release 2014.11.23

Update jsinterp.py

AUTHORS

@@ -81,3 +81,5 @@ winwon
 Xavier Beynon
 Gabriel Schubiner
 xantares
+Jan Matějka
+Mauroy Sébastien

test/swftests/ConstArrayAccess.as

@@ -0,0 +1,18 @@
+// input: []
+// output: 4
+
+package {
+public class ConstArrayAccess {
+	private static const x:int = 2;
+	private static const ar:Array = ["42", "3411"];
+
+    public static function main():int{
+        var c:ConstArrayAccess = new ConstArrayAccess();
+        return c.f();
+    }
+
+    public function f(): int {
+    	return ar[1].length;
+    }
+}
+}

test/swftests/ConstantInt.as

@@ -0,0 +1,12 @@
+// input: []
+// output: 2
+
+package {
+public class ConstantInt {
+	private static const x:int = 2;
+
+    public static function main():int{
+        return x;
+    }
+}
+}

test/swftests/DictCall.as

@@ -0,0 +1,10 @@
+// input: [{"x": 1, "y": 2}]
+// output: 3
+
+package {
+public class DictCall {
+    public static function main(d:Object):int{
+        return d.x + d.y;
+    }
+}
+}

test/swftests/EqualsOperator.as

@@ -0,0 +1,10 @@
+// input: []
+// output: false
+
+package {
+public class EqualsOperator {
+    public static function main():Boolean{
+        return 1 == 2;
+    }
+}
+}

test/swftests/MemberAssignment.as

@@ -0,0 +1,22 @@
+// input: [1]
+// output: 2
+
+package {
+public class MemberAssignment {
+    public var v:int;
+
+    public function g():int {
+        return this.v;
+    }
+
+    public function f(a:int):int{
+        this.v = a;
+        return this.v + this.g();
+    }
+
+    public static function main(a:int): int {
+        var v:MemberAssignment = new MemberAssignment();
+        return v.f(a);
+    }
+}
+}

test/swftests/NeOperator.as

@@ -0,0 +1,24 @@
+// input: []
+// output: 123
+
+package {
+public class NeOperator {
+    public static function main(): int {
+        var res:int = 0;
+        if (1 != 2) {
+            res += 3;
+        } else {
+            res += 4;
+        }
+        if (2 != 2) {
+            res += 10;
+        } else {
+            res += 20;
+        }
+        if (9 == 9) {
+            res += 100;
+        }
+        return res;
+    }
+}
+}

test/swftests/PrivateVoidCall.as

@@ -0,0 +1,22 @@
+// input: []
+// output: 9
+
+package {
+public class PrivateVoidCall {
+    public static function main():int{
+        var f:OtherClass = new OtherClass();
+        f.func();
+        return 9;
+    }
+}
+}
+
+class OtherClass {
+    private function pf():void {
+        ;
+    }
+
+    public function func():void {
+        this.pf();
+    }
+}

test/swftests/StringBasics.as

@@ -0,0 +1,11 @@
+// input: []
+// output: 3
+
+package {
+public class StringBasics {
+    public static function main():int{
+        var s:String = "abc";
+        return s.length;
+    }
+}
+}

test/swftests/StringCharCodeAt.as

@@ -0,0 +1,11 @@
+// input: []
+// output: 9897
+
+package {
+public class StringCharCodeAt {
+    public static function main():int{
+        var s:String = "abc";
+        return s.charCodeAt(1) * 100 + s.charCodeAt();
+    }
+}
+}

test/swftests/StringConversion.as

@@ -0,0 +1,11 @@
+// input: []
+// output: 2
+
+package {
+public class StringConversion {
+    public static function main():int{
+        var s:String = String(99);
+        return s.length;
+    }
+}
+}

test/test_age_restriction.py

@@ -1,4 +1,5 @@
 #!/usr/bin/env python
+from __future__ import unicode_literals
 
 # Allow direct execution
 import os
@@ -19,7 +20,7 @@ def _download_restricted(url, filename, age):
         'age_limit': age,
         'skip_download': True,
         'writeinfojson': True,
-        "outtmpl": "%(id)s.%(ext)s",
+        'outtmpl': '%(id)s.%(ext)s',
     }
     ydl = YoutubeDL(params)
     ydl.add_default_info_extractors()

test/test_compat.py

@@ -26,11 +26,13 @@ class TestCompat(unittest.TestCase):
         self.assertEqual(compat_getenv('YOUTUBE-DL-TEST'), test_str)
 
     def test_compat_expanduser(self):
+        old_home = os.environ.get('HOME')
         test_str = 'C:\Documents and Settings\тест\Application Data'
         os.environ['HOME'] = (
             test_str if sys.version_info >= (3, 0)
             else test_str.encode(get_filesystem_encoding()))
         self.assertEqual(compat_expanduser('~'), test_str)
+        os.environ['HOME'] = old_home
 
     def test_all_present(self):
         import youtube_dl.compat

test/test_download.py

@@ -1,5 +1,7 @@
 #!/usr/bin/env python
 
+from __future__ import unicode_literals
+
 # Allow direct execution
 import os
 import sys
@@ -210,9 +212,9 @@ for n, test_case in enumerate(defs):
     tname = 'test_' + str(test_case['name'])
     i = 1
     while hasattr(TestDownload, tname):
-        tname = 'test_'  + str(test_case['name']) + '_' + str(i)
+        tname = 'test_%s_%d' % (test_case['name'], i)
         i += 1
-    test_method.__name__ = tname
+    test_method.__name__ = str(tname)
     setattr(TestDownload, test_method.__name__, test_method)
     del test_method
 

test/test_execution.py

@@ -1,3 +1,6 @@
+#!/usr/bin/env python
+from __future__ import unicode_literals
+
 import unittest
 
 import sys
@@ -6,17 +9,19 @@ import subprocess
 
 rootDir = os.path.dirname(os.path.dirname(os.path.abspath(__file__)))
 
+
 try:
     _DEV_NULL = subprocess.DEVNULL
 except AttributeError:
     _DEV_NULL = open(os.devnull, 'wb')
 
+
 class TestExecution(unittest.TestCase):
     def test_import(self):
         subprocess.check_call([sys.executable, '-c', 'import youtube_dl'], cwd=rootDir)
 
     def test_module_exec(self):
-        if sys.version_info >= (2,7): # Python 2.6 doesn't support package execution
+        if sys.version_info >= (2, 7):  # Python 2.6 doesn't support package execution
             subprocess.check_call([sys.executable, '-m', 'youtube_dl', '--version'], cwd=rootDir, stdout=_DEV_NULL)
 
     def test_main_exec(self):

test/test_subtitles.py

@@ -1,4 +1,5 @@
 #!/usr/bin/env python
+from __future__ import unicode_literals
 
 # Allow direct execution
 import os
@@ -74,7 +75,7 @@ class TestYoutubeSubtitles(BaseTestSubtitles):
         self.assertEqual(md5(subtitles['en']), '3cb210999d3e021bd6c7f0ea751eab06')
 
     def test_youtube_list_subtitles(self):
-        self.DL.expect_warning(u'Video doesn\'t have automatic captions')
+        self.DL.expect_warning('Video doesn\'t have automatic captions')
         self.DL.params['listsubtitles'] = True
         info_dict = self.getInfoDict()
         self.assertEqual(info_dict, None)
@@ -87,7 +88,7 @@ class TestYoutubeSubtitles(BaseTestSubtitles):
         self.assertTrue(subtitles['it'] is not None)
 
     def test_youtube_nosubtitles(self):
-        self.DL.expect_warning(u'video doesn\'t have subtitles')
+        self.DL.expect_warning('video doesn\'t have subtitles')
         self.url = 'n5BB19UTcdA'
         self.DL.params['writesubtitles'] = True
         self.DL.params['allsubtitles'] = True
@@ -101,7 +102,7 @@ class TestYoutubeSubtitles(BaseTestSubtitles):
         self.DL.params['subtitleslangs'] = langs
         subtitles = self.getSubtitles()
         for lang in langs:
-            self.assertTrue(subtitles.get(lang) is not None, u'Subtitles for \'%s\' not extracted' % lang)
+            self.assertTrue(subtitles.get(lang) is not None, 'Subtitles for \'%s\' not extracted' % lang)
 
 
 class TestDailymotionSubtitles(BaseTestSubtitles):
@@ -130,20 +131,20 @@ class TestDailymotionSubtitles(BaseTestSubtitles):
         self.assertEqual(len(subtitles.keys()), 5)
 
     def test_list_subtitles(self):
-        self.DL.expect_warning(u'Automatic Captions not supported by this server')
+        self.DL.expect_warning('Automatic Captions not supported by this server')
         self.DL.params['listsubtitles'] = True
         info_dict = self.getInfoDict()
         self.assertEqual(info_dict, None)
 
     def test_automatic_captions(self):
-        self.DL.expect_warning(u'Automatic Captions not supported by this server')
+        self.DL.expect_warning('Automatic Captions not supported by this server')
         self.DL.params['writeautomaticsub'] = True
         self.DL.params['subtitleslang'] = ['en']
         subtitles = self.getSubtitles()
         self.assertTrue(len(subtitles.keys()) == 0)
 
     def test_nosubtitles(self):
-        self.DL.expect_warning(u'video doesn\'t have subtitles')
+        self.DL.expect_warning('video doesn\'t have subtitles')
         self.url = 'http://www.dailymotion.com/video/x12u166_le-zapping-tele-star-du-08-aout-2013_tv'
         self.DL.params['writesubtitles'] = True
         self.DL.params['allsubtitles'] = True
@@ -156,7 +157,7 @@ class TestDailymotionSubtitles(BaseTestSubtitles):
         self.DL.params['subtitleslangs'] = langs
         subtitles = self.getSubtitles()
         for lang in langs:
-            self.assertTrue(subtitles.get(lang) is not None, u'Subtitles for \'%s\' not extracted' % lang)
+            self.assertTrue(subtitles.get(lang) is not None, 'Subtitles for \'%s\' not extracted' % lang)
 
 
 class TestTedSubtitles(BaseTestSubtitles):
@@ -185,13 +186,13 @@ class TestTedSubtitles(BaseTestSubtitles):
         self.assertTrue(len(subtitles.keys()) >= 28)
 
     def test_list_subtitles(self):
-        self.DL.expect_warning(u'Automatic Captions not supported by this server')
+        self.DL.expect_warning('Automatic Captions not supported by this server')
         self.DL.params['listsubtitles'] = True
         info_dict = self.getInfoDict()
         self.assertEqual(info_dict, None)
 
     def test_automatic_captions(self):
-        self.DL.expect_warning(u'Automatic Captions not supported by this server')
+        self.DL.expect_warning('Automatic Captions not supported by this server')
         self.DL.params['writeautomaticsub'] = True
         self.DL.params['subtitleslang'] = ['en']
         subtitles = self.getSubtitles()
@@ -203,7 +204,7 @@ class TestTedSubtitles(BaseTestSubtitles):
         self.DL.params['subtitleslangs'] = langs
         subtitles = self.getSubtitles()
         for lang in langs:
-            self.assertTrue(subtitles.get(lang) is not None, u'Subtitles for \'%s\' not extracted' % lang)
+            self.assertTrue(subtitles.get(lang) is not None, 'Subtitles for \'%s\' not extracted' % lang)
 
 
 class TestBlipTVSubtitles(BaseTestSubtitles):
@@ -211,13 +212,13 @@ class TestBlipTVSubtitles(BaseTestSubtitles):
     IE = BlipTVIE
 
     def test_list_subtitles(self):
-        self.DL.expect_warning(u'Automatic Captions not supported by this server')
+        self.DL.expect_warning('Automatic Captions not supported by this server')
         self.DL.params['listsubtitles'] = True
         info_dict = self.getInfoDict()
         self.assertEqual(info_dict, None)
 
     def test_allsubtitles(self):
-        self.DL.expect_warning(u'Automatic Captions not supported by this server')
+        self.DL.expect_warning('Automatic Captions not supported by this server')
         self.DL.params['writesubtitles'] = True
         self.DL.params['allsubtitles'] = True
         subtitles = self.getSubtitles()
@@ -251,20 +252,20 @@ class TestVimeoSubtitles(BaseTestSubtitles):
         self.assertEqual(set(subtitles.keys()), set(['de', 'en', 'es', 'fr']))
 
     def test_list_subtitles(self):
-        self.DL.expect_warning(u'Automatic Captions not supported by this server')
+        self.DL.expect_warning('Automatic Captions not supported by this server')
         self.DL.params['listsubtitles'] = True
         info_dict = self.getInfoDict()
         self.assertEqual(info_dict, None)
 
     def test_automatic_captions(self):
-        self.DL.expect_warning(u'Automatic Captions not supported by this server')
+        self.DL.expect_warning('Automatic Captions not supported by this server')
         self.DL.params['writeautomaticsub'] = True
         self.DL.params['subtitleslang'] = ['en']
         subtitles = self.getSubtitles()
         self.assertTrue(len(subtitles.keys()) == 0)
 
     def test_nosubtitles(self):
-        self.DL.expect_warning(u'video doesn\'t have subtitles')
+        self.DL.expect_warning('video doesn\'t have subtitles')
         self.url = 'http://vimeo.com/56015672'
         self.DL.params['writesubtitles'] = True
         self.DL.params['allsubtitles'] = True
@@ -277,7 +278,7 @@ class TestVimeoSubtitles(BaseTestSubtitles):
         self.DL.params['subtitleslangs'] = langs
         subtitles = self.getSubtitles()
         for lang in langs:
-            self.assertTrue(subtitles.get(lang) is not None, u'Subtitles for \'%s\' not extracted' % lang)
+            self.assertTrue(subtitles.get(lang) is not None, 'Subtitles for \'%s\' not extracted' % lang)
 
 
 class TestWallaSubtitles(BaseTestSubtitles):
@@ -285,13 +286,13 @@ class TestWallaSubtitles(BaseTestSubtitles):
     IE = WallaIE
 
     def test_list_subtitles(self):
-        self.DL.expect_warning(u'Automatic Captions not supported by this server')
+        self.DL.expect_warning('Automatic Captions not supported by this server')
         self.DL.params['listsubtitles'] = True
         info_dict = self.getInfoDict()
         self.assertEqual(info_dict, None)
 
     def test_allsubtitles(self):
-        self.DL.expect_warning(u'Automatic Captions not supported by this server')
+        self.DL.expect_warning('Automatic Captions not supported by this server')
         self.DL.params['writesubtitles'] = True
         self.DL.params['allsubtitles'] = True
         subtitles = self.getSubtitles()
@@ -299,7 +300,7 @@ class TestWallaSubtitles(BaseTestSubtitles):
         self.assertEqual(md5(subtitles['heb']), 'e758c5d7cb982f6bef14f377ec7a3920')
 
     def test_nosubtitles(self):
-        self.DL.expect_warning(u'video doesn\'t have subtitles')
+        self.DL.expect_warning('video doesn\'t have subtitles')
         self.url = 'http://vod.walla.co.il/movie/2642630/one-direction-all-for-one'
         self.DL.params['writesubtitles'] = True
         self.DL.params['allsubtitles'] = True

test/test_swfinterp.py

@@ -1,4 +1,5 @@
 #!/usr/bin/env python
+from __future__ import unicode_literals
 
 # Allow direct execution
 import os

test/test_utils.py

@@ -218,6 +218,7 @@ class TestUtil(unittest.TestCase):
         self.assertEqual(parse_duration('0m0s'), 0)
         self.assertEqual(parse_duration('0s'), 0)
         self.assertEqual(parse_duration('01:02:03.05'), 3723.05)
+        self.assertEqual(parse_duration('T30M38S'), 1838)
 
     def test_fix_xml_ampersands(self):
         self.assertEqual(
@@ -284,6 +285,10 @@ class TestUtil(unittest.TestCase):
         d = json.loads(stripped)
         self.assertEqual(d, [{"id": "532cb", "x": 3}])
 
