list_title)

youtube_dl/extractor/kankan.py

@@ -1,21 +1,24 @@
+from __future__ import unicode_literals
+
 import re
 import hashlib
 
 from .common import InfoExtractor
-from ..utils import determine_ext
 
 _md5 = lambda s: hashlib.md5(s.encode('utf-8')).hexdigest()
 
+
 class KankanIE(InfoExtractor):
     _VALID_URL = r'https?://(?:.*?\.)?kankan\.com/.+?/(?P<id>\d+)\.shtml'
     
     _TEST = {
-        u'url': u'http://yinyue.kankan.com/vod/48/48863.shtml',
+        'url': 'http://yinyue.kankan.com/vod/48/48863.shtml',
-        u'file': u'48863.flv',
+        'file': '48863.flv',
-        u'md5': u'29aca1e47ae68fc28804aca89f29507e',
+        'md5': '29aca1e47ae68fc28804aca89f29507e',
-        u'info_dict': {
+        'info_dict': {
-            u'title': u'Ready To Go',
+            'title': 'Ready To Go',
         },
+        'skip': 'Only available from China',
     }
 
     def _real_extract(self, url):
@@ -23,22 +26,23 @@ class KankanIE(InfoExtractor):
         video_id = mobj.group('id')
         webpage = self._download_webpage(url, video_id)
 
-        title = self._search_regex(r'(?:G_TITLE=|G_MOVIE_TITLE = )[\'"](.+?)[\'"]', webpage, u'video title')
+        title = self._search_regex(r'(?:G_TITLE=|G_MOVIE_TITLE = )[\'"](.+?)[\'"]', webpage, 'video title')
         surls = re.search(r'surls:\[\'.+?\'\]|lurl:\'.+?\.flv\'', webpage).group(0)
         gcids = re.findall(r"http://.+?/.+?/(.+?)/", surls)
         gcid = gcids[-1]
 
-        video_info_page = self._download_webpage('http://p2s.cl.kankan.com/getCdnresource_flv?gcid=%s' % gcid,
+        info_url = 'http://p2s.cl.kankan.com/getCdnresource_flv?gcid=%s' % gcid
-                                                 video_id, u'Downloading video url info')
+        video_info_page = self._download_webpage(
-        ip = self._search_regex(r'ip:"(.+?)"', video_info_page, u'video url ip')
+            info_url, video_id, 'Downloading video url info')
-        path = self._search_regex(r'path:"(.+?)"', video_info_page, u'video url path')
+        ip = self._search_regex(r'ip:"(.+?)"', video_info_page, 'video url ip')
-        param1 = self._search_regex(r'param1:(\d+)', video_info_page, u'param1')
+        path = self._search_regex(r'path:"(.+?)"', video_info_page, 'video url path')
-        param2 = self._search_regex(r'param2:(\d+)', video_info_page, u'param2')
+        param1 = self._search_regex(r'param1:(\d+)', video_info_page, 'param1')
+        param2 = self._search_regex(r'param2:(\d+)', video_info_page, 'param2')
         key = _md5('xl_mp43651' + param1 + param2)
         video_url = 'http://%s%s?key=%s&key1=%s' % (ip, path, key, param2)
 
-        return {'id': video_id,
+        return {
-                'title': title,
+            'id': video_id,
-                'url': video_url,
+            'title': title,
-                'ext': determine_ext(video_url),
+            'url': video_url,
-                }
+        }

youtube_dl/extractor/metacritic.py

@@ -4,7 +4,7 @@ import re
 
 from .common import InfoExtractor
 from ..utils import (
-    fix_xml_all_ampersand,
+    fix_xml_ampersands,
 )
 
 
@@ -27,7 +27,7 @@ class MetacriticIE(InfoExtractor):
         webpage = self._download_webpage(url, video_id)
         # The xml is not well formatted, there are raw '&'
         info = self._download_xml('http://www.metacritic.com/video_data?video=' + video_id,
-            video_id, 'Downloading info xml', transform_source=fix_xml_all_ampersand)
+            video_id, 'Downloading info xml', transform_source=fix_xml_ampersands)
 
         clip = next(c for c in info.findall('playList/clip') if c.find('id').text == video_id)
         formats = []

youtube_dl/extractor/mixcloud.py

@@ -1,4 +1,5 @@
-import json
+from __future__ import unicode_literals
+
 import re
 
 from .common import InfoExtractor
@@ -10,17 +11,17 @@ from ..utils import (
 
 class MixcloudIE(InfoExtractor):
     _VALID_URL = r'^(?:https?://)?(?:www\.)?mixcloud\.com/([\w\d-]+)/([\w\d-]+)'
-    IE_NAME = u'mixcloud'
+    IE_NAME = 'mixcloud'
 
     _TEST = {
-        u'url': u'http://www.mixcloud.com/dholbach/cryptkeeper/',
+        'url': 'http://www.mixcloud.com/dholbach/cryptkeeper/',
-        u'file': u'dholbach-cryptkeeper.mp3',
+        'file': 'dholbach-cryptkeeper.mp3',
-        u'info_dict': {
+        'info_dict': {
-            u'title': u'Cryptkeeper',
+            'title': 'Cryptkeeper',
-            u'description': u'After quite a long silence from myself, finally another Drum\'n\'Bass mix with my favourite current dance floor bangers.',
+            'description': 'After quite a long silence from myself, finally another Drum\'n\'Bass mix with my favourite current dance floor bangers.',
-            u'uploader': u'Daniel Holbach',
+            'uploader': 'Daniel Holbach',
-            u'uploader_id': u'dholbach',
+            'uploader_id': 'dholbach',
-            u'upload_date': u'20111115',
+            'upload_date': '20111115',
         },
     }
 
@@ -42,17 +43,18 @@ class MixcloudIE(InfoExtractor):
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
-
         uploader = mobj.group(1)
         cloudcast_name = mobj.group(2)
         track_id = '-'.join((uploader, cloudcast_name))
-        api_url = 'http://api.mixcloud.com/%s/%s/' % (uploader, cloudcast_name)
-        webpage = self._download_webpage(url, track_id)
-        json_data = self._download_webpage(api_url, track_id,
-            u'Downloading cloudcast info')
-        info = json.loads(json_data)
 
-        preview_url = self._search_regex(r'data-preview-url="(.+?)"', webpage, u'preview url')
+        webpage = self._download_webpage(url, track_id)
+
+        api_url = 'http://api.mixcloud.com/%s/%s/' % (uploader, cloudcast_name)
+        info = self._download_json(
+            api_url, track_id, 'Downloading cloudcast info')
+
+        preview_url = self._search_regex(
+            r'\s(?:data-preview-url|m-preview)="(.+?)"', webpage, 'preview url')
         song_url = preview_url.replace('/previews/', '/c/originals/')
         template_url = re.sub(r'(stream\d*)', 'stream%d', song_url)
         final_song_url = self._get_url(template_url)

youtube_dl/extractor/mpora.py

@@ -34,7 +34,7 @@ class MporaIE(InfoExtractor):
 
         data = json.loads(data_json)
 
