release 2014.11.23

Update jsinterp.py

AUTHORS

@@ -80,3 +80,6 @@ Damon Timm
 winwon
 Xavier Beynon
 Gabriel Schubiner
+xantares
+Jan Matějka
+Mauroy Sébastien

README.md

@@ -131,17 +131,19 @@ which means you can modify it, redistribute it or use it however you like.
                                      %(upload_date)s for the upload date
                                      (YYYYMMDD), %(extractor)s for the provider
                                      (youtube, metacafe, etc), %(id)s for the
-                                     video id, %(playlist)s for the playlist the
+                                     video id, %(playlist_title)s,
+                                     %(playlist_id)s, or %(playlist)s (=title if
+                                     present, ID otherwise) for the playlist the
                                      video is in, %(playlist_index)s for the
-                                     position in the playlist and %% for a
-                                     literal percent. %(height)s and %(width)s
-                                     for the width and height of the video
-                                     format. %(resolution)s for a textual
+                                     position in the playlist. %(height)s and
+                                     %(width)s for the width and height of the
+                                     video format. %(resolution)s for a textual
                                      description of the resolution of the video
-                                     format. Use - to output to stdout. Can also
-                                     be used to download to a different
-                                     directory, for example with -o '/my/downloa
-                                     ds/%(uploader)s/%(title)s-%(id)s.%(ext)s' .
+                                     format. %% for a literal percent. Use - to
+                                     output to stdout. Can also be used to
+                                     download to a different directory, for
+                                     example with -o '/my/downloads/%(uploader)s
+                                     /%(title)s-%(id)s.%(ext)s' .
     --autonumber-size NUMBER         Specifies the number of digits in
                                      %(autonumber)s when it is present in output
                                      filename template or --auto-number option
@@ -239,8 +241,13 @@ which means you can modify it, redistribute it or use it however you like.
                                      "worst", "worstvideo" and "worstaudio". By
                                      default, youtube-dl will pick the best
                                      quality. Use commas to download multiple
-                                     audio formats, such as  -f
-                                     136/137/mp4/bestvideo,140/m4a/bestaudio
+                                     audio formats, such as -f
+                                     136/137/mp4/bestvideo,140/m4a/bestaudio.
+                                     You can merge the video and audio of two
+                                     formats into a single file using -f <video-
+                                     format>+<audio-format> (requires ffmpeg or
+                                     avconv), for example -f
+                                     bestvideo+bestaudio.
     --all-formats                    download all available video formats
     --prefer-free-formats            prefer free video formats unless a specific
                                      one is requested
@@ -500,7 +507,7 @@ If you want to add support for a new site, you can follow this quick list (assum
 6. Run `python test/test_download.py TestDownload.test_YourExtractor`. This *should fail* at first, but you can continually re-run it until you're done. If you decide to add more than one test, then rename ``_TEST`` to ``_TESTS`` and make it into a list of dictionaries. The tests will be then be named `TestDownload.test_YourExtractor`, `TestDownload.test_YourExtractor_1`, `TestDownload.test_YourExtractor_2`, etc.
 7. Have a look at [`youtube_dl/common/extractor/common.py`](https://github.com/rg3/youtube-dl/blob/master/youtube_dl/extractor/common.py) for possible helper methods and a [detailed description of what your extractor should return](https://github.com/rg3/youtube-dl/blob/master/youtube_dl/extractor/common.py#L38). Add tests and code for as many as you want.
 8. If you can, check the code with [pyflakes](https://pypi.python.org/pypi/pyflakes) (a good idea) and [pep8](https://pypi.python.org/pypi/pep8) (optional, ignore E501).
-9. When the tests pass, [add](https://www.kernel.org/pub/software/scm/git/docs/git-add.html) the new files and [commit](https://www.kernel.org/pub/software/scm/git/docs/git-commit.html) them and [push](https://www.kernel.org/pub/software/scm/git/docs/git-push.html) the result, like this:
+9. When the tests pass, [add](http://git-scm.com/docs/git-add) the new files and [commit](http://git-scm.com/docs/git-commit) them and [push](http://git-scm.com/docs/git-push) the result, like this:
 
         $ git add youtube_dl/extractor/__init__.py
         $ git add youtube_dl/extractor/yourextractor.py

test/helper.py

@@ -57,7 +57,7 @@ class FakeYDL(YoutubeDL):
         # Different instances of the downloader can't share the same dictionary
         # some test set the "sublang" parameter, which would break the md5 checks.
         params = get_params(override=override)
-        super(FakeYDL, self).__init__(params)
+        super(FakeYDL, self).__init__(params, auto_init=False)
         self.result = []
         
     def to_screen(self, s, skip_eol=None):
@@ -145,7 +145,8 @@ def expect_info_dict(self, expected_dict, got_dict):
         info_dict_str = ''.join(
             '    %s: %s,\n' % (_repr(k), _repr(v))
             for k, v in test_info_dict.items())
-        write_string('\n"info_dict": {\n' + info_dict_str + '}\n', out=sys.stderr)
+        write_string(
+            '\n\'info_dict\': {\n' + info_dict_str + '}\n', out=sys.stderr)
         self.assertFalse(
             missing_keys,
             'Missing keys in test definition: %s' % (

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

@@ -0,0 +1,46 @@
+#!/usr/bin/env python
+# coding: utf-8
+
+from __future__ import unicode_literals
+
+# Allow direct execution
+import os
+import sys
+import unittest
+sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
+
+
+from youtube_dl.utils import get_filesystem_encoding
+from youtube_dl.compat import (
+    compat_getenv,
+    compat_expanduser,
+)
+
+
+class TestCompat(unittest.TestCase):
+    def test_compat_getenv(self):
+        test_str = 'тест'
+        os.environ['YOUTUBE-DL-TEST'] = (
+            test_str if sys.version_info >= (3, 0)
+            else test_str.encode(get_filesystem_encoding()))
+        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
+        all_names = youtube_dl.compat.__all__
+        present_names = set(filter(
+            lambda c: '_' in c and not c.startswith('_'),
+            dir(youtube_dl.compat))) - set(['unicode_literals'])
+        self.assertEqual(all_names, sorted(present_names))
+
+if __name__ == '__main__':
+    unittest.main()

test/test_download.py

@@ -1,5 +1,7 @@
 #!/usr/bin/env python
 
+from __future__ import unicode_literals
+
 # Allow direct execution
 import os
 import sys
@@ -23,10 +25,12 @@ import json
 import socket
 
 import youtube_dl.YoutubeDL
-from youtube_dl.utils import (
+from youtube_dl.compat import (
     compat_http_client,
     compat_urllib_error,
     compat_HTTPError,
+)
+from youtube_dl.utils import (
     DownloadError,
     ExtractorError,
     format_bytes,
@@ -94,7 +98,7 @@ def generator(test_case):
             params.setdefault('extract_flat', True)
             params.setdefault('skip_download', True)
 