+        stripped = strip_jsonp('parseMetadata({"STATUS":"OK"})\n\n\n//epc')
+        d = json.loads(stripped)
+        self.assertEqual(d, {'STATUS': 'OK'})
+
     def test_uppercase_escape(self):
         self.assertEqual(uppercase_escape('aä'), 'aä')
         self.assertEqual(uppercase_escape('\\U0001d550'), '𝕐')

youtube_dl/YoutubeDL.py

@@ -624,7 +624,7 @@ class YoutubeDL(object):
 
             return self.process_ie_result(
                 new_result, download=download, extra_info=extra_info)
-        elif result_type == 'playlist':
+        elif result_type == 'playlist' or result_type == 'multi_video':
             # We process each entry in the playlist
             playlist = ie_result.get('title', None) or ie_result.get('id', None)
             self.to_screen('[download] Downloading playlist: %s' % playlist)
@@ -679,6 +679,9 @@ class YoutubeDL(object):
             ie_result['entries'] = playlist_results
             return ie_result
         elif result_type == 'compat_list':
+            self.report_warning(
+                'Extractor %s returned a compat_list result. '
+                'It needs to be updated.' % ie_result.get('extractor'))
             def _fixup(r):
                 self.add_extra_info(r,
                     {
@@ -1001,7 +1004,7 @@ class YoutubeDL(object):
             else:
                 self.to_screen('[info] Writing video description metadata as JSON to: ' + infofn)
                 try:
-                    write_json_file(info_dict, encodeFilename(infofn))
+                    write_json_file(info_dict, infofn)
                 except (OSError, IOError):
                     self.report_error('Cannot write metadata to JSON file ' + infofn)
                     return

youtube_dl/__init__.py

@@ -1,6 +1,8 @@
 #!/usr/bin/env python
 # -*- coding: utf-8 -*-
 
+from __future__ import unicode_literals
+
 __license__ = 'Public Domain'
 
 import codecs
@@ -17,6 +19,7 @@ from .compat import (
     compat_expanduser,
     compat_getpass,
     compat_print,
+    workaround_optparse_bug9161,
 )
 from .utils import (
     DateRange,
@@ -55,7 +58,9 @@ def _real_main(argv=None):
         # https://github.com/rg3/youtube-dl/issues/820
         codecs.register(lambda name: codecs.lookup('utf-8') if name == 'cp65001' else None)
 
-    setproctitle(u'youtube-dl')
+    workaround_optparse_bug9161()
+
+    setproctitle('youtube-dl')
 
     parser, opts, args = parseOpts(argv)
 
@@ -71,10 +76,10 @@ def _real_main(argv=None):
     if opts.headers is not None:
         for h in opts.headers:
             if h.find(':', 1) < 0:
-                parser.error(u'wrong header formatting, it should be key:value, not "%s"'%h)
+                parser.error('wrong header formatting, it should be key:value, not "%s"'%h)
             key, value = h.split(':', 2)
             if opts.verbose:
-                write_string(u'[debug] Adding header from command line option %s:%s\n'%(key, value))
+                write_string('[debug] Adding header from command line option %s:%s\n'%(key, value))
             std_headers[key] = value
 
     # Dump user agent
@@ -92,9 +97,9 @@ def _real_main(argv=None):
                 batchfd = io.open(opts.batchfile, 'r', encoding='utf-8', errors='ignore')
             batch_urls = read_batch_urls(batchfd)
             if opts.verbose:
-                write_string(u'[debug] Batch file urls: ' + repr(batch_urls) + u'\n')
+                write_string('[debug] Batch file urls: ' + repr(batch_urls) + '\n')
         except IOError:
-            sys.exit(u'ERROR: batch file could not be read')
+            sys.exit('ERROR: batch file could not be read')
     all_urls = batch_urls + args
     all_urls = [url.strip() for url in all_urls]
     _enc = preferredencoding()
@@ -107,7 +112,7 @@ def _real_main(argv=None):
             compat_print(ie.IE_NAME + (' (CURRENTLY BROKEN)' if not ie._WORKING else ''))
             matchedUrls = [url for url in all_urls if ie.suitable(url)]
             for mu in matchedUrls:
-                compat_print(u'  ' + mu)
+                compat_print('  ' + mu)
         sys.exit(0)
     if opts.list_extractor_descriptions:
         for ie in sorted(extractors, key=lambda ie: ie.IE_NAME.lower()):
@@ -117,63 +122,63 @@ def _real_main(argv=None):
             if desc is False:
                 continue
             if hasattr(ie, 'SEARCH_KEY'):
-                _SEARCHES = (u'cute kittens', u'slithering pythons', u'falling cat', u'angry poodle', u'purple fish', u'running tortoise', u'sleeping bunny')
+                _SEARCHES = ('cute kittens', 'slithering pythons', 'falling cat', 'angry poodle', 'purple fish', 'running tortoise', 'sleeping bunny')
-                _COUNTS = (u'', u'5', u'10', u'all')
+                _COUNTS = ('', '5', '10', 'all')
-                desc += u' (Example: "%s%s:%s" )' % (ie.SEARCH_KEY, random.choice(_COUNTS), random.choice(_SEARCHES))
+                desc += ' (Example: "%s%s:%s" )' % (ie.SEARCH_KEY, random.choice(_COUNTS), random.choice(_SEARCHES))
             compat_print(desc)
         sys.exit(0)
 
 
     # Conflicting, missing and erroneous options
     if opts.usenetrc and (opts.username is not None or opts.password is not None):
-        parser.error(u'using .netrc conflicts with giving username/password')
+        parser.error('using .netrc conflicts with giving username/password')
     if opts.password is not None and opts.username is None:
-        parser.error(u'account username missing\n')
+        parser.error('account username missing\n')
     if opts.outtmpl is not None and (opts.usetitle or opts.autonumber or opts.useid):
-        parser.error(u'using output template conflicts with using title, video ID or auto number')
+        parser.error('using output template conflicts with using title, video ID or auto number')
     if opts.usetitle and opts.useid:
-        parser.error(u'using title conflicts with using video ID')
+        parser.error('using title conflicts with using video ID')
     if opts.username is not None and opts.password is None:
-        opts.password = compat_getpass(u'Type account password and press [Return]: ')
+        opts.password = compat_getpass('Type account password and press [Return]: ')
     if opts.ratelimit is not None:
         numeric_limit = FileDownloader.parse_bytes(opts.ratelimit)
         if numeric_limit is None:
-            parser.error(u'invalid rate limit specified')
+            parser.error('invalid rate limit specified')
         opts.ratelimit = numeric_limit
     if opts.min_filesize is not None:
         numeric_limit = FileDownloader.parse_bytes(opts.min_filesize)
         if numeric_limit is None:
-            parser.error(u'invalid min_filesize specified')
+            parser.error('invalid min_filesize specified')
         opts.min_filesize = numeric_limit
     if opts.max_filesize is not None:
         numeric_limit = FileDownloader.parse_bytes(opts.max_filesize)
         if numeric_limit is None:
-            parser.error(u'invalid max_filesize specified')
+            parser.error('invalid max_filesize specified')
         opts.max_filesize = numeric_limit
     if opts.retries is not None:
         try:
             opts.retries = int(opts.retries)
         except (TypeError, ValueError):
-            parser.error(u'invalid retry count specified')
+            parser.error('invalid retry count specified')
     if opts.buffersize is not None:
         numeric_buffersize = FileDownloader.parse_bytes(opts.buffersize)
         if numeric_buffersize is None:
-            parser.error(u'invalid buffer size specified')
+            parser.error('invalid buffer size specified')
         opts.buffersize = numeric_buffersize
     if opts.playliststart <= 0:
-        raise ValueError(u'Playlist start must be positive')
+        raise ValueError('Playlist start must be positive')
     if opts.playlistend not in (-1, None) and opts.playlistend < opts.playliststart:
-        raise ValueError(u'Playlist end must be greater than playlist start')
+        raise ValueError('Playlist end must be greater than playlist start')
     if opts.extractaudio:
         if opts.audioformat not in ['best', 'aac', 'mp3', 'm4a', 'opus', 'vorbis', 'wav']:
-            parser.error(u'invalid audio format specified')
+            parser.error('invalid audio format specified')
     if opts.audioquality:
         opts.audioquality = opts.audioquality.strip('k').strip('K')
         if not opts.audioquality.isdigit():
-            parser.error(u'invalid audio quality specified')
+            parser.error('invalid audio quality specified')
     if opts.recodevideo is not None:
         if opts.recodevideo not in ['mp4', 'flv', 'webm', 'ogg', 'mkv']:
-            parser.error(u'invalid video recode format specified')
+            parser.error('invalid video recode format specified')
     if opts.date is not None:
         date = DateRange.day(opts.date)
     else:
@@ -193,17 +198,17 @@ def _real_main(argv=None):
         if opts.outtmpl is not None:
             opts.outtmpl = opts.outtmpl.decode(preferredencoding())
     outtmpl =((opts.outtmpl is not None and opts.outtmpl)
-            or (opts.format == '-1' and opts.usetitle and u'%(title)s-%(id)s-%(format)s.%(ext)s')
+            or (opts.format == '-1' and opts.usetitle and '%(title)s-%(id)s-%(format)s.%(ext)s')
-            or (opts.format == '-1' and u'%(id)s-%(format)s.%(ext)s')
+            or (opts.format == '-1' and '%(id)s-%(format)s.%(ext)s')
-            or (opts.usetitle and opts.autonumber and u'%(autonumber)s-%(title)s-%(id)s.%(ext)s')
+            or (opts.usetitle and opts.autonumber and '%(autonumber)s-%(title)s-%(id)s.%(ext)s')
-            or (opts.usetitle and u'%(title)s-%(id)s.%(ext)s')
+            or (opts.usetitle and '%(title)s-%(id)s.%(ext)s')
-            or (opts.useid and u'%(id)s.%(ext)s')
+            or (opts.useid and '%(id)s.%(ext)s')
-            or (opts.autonumber and u'%(autonumber)s-%(id)s.%(ext)s')
+            or (opts.autonumber and '%(autonumber)s-%(id)s.%(ext)s')
             or DEFAULT_OUTTMPL)
     if not os.path.splitext(outtmpl)[1] and opts.extractaudio:
-        parser.error(u'Cannot download a video and extract audio into the same'
+        parser.error('Cannot download a video and extract audio into the same'
-                     u' file! Use "{0}.%(ext)s" instead of "{0}" as the output'
+                     ' file! Use "{0}.%(ext)s" instead of "{0}" as the output'
-                     u' template'.format(outtmpl))
+                     ' template'.format(outtmpl))
 
     any_printing = opts.geturl or opts.gettitle or opts.getid or opts.getthumbnail or opts.getdescription or opts.getfilename or opts.getformat or opts.getduration or opts.dumpjson or opts.dump_single_json
     download_archive_fn = compat_expanduser(opts.download_archive) if opts.download_archive is not None else opts.download_archive
@@ -330,7 +335,7 @@ def _real_main(argv=None):
         # Maybe do nothing
         if (len(all_urls) < 1) and (opts.load_info_filename is None):
             if not (opts.update_self or opts.rm_cachedir):
-                parser.error(u'you must provide at least one URL')
+                parser.error('you must provide at least one URL')
             else:
                 sys.exit()
 
@@ -340,7 +345,7 @@ def _real_main(argv=None):
             else:
                 retcode = ydl.download(all_urls)
         except MaxDownloadsReached:
-            ydl.to_screen(u'--max-download limit reached, aborting.')
+            ydl.to_screen('--max-download limit reached, aborting.')
             retcode = 101
 
     sys.exit(retcode)
@@ -352,6 +357,6 @@ def main(argv=None):
     except DownloadError:
         sys.exit(1)
     except SameFileError:
-        sys.exit(u'ERROR: fixed output name but more than one file to download')
+        sys.exit('ERROR: fixed output name but more than one file to download')
     except KeyboardInterrupt:
-        sys.exit(u'\nERROR: Interrupted by user')
+        sys.exit('\nERROR: Interrupted by user')

youtube_dl/cache.py

@@ -8,7 +8,7 @@ import re
 import shutil
 import traceback
 
-from .compat import compat_expanduser
+from .compat import compat_expanduser, compat_getenv
 from .utils import write_json_file
 
 
@@ -19,7 +19,7 @@ class Cache(object):
     def _get_root_dir(self):
         res = self._ydl.params.get('cachedir')
         if res is None:
-            cache_root = os.environ.get('XDG_CACHE_HOME', '~/.cache')
+            cache_root = compat_getenv('XDG_CACHE_HOME', '~/.cache')
             res = os.path.join(cache_root, 'youtube-dl')
         return compat_expanduser(res)
 

youtube_dl/compat.py

@@ -1,6 +1,7 @@
 from __future__ import unicode_literals
 
 import getpass
+import optparse
 import os
 import subprocess
 import sys
@@ -288,6 +289,36 @@ if sys.version_info < (3, 0) and sys.platform == 'win32':
 else:
     compat_getpass = getpass.getpass
 