-        uploader = data['info_overlay']['name']
+        uploader = data['info_overlay'].get('username')
         duration = data['video']['duration'] // 1000
         thumbnail = data['video']['encodings']['sd']['poster']
         title = data['info_overlay']['title']

youtube_dl/extractor/mtv.py

@@ -1,12 +1,18 @@
+from __future__ import unicode_literals
+
 import re
-import xml.etree.ElementTree
 
 from .common import InfoExtractor
 from ..utils import (
     compat_urllib_parse,
     ExtractorError,
+    find_xpath_attr,
+    fix_xml_ampersands,
+    url_basename,
+    RegexNotFoundError,
 )
 
+
 def _media_xml_tag(tag):
     return '{http://search.yahoo.com/mrss/}%s' % tag
 
@@ -33,10 +39,9 @@ class MTVServicesInfoExtractor(InfoExtractor):
         else:
             return thumb_node.attrib['url']
 
-    def _extract_video_formats(self, metadataXml):
+    def _extract_video_formats(self, mdoc):
-        if '/error_country_block.swf' in metadataXml:
+        if re.match(r'.*/error_country_block\.swf$', mdoc.find('.//src').text) is not None:
-            raise ExtractorError(u'This video is not available from your country.', expected=True)
+            raise ExtractorError('This video is not available from your country.', expected=True)
-        mdoc = xml.etree.ElementTree.fromstring(metadataXml.encode('utf-8'))
 
         formats = []
         for rendition in mdoc.findall('.//rendition'):
@@ -59,11 +64,12 @@ class MTVServicesInfoExtractor(InfoExtractor):
         self.report_extraction(video_id)
         mediagen_url = itemdoc.find('%s/%s' % (_media_xml_tag('group'), _media_xml_tag('content'))).attrib['url']
         # Remove the templates, like &device={device}
-        mediagen_url = re.sub(r'&[^=]*?={.*?}(?=(&|$))', u'', mediagen_url)
+        mediagen_url = re.sub(r'&[^=]*?={.*?}(?=(&|$))', '', mediagen_url)
         if 'acceptMethods' not in mediagen_url:
             mediagen_url += '&acceptMethods=fms'
-        mediagen_page = self._download_webpage(mediagen_url, video_id,
+
-                                               u'Downloading video urls')
+        mediagen_doc = self._download_xml(mediagen_url, video_id,
+            'Downloading video urls')
 
         description_node = itemdoc.find('description')
         if description_node is not None:
@@ -71,9 +77,23 @@ class MTVServicesInfoExtractor(InfoExtractor):
         else:
             description = None
 
+        title_el = None
+        if title_el is None:
+            title_el = find_xpath_attr(
+                itemdoc, './/{http://search.yahoo.com/mrss/}category',
+                'scheme', 'urn:mtvn:video_title')
+        if title_el is None:
+            title_el = itemdoc.find('.//{http://search.yahoo.com/mrss/}title')
+        if title_el is None:
+            title_el = itemdoc.find('.//title')
+        title = title_el.text
+        if title is None:
+            raise ExtractorError('Could not find video title')
+        title = title.strip()
+
         return {
-            'title': itemdoc.find('title').text,
+            'title': title,
-            'formats': self._extract_video_formats(mediagen_page),
+            'formats': self._extract_video_formats(mediagen_doc),
             'id': video_id,
             'thumbnail': self._get_thumbnail_url(uri, itemdoc),
             'description': description,
@@ -83,14 +103,25 @@ class MTVServicesInfoExtractor(InfoExtractor):
         video_id = self._id_from_uri(uri)
         data = compat_urllib_parse.urlencode({'uri': uri})
 
-        def fix_ampersand(s):
-            """ Fix unencoded ampersand in XML """
-            return s.replace(u'& ', '& ')
         idoc = self._download_xml(
             self._FEED_URL + '?' + data, video_id,
-            u'Downloading info', transform_source=fix_ampersand)
+            'Downloading info', transform_source=fix_xml_ampersands)
         return [self._get_video_info(item) for item in idoc.findall('.//item')]
 
+    def _real_extract(self, url):
+        title = url_basename(url)
+        webpage = self._download_webpage(url, title)
+        try:
+            # the url can be http://media.mtvnservices.com/fb/{mgid}.swf
+            # or http://media.mtvnservices.com/{mgid}
+            og_url = self._og_search_video_url(webpage)
+            mgid = url_basename(og_url)
+            if mgid.endswith('.swf'):
+                mgid = mgid[:-4]
+        except RegexNotFoundError:
+            mgid = self._search_regex(r'data-mgid="(.*?)"', webpage, u'mgid')
+        return self._get_videos_info(mgid)
+
 
 class MTVIE(MTVServicesInfoExtractor):
     _VALID_URL = r'''(?x)^https?://
@@ -101,25 +132,25 @@ class MTVIE(MTVServicesInfoExtractor):
 
     _TESTS = [
         {
-            u'url': u'http://www.mtv.com/videos/misc/853555/ours-vh1-storytellers.jhtml',
+            'url': 'http://www.mtv.com/videos/misc/853555/ours-vh1-storytellers.jhtml',
-            u'file': u'853555.mp4',
+            'file': '853555.mp4',
-            u'md5': u'850f3f143316b1e71fa56a4edfd6e0f8',
+            'md5': '850f3f143316b1e71fa56a4edfd6e0f8',
-            u'info_dict': {
+            'info_dict': {
-                u'title': u'Taylor Swift - "Ours (VH1 Storytellers)"',
+                'title': 'Taylor Swift - "Ours (VH1 Storytellers)"',
-                u'description': u'Album: Taylor Swift performs "Ours" for VH1 Storytellers at Harvey Mudd College.',
+                'description': 'Album: Taylor Swift performs "Ours" for VH1 Storytellers at Harvey Mudd College.',
             },
         },
         {
-            u'add_ie': ['Vevo'],
+            'add_ie': ['Vevo'],
-            u'url': u'http://www.mtv.com/videos/taylor-swift/916187/everything-has-changed-ft-ed-sheeran.jhtml',
+            'url': 'http://www.mtv.com/videos/taylor-swift/916187/everything-has-changed-ft-ed-sheeran.jhtml',
-            u'file': u'USCJY1331283.mp4',
+            'file': 'USCJY1331283.mp4',
-            u'md5': u'73b4e7fcadd88929292fe52c3ced8caf',
+            'md5': '73b4e7fcadd88929292fe52c3ced8caf',
-            u'info_dict': {
+            'info_dict': {
-                u'title': u'Everything Has Changed',
+                'title': 'Everything Has Changed',
-                u'upload_date': u'20130606',
+                'upload_date': '20130606',
-                u'uploader': u'Taylor Swift',
+                'uploader': 'Taylor Swift',
             },
-            u'skip': u'VEVO is only available in some countries',
+            'skip': 'VEVO is only available in some countries',
         },
     ]
 