-        ydl = YoutubeDL(params)
+        ydl = YoutubeDL(params, auto_init=False)
         ydl.add_default_info_extractors()
         finished_hook_called = set()
         def _hook(status):
@@ -208,9 +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
@@ -37,7 +38,9 @@ def _make_testfunc(testfile):
                 or os.path.getmtime(swf_file) < os.path.getmtime(as_file)):
             # Recompile
             try:
-                subprocess.check_call(['mxmlc', '-output', swf_file, as_file])
+                subprocess.check_call([
+                    'mxmlc', '-output', swf_file,
+                    '-static-link-runtime-shared-libraries', as_file])
             except OSError as ose:
                 if ose.errno == errno.ENOENT:
                     print('mxmlc not found! Skipping test.')

test/test_utils.py

@@ -16,11 +16,11 @@ import json
 import xml.etree.ElementTree
 
 from youtube_dl.utils import (
+    clean_html,
     DateRange,
     encodeFilename,
     find_xpath_attr,
     fix_xml_ampersands,
-    get_meta_content,
     orderedSet,
     OnDemandPagedList,
     InAdvancePagedList,
@@ -46,8 +46,7 @@ from youtube_dl.utils import (
     escape_url,
     js_to_json,
     get_filesystem_encoding,
-    compat_getenv,
-    compat_expanduser,
+    intlist_to_bytes,
 )
 
 
@@ -157,17 +156,6 @@ class TestUtil(unittest.TestCase):
         self.assertEqual(find_xpath_attr(doc, './/node', 'x', 'a'), doc[1])
         self.assertEqual(find_xpath_attr(doc, './/node', 'y', 'c'), doc[2])
 
-    def test_meta_parser(self):
-        testhtml = '''
-        <head>
-            <meta name="description" content="foo &amp; bar">
-            <meta content='Plato' name='author'/>
-        </head>
-        '''
-        get_meta = lambda name: get_meta_content(name, testhtml)
-        self.assertEqual(get_meta('description'), 'foo & bar')
-        self.assertEqual(get_meta('author'), 'Plato')
-
     def test_xpath_with_ns(self):
         testxml = '''<root xmlns:media="http://example.com/">
             <media:song>
@@ -230,6 +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(
@@ -296,6 +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')
@@ -359,17 +352,14 @@ class TestUtil(unittest.TestCase):
         on = js_to_json('{"abc": true}')
         self.assertEqual(json.loads(on), {'abc': True})
 
-    def test_compat_getenv(self):
-        test_str = 'тест'
-        os.environ['YOUTUBE-DL-TEST'] = (test_str if sys.version_info >= (3, 0)
-            else test_str.encode(get_filesystem_encoding()))
-        self.assertEqual(compat_getenv('YOUTUBE-DL-TEST'), test_str)
+    def test_clean_html(self):
+        self.assertEqual(clean_html('a:\nb'), 'a: b')
+        self.assertEqual(clean_html('a:\n   "b"'), 'a:    "b"')
 
-    def test_compat_expanduser(self):
-        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)
+    def test_intlist_to_bytes(self):
+        self.assertEqual(
+            intlist_to_bytes([0, 1, 127, 128, 255]),
+            b'\x00\x01\x7f\x80\xff')
 
 if __name__ == '__main__':
     unittest.main()

test/test_youtube_signature.py

@@ -14,7 +14,7 @@ import re
 import string
 
 from youtube_dl.extractor import YoutubeIE
-from youtube_dl.utils import compat_str, compat_urlretrieve
+from youtube_dl.compat import compat_str, compat_urlretrieve
 
 _TESTS = [
     (

youtube_dl/YoutubeDL.py

@@ -22,13 +22,15 @@ import traceback
 if os.name == 'nt':
     import ctypes
 
-from .utils import (
+from .compat import (
     compat_cookiejar,
     compat_expanduser,
     compat_http_client,
     compat_str,
     compat_urllib_error,
     compat_urllib_request,
+)
+from .utils import (
     escape_url,
     ContentTooShortError,
     date_from_str,
@@ -62,6 +64,7 @@ from .utils import (
 from .cache import Cache
 from .extractor import get_info_extractor, gen_extractors
 from .downloader import get_suitable_downloader
+from .downloader.rtmp import rtmpdump_version
 from .postprocessor import FFmpegMergerPP, FFmpegPostProcessor
 from .version import __version__
 
@@ -621,7 +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)
@@ -655,6 +658,8 @@ class YoutubeDL(object):
                 extra = {
                     'n_entries': n_entries,
                     'playlist': playlist,
+                    'playlist_id': ie_result.get('id'),
+                    'playlist_title': ie_result.get('title'),
                     'playlist_index': i + playliststart,
                     'extractor': ie_result['extractor'],
                     'webpage_url': ie_result['webpage_url'],
@@ -674,6 +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,
                     {
@@ -833,6 +841,13 @@ class YoutubeDL(object):
                         formats_info = (self.select_format(format_1, formats),
                             self.select_format(format_2, formats))
                         if all(formats_info):
+                            # The first format must contain the video and the
+                            # second the audio
+                            if formats_info[0].get('vcodec') == 'none':
+                                self.report_error('The first format must '
+                                    'contain the video, try using '
+                                    '"-f %s+%s"' % (format_2, format_1))
+                                return
                             selected_format = {
                                 'requested_formats': formats_info,
                                 'format': rf,
@@ -989,7 +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
@@ -1294,11 +1309,13 @@ class YoutubeDL(object):
             self.report_warning(
                 'Your Python is broken! Update to a newer and supported version')
 
+        stdout_encoding = getattr(
+            sys.stdout, 'encoding', 'missing (%s)' % type(sys.stdout).__name__)
         encoding_str = (
             '[debug] Encodings: locale %s, fs %s, out %s, pref %s\n' % (
                 locale.getpreferredencoding(),
                 sys.getfilesystemencoding(),
-                sys.stdout.encoding,
+                stdout_encoding,
                 self.get_encoding()))
         write_string(encoding_str, encoding=None)
 