+# Old 2.6 and 2.7 releases require kwargs to be bytes
+try:
+    (lambda x: x)(**{'x': 0})
+except TypeError:
+    def compat_kwargs(kwargs):
+        return dict((bytes(k), v) for k, v in kwargs.items())
+else:
+    compat_kwargs = lambda kwargs: kwargs
+
+
+# Fix https://github.com/rg3/youtube-dl/issues/4223
+# See http://bugs.python.org/issue9161 for what is broken
+def workaround_optparse_bug9161():
+    op = optparse.OptionParser()
+    og = optparse.OptionGroup(op, 'foo')
+    try:
+        og.add_option('-t')
+    except TypeError:
+        real_add_option = optparse.OptionGroup.add_option
+
+        def _compat_add_option(self, *args, **kwargs):
+            enc = lambda v: (
+                v.encode('ascii', 'replace') if isinstance(v, compat_str)
+                else v)
+            bargs = [enc(a) for a in args]
+            bkwargs = dict(
+                (k, enc(v)) for k, v in kwargs.items())
+            return real_add_option(self, *bargs, **bkwargs)
+        optparse.OptionGroup.add_option = _compat_add_option
+
 
 __all__ = [
     'compat_HTTPError',
@@ -299,6 +330,7 @@ __all__ = [
     'compat_html_entities',
     'compat_html_parser',
     'compat_http_client',
+    'compat_kwargs',
     'compat_ord',
     'compat_parse_qs',
     'compat_print',
@@ -314,4 +346,5 @@ __all__ = [
     'compat_xml_parse_error',
     'shlex_quote',
     'subprocess_check_output',
+    'workaround_optparse_bug9161',
 ]

youtube_dl/downloader/common.py

@@ -1,3 +1,5 @@
+from __future__ import unicode_literals
+
 import os
 import re
 import sys
@@ -159,14 +161,14 @@ class FileDownloader(object):
 
     def temp_name(self, filename):
         """Returns a temporary filename for the given filename."""
-        if self.params.get('nopart', False) or filename == u'-' or \
+        if self.params.get('nopart', False) or filename == '-' or \
                 (os.path.exists(encodeFilename(filename)) and not os.path.isfile(encodeFilename(filename))):
             return filename
-        return filename + u'.part'
+        return filename + '.part'
 
     def undo_temp_name(self, filename):
-        if filename.endswith(u'.part'):
+        if filename.endswith('.part'):
-            return filename[:-len(u'.part')]
+            return filename[:-len('.part')]
         return filename
 
     def try_rename(self, old_filename, new_filename):
@@ -175,7 +177,7 @@ class FileDownloader(object):
                 return
             os.rename(encodeFilename(old_filename), encodeFilename(new_filename))
         except (IOError, OSError) as err:
-            self.report_error(u'unable to rename file: %s' % compat_str(err))
+            self.report_error('unable to rename file: %s' % compat_str(err))
 
     def try_utime(self, filename, last_modified_hdr):
         """Try to set the last-modified time of the given file."""
@@ -200,10 +202,10 @@ class FileDownloader(object):
 
     def report_destination(self, filename):
         """Report destination filename."""
-        self.to_screen(u'[download] Destination: ' + filename)
+        self.to_screen('[download] Destination: ' + filename)
 
     def _report_progress_status(self, msg, is_last_line=False):
-        fullmsg = u'[download] ' + msg
+        fullmsg = '[download] ' + msg
         if self.params.get('progress_with_newline', False):
             self.to_screen(fullmsg)
         else:
@@ -211,13 +213,13 @@ class FileDownloader(object):
                 prev_len = getattr(self, '_report_progress_prev_line_length',
                                    0)
                 if prev_len > len(fullmsg):
-                    fullmsg += u' ' * (prev_len - len(fullmsg))
+                    fullmsg += ' ' * (prev_len - len(fullmsg))
                 self._report_progress_prev_line_length = len(fullmsg)
-                clear_line = u'\r'
+                clear_line = '\r'
             else:
-                clear_line = (u'\r\x1b[K' if sys.stderr.isatty() else u'\r')
+                clear_line = ('\r\x1b[K' if sys.stderr.isatty() else '\r')
             self.to_screen(clear_line + fullmsg, skip_eol=not is_last_line)
-        self.to_console_title(u'youtube-dl ' + msg)
+        self.to_console_title('youtube-dl ' + msg)
 
     def report_progress(self, percent, data_len_str, speed, eta):
         """Report download progress."""
@@ -233,7 +235,7 @@ class FileDownloader(object):
             percent_str = 'Unknown %'
         speed_str = self.format_speed(speed)
 
-        msg = (u'%s of %s at %s ETA %s' %
+        msg = ('%s of %s at %s ETA %s' %
                (percent_str, data_len_str, speed_str, eta_str))
         self._report_progress_status(msg)
 
@@ -243,37 +245,37 @@ class FileDownloader(object):
         downloaded_str = format_bytes(downloaded_data_len)
         speed_str = self.format_speed(speed)
         elapsed_str = FileDownloader.format_seconds(elapsed)
-        msg = u'%s at %s (%s)' % (downloaded_str, speed_str, elapsed_str)
+        msg = '%s at %s (%s)' % (downloaded_str, speed_str, elapsed_str)
         self._report_progress_status(msg)
 
     def report_finish(self, data_len_str, tot_time):
         """Report download finished."""
         if self.params.get('noprogress', False):
-            self.to_screen(u'[download] Download completed')
+            self.to_screen('[download] Download completed')
         else:
             self._report_progress_status(
-                (u'100%% of %s in %s' %
+                ('100%% of %s in %s' %
                  (data_len_str, self.format_seconds(tot_time))),
                 is_last_line=True)
 
     def report_resuming_byte(self, resume_len):
         """Report attempt to resume at given byte."""
-        self.to_screen(u'[download] Resuming download at byte %s' % resume_len)
+        self.to_screen('[download] Resuming download at byte %s' % resume_len)
 
     def report_retry(self, count, retries):
         """Report retry in case of HTTP error 5xx"""
-        self.to_screen(u'[download] Got server HTTP error. Retrying (attempt %d of %d)...' % (count, retries))
+        self.to_screen('[download] Got server HTTP error. Retrying (attempt %d of %d)...' % (count, retries))
 
     def report_file_already_downloaded(self, file_name):
         """Report file has already been fully downloaded."""
         try:
-            self.to_screen(u'[download] %s has already been downloaded' % file_name)
+            self.to_screen('[download] %s has already been downloaded' % file_name)
         except UnicodeEncodeError:
-            self.to_screen(u'[download] The file has already been downloaded')
+            self.to_screen('[download] The file has already been downloaded')
 
     def report_unable_to_resume(self):
         """Report it was impossible to resume download."""
-        self.to_screen(u'[download] Unable to resume')
+        self.to_screen('[download] Unable to resume')
 
     def download(self, filename, info_dict):
         """Download to a filename using the info from info_dict
@@ -293,7 +295,7 @@ class FileDownloader(object):
 
     def real_download(self, filename, info_dict):
         """Real download process. Redefine in subclasses."""
-        raise NotImplementedError(u'This method must be implemented by subclasses')
+        raise NotImplementedError('This method must be implemented by subclasses')
 
     def _hook_progress(self, status):
         for ph in self._progress_hooks:

youtube_dl/extractor/__init__.py

@@ -115,6 +115,7 @@ from .fktv import (
     FKTVPosteckeIE,
 )
 from .flickr import FlickrIE
+from .folketinget import FolketingetIE
 from .fourtube import FourTubeIE
 from .franceculture import FranceCultureIE
 from .franceinter import FranceInterIE
@@ -127,6 +128,7 @@ from .francetv import (
 )
 from .freesound import FreesoundIE
 from .freespeech import FreespeechIE
+from .freevideo import FreeVideoIE
 from .funnyordie import FunnyOrDieIE
 from .gamekings import GamekingsIE
 from .gameone import (
@@ -378,6 +380,7 @@ from .teachingchannel import TeachingChannelIE
 from .teamcoco import TeamcocoIE
 from .techtalks import TechTalksIE
 from .ted import TEDIE
+from .telebruxelles import TeleBruxellesIE
 from .telecinco import TelecincoIE
 from .telemb import TeleMBIE
 from .tenplay import TenPlayIE

youtube_dl/extractor/arte.py

@@ -5,13 +5,12 @@ import re
 
 from .common import InfoExtractor
 from ..utils import (
-    ExtractorError,
     find_xpath_attr,
     unified_strdate,
-    determine_ext,
     get_element_by_id,
     get_element_by_attribute,
     int_or_none,
+    qualities,
 )
 
 # There are different sources of video in arte.tv, the extraction process 
@@ -102,79 +101,54 @@ class ArteTVPlus7IE(InfoExtractor):
             'upload_date': unified_strdate(upload_date_str),
             'thumbnail': player_info.get('programImage') or player_info.get('VTU', {}).get('IUR'),
         }
+        qfunc = qualities(['HQ', 'MQ', 'EQ', 'SQ'])
 
-        all_formats = []
+        formats = []
         for format_id, format_dict in player_info['VSR'].items():
-            fmt = dict(format_dict)
+            f = dict(format_dict)
-            fmt['format_id'] = format_id
+            versionCode = f.get('versionCode')
-            all_formats.append(fmt)
-        # Some formats use the m3u8 protocol
-        all_formats = list(filter(lambda f: f.get('videoFormat') != 'M3U8', all_formats))
-        def _match_lang(f):
-            if f.get('versionCode') is None:
-                return True
-            # Return true if that format is in the language of the url
-            if lang == 'fr':
-                l = 'F'
-            elif lang == 'de':
-                l = 'A'
-            else:
-                l = lang
-            regexes = [r'VO?%s' % l, r'VO?.-ST%s' % l]
-            return any(re.match(r, f['versionCode']) for r in regexes)
-        # Some formats may not be in the same language as the url
-        # TODO: Might want not to drop videos that does not match requested language
-        # but to process those formats with lower precedence
-        formats = filter(_match_lang, all_formats)
-        formats = list(formats)  # in python3 filter returns an iterator
-        if not formats:
-            # Some videos are only available in the 'Originalversion'
-            # they aren't tagged as being in French or German
-            # Sometimes there are neither videos of requested lang code
-            # nor original version videos available
-            # For such cases we just take all_formats as is
-            formats = all_formats
-            if not formats:
-                raise ExtractorError('The formats list is empty')
 
-        if re.match(r'[A-Z]Q', formats[0]['quality']) is not None:
+            langcode = {
-            def sort_key(f):
+                'fr': 'F',
-                return ['HQ', 'MQ', 'EQ', 'SQ'].index(f['quality'])
+                'de': 'A',
-        else:
+            }.get(lang, lang)
-            def sort_key(f):
+            lang_rexs = [r'VO?%s' % langcode, r'VO?.-ST%s' % langcode]
-                versionCode = f.get('versionCode')
+            lang_pref = (
-                if versionCode is None:
+                None if versionCode is None else (
-                    versionCode = ''
+                    10 if any(re.match(r, versionCode) for r in lang_rexs)
-                return (
+                    else -10))
-                    # Sort first by quality
+            source_pref = 0
-                    int(f.get('height', -1)),
+            if versionCode is not None:
-                    int(f.get('bitrate', -1)),
+                # The original version with subtitles has lower relevance
-                    # The original version with subtitles has lower relevance
+                if re.match(r'VO-ST(F|A)', versionCode):
-                    re.match(r'VO-ST(F|A)', versionCode) is None,
+                    source_pref -= 10
-                    # The version with sourds/mal subtitles has also lower relevance
+                # The version with sourds/mal subtitles has also lower relevance
-                    re.match(r'VO?(F|A)-STM\1', versionCode) is None,
+                elif re.match(r'VO?(F|A)-STM\1', versionCode):
-                    # Prefer http downloads over m3u8
+                    source_pref -= 9
-                    0 if f['url'].endswith('m3u8') else 1,
+            format = {
-                )
+                'format_id': format_id,
-        formats = sorted(formats, key=sort_key)
+                'preference': -10 if f.get('videoFormat') == 'M3U8' else None,
-        def _format(format_info):
+                'language_preference': lang_pref,
-            info = {
+                'format_note': '%s, %s' % (f.get('versionCode'), f.get('versionLibelle')),
-                'format_id': format_info['format_id'],
+                'width': int_or_none(f.get('width')),
-                'format_note': '%s, %s' % (format_info.get('versionCode'), format_info.get('versionLibelle')),
+                'height': int_or_none(f.get('height')),
-                'width': int_or_none(format_info.get('width')),
+                'tbr': int_or_none(f.get('bitrate')),
-                'height': int_or_none(format_info.get('height')),
+                'quality': qfunc(f['quality']),
-                'tbr': int_or_none(format_info.get('bitrate')),
+                'source_preference': source_pref,
             }
-            if format_info['mediaType'] == 'rtmp':
-                info['url'] = format_info['streamer']
-                info['play_path'] = 'mp4:' + format_info['url']
-                info['ext'] = 'flv'
-            else:
-                info['url'] = format_info['url']
-                info['ext'] = determine_ext(info['url'])
-            return info
-        info_dict['formats'] = [_format(f) for f in formats]
 
+            if f.get('mediaType') == 'rtmp':
+                format['url'] = f['streamer']
+                format['play_path'] = 'mp4:' + f['url']
+                format['ext'] = 'flv'
+            else:
+                format['url'] = f['url']
+
+            formats.append(format)
+
+        self._sort_formats(formats)
+
+        info_dict['formats'] = formats
         return info_dict
 
 

youtube_dl/extractor/bliptv.py

@@ -71,11 +71,12 @@ class BlipTVIE(SubtitlesInfoExtractor):
         mobj = re.match(self._VALID_URL, url)
         lookup_id = mobj.group('lookup_id')
 