@@ -138,8 +169,22 @@ class MTVIE(MTVServicesInfoExtractor):
                                webpage, re.DOTALL)
             if m_vevo:
                 vevo_id = m_vevo.group(1);
-                self.to_screen(u'Vevo video detected: %s' % vevo_id)
+                self.to_screen('Vevo video detected: %s' % vevo_id)
                 return self.url_result('vevo:%s' % vevo_id, ie='Vevo')
     
-            uri = self._html_search_regex(r'/uri/(.*?)\?', webpage, u'uri')
+            uri = self._html_search_regex(r'/uri/(.*?)\?', webpage, 'uri')
         return self._get_videos_info(uri)
+
+
+class MTVIggyIE(MTVServicesInfoExtractor):
+    IE_NAME = 'mtviggy.com'
+    _VALID_URL = r'https?://www\.mtviggy\.com/videos/.+'
+    _TEST = {
+        'url': 'http://www.mtviggy.com/videos/arcade-fire-behind-the-scenes-at-the-biggest-music-experiment-yet/',
+        'info_dict': {
+            'id': '984696',
+            'ext': 'mp4',
+            'title': 'Arcade Fire: Behind the Scenes at the Biggest Music Experiment Yet',
+        }
+    }
+    _FEED_URL = 'http://all.mtvworldverticals.com/feed-xml/'

youtube_dl/extractor/myspace.py

@@ -1,3 +1,5 @@
+from __future__ import unicode_literals
+
 import re
 import json
 
@@ -8,41 +10,75 @@ from ..utils import (
 
 
 class MySpaceIE(InfoExtractor):
-    _VALID_URL = r'https?://myspace\.com/([^/]+)/video/[^/]+/(?P<id>\d+)'
+    _VALID_URL = r'https?://myspace\.com/([^/]+)/(?P<mediatype>video/[^/]+/|music/song/.*?)(?P<id>\d+)'
 
-    _TEST = {
+    _TESTS = [
-        u'url': u'https://myspace.com/coldplay/video/viva-la-vida/100008689',
+        {
-        u'info_dict': {
+            'url': 'https://myspace.com/coldplay/video/viva-la-vida/100008689',
-            u'id': u'100008689',
+            'info_dict': {
-            u'ext': u'flv',
+                'id': '100008689',
-            u'title': u'Viva La Vida',
+                'ext': 'flv',
-            u'description': u'The official Viva La Vida video, directed by Hype Williams',
+                'title': 'Viva La Vida',
-            u'uploader': u'Coldplay',
+                'description': 'The official Viva La Vida video, directed by Hype Williams',
-            u'uploader_id': u'coldplay',
+                'uploader': 'Coldplay',
+                'uploader_id': 'coldplay',
+            },
+            'params': {
+                # rtmp download
+                'skip_download': True,
+            },
         },
-        u'params': {
+        # song
-            # rtmp download
+        {
-            u'skip_download': True,
+            'url': 'https://myspace.com/spiderbags/music/song/darkness-in-my-heart-39008454-27041242',
+            'info_dict': {
+                'id': '39008454',
+                'ext': 'flv',
+                'title': 'Darkness In My Heart',
+                'uploader_id': 'spiderbags',
+            },
+            'params': {
+                # rtmp download
+                'skip_download': True,
+            },
         },
-    }
+    ]
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
         video_id = mobj.group('id')
         webpage = self._download_webpage(url, video_id)
-        context = json.loads(self._search_regex(r'context = ({.*?});', webpage,
-            u'context'))
-        video = context['video']
-        rtmp_url, play_path = video['streamUrl'].split(';', 1)
 
-        return {
+        if mobj.group('mediatype').startswith('music/song'):
-            'id': compat_str(video['mediaId']),
+            # songs don't store any useful info in the 'context' variable
-            'title': video['title'],
+            def search_data(name):
+                return self._search_regex(r'data-%s="(.*?)"' % name, webpage,
+                    name)
+            streamUrl = search_data('stream-url')
+            info = {
+                'id': video_id,
+                'title': self._og_search_title(webpage),
+                'uploader_id': search_data('artist-username'),
+                'thumbnail': self._og_search_thumbnail(webpage),
+            }
+        else:
+            context = json.loads(self._search_regex(r'context = ({.*?});', webpage,
+                u'context'))
+            video = context['video']
+            streamUrl = video['streamUrl']
+            info = {
+                'id': compat_str(video['mediaId']),
+                'title': video['title'],
+                'description': video['description'],
+                'thumbnail': video['imageUrl'],
+                'uploader': video['artistName'],
+                'uploader_id': video['artistUsername'],
+            }
+
+        rtmp_url, play_path = streamUrl.split(';', 1)
+        info.update({
             'url': rtmp_url,
             'play_path': play_path,
             'ext': 'flv',
-            'description': video['description'],
+        })
-            'thumbnail': video['imageUrl'],
+        return info
-            'uploader': video['artistName'],
-            'uploader_id': video['artistUsername'],
-        }

youtube_dl/extractor/novamov.py

@@ -19,7 +19,8 @@ class NovamovIE(InfoExtractor):
         'info_dict': {
             'title': 'search engine optimization',
             'description': 'search engine optimization is used to rank the web page in the google search engine'
-        }
+        },
+        'skip': '"Invalid token" errors abound (in web interface as well as youtube-dl, there is nothing we can do about it.)'
     }
 
     def _real_extract(self, url):

youtube_dl/extractor/redtube.py

@@ -1,3 +1,5 @@
+from __future__ import unicode_literals
+
 import re
 
 from .common import InfoExtractor
@@ -6,14 +8,14 @@ from .common import InfoExtractor
 class RedTubeIE(InfoExtractor):
     _VALID_URL = r'http://(?:www\.)?redtube\.com/(?P<id>[0-9]+)'
     _TEST = {
-        u'url': u'http://www.redtube.com/66418',
+        'url': 'http://www.redtube.com/66418',
-        u'file': u'66418.mp4',
+        'file': '66418.mp4',
         # md5 varies from time to time, as in
         # https://travis-ci.org/rg3/youtube-dl/jobs/14052463#L295
-        #u'md5': u'7b8c22b5e7098a3e1c09709df1126d2d',
+        #'md5': u'7b8c22b5e7098a3e1c09709df1126d2d',
-        u'info_dict': {
+        'info_dict': {
-            u"title": u"Sucked on a toilet",
+            "title": "Sucked on a toilet",
-            u"age_limit": 18,
+            "age_limit": 18,
         }
     }
 
@@ -33,14 +35,19 @@ class RedTubeIE(InfoExtractor):
             r'<h1 class="videoTitle[^"]*">(.+?)</h1>',
             webpage, u'title')
 