@@ -1321,6 +1338,7 @@ class YoutubeDL(object):
             platform.python_version(), platform_name()))
 
         exe_versions = FFmpegPostProcessor.get_versions()
+        exe_versions['rtmpdump'] = rtmpdump_version()
         exe_str = ', '.join(
             '%s %s' % (exe, v)
             for exe, v in sorted(exe_versions.items())

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
@@ -13,10 +15,13 @@ import sys
 from .options import (
     parseOpts,
 )
-from .utils import (
+from .compat import (
     compat_expanduser,
     compat_getpass,
     compat_print,
+    workaround_optparse_bug9161,
+)
+from .utils import (
     DateRange,
     DEFAULT_OUTTMPL,
     decodeOption,
@@ -53,7 +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)
 
@@ -69,10 +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
@@ -90,9 +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()
@@ -105,7 +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()):
@@ -115,63 +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')
-                _COUNTS = (u'', u'5', u'10', u'all')
-                desc += u' (Example: "%s%s:%s" )' % (ie.SEARCH_KEY, random.choice(_COUNTS), random.choice(_SEARCHES))
+                _SEARCHES = ('cute kittens', 'slithering pythons', 'falling cat', 'angry poodle', 'purple fish', 'running tortoise', 'sleeping bunny')
+                _COUNTS = ('', '5', '10', 'all')
+                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:
@@ -191,17 +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 u'%(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 u'%(title)s-%(id)s.%(ext)s')
-            or (opts.useid and u'%(id)s.%(ext)s')
-            or (opts.autonumber and u'%(autonumber)s-%(id)s.%(ext)s')
+            or (opts.format == '-1' and opts.usetitle and '%(title)s-%(id)s-%(format)s.%(ext)s')
+            or (opts.format == '-1' and '%(id)s-%(format)s.%(ext)s')
+            or (opts.usetitle and opts.autonumber and '%(autonumber)s-%(title)s-%(id)s.%(ext)s')
+            or (opts.usetitle and '%(title)s-%(id)s.%(ext)s')
+            or (opts.useid and '%(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'
-                     u' file! Use "{0}.%(ext)s" instead of "{0}" as the output'
-                     u' template'.format(outtmpl))
+        parser.error('Cannot download a video and extract audio into the same'
+                     ' file! Use "{0}.%(ext)s" instead of "{0}" as the output'
+                     ' 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
@@ -328,7 +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()
 
@@ -338,7 +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)
@@ -350,6 +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

@@ -8,10 +8,8 @@ import re
 import shutil
 import traceback
 
-from .utils import (
-    compat_expanduser,
-    write_json_file,
-)
+from .compat import compat_expanduser, compat_getenv
+from .utils import write_json_file
 
 
 class Cache(object):
@@ -21,7 +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