-        # See https://github.com/rg3/youtube-dl/issues/857
+        # See https://github.com/rg3/youtube-dl/issues/857 and
+        # https://github.com/rg3/youtube-dl/issues/4197
         if lookup_id:
             info_page = self._download_webpage(
                 'http://blip.tv/play/%s.x?p=1' % lookup_id, lookup_id, 'Resolving lookup id')
-            video_id = self._search_regex(r'data-episode-id="([0-9]+)', info_page, 'video_id')
+            video_id = self._search_regex(r'config\.id\s*=\s*"([0-9]+)', info_page, 'video_id')
         else:
             video_id = mobj.group('id')
 
@@ -165,9 +166,17 @@ class BlipTVIE(SubtitlesInfoExtractor):
 
 
 class BlipTVUserIE(InfoExtractor):
-    _VALID_URL = r'(?:(?:(?:https?://)?(?:\w+\.)?blip\.tv/)|bliptvuser:)(?!api\.swf)([^/]+)/*$'
+    _VALID_URL = r'(?:(?:https?://(?:\w+\.)?blip\.tv/)|bliptvuser:)(?!api\.swf)([^/]+)/*$'
     _PAGE_SIZE = 12
     IE_NAME = 'blip.tv:user'
+    _TEST = {
+        'url': 'http://blip.tv/actone',
+        'info_dict': {
+            'id': 'actone',
+            'title': 'Act One: The Series',
+        },
+        'playlist_count': 5,
+    }
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
@@ -178,6 +187,7 @@ class BlipTVUserIE(InfoExtractor):
         page = self._download_webpage(url, username, 'Downloading user page')
         mobj = re.search(r'data-users-id="([^"]+)"', page)
         page_base = page_base % mobj.group(1)
+        title = self._og_search_title(page)
 
         # Download video ids using BlipTV Ajax calls. Result size per
         # query is limited (currently to 12 videos) so we need to query
@@ -214,4 +224,5 @@ class BlipTVUserIE(InfoExtractor):
 
         urls = ['http://blip.tv/%s' % video_id for video_id in video_ids]
         url_entries = [self.url_result(vurl, 'BlipTV') for vurl in urls]
-        return [self.playlist_result(url_entries, playlist_title=username)]
+        return self.playlist_result(
+            url_entries, playlist_title=title, playlist_id=username)

youtube_dl/extractor/brightcove.py

@@ -111,6 +111,8 @@ 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('<--', '<!--')
+        # remove namespace to simplify extraction
+        object_str = re.sub(r'(<object[^>]*)(xmlns=".*?")', r'\1', object_str)
         object_str = fix_xml_ampersands(object_str)
 
         object_doc = xml.etree.ElementTree.fromstring(object_str.encode('utf-8'))
@@ -219,7 +221,7 @@ class BrightcoveIE(InfoExtractor):
         webpage = self._download_webpage(req, video_id)
 
         error_msg = self._html_search_regex(
-            r"<h1>We're sorry.</h1>\s*<p>(.*?)</p>", webpage,
+            r"<h1>We're sorry.</h1>([\s\n]*<p>.*?</p>)+", webpage,
             'error message', default=None)
         if error_msg is not None:
             raise ExtractorError(

youtube_dl/extractor/comedycentral.py

@@ -2,7 +2,6 @@ from __future__ import unicode_literals
 
 import re
 
-from .common import InfoExtractor
 from .mtv import MTVServicesInfoExtractor
 from ..utils import (
     compat_str,
@@ -31,7 +30,7 @@ class ComedyCentralIE(MTVServicesInfoExtractor):
     }
 
 
-class ComedyCentralShowsIE(InfoExtractor):
+class ComedyCentralShowsIE(MTVServicesInfoExtractor):
     IE_DESC = 'The Daily Show / The Colbert Report'
     # urls can be abbreviations like :thedailyshow or :colbert
     # urls for episodes like:
@@ -109,18 +108,8 @@ class ComedyCentralShowsIE(InfoExtractor):
         '400': (384, 216),
     }
 
-    @staticmethod
-    def _transform_rtmp_url(rtmp_video_url):
-        m = re.match(r'^rtmpe?://.*?/(?P<finalid>gsp\.comedystor/.*)$', rtmp_video_url)
-        if not m:
-            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)
+        mobj = re.match(self._VALID_URL, url)
-        if mobj is None:
-            raise ExtractorError('Invalid URL: %s' % url)
 
         if mobj.group('shortname'):
             if mobj.group('shortname') in ('tds', 'thedailyshow'):
@@ -212,9 +201,6 @@ class ComedyCentralShowsIE(InfoExtractor):
                     'ext': self._video_extensions.get(format, 'mp4'),
                     'height': h,
                     'width': w,
-
-                    'format_note': 'HTTP 400 at the moment (patches welcome!)',
-                    'preference': -100,
                 })
                 formats.append({
                     'format_id': 'rtmp-%s' % format,

youtube_dl/extractor/common.py

@@ -43,7 +43,11 @@ class InfoExtractor(object):
     information possibly downloading the video to the file system, among
     other possible outcomes.
 
-    The dictionaries must include the following fields:
+    The type field determines the the type of the result.
+    By far the most common value (and the default if _type is missing) is
+    "video", which indicates a single video.
+
+    For a video, the dictionaries must include the following fields:
 
     id:             Video identifier.
     title:          Video title, unescaped.
@@ -87,6 +91,11 @@ class InfoExtractor(object):
                                  by this field, regardless of all other values.
                                  -1 for default (order by other properties),
                                  -2 or smaller for less than default.
+                    * language_preference  Is this in the correct requested
+                                 language?
+                                 10 if it's what the URL is about,
+                                 -1 for default (don't know),
+                                 -10 otherwise, other values reserved for now.
                     * quality    Order number of the video quality of this
                                  format, irrespective of the file format.
                                  -1 for default (order by other properties),
@@ -146,6 +155,38 @@ class InfoExtractor(object):
 
     Unless mentioned otherwise, None is equivalent to absence of information.
 
+
+    _type "playlist" indicates multiple videos.
+    There must be a key "entries", which is a list or a PagedList object, each
+    element of which is a valid dictionary under this specfication.
+
+    Additionally, playlists can have "title" and "id" attributes with the same
+    semantics as videos (see above).
+
+
+    _type "multi_video" indicates that there are multiple videos that
+    form a single show, for examples multiple acts of an opera or TV episode.
+    It must have an entries key like a playlist and contain all the keys
+    required for a video at the same time.
+
+
+    _type "url" indicates that the video must be extracted from another
+    location, possibly by a different extractor. Its only required key is:
+    "url" - the next URL to extract.
+
+    Additionally, it may have properties believed to be identical to the
+    resolved entity, for example "title" if the title of the referred video is
+    known ahead of time.
+
+
+    _type "url_transparent" entities have the same specification as "url", but
+    indicate that the given additional information is more precise than the one
+    associated with the resolved URL.
+    This is useful when a site employs a video service that hosts the video and
+    its technical metadata, but that video service does not embed a useful
+    title, description etc.
+
+
     Subclasses of this one should re-define the _real_initialize() and
     _real_extract() methods and define a _VALID_URL regexp.
     Probably, they should also be added to the list of extractors.
@@ -615,6 +656,7 @@ class InfoExtractor(object):
 
             return (
                 preference,
+                f.get('language_preference') if f.get('language_preference') is not None else -1,
                 f.get('quality') if f.get('quality') is not None else -1,
                 f.get('height') if f.get('height') is not None else -1,
                 f.get('width') if f.get('width') is not None else -1,

youtube_dl/extractor/crunchyroll.py

@@ -264,8 +264,6 @@ Format: Layer, Start, End, Style, Name, MarginL, MarginR, MarginV, Effect, Text
             if not lang_code:
                 continue
             sub_root = xml.etree.ElementTree.fromstring(subtitle)
-            if not sub_root:
-                subtitles[lang_code] = ''
             if sub_format == 'ass':
                 subtitles[lang_code] = self._convert_subtitles_to_ass(sub_root)
             else:

youtube_dl/extractor/eporner.py

@@ -20,7 +20,7 @@ class EpornerIE(InfoExtractor):
             'display_id': 'Infamous-Tiffany-Teen-Strip-Tease-Video',
             'ext': 'mp4',
             'title': 'Infamous Tiffany Teen Strip Tease Video',
-            'duration': 194,
+            'duration': 1838,
             'view_count': int,
             'age_limit': 18,
         }
@@ -57,9 +57,7 @@ class EpornerIE(InfoExtractor):
             formats.append(fmt)
         self._sort_formats(formats)
 
-        duration = parse_duration(self._search_regex(
+        duration = parse_duration(self._html_search_meta('duration', webpage))
-            r'class="mbtim">([0-9:]+)</div>', webpage, 'duration',
-            fatal=False))
         view_count = str_to_int(self._search_regex(
             r'id="cinemaviews">\s*([0-9,]+)\s*<small>views',
             webpage, 'view count', fatal=False))

youtube_dl/extractor/folketinget.py

@@ -0,0 +1,75 @@
+# -*- coding: utf-8 -*-
+from __future__ import unicode_literals
+
+from .common import InfoExtractor
+from ..compat import compat_parse_qs
+from ..utils import (
+    int_or_none,
+    parse_duration,
+    parse_iso8601,
+    xpath_text,
+)
+
+
+class FolketingetIE(InfoExtractor):
+    IE_DESC = 'Folketinget (ft.dk; Danish parliament)'
+    _VALID_URL = r'https?://(?:www\.)?ft\.dk/webtv/video/[^?#]*?\.(?P<id>[0-9]+)\.aspx'
+    _TEST = {
+        'url': 'http://www.ft.dk/webtv/video/20141/eru/td.1165642.aspx?as=1#player',
+        'info_dict': {
+            'id': '1165642',
+            'ext': 'mp4',
+            'title': 'Åbent samråd i Erhvervsudvalget',
+            'description': 'Åbent samråd med erhvervs- og vækstministeren om regeringens politik på teleområdet',
+            'view_count': int,
+            'width': 768,
+            'height': 432,
+            'tbr': 928000,
+            'timestamp': 1416493800,
+            'upload_date': '20141120',
+            'duration': 3960,
+        },
+        'params': {
+            'skip_download': 'rtmpdump required',
+        }
+    }
+
+    def _real_extract(self, url):
+        video_id = self._match_id(url)
+        webpage = self._download_webpage(url, video_id)
+
+        title = self._og_search_title(webpage)
+        description = self._html_search_regex(
+            r'(?s)<div class="video-item-agenda"[^>]*>(.*?)<',
+            webpage, 'description', fatal=False)
+
+        player_params = compat_parse_qs(self._search_regex(
+            r'<embed src="http://ft\.arkena\.tv/flash/ftplayer\.swf\?([^"]+)"',
+            webpage, 'player params'))
+        xml_url = player_params['xml'][0]
+        doc = self._download_xml(xml_url, video_id)
+
+        timestamp = parse_iso8601(xpath_text(doc, './/date'))
+        duration = parse_duration(xpath_text(doc, './/duration'))
+        width = int_or_none(xpath_text(doc, './/width'))
+        height = int_or_none(xpath_text(doc, './/height'))
+        view_count = int_or_none(xpath_text(doc, './/views'))
+
+        formats = [{
+            'format_id': n.attrib['bitrate'],
+            'url': xpath_text(n, './url', fatal=True),
+            'tbr': int_or_none(n.attrib['bitrate']),
+        } for n in doc.findall('.//streams/stream')]
+        self._sort_formats(formats)
+
+        return {
+            'id': video_id,
+            'title': title,
+            'formats': formats,
+            'description': description,
+            'timestamp': timestamp,
+            'width': width,
+            'height': height,
+            'duration': duration,
+            'view_count': view_count,
+        }

youtube_dl/extractor/freevideo.py

@@ -0,0 +1,38 @@
+from __future__ import unicode_literals
+
+from .common import InfoExtractor
+from ..utils import ExtractorError
+
+
+class FreeVideoIE(InfoExtractor):
+    _VALID_URL = r'^http://www.freevideo.cz/vase-videa/(?P<id>[^.]+)\.html(?:$|[?#])'
+
+    _TEST = {
+        'url': 'http://www.freevideo.cz/vase-videa/vysukany-zadecek-22033.html',
+        'info_dict': {
+            'id': 'vysukany-zadecek-22033',
+            'ext': 'mp4',
+            "title": "vysukany-zadecek-22033",
+            "age_limit": 18,
+        },
+        'skip': 'Blocked outside .cz',
+    }
+
+    def _real_extract(self, url):
+        video_id = self._match_id(url)
+        webpage, handle = self._download_webpage_handle(url, video_id)
+        if '//www.czechav.com/' in handle.geturl():
+            raise ExtractorError(
+                'Access to freevideo is blocked from your location',
+                expected=True)
+
+        video_url = self._search_regex(
+            r'\s+url: "(http://[a-z0-9-]+.cdn.freevideo.cz/stream/.*?/video.mp4)"',
+            webpage, 'video URL')
+
+        return {
+            'id': video_id,
+            'url': video_url,
+            'title': video_id,
+            'age_limit': 18,
+        }

youtube_dl/extractor/generic.py

@@ -434,7 +434,17 @@ class GenericIE(InfoExtractor):
                 'title': 'Chet Chat 171 - Oct 29, 2014',
                 'upload_date': '20141029',
             }
-        }
+        },
+        # Livestream embed
+        {
+            'url': 'http://www.esa.int/Our_Activities/Space_Science/Rosetta/Philae_comet_touch-down_webcast',
+            'info_dict': {
+                'id': '67864563',
+                'ext': 'flv',
+                'upload_date': '20141112',
+                'title': 'Rosetta #CometLanding webcast HL 10',
+            }
+        },
     ]
 
     def report_following_redirect(self, new_url):
@@ -916,6 +926,12 @@ class GenericIE(InfoExtractor):
         if mobj is not None:
             return self.url_result(self._proto_relative_url(mobj.group('url'), scheme='http:'), 'CondeNast')
 