+        video_thumbnail = self._html_search_regex(
+            r'playerInnerHTML.+?<img\s+src="(.+?)"',
+            webpage, u'thumbnail', fatal=False)
+
         # No self-labeling, but they describe themselves as
         # "Home of Videos Porno"
         age_limit = 18
 
         return {
-            'id':        video_id,
+            'id': video_id,
-            'url':       video_url,
+            'url': video_url,
-            'ext':       video_extension,
+            'ext': video_extension,
-            'title':     video_title,
+            'title': video_title,
+            'thumbnail': video_thumbnail,
             'age_limit': age_limit,
         }

youtube_dl/extractor/ringtv.py

@@ -1,37 +1,44 @@
+from __future__ import unicode_literals
+
 import re
 
 from .common import InfoExtractor
 
 
 class RingTVIE(InfoExtractor):
-    _VALID_URL = r'(?:http://)?(?:www\.)?ringtv\.craveonline\.com/videos/video/([^/]+)'
+    _VALID_URL = r'(?:http://)?(?:www\.)?ringtv\.craveonline\.com/(?P<type>news|videos/video)/(?P<id>[^/?#]+)'
     _TEST = {
-        u"url": u"http://ringtv.craveonline.com/videos/video/746619-canelo-alvarez-talks-about-mayweather-showdown",
+        "url": "http://ringtv.craveonline.com/news/310833-luis-collazo-says-victor-ortiz-better-not-quit-on-jan-30",
-        u"file": u"746619.mp4",
+        "file": "857645.mp4",
-        u"md5": u"7c46b4057d22de32e0a539f017e64ad3",
+        "md5": "d25945f5df41cdca2d2587165ac28720",
-        u"info_dict": {
+        "info_dict": {
-            u"title": u"Canelo Alvarez talks about Mayweather showdown",
+            "title": 'Video: Luis Collazo says Victor Ortiz "better not quit on Jan. 30" - Ring TV',
-            u"description": u"Saul \\\"Canelo\\\" Alvarez spoke to the media about his Sept. 14 showdown with Floyd Mayweather after their kick-off presser in NYC. Canelo is motivated and confident that he will have the speed and gameplan to beat the pound-for-pound king."
+            "description": 'Luis Collazo is excited about his Jan. 30 showdown with fellow former welterweight titleholder Victor Ortiz at Barclays Center in his hometown of Brooklyn. The SuperBowl week fight headlines a Golden Boy Live! card on Fox Sports 1.',
         }
     }
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
-        video_id = mobj.group(1).split('-')[0]
+        video_id = mobj.group('id').split('-')[0]
         webpage = self._download_webpage(url, video_id)
-        title = self._search_regex(r'<title>(.+?)</title>',
+
-        		webpage, 'video title').replace(' | RingTV','')
+        if mobj.group('type') == 'news':
-        description = self._search_regex(r'<div class="blurb">(.+?)</div>',
+            video_id = self._search_regex(
-        		webpage, 'Description')
+                r'''(?x)<iframe[^>]+src="http://cms\.springboardplatform\.com/
-        final_url = "http://ringtv.craveonline.springboardplatform.com/storage/ringtv.craveonline.com/conversion/%s.mp4" %(str(video_id))
+                        embed_iframe/[0-9]+/video/([0-9]+)/''',
-        thumbnail_url = "http://ringtv.craveonline.springboardplatform.com/storage/ringtv.craveonline.com/snapshots/%s.jpg" %(str(video_id))
+                webpage, 'real video ID')
-        ext = final_url.split('.')[-1]
+        title = self._og_search_title(webpage)
-        return [{
+        description = self._html_search_regex(
-            'id'          : video_id,
+            r'addthis:description="([^"]+)"',
-            'url'         : final_url,
+            webpage, 'description', fatal=False)
-            'ext'         : ext,
+        final_url = "http://ringtv.craveonline.springboardplatform.com/storage/ringtv.craveonline.com/conversion/%s.mp4" % video_id
-            'title'       : title,
+        thumbnail_url = "http://ringtv.craveonline.springboardplatform.com/storage/ringtv.craveonline.com/snapshots/%s.jpg" % video_id
-            'thumbnail'   : thumbnail_url,
+
-            'description' : description,
+        return {
-        }]
+            'id': video_id,
+            'url': final_url,
+            'title': title,
+            'thumbnail': thumbnail_url,
+            'description': description,
+        }
 

youtube_dl/extractor/servingsys.py

@@ -0,0 +1,71 @@
+from __future__ import unicode_literals
+
+import re
+
+from .common import InfoExtractor
+from ..utils import (
+    int_or_none,
+)
+
+
+class ServingSysIE(InfoExtractor):
+    _VALID_URL = r'https?://(?:[^.]+\.)?serving-sys\.com/BurstingPipe/adServer\.bs\?.*?&pli=(?P<id>[0-9]+)'
+
+    _TEST = {
+        'url': 'http://bs.serving-sys.com/BurstingPipe/adServer.bs?cn=is&c=23&pl=VAST&pli=5349193&PluID=0&pos=7135&ord=[timestamp]&cim=1?',
+        'playlist': [{
+            'file': '29955898.flv',
+            'md5': 'baed851342df6846eb8677a60a011a0f',
+            'info_dict': {
+                'title': 'AdAPPter_Hyundai_demo (1)',
+                'duration': 74,
+                'tbr': 1378,
+                'width': 640,
+                'height': 400,
+            },
+        }, {
+            'file': '29907998.flv',
+            'md5': '979b4da2655c4bc2d81aeb915a8c5014',
+            'info_dict': {
+                'title': 'AdAPPter_Hyundai_demo (2)',
+                'duration': 34,
+                'width': 854,
+                'height': 480,
+                'tbr': 516,
+            },
+        }],
+        'params': {
+            'playlistend': 2,
+        },
+        'skip': 'Blocked in the US [sic]',
+    }
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        pl_id = mobj.group('id')
+
+        vast_doc = self._download_xml(url, pl_id)
+        title = vast_doc.find('.//AdTitle').text
+        media = vast_doc.find('.//MediaFile').text
+        info_url = self._search_regex(r'&adData=([^&]+)&', media, 'info URL')
+
+        doc = self._download_xml(info_url, pl_id, 'Downloading video info')
+        entries = [{
+            '_type': 'video',
+            'id': a.attrib['id'],
+            'title': '%s (%s)' % (title, a.attrib['assetID']),
+            'url': a.attrib['URL'],
+            'duration': int_or_none(a.attrib.get('length')),
+            'tbr': int_or_none(a.attrib.get('bitrate')),
+            'height': int_or_none(a.attrib.get('height')),
+            'width': int_or_none(a.attrib.get('width')),
+        } for a in doc.findall('.//AdditionalAssets/asset')]
+
+        return {
+            '_type': 'playlist',
+            'id': pl_id,
+            'title': title,
+            'entries': entries,
+        }
+
+ 