@@ -0,0 +1,350 @@
+from __future__ import unicode_literals
+
+import getpass
+import optparse
+import os
+import subprocess
+import sys
+
+
+try:
+    import urllib.request as compat_urllib_request
+except ImportError: # Python 2
+    import urllib2 as compat_urllib_request
+
+try:
+    import urllib.error as compat_urllib_error
+except ImportError: # Python 2
+    import urllib2 as compat_urllib_error
+
+try:
+    import urllib.parse as compat_urllib_parse
+except ImportError: # Python 2
+    import urllib as compat_urllib_parse
+
+try:
+    from urllib.parse import urlparse as compat_urllib_parse_urlparse
+except ImportError: # Python 2
+    from urlparse import urlparse as compat_urllib_parse_urlparse
+
+try:
+    import urllib.parse as compat_urlparse
+except ImportError: # Python 2
+    import urlparse as compat_urlparse
+
+try:
+    import http.cookiejar as compat_cookiejar
+except ImportError: # Python 2
+    import cookielib as compat_cookiejar
+
+try:
+    import html.entities as compat_html_entities
+except ImportError: # Python 2
+    import htmlentitydefs as compat_html_entities
+
+try:
+    import html.parser as compat_html_parser
+except ImportError: # Python 2
+    import HTMLParser as compat_html_parser
+
+try:
+    import http.client as compat_http_client
+except ImportError: # Python 2
+    import httplib as compat_http_client
+
+try:
+    from urllib.error import HTTPError as compat_HTTPError
+except ImportError:  # Python 2
+    from urllib2 import HTTPError as compat_HTTPError
+
+try:
+    from urllib.request import urlretrieve as compat_urlretrieve
+except ImportError:  # Python 2
+    from urllib import urlretrieve as compat_urlretrieve
+
+
+try:
+    from subprocess import DEVNULL
+    compat_subprocess_get_DEVNULL = lambda: DEVNULL
+except ImportError:
+    compat_subprocess_get_DEVNULL = lambda: open(os.path.devnull, 'w')
+
+try:
+    from urllib.parse import unquote as compat_urllib_parse_unquote
+except ImportError:
+    def compat_urllib_parse_unquote(string, encoding='utf-8', errors='replace'):
+        if string == '':
+            return string
+        res = string.split('%')
+        if len(res) == 1:
+            return string
+        if encoding is None:
+            encoding = 'utf-8'
+        if errors is None:
+            errors = 'replace'
+        # pct_sequence: contiguous sequence of percent-encoded bytes, decoded
+        pct_sequence = b''
+        string = res[0]
+        for item in res[1:]:
+            try:
+                if not item:
+                    raise ValueError
+                pct_sequence += item[:2].decode('hex')
+                rest = item[2:]
+                if not rest:
+                    # This segment was just a single percent-encoded character.
+                    # May be part of a sequence of code units, so delay decoding.
+                    # (Stored in pct_sequence).
+                    continue
+            except ValueError:
+                rest = '%' + item
+            # Encountered non-percent-encoded characters. Flush the current
+            # pct_sequence.
+            string += pct_sequence.decode(encoding, errors) + rest
+            pct_sequence = b''
+        if pct_sequence:
+            # Flush the final pct_sequence
+            string += pct_sequence.decode(encoding, errors)
+        return string
+
+
+try:
+    from urllib.parse import parse_qs as compat_parse_qs
+except ImportError: # Python 2
+    # HACK: The following is the correct parse_qs implementation from cpython 3's stdlib.
+    # Python 2's version is apparently totally broken
+
+    def _parse_qsl(qs, keep_blank_values=False, strict_parsing=False,
+                encoding='utf-8', errors='replace'):
+        qs, _coerce_result = qs, unicode
+        pairs = [s2 for s1 in qs.split('&') for s2 in s1.split(';')]
+        r = []
+        for name_value in pairs:
+            if not name_value and not strict_parsing:
+                continue
+            nv = name_value.split('=', 1)
+            if len(nv) != 2:
+                if strict_parsing:
+                    raise ValueError("bad query field: %r" % (name_value,))
+                # Handle case of a control-name with no equal sign
+                if keep_blank_values:
+                    nv.append('')
+                else:
+                    continue
+            if len(nv[1]) or keep_blank_values:
+                name = nv[0].replace('+', ' ')
+                name = compat_urllib_parse_unquote(
+                    name, encoding=encoding, errors=errors)
+                name = _coerce_result(name)
+                value = nv[1].replace('+', ' ')
+                value = compat_urllib_parse_unquote(
+                    value, encoding=encoding, errors=errors)
+                value = _coerce_result(value)
+                r.append((name, value))
+        return r
+
+    def compat_parse_qs(qs, keep_blank_values=False, strict_parsing=False,
+                encoding='utf-8', errors='replace'):
+        parsed_result = {}
+        pairs = _parse_qsl(qs, keep_blank_values, strict_parsing,
+                        encoding=encoding, errors=errors)
+        for name, value in pairs:
+            if name in parsed_result:
+                parsed_result[name].append(value)
+            else:
+                parsed_result[name] = [value]
+        return parsed_result
+
+try:
+    compat_str = unicode # Python 2
+except NameError:
+    compat_str = str
+
+try:
+    compat_chr = unichr # Python 2
+except NameError:
+    compat_chr = chr
+
+try:
+    from xml.etree.ElementTree import ParseError as compat_xml_parse_error
+except ImportError:  # Python 2.6
+    from xml.parsers.expat import ExpatError as compat_xml_parse_error
+
+try:
+    from shlex import quote as shlex_quote
+except ImportError:  # Python < 3.3
+    def shlex_quote(s):
+        return "'" + s.replace("'", "'\"'\"'") + "'"
+
+
+def compat_ord(c):
+    if type(c) is int: return c
+    else: return ord(c)
+
+
+if sys.version_info >= (3, 0):
+    compat_getenv = os.getenv
+    compat_expanduser = os.path.expanduser
+else:
+    # Environment variables should be decoded with filesystem encoding.
+    # Otherwise it will fail if any non-ASCII characters present (see #3854 #3217 #2918)
+
+    def compat_getenv(key, default=None):
+        from .utils import get_filesystem_encoding
+        env = os.getenv(key, default)
+        if env:
+            env = env.decode(get_filesystem_encoding())
+        return env
+
+    # HACK: The default implementations of os.path.expanduser from cpython do not decode
+    # environment variables with filesystem encoding. We will work around this by
+    # providing adjusted implementations.
+    # The following are os.path.expanduser implementations from cpython 2.7.8 stdlib
+    # for different platforms with correct environment variables decoding.
+
+    if os.name == 'posix':
+        def compat_expanduser(path):
+            """Expand ~ and ~user constructions.  If user or $HOME is unknown,
+            do nothing."""
+            if not path.startswith('~'):
+                return path
+            i = path.find('/', 1)
+            if i < 0:
+                i = len(path)
+            if i == 1:
+                if 'HOME' not in os.environ:
+                    import pwd
+                    userhome = pwd.getpwuid(os.getuid()).pw_dir
+                else:
+                    userhome = compat_getenv('HOME')
+            else:
+                import pwd
+                try:
+                    pwent = pwd.getpwnam(path[1:i])
+                except KeyError:
+                    return path
+                userhome = pwent.pw_dir
+            userhome = userhome.rstrip('/')
+            return (userhome + path[i:]) or '/'
+    elif os.name == 'nt' or os.name == 'ce':
+        def compat_expanduser(path):
+            """Expand ~ and ~user constructs.
+
+            If user or $HOME is unknown, do nothing."""
+            if path[:1] != '~':
+                return path
+            i, n = 1, len(path)
+            while i < n and path[i] not in '/\\':
+                i = i + 1
+
+            if 'HOME' in os.environ:
+                userhome = compat_getenv('HOME')
+            elif 'USERPROFILE' in os.environ:
+                userhome = compat_getenv('USERPROFILE')
+            elif not 'HOMEPATH' in os.environ:
+                return path
+            else:
+                try:
+                    drive = compat_getenv('HOMEDRIVE')
+                except KeyError:
+                    drive = ''
+                userhome = os.path.join(drive, compat_getenv('HOMEPATH'))
+
+            if i != 1: #~user
+                userhome = os.path.join(os.path.dirname(userhome), path[1:i])
+
+            return userhome + path[i:]
+    else:
+        compat_expanduser = os.path.expanduser
+
+
+if sys.version_info < (3, 0):
+    def compat_print(s):
+        from .utils import preferredencoding
+        print(s.encode(preferredencoding(), 'xmlcharrefreplace'))
+else:
+    def compat_print(s):
+        assert type(s) == type(u'')
+        print(s)
+
+
+try:
+    subprocess_check_output = subprocess.check_output
+except AttributeError:
+    def subprocess_check_output(*args, **kwargs):
+        assert 'input' not in kwargs
+        p = subprocess.Popen(*args, stdout=subprocess.PIPE, **kwargs)
+        output, _ = p.communicate()
+        ret = p.poll()
+        if ret:
+            raise subprocess.CalledProcessError(ret, p.args, output=output)
+        return output
+
+if sys.version_info < (3, 0) and sys.platform == 'win32':
+    def compat_getpass(prompt, *args, **kwargs):
+        if isinstance(prompt, compat_str):
+            from .utils import preferredencoding
+            prompt = prompt.encode(preferredencoding())
+        return getpass.getpass(prompt, *args, **kwargs)
+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',
+    'compat_chr',
+    'compat_cookiejar',
+    'compat_expanduser',
+    'compat_getenv',
+    'compat_getpass',
+    'compat_html_entities',
+    'compat_html_parser',
+    'compat_http_client',
+    'compat_kwargs',
+    'compat_ord',
+    'compat_parse_qs',
+    'compat_print',
+    'compat_str',
+    'compat_subprocess_get_DEVNULL',
+    'compat_urllib_error',
+    'compat_urllib_parse',
+    'compat_urllib_parse_unquote',
+    'compat_urllib_parse_urlparse',
+    'compat_urllib_request',
+    'compat_urlparse',
+    'compat_urlretrieve',
+    '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'):
-            return filename[:-len(u'.part')]
+        if filename.endswith('.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/downloader/rtmp.py

@@ -12,9 +12,15 @@ from ..utils import (
     compat_str,
     encodeFilename,
     format_bytes,
+    get_exe_version,
 )
 