+        mobj = re.search(
+            r'<iframe[^>]+src="(?P<url>https?://new\.livestream\.com/[^"]+/player[^"]+)"',
+            webpage)
+        if mobj is not None:
+            return self.url_result(mobj.group('url'), 'Livestream')
+
         def check_video(vurl):
             vpath = compat_urlparse.urlparse(vurl).path
             vext = determine_ext(vpath)
@@ -963,7 +979,7 @@ class GenericIE(InfoExtractor):
                 found = filter_video(re.findall(r'<meta.*?property="og:video".*?content="(.*?)"', webpage))
         if not found:
             # HTML5 video
-            found = re.findall(r'(?s)<video[^<]*(?:>.*?<source[^>]*)?\s+src="([^"]+)"', webpage)
+            found = re.findall(r'(?s)<video[^<]*(?:>.*?<source[^>]*)?\s+src=["\'](.*?)["\']', webpage)
         if not found:
             found = re.search(
                 r'(?i)<meta\s+(?=(?:[a-z-]+="[^"]+"\s+)*http-equiv="refresh")'

youtube_dl/extractor/goldenmoustache.py

@@ -1,16 +1,14 @@
 from __future__ import unicode_literals
 
-import re
 from .common import InfoExtractor
 from ..utils import (
-    parse_duration,
     int_or_none,
 )
 
 
 class GoldenMoustacheIE(InfoExtractor):
     _VALID_URL = r'https?://(?:www\.)?goldenmoustache\.com/(?P<display_id>[\w-]+)-(?P<id>\d+)'
-    _TEST = {
+    _TESTS = [{
         'url': 'http://www.goldenmoustache.com/suricate-le-poker-3700/',
         'md5': '0f904432fa07da5054d6c8beb5efb51a',
         'info_dict': {
@@ -21,7 +19,18 @@ class GoldenMoustacheIE(InfoExtractor):
             'thumbnail': 're:^https?://.*\.jpg$',
             'view_count': int,
         }
-    }
+    }, {
+        'url': 'http://www.goldenmoustache.com/le-lab-tout-effacer-mc-fly-et-carlito-55249/',
+        'md5': '27f0c50fb4dd5f01dc9082fc67cd5700',
+        'info_dict': {
+            'id': '55249',
+            'ext': 'mp4',
+            'title': 'Le LAB - Tout Effacer (Mc Fly et Carlito)',
+            'description': 'md5:9b7fbf11023fb2250bd4b185e3de3b2a',
+            'thumbnail': 're:^https?://.*\.(?:png|jpg)$',
+            'view_count': int,
+        }
+    }]
 
     def _real_extract(self, url):
         video_id = self._match_id(url)
@@ -30,7 +39,7 @@ class GoldenMoustacheIE(InfoExtractor):
         video_url = self._html_search_regex(
             r'data-src-type="mp4" data-src="([^"]+)"', webpage, 'video URL')
         title = self._html_search_regex(
-            r'<title>(.*?) - Golden Moustache</title>', webpage, 'title')
+            r'<title>(.*?)(?: - Golden Moustache)?</title>', webpage, 'title')
         thumbnail = self._og_search_thumbnail(webpage)
         description = self._og_search_description(webpage)
         view_count = int_or_none(self._html_search_regex(

youtube_dl/extractor/livestream.py

@@ -18,7 +18,7 @@ from ..utils import (
 
 class LivestreamIE(InfoExtractor):
     IE_NAME = 'livestream'
-    _VALID_URL = r'http://new\.livestream\.com/.*?/(?P<event_name>.*?)(/videos/(?P<id>\d+))?/?$'
+    _VALID_URL = r'https?://new\.livestream\.com/.*?/(?P<event_name>.*?)(/videos/(?P<id>[0-9]+)(?:/player)?)?/?(?:$|[?#])'
     _TESTS = [{
         'url': 'http://new.livestream.com/CoheedandCambria/WebsterHall/videos/4719370',
         'md5': '53274c76ba7754fb0e8d072716f2292b',
@@ -37,6 +37,9 @@ class LivestreamIE(InfoExtractor):
             'title': 'TEDCity2.0 (English)',
         },
         'playlist_mincount': 4,
+    }, {
+        'url': 'https://new.livestream.com/accounts/362/events/3557232/videos/67864563/player?autoPlay=false&height=360&mute=false&width=640',
+        'only_matching': True,
     }]
 
     def _parse_smil(self, video_id, smil_url):

youtube_dl/extractor/mailru.py

@@ -16,7 +16,7 @@ class MailRuIE(InfoExtractor):
             'url': 'http://my.mail.ru/video/top#video=/mail/sonypicturesrus/75/76',
             'md5': 'dea205f03120046894db4ebb6159879a',
             'info_dict': {
-                'id': '46301138',
+                'id': '46301138_76',
                 'ext': 'mp4',
                 'title': 'Новый Человек-Паук. Высокое напряжение. Восстание Электро',
                 'timestamp': 1393232740,
@@ -30,7 +30,7 @@ class MailRuIE(InfoExtractor):
             'url': 'http://my.mail.ru/corp/hitech/video/news_hi-tech_mail_ru/1263.html',
             'md5': '00a91a58c3402204dcced523777b475f',
             'info_dict': {
-                'id': '46843144',
+                'id': '46843144_1263',
                 'ext': 'mp4',
                 'title': 'Samsung Galaxy S5 Hammer Smash Fail Battery Explosion',
                 'timestamp': 1397217632,
@@ -54,30 +54,33 @@ class MailRuIE(InfoExtractor):
 
         author = video_data['author']
         uploader = author['name']
-        uploader_id = author['id']
+        uploader_id = author.get('id') or author.get('email')
+        view_count = video_data.get('views_count')
 
-        movie = video_data['movie']
+        meta_data = video_data['meta']
-        content_id = str(movie['contentId'])
+        content_id = '%s_%s' % (
-        title = movie['title']
+            meta_data.get('accId', ''), meta_data['itemId'])
+        title = meta_data['title']
         if title.endswith('.mp4'):
             title = title[:-4]
-        thumbnail = movie['poster']
+        thumbnail = meta_data['poster']
-        duration = movie['duration']
+        duration = meta_data['duration']
-
+        timestamp = meta_data['timestamp']
-        view_count = video_data['views_count']
 
         formats = [
             {
                 'url': video['url'],
-                'format_id': video['name'],
+                'format_id': video['key'],
+                'height': int(video['key'].rstrip('p'))
             } for video in video_data['videos']
         ]
+        self._sort_formats(formats)
 
         return {
             'id': content_id,
             'title': title,
             'thumbnail': thumbnail,
-            'timestamp': video_data['timestamp'],
+            'timestamp': timestamp,
             'uploader': uploader,
             'uploader_id': uploader_id,
             'duration': duration,

youtube_dl/extractor/mtv.py

@@ -145,7 +145,8 @@ class MTVServicesInfoExtractor(InfoExtractor):
         idoc = self._download_xml(
             feed_url + '?' + data, video_id,
             'Downloading info', transform_source=fix_xml_ampersands)
-        return [self._get_video_info(item) for item in idoc.findall('.//item')]
+        return self.playlist_result(
+            [self._get_video_info(item) for item in idoc.findall('.//item')])
 
     def _real_extract(self, url):
         title = url_basename(url)
@@ -186,7 +187,8 @@ class MTVServicesEmbeddedIE(MTVServicesInfoExtractor):
     def _get_feed_url(self, uri):
         video_id = self._id_from_uri(uri)
         site_id = uri.replace(video_id, '')
-        config_url = 'http://media.mtvnservices.com/pmt/e1/players/{0}/config.xml'.format(site_id)
+        config_url = ('http://media.mtvnservices.com/pmt/e1/players/{0}/'
+            'context4/context5/config.xml'.format(site_id))
         config_doc = self._download_xml(config_url, video_id)
         feed_node = config_doc.find('.//feed')
         feed_url = feed_node.text.strip().split('?')[0]

youtube_dl/extractor/npo.py

@@ -7,6 +7,7 @@ from ..utils import (
     unified_strdate,
     parse_duration,
     qualities,
+    strip_jsonp,
     url_basename,
 )
 
@@ -63,7 +64,7 @@ class NPOIE(InfoExtractor):
             'http://e.omroep.nl/metadata/aflevering/%s' % video_id,
             video_id,
             # We have to remove the javascript callback
-            transform_source=lambda j: re.sub(r'parseMetadata\((.*?)\);\n//.*$', r'\1', j)
+            transform_source=strip_jsonp,
         )
         token_page = self._download_webpage(
             'http://ida.omroep.nl/npoplayer/i.js',

youtube_dl/extractor/rtlnl.py

@@ -38,10 +38,11 @@ class RtlXlIE(InfoExtractor):
         progname = info['abstracts'][0]['name']
         subtitle = material['title'] or info['episodes'][0]['name']
 
-        videopath = material['videopath']
+        # Use unencrypted m3u8 streams (See https://github.com/rg3/youtube-dl/issues/4118)
-        f4m_url = 'http://manifest.us.rtl.nl' + videopath
+        videopath = material['videopath'].replace('.f4m', '.m3u8')
+        m3u8_url = 'http://manifest.us.rtl.nl' + videopath
 
-        formats = self._extract_f4m_formats(f4m_url, uuid)
+        formats = self._extract_m3u8_formats(m3u8_url, uuid, ext='mp4')
 
         video_urlpart = videopath.split('/flash/')[1][:-4]
         PG_URL_TEMPLATE = 'http://pg.us.rtl.nl/rtlxl/network/%s/progressive/%s.mp4'
@@ -54,9 +55,12 @@ class RtlXlIE(InfoExtractor):
             {
                 'url': PG_URL_TEMPLATE % ('a3m', video_urlpart),
                 'format_id': 'pg-hd',
+                'quality': 0,
             }
         ])
 
+        self._sort_formats(formats)
+
         return {
             'id': uuid,
             'title': '%s - %s' % (progname, subtitle),

youtube_dl/extractor/spiegel.py

@@ -5,6 +5,7 @@ import re
 
 from .common import InfoExtractor
 from ..compat import compat_urlparse
+from .spiegeltv import SpiegeltvIE
 
 
 class SpiegelIE(InfoExtractor):
@@ -42,7 +43,11 @@ class SpiegelIE(InfoExtractor):
 
     def _real_extract(self, url):
         video_id = self._match_id(url)
-        webpage = self._download_webpage(url, video_id)
+        webpage, handle = self._download_webpage_handle(url, video_id)
+
+        # 302 to spiegel.tv, like http://www.spiegel.de/video/der-film-zum-wochenende-die-wahrheit-ueber-maenner-video-99003272.html
+        if SpiegeltvIE.suitable(handle.geturl()):
+            return self.url_result(handle.geturl(), 'Spiegeltv')
 
         title = re.sub(r'\s+', ' ', self._html_search_regex(
             r'(?s)<(?:h1|div) class="module-title"[^>]*>(.*?)</(?:h1|div)>',

youtube_dl/extractor/spiegeltv.py

@@ -1,13 +1,13 @@
 # coding: utf-8
 from __future__ import unicode_literals
 
-import re
 from .common import InfoExtractor
+from ..utils import float_or_none
 
 
 class SpiegeltvIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:www\.)?spiegel\.tv/filme/(?P<id>[\-a-z0-9]+)'
+    _VALID_URL = r'https?://(?:www\.)?spiegel\.tv/(?:#/)?filme/(?P<id>[\-a-z0-9]+)'
-    _TEST = {
+    _TESTS = [{
         'url': 'http://www.spiegel.tv/filme/flug-mh370/',
         'info_dict': {
             'id': 'flug-mh370',
@@ -20,12 +20,15 @@ class SpiegeltvIE(InfoExtractor):
             # rtmp download
             'skip_download': True,
         }
-    }
+    }, {
+        'url': 'http://www.spiegel.tv/#/filme/alleskino-die-wahrheit-ueber-maenner/',
+        'only_matching': True,
+    }]
 
     def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
+        if '/#/' in url:
-        video_id = mobj.group('id')
+            url = url.replace('/#/', '/')
-
+        video_id = self._match_id(url)
         webpage = self._download_webpage(url, video_id)
         title = self._html_search_regex(r'<h1.*?>(.*?)</h1>', webpage, 'title')
 
@@ -61,12 +64,8 @@ class SpiegeltvIE(InfoExtractor):
             })
 
         description = media_json['subtitle']
-        duration = media_json['duration_in_ms'] / 1000.
+        duration = float_or_none(media_json.get('duration_in_ms'), scale=1000)
-
+        format = '16x9' if is_wide else '4x3'
-        if is_wide:
-            format = '16x9'
-        else:
-            format = '4x3'
 
         url = server + 'mp4:' + uuid + '_spiegeltv_0500_' + format + '.m4v'
 

youtube_dl/extractor/stanfordoc.py

@@ -1,3 +1,5 @@
+from __future__ import unicode_literals
+
 import re
 
 from .common import InfoExtractor
@@ -9,24 +11,23 @@ from ..utils import (
 
 
 class StanfordOpenClassroomIE(InfoExtractor):
-    IE_NAME = u'stanfordoc'
+    IE_NAME = 'stanfordoc'
-    IE_DESC = u'Stanford Open ClassRoom'
+    IE_DESC = 'Stanford Open ClassRoom'
-    _VALID_URL = r'^(?:https?://)?openclassroom\.stanford\.edu(?P<path>/?|(/MainFolder/(?:HomePage|CoursePage|VideoPage)\.php([?]course=(?P<course>[^&]+)(&video=(?P<video>[^&]+))?(&.*)?)?))$'
+    _VALID_URL = r'https?://openclassroom\.stanford\.edu(?P<path>/?|(/MainFolder/(?:HomePage|CoursePage|VideoPage)\.php([?]course=(?P<course>[^&]+)(&video=(?P<video>[^&]+))?(&.*)?)?))$'
     _TEST = {
-        u'url': u'http://openclassroom.stanford.edu/MainFolder/VideoPage.php?course=PracticalUnix&video=intro-environment&speed=100',
+        'url': 'http://openclassroom.stanford.edu/MainFolder/VideoPage.php?course=PracticalUnix&video=intro-environment&speed=100',
-        u'file': u'PracticalUnix_intro-environment.mp4',
+        'md5': '544a9468546059d4e80d76265b0443b8',
-        u'md5': u'544a9468546059d4e80d76265b0443b8',
+        'info_dict': {
-        u'info_dict': {
+            'id': 'PracticalUnix_intro-environment',
-            u"title": u"Intro Environment"
+            'ext': 'mp4',
+            'title': 'Intro Environment',
         }
     }
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
-        if mobj is None:
-            raise ExtractorError(u'Invalid URL: %s' % url)
 