youtube_dl/extractor/soundcloud.py

@@ -1,4 +1,6 @@
 # encoding: utf-8
+from __future__ import unicode_literals
+
 import json
 import re
 import itertools
@@ -32,58 +34,58 @@ class SoundcloudIE(InfoExtractor):
                        |(?P<player>(?:w|player|p.)\.soundcloud\.com/player/?.*?url=.*)
                     )
                     '''
-    IE_NAME = u'soundcloud'
+    IE_NAME = 'soundcloud'
     _TESTS = [
         {
-            u'url': u'http://soundcloud.com/ethmusic/lostin-powers-she-so-heavy',
+            'url': 'http://soundcloud.com/ethmusic/lostin-powers-she-so-heavy',
-            u'file': u'62986583.mp3',
+            'file': '62986583.mp3',
-            u'md5': u'ebef0a451b909710ed1d7787dddbf0d7',
+            'md5': 'ebef0a451b909710ed1d7787dddbf0d7',
-            u'info_dict': {
+            'info_dict': {
-                u"upload_date": u"20121011", 
+                "upload_date": "20121011",
-                u"description": u"No Downloads untill we record the finished version this weekend, i was too pumped n i had to post it , earl is prolly gonna b hella p.o'd", 
+                "description": "No Downloads untill we record the finished version this weekend, i was too pumped n i had to post it , earl is prolly gonna b hella p.o'd",
-                u"uploader": u"E.T. ExTerrestrial Music", 
+                "uploader": "E.T. ExTerrestrial Music",
-                u"title": u"Lostin Powers - She so Heavy (SneakPreview) Adrian Ackers Blueprint 1"
+                "title": "Lostin Powers - She so Heavy (SneakPreview) Adrian Ackers Blueprint 1"
             }
         },
         # not streamable song
         {
-            u'url': u'https://soundcloud.com/the-concept-band/goldrushed-mastered?in=the-concept-band/sets/the-royal-concept-ep',
+            'url': 'https://soundcloud.com/the-concept-band/goldrushed-mastered?in=the-concept-band/sets/the-royal-concept-ep',
-            u'info_dict': {
+            'info_dict': {
-                u'id': u'47127627',
+                'id': '47127627',
-                u'ext': u'mp3',
+                'ext': 'mp3',
-                u'title': u'Goldrushed',
+                'title': 'Goldrushed',
-                u'uploader': u'The Royal Concept',
+                'uploader': 'The Royal Concept',
-                u'upload_date': u'20120521',
+                'upload_date': '20120521',
             },
-            u'params': {
+            'params': {
                 # rtmp
-                u'skip_download': True,
+                'skip_download': True,
             },
         },
         # private link
         {
-            u'url': u'https://soundcloud.com/jaimemf/youtube-dl-test-video-a-y-baw/s-8Pjrp',
+            'url': 'https://soundcloud.com/jaimemf/youtube-dl-test-video-a-y-baw/s-8Pjrp',
-            u'md5': u'aa0dd32bfea9b0c5ef4f02aacd080604',
+            'md5': 'aa0dd32bfea9b0c5ef4f02aacd080604',
-            u'info_dict': {
+            'info_dict': {
-                u'id': u'123998367',
+                'id': '123998367',
-                u'ext': u'mp3',
+                'ext': 'mp3',
-                u'title': u'Youtube - Dl Test Video \'\' Ä↭',
+                'title': 'Youtube - Dl Test Video \'\' Ä↭',
-                u'uploader': u'jaimeMF',
+                'uploader': 'jaimeMF',
-                u'description': u'test chars:  \"\'/\\ä↭',
+                'description': 'test chars:  \"\'/\\ä↭',
-                u'upload_date': u'20131209',
+                'upload_date': '20131209',
             },
         },
         # downloadable song
         {
-            u'url': u'https://soundcloud.com/simgretina/just-your-problem-baby-1',
+            'url': 'https://soundcloud.com/simgretina/just-your-problem-baby-1',
-            u'md5': u'56a8b69568acaa967b4c49f9d1d52d19',
+            'md5': '56a8b69568acaa967b4c49f9d1d52d19',
-            u'info_dict': {
+            'info_dict': {
-                u'id': u'105614606',
+                'id': '105614606',
-                u'ext': u'wav',
+                'ext': 'wav',
-                u'title': u'Just Your Problem Baby (Acapella)',
+                'title': 'Just Your Problem Baby (Acapella)',
-                u'description': u'Vocals',
+                'description': 'Vocals',
-                u'uploader': u'Sim Gretina',
+                'uploader': 'Sim Gretina',
-                u'upload_date': u'20130815',
+                'upload_date': '20130815',
             },
         },
     ]
@@ -112,7 +114,7 @@ class SoundcloudIE(InfoExtractor):
         thumbnail = info['artwork_url']
         if thumbnail is not None:
             thumbnail = thumbnail.replace('-large', '-t500x500')
-        ext = u'mp3'
+        ext = 'mp3'
         result = {
             'id': track_id,
             'uploader': info['user']['username'],
@@ -124,11 +126,11 @@ class SoundcloudIE(InfoExtractor):
         if info.get('downloadable', False):
             # We can build a direct link to the song
             format_url = (
-                u'https://api.soundcloud.com/tracks/{0}/download?client_id={1}'.format(
+                'https://api.soundcloud.com/tracks/{0}/download?client_id={1}'.format(
                     track_id, self._CLIENT_ID))
             result['formats'] = [{
                 'format_id': 'download',
-                'ext': info.get('original_format', u'mp3'),
+                'ext': info.get('original_format', 'mp3'),
                 'url': format_url,
                 'vcodec': 'none',
             }]
@@ -138,7 +140,7 @@ class SoundcloudIE(InfoExtractor):
                 'client_id={1}&secret_token={2}'.format(track_id, self._IPHONE_CLIENT_ID, secret_token))
             stream_json = self._download_webpage(
                 streams_url,
-                track_id, u'Downloading track url')
+                track_id, 'Downloading track url')
 
             formats = []
             format_dict = json.loads(stream_json)
@@ -165,20 +167,19 @@ class SoundcloudIE(InfoExtractor):
                 # We fallback to the stream_url in the original info, this
                 # cannot be always used, sometimes it can give an HTTP 404 error
                 formats.append({
-                    'format_id': u'fallback',
+                    'format_id': 'fallback',
                     'url': info['stream_url'] + '?client_id=' + self._CLIENT_ID,
                     'ext': ext,
                     'vcodec': 'none',
                 })
 
-            def format_pref(f):
+            for f in formats:
                 if f['format_id'].startswith('http'):
-                    return 2
+                    f['protocol'] = 'http'
                 if f['format_id'].startswith('rtmp'):
-                    return 1
+                    f['protocol'] = 'rtmp'
-                return 0
 
-            formats.sort(key=format_pref)
+            self._sort_formats(formats)
             result['formats'] = formats
 
         return result
@@ -210,14 +211,14 @@ class SoundcloudIE(InfoExtractor):
     
             url = 'http://soundcloud.com/%s' % resolve_title
             info_json_url = self._resolv_url(url)
-        info_json = self._download_webpage(info_json_url, full_title, u'Downloading info JSON')
+        info_json = self._download_webpage(info_json_url, full_title, 'Downloading info JSON')
 
         info = json.loads(info_json)
         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-]+)(?:[?].*)?$'
-    IE_NAME = u'soundcloud:set'
+    IE_NAME = 'soundcloud:set'
     # it's in tests/test_playlists.py
     _TESTS = []
 