 
+def rtmpdump_version():
+    return get_exe_version(
+        'rtmpdump', ['--help'], r'(?i)RTMPDump\s*v?([0-9a-zA-Z._-]+)')
+
+
 class RtmpFD(FileDownloader):
     def real_download(self, filename, info_dict):
         def run_rtmpdump(args):

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 (
@@ -141,6 +143,7 @@ from .generic import GenericIE
 from .glide import GlideIE
 from .globo import GloboIE
 from .godtube import GodTubeIE
+from .goldenmoustache import GoldenMoustacheIE
 from .golem import GolemIE
 from .googleplus import GooglePlusIE
 from .googlesearch import GoogleSearchIE
@@ -323,6 +326,7 @@ from .sbs import SBSIE
 from .scivee import SciVeeIE
 from .screencast import ScreencastIE
 from .servingsys import ServingSysIE
+from .sexu import SexuIE
 from .sexykarma import SexyKarmaIE
 from .shared import SharedIE
 from .sharesix import ShareSixIE
@@ -376,6 +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
@@ -421,6 +426,7 @@ from .vesti import VestiIE
 from .vevo import VevoIE
 from .vgtv import VGTVIE
 from .vh1 import VH1IE
+from .vice import ViceIE
 from .viddler import ViddlerIE
 from .videobam import VideoBamIE
 from .videodetective import VideoDetectiveIE

youtube_dl/extractor/abc.py

@@ -11,13 +11,13 @@ class ABCIE(InfoExtractor):
     _VALID_URL = r'http://www\.abc\.net\.au/news/[^/]+/[^/]+/(?P<id>\d+)'
 
     _TEST = {
-        'url': 'http://www.abc.net.au/news/2014-07-25/bringing-asylum-seekers-to-australia-would-give/5624716',
-        'md5': 'dad6f8ad011a70d9ddf887ce6d5d0742',
+        'url': 'http://www.abc.net.au/news/2014-11-05/australia-to-staff-ebola-treatment-centre-in-sierra-leone/5868334',
+        'md5': 'cb3dd03b18455a661071ee1e28344d9f',
         'info_dict': {
-            'id': '5624716',
+            'id': '5868334',
             'ext': 'mp4',
-            'title': 'Bringing asylum seekers to Australia would give them right to asylum claims: professor',
-            'description': 'md5:ba36fa5e27e5c9251fd929d339aea4af',
+            'title': 'Australia to help staff Ebola treatment centre in Sierra Leone',
+            'description': 'md5:809ad29c67a05f54eb41f2a105693a67',
         },
     }
 

youtube_dl/extractor/addanime.py

@@ -3,12 +3,13 @@ from __future__ import unicode_literals
 import re
 
 from .common import InfoExtractor
-from ..utils import (
+from ..compat import (
     compat_HTTPError,
     compat_str,
     compat_urllib_parse,
     compat_urllib_parse_urlparse,
-
+)
+from ..utils import (
     ExtractorError,
 )
 

youtube_dl/extractor/allocine.py

@@ -22,7 +22,7 @@ class AllocineIE(InfoExtractor):
             'id': '19546517',
             'ext': 'mp4',
             'title': 'Astérix - Le Domaine des Dieux Teaser VF',
-            'description': 'md5:4a754271d9c6f16c72629a8a993ee884',
+            'description': 'md5:abcd09ce503c6560512c14ebfdb720d2',
             'thumbnail': 're:http://.*\.jpg',
         },
     }, {

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)
-            fmt['format_id'] = format_id
-            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')
+            f = dict(format_dict)
+            versionCode = f.get('versionCode')
 
-        if re.match(r'[A-Z]Q', formats[0]['quality']) is not None:
-            def sort_key(f):
-                return ['HQ', 'MQ', 'EQ', 'SQ'].index(f['quality'])
-        else:
-            def sort_key(f):
-                versionCode = f.get('versionCode')
-                if versionCode is None:
-                    versionCode = ''
-                return (
-                    # Sort first by quality
-                    int(f.get('height', -1)),
-                    int(f.get('bitrate', -1)),
-                    # The original version with subtitles has lower relevance
-                    re.match(r'VO-ST(F|A)', versionCode) is None,
-                    # The version with sourds/mal subtitles has also lower relevance
-                    re.match(r'VO?(F|A)-STM\1', versionCode) is None,
-                    # Prefer http downloads over m3u8
-                    0 if f['url'].endswith('m3u8') else 1,
-                )
-        formats = sorted(formats, key=sort_key)
-        def _format(format_info):
-            info = {
-                'format_id': format_info['format_id'],
-                'format_note': '%s, %s' % (format_info.get('versionCode'), format_info.get('versionLibelle')),
-                'width': int_or_none(format_info.get('width')),
-                'height': int_or_none(format_info.get('height')),
-                'tbr': int_or_none(format_info.get('bitrate')),
+            langcode = {
+                'fr': 'F',
+                'de': 'A',
+            }.get(lang, lang)
+            lang_rexs = [r'VO?%s' % langcode, r'VO?.-ST%s' % langcode]
+            lang_pref = (
+                None if versionCode is None else (
+                    10 if any(re.match(r, versionCode) for r in lang_rexs)
+                    else -10))
+            source_pref = 0
+            if versionCode is not None:
+                # The original version with subtitles has lower relevance
+                if re.match(r'VO-ST(F|A)', versionCode):
+                    source_pref -= 10
+                # The version with sourds/mal subtitles has also lower relevance
+                elif re.match(r'VO?(F|A)-STM\1', versionCode):
+                    source_pref -= 9
+            format = {
+                'format_id': format_id,
+                'preference': -10 if f.get('videoFormat') == 'M3U8' else None,
+                'language_preference': lang_pref,
+                'format_note': '%s, %s' % (f.get('versionCode'), f.get('versionLibelle')),
+                'width': int_or_none(f.get('width')),
+                'height': int_or_none(f.get('height')),
+                'tbr': int_or_none(f.get('bitrate')),
+                'quality': qfunc(f['quality']),
+                '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/bandcamp.py

@@ -110,20 +110,25 @@ class BandcampAlbumIE(InfoExtractor):
         'url': 'http://blazo.bandcamp.com/album/jazz-format-mixtape-vol-1',
         'playlist': [
             {
-                'file': '1353101989.mp3',
                 'md5': '39bc1eded3476e927c724321ddf116cf',
                 'info_dict': {
+                    'id': '1353101989',
+                    'ext': 'mp3',
                     'title': 'Intro',
                 }
             },
             {
-                'file': '38097443.mp3',
                 'md5': '1a2c32e2691474643e912cc6cd4bffaa',
                 'info_dict': {
+                    'id': '38097443',
+                    'ext': 'mp3',
                     'title': 'Kero One - Keep It Alive (Blazo remix)',
                 }
             },
         ],
+        'info_dict': {
+            'title': 'Jazz Format Mixtape vol.1',
+        },
         'params': {
             'playlistend': 2
         },