-        if mobj.group('course') and mobj.group('video'): # A specific video
+        if mobj.group('course') and mobj.group('video'):  # A specific video
             course = mobj.group('course')
             video = mobj.group('video')
             info = {
@@ -35,7 +36,6 @@ class StanfordOpenClassroomIE(InfoExtractor):
                 'upload_date': None,
             }
 
-            self.report_extraction(info['id'])
             baseUrl = 'http://openclassroom.stanford.edu/MainFolder/courses/' + course + '/videos/'
             xmlUrl = baseUrl + video + '.xml'
             mdoc = self._download_xml(xmlUrl, info['id'])
@@ -43,63 +43,49 @@ class StanfordOpenClassroomIE(InfoExtractor):
                 info['title'] = mdoc.findall('./title')[0].text
                 info['url'] = baseUrl + mdoc.findall('./videoFile')[0].text
             except IndexError:
-                raise ExtractorError(u'Invalid metadata XML file')
+                raise ExtractorError('Invalid metadata XML file')
-            info['ext'] = info['url'].rpartition('.')[2]
+            return info
-            return [info]
+        elif mobj.group('course'):  # A course page
-        elif mobj.group('course'): # A course page
             course = mobj.group('course')
             info = {
                 'id': course,
-                'type': 'playlist',
+                '_type': 'playlist',
                 'uploader': None,
                 'upload_date': None,
             }
 
-            coursepage = self._download_webpage(url, info['id'],
+            coursepage = self._download_webpage(
-                                        note='Downloading course info page',
+                url, info['id'],
-                                        errnote='Unable to download course info page')
+                note='Downloading course info page',
+                errnote='Unable to download course info page')
 
-            info['title'] = self._html_search_regex('<h1>([^<]+)</h1>', coursepage, 'title', default=info['id'])
+            info['title'] = self._html_search_regex(
+                r'<h1>([^<]+)</h1>', coursepage, 'title', default=info['id'])
 
-            info['description'] = self._html_search_regex('<description>([^<]+)</description>',
+            info['description'] = self._html_search_regex(
-                coursepage, u'description', fatal=False)
+                r'(?s)<description>([^<]+)</description>',
+                coursepage, 'description', fatal=False)
 
             links = orderedSet(re.findall('<a href="(VideoPage.php\?[^"]+)">', coursepage))
-            info['list'] = [
+            info['entries'] = [self.url_result(
-                {
+                'http://openclassroom.stanford.edu/MainFolder/%s' % unescapeHTML(l)
-                    'type': 'reference',
+            ) for l in links]
-                    'url': 'http://openclassroom.stanford.edu/MainFolder/' + unescapeHTML(vpage),
+            return info
-                }
+        else:  # Root page
-                    for vpage in links]
-            results = []
-            for entry in info['list']:
-                assert entry['type'] == 'reference'
-                results += self.extract(entry['url'])
-            return results
-        else: # Root page
             info = {
                 'id': 'Stanford OpenClassroom',
-                'type': 'playlist',
+                '_type': 'playlist',
                 'uploader': None,
                 'upload_date': None,
             }
+            info['title'] = info['id']
 
             rootURL = 'http://openclassroom.stanford.edu/MainFolder/HomePage.php'
             rootpage = self._download_webpage(rootURL, info['id'],
-                errnote=u'Unable to download course info page')
+                errnote='Unable to download course info page')
-
-            info['title'] = info['id']
 
             links = orderedSet(re.findall('<a href="(CoursePage.php\?[^"]+)">', rootpage))
-            info['list'] = [
+            info['entries'] = [self.url_result(
-                {
+                'http://openclassroom.stanford.edu/MainFolder/%s' % unescapeHTML(l)
-                    'type': 'reference',
+            ) for l in links]
-                    'url': 'http://openclassroom.stanford.edu/MainFolder/' + unescapeHTML(cpage),
+            return info
-                }
-                    for cpage in links]
-
-            results = []
-            for entry in info['list']:
-                assert entry['type'] == 'reference'
-                results += self.extract(entry['url'])
-            return results

youtube_dl/extractor/sztvhu.py

@@ -1,27 +1,24 @@
 # -*- coding: utf-8 -*-
-
+from __future__ import unicode_literals
-import re
 
 from .common import InfoExtractor
-from ..utils import determine_ext
 
 
 class SztvHuIE(InfoExtractor):
-    _VALID_URL = r'(?:http://)?(?:(?:www\.)?sztv\.hu|www\.tvszombathely\.hu)/(?:[^/]+)/.+-(?P<id>[0-9]+)'
+    _VALID_URL = r'http://(?:(?:www\.)?sztv\.hu|www\.tvszombathely\.hu)/(?:[^/]+)/.+-(?P<id>[0-9]+)'
     _TEST = {
-        u'url': u'http://sztv.hu/hirek/cserkeszek-nepszerusitettek-a-kornyezettudatos-eletmodot-a-savaria-teren-20130909',
+        'url': 'http://sztv.hu/hirek/cserkeszek-nepszerusitettek-a-kornyezettudatos-eletmodot-a-savaria-teren-20130909',
-        u'file': u'20130909.mp4',
+        'md5': 'a6df607b11fb07d0e9f2ad94613375cb',
-        u'md5': u'a6df607b11fb07d0e9f2ad94613375cb',
+        'info_dict': {
-        u'info_dict': {
+            'id': '20130909',
-            u"title": u"Cserkészek népszerűsítették a környezettudatos életmódot a Savaria téren",
+            'ext': 'mp4',
-            u"description": u'A zöld nap játékos ismeretterjesztő programjait a Magyar Cserkész Szövetség szervezte, akik az ország nyolc városában adják át tudásukat az érdeklődőknek. A PET...',
+            'title': 'Cserkészek népszerűsítették a környezettudatos életmódot a Savaria téren',
+            'description': 'A zöld nap játékos ismeretterjesztő programjait a Magyar Cserkész Szövetség szervezte, akik az ország nyolc városában adják át tudásukat az érdeklődőknek. A PET...',
         },
-        u'skip': u'Service temporarily disabled as of 2013-11-20'
     }
 
     def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
+        video_id = self._match_id(url)
-        video_id = mobj.group('id')
         webpage = self._download_webpage(url, video_id)
         video_file = self._search_regex(
             r'file: "...:(.*?)",', webpage, 'video file')
@@ -39,7 +36,6 @@ class SztvHuIE(InfoExtractor):
             'id': video_id,
             'url': video_url,
             'title': title,
-            'ext': determine_ext(video_url),
             'description': description,
             'thumbnail': thumbnail,
         }

youtube_dl/extractor/telebruxelles.py

@@ -0,0 +1,60 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+from .common import InfoExtractor
+
+
+class TeleBruxellesIE(InfoExtractor):
+    _VALID_URL = r'https?://(?:www\.)?telebruxelles\.be/(news|sport|dernier-jt)/?(?P<id>[^/#?]+)'
+    _TESTS = [{
+        'url': 'http://www.telebruxelles.be/news/auditions-devant-parlement-francken-galant-tres-attendus/',
+        'md5': '59439e568c9ee42fb77588b2096b214f',
+        'info_dict': {
+            'id': '11942',
+            'display_id': 'auditions-devant-parlement-francken-galant-tres-attendus',
+            'ext': 'flv',
+            'title': 'Parlement : Francken et Galant répondent aux interpellations de l’opposition',
+            'description': 're:Les auditions des ministres se poursuivent*'
+        },
+        'params': {
+            'skip_download': 'requires rtmpdump'
+        },
+    }, {
+        'url': 'http://www.telebruxelles.be/sport/basket-brussels-bat-mons-80-74/',
+        'md5': '181d3fbdcf20b909309e5aef5c6c6047',
+        'info_dict': {
+            'id': '10091',
+            'display_id': 'basket-brussels-bat-mons-80-74',
+            'ext': 'flv',
+            'title': 'Basket : le Brussels bat Mons 80-74',
+            'description': 're:^Ils l\u2019on fait ! En basket, le B*',
+        },
+        'params': {
+            'skip_download': 'requires rtmpdump'
+        },
+    }]
+
+    def _real_extract(self, url):
+        display_id = self._match_id(url)
+        webpage = self._download_webpage(url, display_id)
+
+        article_id = self._html_search_regex(
+            r"<article id=\"post-(\d+)\"", webpage, 'article ID')
+        title = self._html_search_regex(
+            r'<h1 class=\"entry-title\">(.*?)</h1>', webpage, 'title')
+        description = self._og_search_description(webpage)
+
+        rtmp_url = self._html_search_regex(
+            r"file: \"(rtmp://\d{1,3}\.\d{1,3}\.\d{1,3}\.\d{1,3}:\d{1,5}/vod/mp4:\" \+ \"\w+\" \+ \".mp4)\"",
+            webpage, 'RTMP url')
+        rtmp_url = rtmp_url.replace("\" + \"", "")
+
+        return {
+            'id': article_id,
+            'display_id': display_id,
+            'title': title,
+            'description': description,
+            'url': rtmp_url,
+            'ext': 'flv',
+            'rtmp_live': True  # if rtmpdump is not called with "--live" argument, the download is blocked and can be completed
+        }

youtube_dl/extractor/vh1.py

@@ -121,4 +121,7 @@ class VH1IE(MTVIE):
         idoc = self._download_xml(
             doc_url, video_id,
             'Downloading info', transform_source=fix_xml_ampersands)
-        return [self._get_video_info(item) for item in idoc.findall('.//item')]
+        return self.playlist_result(
+            [self._get_video_info(item) for item in idoc.findall('.//item')],
+            playlist_id=video_id,
+        )

youtube_dl/extractor/youtube.py

@@ -307,6 +307,11 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor):
         '171': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH audio', 'abr': 128, 'preference': -50},
         '172': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH audio', 'abr': 256, 'preference': -50},
 
+        # Dash webm audio with opus inside
+        '249': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 50, 'preference': -50},
+        '250': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 70, 'preference': -50},
+        '251': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 160, 'preference': -50},
+
         # RTMP (unnamed)
         '_rtmp': {'protocol': 'rtmp'},
     }
@@ -401,6 +406,19 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor):
                 'format': '141',
             },
         },
+        # Controversy video
+        {
+            'url': 'https://www.youtube.com/watch?v=T4XJQO3qol8',
+            'info_dict': {
+                'id': 'T4XJQO3qol8',
+                'ext': 'mp4',
+                'upload_date': '20100909',
+                'uploader': 'The Amazing Atheist',
+                'uploader_id': 'TheAmazingAtheist',
+                'title': 'Burning Everyone\'s Koran',
+                'description': 'SUBSCRIBE: http://www.youtube.com/saturninefilms\n\nEven Obama has taken a stand against freedom on this issue: http://www.huffingtonpost.com/2010/09/09/obama-gma-interview-quran_n_710282.html',
+            }
+        }
     ]
 
     def __init__(self, *args, **kwargs):
@@ -661,7 +679,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor):
         video_id = self.extract_id(url)
 
         # Get video webpage
-        url = proto + '://www.youtube.com/watch?v=%s&gl=US&hl=en&has_verified=1' % video_id
+        url = proto + '://www.youtube.com/watch?v=%s&gl=US&hl=en&has_verified=1&bpctr=9999999999' % video_id
         pref_cookies = [
             c for c in self._downloader.cookiejar
             if c.domain == '.youtube.com' and c.name == 'PREF']
@@ -991,7 +1009,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor):
                         existing_format.update(f)
 
             except (ExtractorError, KeyError) as e:
-                self.report_warning('Skipping DASH manifest: %s' % e, video_id)
+                self.report_warning('Skipping DASH manifest: %r' % e, video_id)
 
         self._sort_formats(formats)
 

youtube_dl/jsinterp.py

@@ -61,7 +61,7 @@ class JSInterpreter(object):
             pass
 
         m = re.match(
-            r'^(?P<var>[a-zA-Z0-9_]+)\.(?P<member>[^(]+)(?:\(+(?P<args>[^()]*)\))?$',
+            r'^(?P<var>[$a-zA-Z0-9_]+)\.(?P<member>[^(]+)(?:\(+(?P<args>[^()]*)\))?$',
             expr)
         if m:
             variable = m.group('var')

youtube_dl/options.py

@@ -8,6 +8,7 @@ import sys
 from .compat import (
     compat_expanduser,
     compat_getenv,
+    compat_kwargs,
 )
 from .utils import (
     get_term_width,
@@ -112,7 +113,7 @@ def parseOpts(overrideArguments=None):
         'conflict_handler': 'resolve',
     }
 
-    parser = optparse.OptionParser(**kw)
+    parser = optparse.OptionParser(**compat_kwargs(kw))
 
     general = optparse.OptionGroup(parser, 'General Options')
     general.add_option(

youtube_dl/swfinterp.py

@@ -62,15 +62,17 @@ class _ScopeDict(dict):
 
 
 class _AVMClass(object):
-    def __init__(self, name_idx, name):
+    def __init__(self, name_idx, name, static_properties=None):
         self.name_idx = name_idx
         self.name = name
         self.method_names = {}
         self.method_idxs = {}
         self.methods = {}
         self.method_pyfunctions = {}
+        self.static_properties = static_properties if static_properties else {}
 
         self.variables = _ScopeDict(self)
+        self.constants = {}
 
     def make_object(self):
         return _AVMClass_Object(self)
@@ -148,8 +150,38 @@ def _read_byte(reader):
     return res
 
 
+StringClass = _AVMClass('(no name idx)', 'String')
+ByteArrayClass = _AVMClass('(no name idx)', 'ByteArray')
+TimerClass = _AVMClass('(no name idx)', 'Timer')
+TimerEventClass = _AVMClass('(no name idx)', 'TimerEvent', {'TIMER': 'timer'})
+_builtin_classes = {
+    StringClass.name: StringClass,
+    ByteArrayClass.name: ByteArrayClass,
+    TimerClass.name: TimerClass,
+    TimerEventClass.name: TimerEventClass,
+}
+
+
+class _Undefined(object):
+    def __bool__(self):
+        return False
+    __nonzero__ = __bool__
+
+    def __hash__(self):
+        return 0
+
+    def __str__(self):
+        return 'undefined'
+    __repr__ = __str__
+
+undefined = _Undefined()
+
+
 class SWFInterpreter(object):
     def __init__(self, file_contents):
+        self._patched_functions = {
+            (TimerClass, 'addEventListener'): lambda params: undefined,
+        }
         code_tag = next(tag
                         for tag_code, tag in _extract_tags(file_contents)
                         if tag_code == 82)
@@ -170,11 +202,13 @@ class SWFInterpreter(object):
 