@@ -254,7 +255,7 @@ class SoundcloudSetIE(SoundcloudIE):
 
 class SoundcloudUserIE(SoundcloudIE):
     _VALID_URL = r'https?://(www\.)?soundcloud\.com/(?P<user>[^/]+)(/?(tracks/)?)?(\?.*)?$'
-    IE_NAME = u'soundcloud:user'
+    IE_NAME = 'soundcloud:user'
 
     # it's in tests/test_playlists.py
     _TESTS = []
@@ -266,7 +267,7 @@ class SoundcloudUserIE(SoundcloudIE):
         url = 'http://soundcloud.com/%s/' % uploader
         resolv_url = self._resolv_url(url)
         user_json = self._download_webpage(resolv_url, uploader,
-            u'Downloading user info')
+            'Downloading user info')
         user = json.loads(user_json)
 
         tracks = []
@@ -276,7 +277,7 @@ class SoundcloudUserIE(SoundcloudIE):
                                                   })
             tracks_url = 'http://api.soundcloud.com/users/%s/tracks.json?' % user['id'] + data
             response = self._download_webpage(tracks_url, uploader, 
-                u'Downloading tracks page %s' % (i+1))
+                'Downloading tracks page %s' % (i+1))
             new_tracks = json.loads(response)
             tracks.extend(self._extract_info_dict(track, quiet=True) for track in new_tracks)
             if len(new_tracks) < 50:

youtube_dl/extractor/southparkstudios.py

@@ -18,14 +18,6 @@ class SouthParkStudiosIE(MTVServicesInfoExtractor):
         },
     }]
 
-    def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
-        url = u'http://www.' + mobj.group(u'url')
-        video_id = mobj.group('id')
-        webpage = self._download_webpage(url, video_id)
-        mgid = self._search_regex(r'swfobject.embedSWF\(".*?(mgid:.*?)"',
-                                  webpage, u'mgid')
-        return self._get_videos_info(mgid)
 
 class SouthparkDeIE(SouthParkStudiosIE):
     IE_NAME = u'southpark.de'

youtube_dl/extractor/space.py

@@ -6,7 +6,7 @@ from ..utils import RegexNotFoundError, ExtractorError
 
 
 class SpaceIE(InfoExtractor):
-    _VALID_URL = r'https?://www\.space\.com/\d+-(?P<title>[^/\.\?]*?)-video\.html'
+    _VALID_URL = r'https?://(?:(?:www|m)\.)?space\.com/\d+-(?P<title>[^/\.\?]*?)-video\.html'
     _TEST = {
         u'add_ie': ['Brightcove'],
         u'url': u'http://www.space.com/23373-huge-martian-landforms-detail-revealed-by-european-probe-video.html',

youtube_dl/extractor/spankwire.py

@@ -1,3 +1,5 @@
+from __future__ import unicode_literals
+
 import os
 import re
 
@@ -11,17 +13,18 @@ from ..aes import (
     aes_decrypt_text
 )
 
+
 class SpankwireIE(InfoExtractor):
     _VALID_URL = r'^(?:https?://)?(?:www\.)?(?P<url>spankwire\.com/[^/]*/video(?P<videoid>[0-9]+)/?)'
     _TEST = {
-        u'url': u'http://www.spankwire.com/Buckcherry-s-X-Rated-Music-Video-Crazy-Bitch/video103545/',
+        'url': 'http://www.spankwire.com/Buckcherry-s-X-Rated-Music-Video-Crazy-Bitch/video103545/',
-        u'file': u'103545.mp4',
+        'file': '103545.mp4',
-        u'md5': u'1b3f55e345500552dbc252a3e9c1af43',
+        'md5': '1b3f55e345500552dbc252a3e9c1af43',
-        u'info_dict': {
+        'info_dict': {
-            u"uploader": u"oreusz", 
+            "uploader": "oreusz",
-            u"title": u"Buckcherry`s X Rated Music Video Crazy Bitch",
+            "title": "Buckcherry`s X Rated Music Video Crazy Bitch",
-            u"description": u"Crazy Bitch X rated music video.",
+            "description": "Crazy Bitch X rated music video.",
-            u"age_limit": 18,
+            "age_limit": 18,
         }
     }
 
@@ -34,17 +37,17 @@ class SpankwireIE(InfoExtractor):
         req.add_header('Cookie', 'age_verified=1')
         webpage = self._download_webpage(req, video_id)
 
-        video_title = self._html_search_regex(r'<h1>([^<]+)', webpage, u'title')
+        video_title = self._html_search_regex(r'<h1>([^<]+)', webpage, 'title')
         video_uploader = self._html_search_regex(
-            r'by:\s*<a [^>]*>(.+?)</a>', webpage, u'uploader', fatal=False)
+            r'by:\s*<a [^>]*>(.+?)</a>', webpage, 'uploader', fatal=False)
         thumbnail = self._html_search_regex(
-            r'flashvars\.image_url = "([^"]+)', webpage, u'thumbnail', fatal=False)
+            r'flashvars\.image_url = "([^"]+)', webpage, 'thumbnail', fatal=False)
         description = self._html_search_regex(
-            r'<div\s+id="descriptionContent">([^<]+)<', webpage, u'description', fatal=False)
+            r'<div\s+id="descriptionContent">([^<]+)<', webpage, 'description', fatal=False)
 
         video_urls = list(map(compat_urllib_parse.unquote , re.findall(r'flashvars\.quality_[0-9]{3}p = "([^"]+)', webpage)))
         if webpage.find('flashvars\.encrypted = "true"') != -1:
-            password = self._html_search_regex(r'flashvars\.video_title = "([^"]+)', webpage, u'password').replace('+', ' ')
+            password = self._html_search_regex(r'flashvars\.video_title = "([^"]+)', webpage, 'password').replace('+', ' ')
             video_urls = list(map(lambda s: aes_decrypt_text(s, password, 32).decode('utf-8'), video_urls))
 
         formats = []
@@ -52,14 +55,21 @@ class SpankwireIE(InfoExtractor):
             path = compat_urllib_parse_urlparse(video_url).path
             extension = os.path.splitext(path)[1][1:]
             format = path.split('/')[4].split('_')[:2]
+            resolution, bitrate_str = format
             format = "-".join(format)
+            height = int(resolution.rstrip('P'))
+            tbr = int(bitrate_str.rstrip('K'))
+
             formats.append({
                 'url': video_url,
                 'ext': extension,
+                'resolution': resolution,
                 'format': format,
+                'tbr': tbr,
+                'height': height,
                 'format_id': format,
             })
-        formats.sort(key=lambda format: list(map(lambda s: s.zfill(6), format['format'].split('-'))))
+        self._sort_formats(formats)
 
         age_limit = self._rta_search(webpage)
 