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

@@ -14,6 +14,7 @@ from ..utils import (
     compat_str,
     compat_urllib_request,
     compat_parse_qs,
+    compat_urllib_parse_urlparse,
 
     determine_ext,
     ExtractorError,
@@ -23,7 +24,7 @@ from ..utils import (
 
 
 class BrightcoveIE(InfoExtractor):
-    _VALID_URL = r'https?://.*brightcove\.com/(services|viewer).*\?(?P<query>.*)'
+    _VALID_URL = r'https?://.*brightcove\.com/(services|viewer).*?\?(?P<query>.*)'
     _FEDERATED_URL_TEMPLATE = 'http://c.brightcove.com/services/viewer/htmlFederated?%s'
 
     _TESTS = [
@@ -110,6 +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'))
@@ -218,7 +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(
@@ -260,11 +263,19 @@ class BrightcoveIE(InfoExtractor):
             formats = []
             for rend in renditions:
                 url = rend['defaultURL']
+                if not url:
+                    continue
                 if rend['remote']:
-                    # This type of renditions are served through akamaihd.net,
-                    # but they don't use f4m manifests
-                    url = url.replace('control/', '') + '?&v=3.3.0&fp=13&r=FEEFJ&g=RTSJIMBMPFPB'
-                    ext = 'flv'
+                    url_comp = compat_urllib_parse_urlparse(url)
+                    if url_comp.path.endswith('.m3u8'):
+                        formats.extend(
+                            self._extract_m3u8_formats(url, info['id'], 'mp4'))
+                        continue
+                    elif 'akamaihd.net' in url_comp.netloc:
+                        # This type of renditions are served through
+                        # akamaihd.net, but they don't use f4m manifests
+                        url = url.replace('control/', '') + '?&v=3.3.0&fp=13&r=FEEFJ&g=RTSJIMBMPFPB'
+                        ext = 'flv'
                 else:
                     ext = determine_ext(url)
                 size = rend.get('size')

youtube_dl/extractor/byutv.py

@@ -10,12 +10,12 @@ from ..utils import ExtractorError
 class BYUtvIE(InfoExtractor):
     _VALID_URL = r'^https?://(?:www\.)?byutv.org/watch/[0-9a-f-]+/(?P<video_id>[^/?#]+)'
     _TEST = {
-        'url': 'http://www.byutv.org/watch/44e80f7b-e3ba-43ba-8c51-b1fd96c94a79/granite-flats-talking',
+        'url': 'http://www.byutv.org/watch/6587b9a3-89d2-42a6-a7f7-fd2f81840a7d/studio-c-season-5-episode-5',
         'info_dict': {
-            'id': 'granite-flats-talking',
+            'id': 'studio-c-season-5-episode-5',
             'ext': 'mp4',
-            'description': 'md5:4e9a7ce60f209a33eca0ac65b4918e1c',
-            'title': 'Talking',
+            'description': 'md5:5438d33774b6bdc662f9485a340401cc',
+            'title': 'Season 5 Episode 5',
             'thumbnail': 're:^https?://.*promo.*'
         },
         'params': {

youtube_dl/extractor/channel9.py

@@ -27,7 +27,7 @@ class Channel9IE(InfoExtractor):
                 'title': 'Developer Kick-Off Session: Stuff We Love',
                 'description': 'md5:c08d72240b7c87fcecafe2692f80e35f',
                 'duration': 4576,
-                'thumbnail': 'http://media.ch9.ms/ch9/9d51/03902f2d-fc97-4d3c-b195-0bfe15a19d51/KOS002_220.jpg',
+                'thumbnail': 'http://video.ch9.ms/ch9/9d51/03902f2d-fc97-4d3c-b195-0bfe15a19d51/KOS002_220.jpg',
                 'session_code': 'KOS002',
                 'session_day': 'Day 1',
                 'session_room': 'Arena 1A',
@@ -43,7 +43,7 @@ class Channel9IE(InfoExtractor):
                 'title': 'Self-service BI with Power BI - nuclear testing',
                 'description': 'md5:d1e6ecaafa7fb52a2cacdf9599829f5b',
                 'duration': 1540,
-                'thumbnail': 'http://media.ch9.ms/ch9/87e1/0300391f-a455-4c72-bec3-4422f19287e1/selfservicenuk_512.jpg',
+                'thumbnail': 'http://video.ch9.ms/ch9/87e1/0300391f-a455-4c72-bec3-4422f19287e1/selfservicenuk_512.jpg',
                 'authors': [ 'Mike Wilmot' ],
             },
         }
@@ -115,7 +115,7 @@ class Channel9IE(InfoExtractor):
         return self._html_search_meta('description', html, 'description')
 
     def _extract_duration(self, html):
-        m = re.search(r'data-video_duration="(?P<hours>\d{2}):(?P<minutes>\d{2}):(?P<seconds>\d{2})"', html)
+        m = re.search(r'"length": *"(?P<hours>\d{2}):(?P<minutes>\d{2}):(?P<seconds>\d{2})"', html)
         return ((int(m.group('hours')) * 60 * 60) + (int(m.group('minutes')) * 60) + int(m.group('seconds'))) if m else None
 
     def _extract_slides(self, html):
@@ -258,16 +258,17 @@ class Channel9IE(InfoExtractor):
 
         webpage = self._download_webpage(url, content_path, 'Downloading web page')
 
-        page_type_m = re.search(r'<meta name="Search.PageType" content="(?P<pagetype>[^"]+)"/>', webpage)
-        if page_type_m is None:
-            raise ExtractorError('Search.PageType not found, don\'t know how to process this page', expected=True)
+        page_type_m = re.search(r'<meta name="WT.entryid" content="(?P<pagetype>[^:]+)[^"]+"/>', webpage)
+        if page_type_m is not None:
+            page_type = page_type_m.group('pagetype')
+            if page_type == 'Entry':      # Any 'item'-like page, may contain downloadable content
+                return self._extract_entry_item(webpage, content_path)
+            elif page_type == 'Session':  # Event session page, may contain downloadable content
+                return self._extract_session(webpage, content_path)
+            elif page_type == 'Event':
+                return self._extract_list(content_path)
+            else:
+                raise ExtractorError('Unexpected WT.entryid %s' % page_type, expected=True)
 