         # Constant pool
         int_count = u30()
+        self.constant_ints = [0]
         for _c in range(1, int_count):
-            s32()
+            self.constant_ints.append(s32())
+        self.constant_uints = [0]
         uint_count = u30()
         for _c in range(1, uint_count):
-            u32()
+            self.constant_uints.append(u32())
         double_count = u30()
         read_bytes(max(0, (double_count - 1)) * 8)
         string_count = u30()
@@ -212,6 +246,10 @@ class SWFInterpreter(object):
                 u30()  # namespace_idx
                 name_idx = u30()
                 self.multinames.append(self.constant_strings[name_idx])
+            elif kind == 0x09:
+                name_idx = u30()
+                u30()
+                self.multinames.append(self.constant_strings[name_idx])
             else:
                 self.multinames.append(_Multiname(kind))
                 for _c2 in range(MULTINAME_SIZES[kind]):
@@ -258,13 +296,28 @@ class SWFInterpreter(object):
             kind = kind_full & 0x0f
             attrs = kind_full >> 4
             methods = {}
-            if kind in [0x00, 0x06]:  # Slot or Const
+            constants = None
+            if kind == 0x00:  # Slot
                 u30()  # Slot id
                 u30()  # type_name_idx
                 vindex = u30()
                 if vindex != 0:
                     read_byte()  # vkind
-            elif kind in [0x01, 0x02, 0x03]:  # Method / Getter / Setter
+            elif kind == 0x06:  # Const
+                u30()  # Slot id
+                u30()  # type_name_idx
+                vindex = u30()
+                vkind = 'any'
+                if vindex != 0:
+                    vkind = read_byte()
+                if vkind == 0x03:  # Constant_Int
+                    value = self.constant_ints[vindex]
+                elif vkind == 0x04:  # Constant_UInt
+                    value = self.constant_uints[vindex]
+                else:
+                    return {}, None  # Ignore silently for now
+                constants = {self.multinames[trait_name_idx]: value}
+            elif kind in (0x01, 0x02, 0x03):  # Method / Getter / Setter
                 u30()  # disp_id
                 method_idx = u30()
                 methods[self.multinames[trait_name_idx]] = method_idx
@@ -283,7 +336,7 @@ class SWFInterpreter(object):
                 for _c3 in range(metadata_count):
                     u30()  # metadata index
 
-            return methods
+            return methods, constants
 
         # Classes
         class_count = u30()
@@ -305,18 +358,22 @@ class SWFInterpreter(object):
             u30()  # iinit
             trait_count = u30()
             for _c2 in range(trait_count):
-                trait_methods = parse_traits_info()
+                trait_methods, trait_constants = parse_traits_info()
                 avm_class.register_methods(trait_methods)
+                if trait_constants:
+                    avm_class.constants.update(trait_constants)
 
         assert len(classes) == class_count
         self._classes_by_name = dict((c.name, c) for c in classes)
 
         for avm_class in classes:
-            u30()  # cinit
+            avm_class.cinit_idx = u30()
             trait_count = u30()
             for _c2 in range(trait_count):
-                trait_methods = parse_traits_info()
+                trait_methods, trait_constants = parse_traits_info()
                 avm_class.register_methods(trait_methods)
+                if trait_constants:
+                    avm_class.constants.update(trait_constants)
 
         # Scripts
         script_count = u30()
@@ -329,6 +386,7 @@ class SWFInterpreter(object):
         # Method bodies
         method_body_count = u30()
         Method = collections.namedtuple('Method', ['code', 'local_count'])
+        self._all_methods = []
         for _c in range(method_body_count):
             method_idx = u30()
             u30()  # max_stack
@@ -337,9 +395,10 @@ class SWFInterpreter(object):
             u30()  # max_scope_depth
             code_length = u30()
             code = read_bytes(code_length)
+            m = Method(code, local_count)
+            self._all_methods.append(m)
             for avm_class in classes:
                 if method_idx in avm_class.method_idxs:
-                    m = Method(code, local_count)
                     avm_class.methods[avm_class.method_idxs[method_idx]] = m
             exception_count = u30()
             for _c2 in range(exception_count):
@@ -354,13 +413,27 @@ class SWFInterpreter(object):
 
         assert p + code_reader.tell() == len(code_tag)
 
-    def extract_class(self, class_name):
+    def patch_function(self, avm_class, func_name, f):
+        self._patched_functions[(avm_class, func_name)] = f
+
+    def extract_class(self, class_name, call_cinit=True):
         try:
-            return self._classes_by_name[class_name]
+            res = self._classes_by_name[class_name]
         except KeyError:
             raise ExtractorError('Class %r not found' % class_name)
 
+        if call_cinit and hasattr(res, 'cinit_idx'):
+            res.register_methods({'$cinit': res.cinit_idx})
+            res.methods['$cinit'] = self._all_methods[res.cinit_idx]
+            cinit = self.extract_function(res, '$cinit')
+            cinit([])
+
+        return res
+
     def extract_function(self, avm_class, func_name):
+        p = self._patched_functions.get((avm_class, func_name))
+        if p:
+            return p
         if func_name in avm_class.method_pyfunctions:
             return avm_class.method_pyfunctions[func_name]
         if func_name in self._classes_by_name:
@@ -379,10 +452,15 @@ class SWFInterpreter(object):
             registers = [avm_class.variables] + list(args) + [None] * m.local_count
             stack = []
             scopes = collections.deque([
-                self._classes_by_name, avm_class.variables])
+                self._classes_by_name, avm_class.constants, avm_class.variables])
             while True:
                 opcode = _read_byte(coder)
-                if opcode == 17:  # iftrue
+                if opcode == 9:  # label
+                    pass  # Spec says: "Do nothing."
+                elif opcode == 16:  # jump
+                    offset = s24()
+                    coder.seek(coder.tell() + offset)
+                elif opcode == 17:  # iftrue
                     offset = s24()
                     value = stack.pop()
                     if value:
@@ -392,9 +470,40 @@ class SWFInterpreter(object):
                     value = stack.pop()
                     if not value:
                         coder.seek(coder.tell() + offset)
+                elif opcode == 19:  # ifeq
+                    offset = s24()
+                    value2 = stack.pop()
+                    value1 = stack.pop()
+                    if value2 == value1:
+                        coder.seek(coder.tell() + offset)
+                elif opcode == 20:  # ifne
+                    offset = s24()
+                    value2 = stack.pop()
+                    value1 = stack.pop()
+                    if value2 != value1:
+                        coder.seek(coder.tell() + offset)
+                elif opcode == 21:  # iflt
+                    offset = s24()
+                    value2 = stack.pop()
+                    value1 = stack.pop()
+                    if value1 < value2:
+                        coder.seek(coder.tell() + offset)
+                elif opcode == 32:  # pushnull
+                    stack.append(None)
+                elif opcode == 33:  # pushundefined
+                    stack.append(undefined)
                 elif opcode == 36:  # pushbyte
                     v = _read_byte(coder)
                     stack.append(v)
+                elif opcode == 37:  # pushshort
+                    v = u30()
+                    stack.append(v)
+                elif opcode == 38:  # pushtrue
+                    stack.append(True)
+                elif opcode == 39:  # pushfalse
+                    stack.append(False)
+                elif opcode == 40:  # pushnan
+                    stack.append(float('NaN'))
                 elif opcode == 42:  # dup
                     value = stack[-1]
                     stack.append(value)
@@ -419,11 +528,31 @@ class SWFInterpreter(object):
                         [stack.pop() for _ in range(arg_count)]))
                     obj = stack.pop()
 
-                    if isinstance(obj, _AVMClass_Object):
+                    if obj == StringClass:
+                        if mname == 'String':
+                            assert len(args) == 1
+                            assert isinstance(args[0], (
+                                int, compat_str, _Undefined))
+                            if args[0] == undefined:
+                                res = 'undefined'
+                            else:
+                                res = compat_str(args[0])
+                            stack.append(res)
+                            continue
+                        else:
+                            raise NotImplementedError(
+                                'Function String.%s is not yet implemented'
+                                % mname)
+                    elif isinstance(obj, _AVMClass_Object):
                         func = self.extract_function(obj.avm_class, mname)
                         res = func(args)
                         stack.append(res)
                         continue
+                    elif isinstance(obj, _AVMClass):
+                        func = self.extract_function(obj, mname)
+                        res = func(args)
+                        stack.append(res)
+                        continue
                     elif isinstance(obj, _ScopeDict):
                         if mname in obj.avm_class.method_names:
                             func = self.extract_function(obj.avm_class, mname)
@@ -442,6 +571,13 @@ class SWFInterpreter(object):
                                 res = obj.split(args[0])
                             stack.append(res)
                             continue
+                        elif mname == 'charCodeAt':
+                            assert len(args) <= 1
+                            idx = 0 if len(args) == 0 else args[0]
+                            assert isinstance(idx, int)
+                            res = ord(obj[idx])
+                            stack.append(res)
+                            continue
                     elif isinstance(obj, list):
                         if mname == 'slice':
                             assert len(args) == 1
@@ -458,9 +594,18 @@ class SWFInterpreter(object):
                     raise NotImplementedError(
                         'Unsupported property %r on %r'
                         % (mname, obj))
+                elif opcode == 71:  # returnvoid
+                    res = undefined
+                    return res
                 elif opcode == 72:  # returnvalue
                     res = stack.pop()
                     return res
+                elif opcode == 73:  # constructsuper
+                    # Not yet implemented, just hope it works without it
+                    arg_count = u30()
+                    args = list(reversed(
+                        [stack.pop() for _ in range(arg_count)]))
+                    obj = stack.pop()
                 elif opcode == 74:  # constructproperty
                     index = u30()
                     arg_count = u30()
@@ -481,6 +626,17 @@ class SWFInterpreter(object):
                     args = list(reversed(
                         [stack.pop() for _ in range(arg_count)]))
                     obj = stack.pop()
+                    if isinstance(obj, _AVMClass_Object):
+                        func = self.extract_function(obj.avm_class, mname)
+                        res = func(args)
+                        assert res is undefined
+                        continue
+                    if isinstance(obj, _ScopeDict):
+                        assert mname in obj.avm_class.method_names
+                        func = self.extract_function(obj.avm_class, mname)
+                        res = func(args)
+                        assert res is undefined
+                        continue
                     if mname == 'reverse':
                         assert isinstance(obj, list)
                         obj.reverse()
@@ -504,7 +660,10 @@ class SWFInterpreter(object):
                             break
                     else:
                         res = scopes[0]
-                    stack.append(res[mname])
+                    if mname not in res and mname in _builtin_classes:
+                        stack.append(_builtin_classes[mname])
+                    else:
+                        stack.append(res[mname])
                 elif opcode == 94:  # findproperty
                     index = u30()
                     mname = self.multinames[index]
@@ -524,9 +683,15 @@ class SWFInterpreter(object):
                             break
                     else:
                         scope = avm_class.variables
-                    # I cannot find where static variables are initialized
+
-                    # so let's just return None
+                    if mname in scope:
-                    res = scope.get(mname)
+                        res = scope[mname]
+                    elif mname in _builtin_classes:
+                        res = _builtin_classes[mname]
+                    else:
+                        # Assume unitialized
+                        # TODO warn here
+                        res = undefined
                     stack.append(res)
                 elif opcode == 97:  # setproperty
                     index = u30()
@@ -548,22 +713,57 @@ class SWFInterpreter(object):
                     pname = self.multinames[index]
                     if pname == 'length':
                         obj = stack.pop()
-                        assert isinstance(obj, list)
+                        assert isinstance(obj, (compat_str, list))
                         stack.append(len(obj))
+                    elif isinstance(pname, compat_str):  # Member access
+                        obj = stack.pop()
+                        if isinstance(obj, _AVMClass):
+                            res = obj.static_properties[pname]
+                            stack.append(res)
+                            continue
+
+                        assert isinstance(obj, (dict, _ScopeDict)),\
+                            'Accessing member %r on %r' % (pname, obj)
+                        res = obj.get(pname, undefined)
+                        stack.append(res)
                     else:  # Assume attribute access
                         idx = stack.pop()
                         assert isinstance(idx, int)
                         obj = stack.pop()
                         assert isinstance(obj, list)
                         stack.append(obj[idx])
+                elif opcode == 104:  # initproperty
+                    index = u30()
+                    value = stack.pop()
+                    idx = self.multinames[index]
+                    if isinstance(idx, _Multiname):
+                        idx = stack.pop()
+                    obj = stack.pop()
+                    obj[idx] = value
                 elif opcode == 115:  # convert_
                     value = stack.pop()
                     intvalue = int(value)
                     stack.append(intvalue)
                 elif opcode == 128:  # coerce
                     u30()
+                elif opcode == 130:  # coerce_a
+                    value = stack.pop()
+                    # um, yes, it's any value
+                    stack.append(value)
                 elif opcode == 133:  # coerce_s
                     assert isinstance(stack[-1], (type(None), compat_str))
+                elif opcode == 147:  # decrement
+                    value = stack.pop()
+                    assert isinstance(value, int)
+                    stack.append(value - 1)
+                elif opcode == 149:  # typeof
+                    value = stack.pop()
+                    return {
+                        _Undefined: 'undefined',
+                        compat_str: 'String',
+                        int: 'Number',
+                        float: 'Number',
+                    }[type(value)]
                 elif opcode == 160:  # add
                     value2 = stack.pop()
                     value1 = stack.pop()
@@ -574,16 +774,37 @@ class SWFInterpreter(object):
                     value1 = stack.pop()
                     res = value1 - value2
                     stack.append(res)
+                elif opcode == 162:  # multiply
+                    value2 = stack.pop()
+                    value1 = stack.pop()
+                    res = value1 * value2
+                    stack.append(res)
                 elif opcode == 164:  # modulo
                     value2 = stack.pop()
                     value1 = stack.pop()
                     res = value1 % value2
                     stack.append(res)
+                elif opcode == 168:  # bitand
+                    value2 = stack.pop()
+                    value1 = stack.pop()
+                    assert isinstance(value1, int)
+                    assert isinstance(value2, int)
+                    res = value1 & value2
+                    stack.append(res)
+                elif opcode == 171:  # equals
+                    value2 = stack.pop()
+                    value1 = stack.pop()
+                    result = value1 == value2
+                    stack.append(result)
                 elif opcode == 175:  # greaterequals
                     value2 = stack.pop()
                     value1 = stack.pop()
                     result = value1 >= value2
                     stack.append(result)
+                elif opcode == 192:  # increment_i
+                    value = stack.pop()
+                    assert isinstance(value, int)
+                    stack.append(value + 1)
                 elif opcode == 208:  # getlocal_0
                     stack.append(registers[0])
                 elif opcode == 209:  # getlocal_1

youtube_dl/utils.py

@@ -63,7 +63,7 @@ def preferredencoding():
     """
     try:
         pref = locale.getpreferredencoding()
-        u'TEST'.encode(pref)
+        'TEST'.encode(pref)
     except:
         pref = 'UTF-8'
 