youtube_dl/extractor/spike.py

@@ -0,0 +1,19 @@
+from __future__ import unicode_literals
+
+from .mtv import MTVServicesInfoExtractor
+
+
+class SpikeIE(MTVServicesInfoExtractor):
+    _VALID_URL = r'https?://www\.spike\.com/(video-clips|episodes)/.+'
+    _TEST = {
+        'url': 'http://www.spike.com/video-clips/lhtu8m/auction-hunters-can-allen-ride-a-hundred-year-old-motorcycle',
+        'md5': '1a9265f32b0c375793d6c4ce45255256',
+        'info_dict': {
+            'id': 'b9c8221a-4e50-479a-b86d-3333323e38ba',
+            'ext': 'mp4',
+            'title': 'Auction Hunters|Can Allen Ride A Hundred Year-Old Motorcycle?',
+            'description': 'md5:fbed7e82ed5fad493615b3094a9499cb',
+        },
+    }
+
+    _FEED_URL = 'http://www.spike.com/feeds/mrss/'

youtube_dl/extractor/teamcoco.py

@@ -1,3 +1,5 @@
+from __future__ import unicode_literals
+
 import re
 
 from .common import InfoExtractor
@@ -9,61 +11,66 @@ from ..utils import (
 class TeamcocoIE(InfoExtractor):
     _VALID_URL = r'http://teamcoco\.com/video/(?P<url_title>.*)'
     _TEST = {
-        u'url': u'http://teamcoco.com/video/louis-ck-interview-george-w-bush',
+        'url': 'http://teamcoco.com/video/louis-ck-interview-george-w-bush',
-        u'file': u'19705.mp4',
+        'file': '19705.mp4',
-        u'md5': u'cde9ba0fa3506f5f017ce11ead928f9a',
+        'md5': 'cde9ba0fa3506f5f017ce11ead928f9a',
-        u'info_dict': {
+        'info_dict': {
-            u"description": u"Louis C.K. got starstruck by George W. Bush, so what? Part one.", 
+            "description": "Louis C.K. got starstruck by George W. Bush, so what? Part one.",
-            u"title": u"Louis C.K. Interview Pt. 1 11/3/11"
+            "title": "Louis C.K. Interview Pt. 1 11/3/11"
         }
     }
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
         if mobj is None:
-            raise ExtractorError(u'Invalid URL: %s' % url)
+            raise ExtractorError('Invalid URL: %s' % url)
         url_title = mobj.group('url_title')
         webpage = self._download_webpage(url, url_title)
 
-        video_id = self._html_search_regex(r'<article class="video" data-id="(\d+?)"',
+        video_id = self._html_search_regex(
-            webpage, u'video id')
+            r'<article class="video" data-id="(\d+?)"',
+            webpage, 'video id')
 
         self.report_extraction(video_id)
 
         data_url = 'http://teamcoco.com/cvp/2.0/%s.xml' % video_id
         data = self._download_xml(data_url, video_id, 'Downloading data webpage')
 
-
         qualities = ['500k', '480p', '1000k', '720p', '1080p']
         formats = []
-        for file in data.findall('files/file'):
+        for filed in data.findall('files/file'):
-            if file.attrib.get('playmode') == 'all':
+            if filed.attrib.get('playmode') == 'all':
                 # it just duplicates one of the entries
                 break
-            file_url = file.text
+            file_url = filed.text
             m_format = re.search(r'(\d+(k|p))\.mp4', file_url)
             if m_format is not None:
                 format_id = m_format.group(1)
             else:
-                format_id = file.attrib['bitrate']
+                format_id = filed.attrib['bitrate']
+            tbr = (
+                int(filed.attrib['bitrate'])
+                if filed.attrib['bitrate'].isdigit()
+                else None)
+
+            try:
+                quality = qualities.index(format_id)
+            except ValueError:
+                quality = -1
             formats.append({
                 'url': file_url,
                 'ext': 'mp4',
+                'tbr': tbr,
                 'format_id': format_id,
+                'quality': quality,
             })
-        def sort_key(f):
+
-            try:
+        self._sort_formats(formats)
-                return qualities.index(f['format_id'])
-            except ValueError:
-                return -1
-        formats.sort(key=sort_key)
-        if not formats:
-            raise ExtractorError(u'Unable to extract video URL')
 
         return {
-            'id':          video_id,
+            'id': video_id,
             'formats': formats,
-            'title':       self._og_search_title(webpage),
+            'title': self._og_search_title(webpage),
-            'thumbnail':   self._og_search_thumbnail(webpage),
+            'thumbnail': self._og_search_thumbnail(webpage),
             'description': self._og_search_description(webpage),
         }

youtube_dl/extractor/ted.py

@@ -1,3 +1,5 @@
+from __future__ import unicode_literals
+
 import json
 import re
 
@@ -7,6 +9,7 @@ from ..utils import (
     RegexNotFoundError,
 )
 
+
 class TEDIE(SubtitlesInfoExtractor):
     _VALID_URL=r'''http://www\.ted\.com/
                    (
@@ -18,12 +21,12 @@ class TEDIE(SubtitlesInfoExtractor):
                    /(?P<name>\w+) # Here goes the name and then ".html"
                    '''
     _TEST = {
-        u'url': u'http://www.ted.com/talks/dan_dennett_on_our_consciousness.html',
+        'url': 'http://www.ted.com/talks/dan_dennett_on_our_consciousness.html',
-        u'file': u'102.mp4',
+        'file': '102.mp4',
-        u'md5': u'2d76ee1576672e0bd8f187513267adf6',
+        'md5': '4ea1dada91e4174b53dac2bb8ace429d',
-        u'info_dict': {
+        'info_dict': {
-            u"description": u"md5:c6fa72e6eedbd938c9caf6b2702f5922", 
+            "description": "md5:c6fa72e6eedbd938c9caf6b2702f5922",
-            u"title": u"Dan Dennett: The illusion of consciousness"
+            "title": "Dan Dennett: The illusion of consciousness"
         }
     }
 
@@ -47,7 +50,7 @@ class TEDIE(SubtitlesInfoExtractor):
         '''Returns the videos of the playlist'''
 
         webpage = self._download_webpage(
-            url, playlist_id, u'Downloading playlist 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)

youtube_dl/extractor/vk.py

@@ -1,4 +1,6 @@
 # encoding: utf-8
+from __future__ import unicode_literals
+
 import re
 import json
 
@@ -10,19 +12,27 @@ from ..utils import (
 
 
 class VKIE(InfoExtractor):
-    IE_NAME = u'vk.com'
+    IE_NAME = 'vk.com'
     _VALID_URL = r'https?://vk\.com/(?:videos.*?\?.*?z=)?video(?P<id>.*?)(?:\?|%2F|$)'
 