-        page_type = page_type_m.group('pagetype')
-        if page_type == 'List':         # List page, may contain list of 'item'-like objects
+        else: # Assuming list
             return self._extract_list(content_path)
-        elif page_type == 'Entry.Item': # Any 'item'-like page, may contain downloadable content
-            return self._extract_entry_item(webpage, content_path)
-        elif page_type == 'Session':    # Event session page, may contain downloadable content
-            return self._extract_session(webpage, content_path)
-        else:
-            raise ExtractorError('Unexpected Search.PageType %s' % page_type, expected=True)

youtube_dl/extractor/cinemassacre.py

@@ -42,11 +42,12 @@ class CinemassacreIE(InfoExtractor):
 
         webpage = self._download_webpage(url, display_id)
         video_date = mobj.group('date_Y') + mobj.group('date_m') + mobj.group('date_d')
-        mobj = re.search(r'src="(?P<embed_url>http://player\.screenwavemedia\.com/play/[a-zA-Z]+\.php\?[^"]*\bid=(?:Cinemassacre-)?(?P<video_id>.+?))"', webpage)
+        mobj = re.search(r'src="(?P<embed_url>http://player\.screenwavemedia\.com/play/[a-zA-Z]+\.php\?[^"]*\bid=(?P<full_video_id>(?:Cinemassacre-)?(?P<video_id>.+?)))"', webpage)
         if not mobj:
             raise ExtractorError('Can\'t extract embed url and video id')
         playerdata_url = mobj.group('embed_url')
         video_id = mobj.group('video_id')
+        full_video_id = mobj.group('full_video_id')
 
         video_title = self._html_search_regex(
             r'<title>(?P<title>.+?)\|', webpage, 'title')
@@ -60,37 +61,52 @@ class CinemassacreIE(InfoExtractor):
         vidurl = self._search_regex(
             r'\'vidurl\'\s*:\s*"([^\']+)"', playerdata, 'vidurl').replace('\\/', '/')
 
-        videolist_url = self._search_regex(
-            r"file\s*:\s*'(http.+?/jwplayer\.smil)'", playerdata, 'jwplayer.smil')
-        videolist = self._download_xml(videolist_url, video_id, 'Downloading videolist XML')
+        videolist_url = None
 
-        formats = []
-        baseurl = vidurl[:vidurl.rfind('/')+1]
-        for video in videolist.findall('.//video'):
-            src = video.get('src')
-            if not src:
-                continue
-            file_ = src.partition(':')[-1]
-            width = int_or_none(video.get('width'))
-            height = int_or_none(video.get('height'))
-            bitrate = int_or_none(video.get('system-bitrate'))
-            format = {
-                'url': baseurl + file_,
-                'format_id': src.rpartition('.')[0].rpartition('_')[-1],
-            }
-            if width or height:
-                format.update({
-                    'tbr': bitrate // 1000 if bitrate else None,
-                    'width': width,
-                    'height': height,
-                })
-            else:
-                format.update({
-                    'abr': bitrate // 1000 if bitrate else None,
-                    'vcodec': 'none',
-                })
-            formats.append(format)
-        self._sort_formats(formats)
+        mobj = re.search(r"'videoserver'\s*:\s*'(?P<videoserver>[^']+)'", playerdata)
+        if mobj:
+            videoserver = mobj.group('videoserver')
+            mobj = re.search(r'\'vidid\'\s*:\s*"(?P<vidid>[^\']+)"', playerdata)
+            vidid = mobj.group('vidid') if mobj else full_video_id
+            videolist_url = 'http://%s/vod/smil:%s.smil/jwplayer.smil' % (videoserver, vidid)
+        else:
+            mobj = re.search(r"file\s*:\s*'(?P<smil>http.+?/jwplayer\.smil)'", playerdata)
+            if mobj:
+                videolist_url = mobj.group('smil')
+
+        if videolist_url:
+            videolist = self._download_xml(videolist_url, video_id, 'Downloading videolist XML')
+            formats = []
+            baseurl = vidurl[:vidurl.rfind('/')+1]
+            for video in videolist.findall('.//video'):
+                src = video.get('src')
+                if not src:
+                    continue
+                file_ = src.partition(':')[-1]
+                width = int_or_none(video.get('width'))
+                height = int_or_none(video.get('height'))
+                bitrate = int_or_none(video.get('system-bitrate'))
+                format = {
+                    'url': baseurl + file_,
+                    'format_id': src.rpartition('.')[0].rpartition('_')[-1],
+                }
+                if width or height:
+                    format.update({
+                        'tbr': bitrate // 1000 if bitrate else None,
+                        'width': width,
+                        'height': height,
+                    })
+                else:
+                    format.update({
+                        'abr': bitrate // 1000 if bitrate else None,
+                        'vcodec': 'none',
+                    })
+                formats.append(format)
+            self._sort_formats(formats)
+        else:
+            formats = [{
+                'url': vidurl,
+            }]
 
         return {
             'id': video_id,

youtube_dl/extractor/cloudy.py

@@ -4,14 +4,16 @@ from __future__ import unicode_literals
 import re
 
 from .common import InfoExtractor
-from ..utils import (
-    ExtractorError,
+from ..compat import (
     compat_parse_qs,
     compat_urllib_parse,
-    remove_end,
-    HEADRequest,
     compat_HTTPError,
 )
+from ..utils import (
+    ExtractorError,
+    HEADRequest,
+    remove_end,
+)
 
 
 class CloudyIE(InfoExtractor):

youtube_dl/extractor/cnn.py

@@ -16,9 +16,10 @@ class CNNIE(InfoExtractor):
 
     _TESTS = [{
         'url': 'http://edition.cnn.com/video/?/video/sports/2013/06/09/nadal-1-on-1.cnn',
-        'file': 'sports_2013_06_09_nadal-1-on-1.cnn.mp4',
         'md5': '3e6121ea48df7e2259fe73a0628605c4',
         'info_dict': {
+            'id': 'sports_2013_06_09_nadal-1-on-1.cnn',
+            'ext': 'mp4',
             'title': 'Nadal wins 8th French Open title',
             'description': 'World Sport\'s Amanda Davies chats with 2013 French Open champion Rafael Nadal.',
             'duration': 135,
@@ -27,9 +28,10 @@ class CNNIE(InfoExtractor):
     },
     {
         "url": "http://edition.cnn.com/video/?/video/us/2013/08/21/sot-student-gives-epic-speech.georgia-institute-of-technology&utm_source=feedburner&utm_medium=feed&utm_campaign=Feed%3A+rss%2Fcnn_topstories+%28RSS%3A+Top+Stories%29",
-        "file": "us_2013_08_21_sot-student-gives-epic-speech.georgia-institute-of-technology.mp4",
         "md5": "b5cc60c60a3477d185af8f19a2a26f4e",
         "info_dict": {
+            'id': 'us/2013/08/21/sot-student-gives-epic-speech.georgia-institute-of-technology',
+            'ext': 'mp4',
             "title": "Student's epic speech stuns new freshmen",
             "description": "A Georgia Tech student welcomes the incoming freshmen with an epic speech backed by music from \"2001: A Space Odyssey.\"",
             "upload_date": "20130821",

youtube_dl/extractor/comedycentral.py

@@ -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)
-        if mobj is None:
-            raise ExtractorError('Invalid URL: %s' % url)
+        mobj = re.match(self._VALID_URL, 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