@@ -71,12 +71,25 @@ def preferredencoding():
 
 
 def write_json_file(obj, fn):
-    """ Encode obj as JSON and write it to fn, atomically """
+    """ Encode obj as JSON and write it to fn, atomically if possible """
+
+    fn = encodeFilename(fn)
+    if sys.version_info < (3, 0) and sys.platform != 'win32':
+        encoding = get_filesystem_encoding()
+        # os.path.basename returns a bytes object, but NamedTemporaryFile
+        # will fail if the filename contains non ascii characters unless we
+        # use a unicode object
+        path_basename = lambda f: os.path.basename(fn).decode(encoding)
+        # the same for os.path.dirname
+        path_dirname = lambda f: os.path.dirname(fn).decode(encoding)
+    else:
+        path_basename = os.path.basename
+        path_dirname = os.path.dirname
 
     args = {
         'suffix': '.tmp',
-        'prefix': os.path.basename(fn) + '.',
+        'prefix': path_basename(fn) + '.',
-        'dir': os.path.dirname(fn),
+        'dir': path_dirname(fn),
         'delete': False,
     }
 
@@ -95,6 +108,13 @@ def write_json_file(obj, fn):
     try:
         with tf:
             json.dump(obj, tf)
+        if sys.platform == 'win32':
+            # Need to remove existing file on Windows, else os.rename raises
+            # WindowsError or FileExistsError.
+            try:
+                os.unlink(fn)
+            except OSError:
+                pass
         os.rename(tf.name, fn)
     except:
         try:
@@ -203,7 +223,7 @@ def sanitize_open(filename, open_mode):
     It returns the tuple (stream, definitive_file_name).
     """
     try:
-        if filename == u'-':
+        if filename == '-':
             if sys.platform == 'win32':
                 import msvcrt
                 msvcrt.setmode(sys.stdout.fileno(), os.O_BINARY)
@@ -216,7 +236,7 @@ def sanitize_open(filename, open_mode):
 
         # In case of error, try to remove win32 forbidden chars
         alt_filename = os.path.join(
-                        re.sub(u'[/<>:"\\|\\\\?\\*]', u'#', path_part)
+                        re.sub('[/<>:"\\|\\\\?\\*]', '#', path_part)
                         for path_part in os.path.split(filename)
                        )
         if alt_filename == filename:
@@ -255,7 +275,7 @@ def sanitize_filename(s, restricted=False, is_id=False):
             return '_'
         return char
 
-    result = u''.join(map(replace_insane, s))
+    result = ''.join(map(replace_insane, s))
     if not is_id:
         while '__' in result:
             result = result.replace('__', '_')
@@ -285,15 +305,15 @@ def _htmlentity_transform(entity):
     mobj = re.match(r'#(x?[0-9]+)', entity)
     if mobj is not None:
         numstr = mobj.group(1)
-        if numstr.startswith(u'x'):
+        if numstr.startswith('x'):
             base = 16
-            numstr = u'0%s' % numstr
+            numstr = '0%s' % numstr
         else:
             base = 10
         return compat_chr(int(numstr, base))
 
     # Unknown entity in name, return its literal representation
-    return (u'&%s;' % entity)
+    return ('&%s;' % entity)
 
 
 def unescapeHTML(s):
@@ -317,7 +337,7 @@ def encodeFilename(s, for_subprocess=False):
         return s
 
     if sys.platform == 'win32' and sys.getwindowsversion()[0] >= 5:
-        # Pass u'' directly to use Unicode APIs on Windows 2000 and up
+        # Pass '' directly to use Unicode APIs on Windows 2000 and up
         # (Detecting Windows NT 4 is tricky because 'major >= 4' would
         # match Windows 9x series as well. Besides, NT 4 is obsolete.)
         if not for_subprocess:
@@ -400,6 +420,7 @@ def make_HTTPS_handler(opts_no_check_certificate, **kwargs):
             pass  # Python < 3.4
         return compat_urllib_request.HTTPSHandler(context=context, **kwargs)
 
+
 class ExtractorError(Exception):
     """Error during info extraction."""
     def __init__(self, msg, tb=None, expected=False, cause=None, video_id=None):
@@ -412,9 +433,15 @@ class ExtractorError(Exception):
         if video_id is not None:
             msg = video_id + ': ' + msg
         if cause:
-            msg += u' (caused by %r)' % cause
+            msg += ' (caused by %r)' % cause
         if not expected:
-            msg = msg + u'; please report this issue on https://yt-dl.org/bug . Be sure to call youtube-dl with the --verbose flag and include its complete output. Make sure you are using the latest version; type  youtube-dl -U  to update.'
+            if ytdl_is_updateable():
+                update_cmd = 'type  youtube-dl -U  to update'
+            else:
+                update_cmd = 'see  https://yt-dl.org/update  on how to update'
+            msg += '; please report this issue on https://yt-dl.org/bug .'
+            msg += ' Make sure you are using the latest version; %s.' % update_cmd
+            msg += ' Be sure to call youtube-dl with the --verbose flag and include its complete output.'
         super(ExtractorError, self).__init__(msg)
 
         self.traceback = tb
@@ -425,7 +452,7 @@ class ExtractorError(Exception):
     def format_traceback(self):
         if self.traceback is None:
             return None
-        return u''.join(traceback.format_tb(self.traceback))
+        return ''.join(traceback.format_tb(self.traceback))
 
 
 class RegexNotFoundError(ExtractorError):
@@ -653,17 +680,17 @@ def unified_strdate(date_str):
             upload_date = datetime.datetime(*timetuple[:6]).strftime('%Y%m%d')
     return upload_date
 
-def determine_ext(url, default_ext=u'unknown_video'):
+def determine_ext(url, default_ext='unknown_video'):
     if url is None:
         return default_ext
-    guess = url.partition(u'?')[0].rpartition(u'.')[2]
+    guess = url.partition('?')[0].rpartition('.')[2]
     if re.match(r'^[A-Za-z0-9]+$', guess):
         return guess
     else:
         return default_ext
 
 def subtitles_filename(filename, sub_lang, sub_format):
-    return filename.rsplit('.', 1)[0] + u'.' + sub_lang + u'.' + sub_format
+    return filename.rsplit('.', 1)[0] + '.' + sub_lang + '.' + sub_format
 
 def date_from_str(date_str):
     """
@@ -843,7 +870,7 @@ def bytes_to_intlist(bs):
 def intlist_to_bytes(xs):
     if not xs:
         return b''
-    return struct.pack('%dB' % len(xs), *xs)
+    return struct_pack('%dB' % len(xs), *xs)
 
 
 # Cross-platform file locking
@@ -955,7 +982,7 @@ def shell_quote(args):
             # We may get a filename encoded with 'encodeFilename'
             a = a.decode(encoding)
         quoted_args.append(pipes.quote(a))
-    return u' '.join(quoted_args)
+    return ' '.join(quoted_args)
 
 
 def takewhile_inclusive(pred, seq):
@@ -971,31 +998,31 @@ def smuggle_url(url, data):
     """ Pass additional data in a URL for internal use. """
 
     sdata = compat_urllib_parse.urlencode(
-        {u'__youtubedl_smuggle': json.dumps(data)})
+        {'__youtubedl_smuggle': json.dumps(data)})
-    return url + u'#' + sdata
+    return url + '#' + sdata
 
 
 def unsmuggle_url(smug_url, default=None):
     if not '#__youtubedl_smuggle' in smug_url:
         return smug_url, default
-    url, _, sdata = smug_url.rpartition(u'#')
+    url, _, sdata = smug_url.rpartition('#')
-    jsond = compat_parse_qs(sdata)[u'__youtubedl_smuggle'][0]
+    jsond = compat_parse_qs(sdata)['__youtubedl_smuggle'][0]
     data = json.loads(jsond)
     return url, data
 
 
 def format_bytes(bytes):
     if bytes is None:
-        return u'N/A'
+        return 'N/A'
     if type(bytes) is str:
         bytes = float(bytes)
     if bytes == 0.0:
         exponent = 0
     else:
         exponent = int(math.log(bytes, 1024.0))
-    suffix = [u'B', u'KiB', u'MiB', u'GiB', u'TiB', u'PiB', u'EiB', u'ZiB', u'YiB'][exponent]
+    suffix = ['B', 'KiB', 'MiB', 'GiB', 'TiB', 'PiB', 'EiB', 'ZiB', 'YiB'][exponent]
     converted = float(bytes) / float(1024 ** exponent)
-    return u'%.2f%s' % (converted, suffix)
+    return '%.2f%s' % (converted, suffix)
 
 
 def get_term_width():
@@ -1018,8 +1045,8 @@ def month_by_name(name):
     """ Return the number of a month by (locale-independently) English name """
 
     ENGLISH_NAMES = [
-        u'January', u'February', u'March', u'April', u'May', u'June',
+        'January', 'February', 'March', 'April', 'May', 'June',
-        u'July', u'August', u'September', u'October', u'November', u'December']
+        'July', 'August', 'September', 'October', 'November', 'December']
     try:
         return ENGLISH_NAMES.index(name) + 1
     except ValueError:
@@ -1030,7 +1057,7 @@ def fix_xml_ampersands(xml_str):
     """Replace all the '&' by '&amp;' in XML"""
     return re.sub(
         r'&(?!amp;|lt;|gt;|apos;|quot;|#x[0-9a-fA-F]{,4};|#[0-9]{,4};)',
-        u'&amp;',
+        '&amp;',
         xml_str)
 
 
@@ -1063,7 +1090,7 @@ def remove_end(s, end):
 
 def url_basename(url):
     path = compat_urlparse.urlparse(url).path
-    return path.strip(u'/').split(u'/')[-1]
+    return path.strip('/').split('/')[-1]
 
 
 class HEADRequest(compat_urllib_request.Request):
@@ -1088,7 +1115,7 @@ def str_to_int(int_str):
     """ A more relaxed version of int_or_none """
     if int_str is None:
         return None
-    int_str = re.sub(r'[,\.\+]', u'', int_str)
+    int_str = re.sub(r'[,\.\+]', '', int_str)
     return int(int_str)
 
 
@@ -1103,7 +1130,12 @@ def parse_duration(s):
     s = s.strip()
 
     m = re.match(
-        r'(?i)(?:(?:(?P<hours>[0-9]+)\s*(?:[:h]|hours?)\s*)?(?P<mins>[0-9]+)\s*(?:[:m]|mins?|minutes?)\s*)?(?P<secs>[0-9]+)(?P<ms>\.[0-9]+)?\s*(?:s|secs?|seconds?)?$', s)
+        r'''(?ix)T?
+            (?:
+                (?:(?P<hours>[0-9]+)\s*(?:[:h]|hours?)\s*)?
+                (?P<mins>[0-9]+)\s*(?:[:m]|mins?|minutes?)\s*
+            )?
+            (?P<secs>[0-9]+)(?P<ms>\.[0-9]+)?\s*(?:s|secs?|seconds?)?$''', s)
     if not m:
         return None
     res = int(m.group('secs'))
@@ -1118,7 +1150,7 @@ def parse_duration(s):
 
 def prepend_extension(filename, ext):
     name, real_ext = os.path.splitext(filename) 
-    return u'{0}.{1}{2}'.format(name, ext, real_ext)
+    return '{0}.{1}{2}'.format(name, ext, real_ext)
 
 
 def check_executable(exe, args=[]):
@@ -1133,7 +1165,7 @@ def check_executable(exe, args=[]):
 
 def get_exe_version(exe, args=['--version'],
                     version_re=r'version\s+([0-9._-a-zA-Z]+)',
-                    unrecognized=u'present'):
+                    unrecognized='present'):
     """ Returns the version of the specified executable,
     or False if the executable is not present """
     try:
@@ -1254,7 +1286,7 @@ def escape_url(url):
     ).geturl()
 
 try:
-    struct.pack(u'!I', 0)
+    struct.pack('!I', 0)
 except TypeError:
     # In Python 2.6 (and some 2.7 versions), struct requires a bytes argument
     def struct_pack(spec, *args):
@@ -1275,7 +1307,7 @@ def read_batch_urls(batch_fd):
     def fixup(url):
         if not isinstance(url, compat_str):
             url = url.decode('utf-8', 'replace')
-        BOM_UTF8 = u'\xef\xbb\xbf'
+        BOM_UTF8 = '\xef\xbb\xbf'
         if url.startswith(BOM_UTF8):
             url = url[len(BOM_UTF8):]
         url = url.strip()
@@ -1331,7 +1363,8 @@ def parse_age_limit(s):
 
 
 def strip_jsonp(code):
-    return re.sub(r'(?s)^[a-zA-Z0-9_]+\s*\(\s*(.*)\);?\s*?\s*$', r'\1', code)
+    return re.sub(
+        r'(?s)^[a-zA-Z0-9_]+\s*\(\s*(.*)\);?\s*?(?://[^\n]*)*$', r'\1', code)
 
 
 def js_to_json(code):
@@ -1393,3 +1426,10 @@ def is_outdated_version(version, limit, assume_new=True):
         return version_tuple(version) < version_tuple(limit)
     except ValueError:
         return not assume_new
+
+
+def ytdl_is_updateable():
+    """ Returns if youtube-dl can be updated with -U """
+    from zipimport import zipimporter
+
+    return isinstance(globals().get('__loader__'), zipimporter) or hasattr(sys, 'frozen')

youtube_dl/version.py

@@ -1,2 +1,2 @@
 
-__version__ = '2014.11.13.1'
+__version__ = '2014.11.23'