-    _TEST = {
+    _TESTS = [{
-        u'url': u'http://vk.com/videos-77521?z=video-77521_162222515%2Fclub77521',
+        'url': 'http://vk.com/videos-77521?z=video-77521_162222515%2Fclub77521',
-        u'md5': u'0deae91935c54e00003c2a00646315f0',
+        'file': '162222515.flv',
-        u'info_dict': {
+        'md5': '0deae91935c54e00003c2a00646315f0',
-            u'id': u'162222515',
+        'info_dict': {
-            u'ext': u'flv',
+            'title': 'ProtivoGunz - Хуёвая песня',
-            u'title': u'ProtivoGunz - Хуёвая песня',
+            'uploader': 'Noize MC',
-            u'uploader': u'Noize MC',
         },
-    }
+    },
+    {
+        'url': 'http://vk.com/video4643923_163339118',
+        'file': '163339118.mp4',
+        'md5': 'f79bccb5cd182b1f43502ca5685b2b36',
+        'info_dict': {
+            'uploader': 'Elvira Dzhonik',
+            'title': 'Dream Theater - Hollow Years Live at Budokan 720*',
+        }
+    }]
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
@@ -33,13 +43,21 @@ class VKIE(InfoExtractor):
         if m_yt is not None:
             self.to_screen(u'Youtube video detected')
             return self.url_result(m_yt.group(1), 'Youtube')
-        vars_json = self._search_regex(r'var vars = ({.*?});', info_page, u'vars')
+        data_json = self._search_regex(r'var vars = ({.*?});', info_page, 'vars')
-        vars = json.loads(vars_json)
+        data = json.loads(data_json)
+
+        formats = [{
+            'format_id': k,
+            'url': v,
+            'width': int(k[len('url'):]),
+        } for k, v in data.items()
+            if k.startswith('url')]
+        self._sort_formats(formats)
 
         return {
-            'id': compat_str(vars['vid']),
+            'id': compat_str(data['vid']),
-            'url': vars['url240'],
+            'formats': formats,
-            'title': unescapeHTML(vars['md_title']),
+            'title': unescapeHTML(data['md_title']),
-            'thumbnail': vars['jpg'],
+            'thumbnail': data.get('jpg'),
-            'uploader': vars['md_author'],
+            'uploader': data.get('md_author'),
         }

youtube_dl/extractor/wimp.py

@@ -1,5 +1,6 @@
+from __future__ import unicode_literals
+
 import re
-import base64
 
 from .common import InfoExtractor
 
@@ -7,12 +8,12 @@ from .common import InfoExtractor
 class WimpIE(InfoExtractor):
     _VALID_URL = r'(?:http://)?(?:www\.)?wimp\.com/([^/]+)/'
     _TEST = {
-        u'url': u'http://www.wimp.com/deerfence/',
+        'url': 'http://www.wimp.com/deerfence/',
-        u'file': u'deerfence.flv',
+        'file': 'deerfence.flv',
-        u'md5': u'8b215e2e0168c6081a1cf84b2846a2b5',
+        'md5': '8b215e2e0168c6081a1cf84b2846a2b5',
-        u'info_dict': {
+        'info_dict': {
-            u"title": u"Watch Till End: Herd of deer jump over a fence.",
+            "title": "Watch Till End: Herd of deer jump over a fence.",
-            u"description": u"These deer look as fluid as running water when they jump over this fence as a herd. This video is one that needs to be watched until the very end for the true majesty to be witnessed, but once it comes, it's sure to take your breath away.",
+            "description": "These deer look as fluid as running water when they jump over this fence as a herd. This video is one that needs to be watched until the very end for the true majesty to be witnessed, but once it comes, it's sure to take your breath away.",
         }
     }
 
@@ -20,13 +21,12 @@ class WimpIE(InfoExtractor):
         mobj = re.match(self._VALID_URL, url)
         video_id = mobj.group(1)
         webpage = self._download_webpage(url, video_id)
-        googleString = self._search_regex("googleCode = '(.*?)'", webpage, 'file url')
+        video_url = self._search_regex(
-        googleString = base64.b64decode(googleString).decode('ascii')
+            r's1\.addVariable\("file",\s*"([^"]+)"\);', webpage, 'video URL')
-        final_url = self._search_regex('","(.*?)"', googleString, u'final video url')
 
         return {
             'id': video_id,
-            'url': final_url,
+            'url': video_url,
             'title': self._og_search_title(webpage),
             'thumbnail': self._og_search_thumbnail(webpage),
             'description': self._og_search_description(webpage),

youtube_dl/extractor/youtube.py

@@ -27,6 +27,7 @@ from ..utils import (
     get_element_by_id,
     get_element_by_attribute,
     ExtractorError,
+    RegexNotFoundError,
     unescapeHTML,
     unified_strdate,
     orderedSet,
@@ -1448,7 +1449,14 @@ class YoutubePlaylistIE(YoutubeBaseInfoExtractor):
             if re.search(self._MORE_PAGES_INDICATOR, page) is None:
                 break
 
-        playlist_title = self._og_search_title(page)
+        try:
+            playlist_title = self._og_search_title(page)
+        except RegexNotFoundError:
+            self.report_warning(
+                u'Playlist page is missing OpenGraph title, falling back ...',
+                playlist_id)
+            playlist_title = self._html_search_regex(
+                r'<h1 class="pl-header-title">(.*?)</h1>', page, u'title')
 
         url_results = self._ids_to_results(ids)
         return self.playlist_result(url_results, playlist_id, playlist_title)

youtube_dl/postprocessor/ffmpeg.py

@@ -479,6 +479,7 @@ class FFmpegMergerPP(FFmpegPostProcessor):
     def run(self, info):
         filename = info['filepath']
         args = ['-c', 'copy']
+        self._downloader.to_screen(u'[ffmpeg] Merging formats into "%s"' % filename)
         self.run_ffmpeg_multiple_files(info['__files_to_merge'], filename, args)
         return True, info
 

youtube_dl/utils.py

@@ -224,7 +224,7 @@ if sys.version_info >= (2,7):
     def find_xpath_attr(node, xpath, key, val):
         """ Find the xpath xpath[@key=val] """
         assert re.match(r'^[a-zA-Z]+$', key)
-        assert re.match(r'^[a-zA-Z0-9@\s]*$', val)
+        assert re.match(r'^[a-zA-Z0-9@\s:._]*$', val)
         expr = xpath + u"[@%s='%s']" % (key, val)
         return node.find(expr)
 else:
@@ -1092,9 +1092,12 @@ def month_by_name(name):
         return None
 
 
-def fix_xml_all_ampersand(xml_str):
+def fix_xml_ampersands(xml_str):
     """Replace all the '&' by '&' in XML"""
-    return xml_str.replace(u'&', u'&')
+    return re.sub(
+        r'&(?!amp;|lt;|gt;|apos;|quot;|#x[0-9a-fA-F]{,4};|#[0-9]{,4};)',
+        u'&',
+        xml_str)
 
 
 def setproctitle(title):

youtube_dl/version.py

@@ -1,2 +1,2 @@
 
-__version__ = '2013.01.17.1'
+__version__ = '2014.01.22.3'