@@ -12,13 +12,14 @@ import sys
 import time
 import xml.etree.ElementTree
 
-from ..utils import (
+from ..compat import (
     compat_http_client,
     compat_urllib_error,
     compat_urllib_parse_urlparse,
     compat_urlparse,
     compat_str,
-
+)
+from ..utils import (
     clean_html,
     compiled_regex_type,
     ExtractorError,
@@ -42,7 +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.
@@ -86,6 +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),
@@ -145,6 +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.
@@ -403,7 +445,7 @@ class InfoExtractor(object):
             video_info['title'] = playlist_title
         return video_info
 
-    def _search_regex(self, pattern, string, name, default=_NO_DEFAULT, fatal=True, flags=0):
+    def _search_regex(self, pattern, string, name, default=_NO_DEFAULT, fatal=True, flags=0, group=None):
         """
         Perform a regex search on the given string, using a single or a list of
         patterns returning the first matching group.
@@ -424,8 +466,11 @@ class InfoExtractor(object):
             _name = name
 
         if mobj:
-            # return the first matching group
-            return next(g for g in mobj.groups() if g is not None)
+            if group is None:
+                # return the first matching group
+                return next(g for g in mobj.groups() if g is not None)
+            else:
+                return mobj.group(group)
         elif default is not _NO_DEFAULT:
             return default
         elif fatal:
@@ -435,11 +480,11 @@ class InfoExtractor(object):
                 'please report this issue on http://yt-dl.org/bug' % _name)
             return None
 
-    def _html_search_regex(self, pattern, string, name, default=_NO_DEFAULT, fatal=True, flags=0):
+    def _html_search_regex(self, pattern, string, name, default=_NO_DEFAULT, fatal=True, flags=0, group=None):
         """
         Like _search_regex, but strips HTML tags and unescapes entities.
         """
-        res = self._search_regex(pattern, string, name, default, fatal, flags)
+        res = self._search_regex(pattern, string, name, default, fatal, flags, group)
         if res:
             return clean_html(res).strip()
         else:
@@ -533,9 +578,9 @@ class InfoExtractor(object):
             display_name = name
         return self._html_search_regex(
             r'''(?ix)<meta
-                    (?=[^>]+(?:itemprop|name|property)=["\']?%s["\']?)
-                    [^>]+content=["\']([^"\']+)["\']''' % re.escape(name),
-            html, display_name, fatal=fatal, **kwargs)
+                    (?=[^>]+(?:itemprop|name|property)=(["\']?)%s\1)
+                    [^>]+content=(["\'])(?P<content>.*?)\1''' % re.escape(name),
+            html, display_name, fatal=fatal, group='content', **kwargs)
 
     def _dc_search_uploader(self, html):
         return self._html_search_meta('dc.creator', html, 'uploader')
@@ -611,6 +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

@@ -17,7 +17,6 @@ from ..utils import (
     bytes_to_intlist,
     intlist_to_bytes,
     unified_strdate,
-    clean_html,
     urlencode_postdata,
 )
 from ..aes import (
@@ -265,8 +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/dailymotion.py

@@ -94,7 +94,7 @@ class DailymotionIE(DailymotionBaseInfoExtractor, SubtitlesInfoExtractor):
 
         # It may just embed a vevo video:
         m_vevo = re.search(
-            r'<link rel="video_src" href="[^"]*?vevo.com[^"]*?videoId=(?P<id>[\w]*)',
+            r'<link rel="video_src" href="[^"]*?vevo.com[^"]*?video=(?P<id>[\w]*)',
             webpage)
         if m_vevo is not None:
             vevo_id = m_vevo.group('id')

youtube_dl/extractor/dropbox.py

@@ -5,7 +5,8 @@ import os.path
 import re
 
 from .common import InfoExtractor
-from ..utils import compat_urllib_parse_unquote, url_basename
+from ..compat import compat_urllib_parse_unquote
+from ..utils import url_basename
 
 
 class DropboxIE(InfoExtractor):

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(
-            r'class="mbtim">([0-9:]+)</div>', webpage, 'duration',
-            fatal=False))
+        duration = parse_duration(self._html_search_meta('duration', webpage))
         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/facebook.py

@@ -5,12 +5,14 @@ import re
 import socket
 
 from .common import InfoExtractor
-from ..utils import (
+from ..compat import (
     compat_http_client,
     compat_str,
     compat_urllib_error,
     compat_urllib_parse,
     compat_urllib_request,
+)
+from ..utils import (
     urlencode_postdata,
     ExtractorError,
     limit_length,

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/funnyordie.py

@@ -21,7 +21,6 @@ class FunnyOrDieIE(InfoExtractor):
         },
     }, {
         'url': 'http://www.funnyordie.com/embed/e402820827',
-        'md5': '29f4c5e5a61ca39dfd7e8348a75d0aad',
         'info_dict': {
             'id': 'e402820827',
             'ext': 'mp4',

youtube_dl/extractor/gamespot.py

@@ -8,12 +8,11 @@ from ..utils import (
     compat_urllib_parse,
     compat_urlparse,
     unescapeHTML,
-    get_meta_content,
 )
 
 
 class GameSpotIE(InfoExtractor):
-    _VALID_URL = r'(?:http://)?(?:www\.)?gamespot\.com/.*-(?P<page_id>\d+)/?'
+    _VALID_URL = r'(?:http://)?(?:www\.)?gamespot\.com/.*-(?P<id>\d+)/?'
     _TEST = {
         'url': 'http://www.gamespot.com/videos/arma-3-community-guide-sitrep-i/2300-6410818/',
         'md5': 'b2a30deaa8654fcccd43713a6b6a4825',
@@ -26,10 +25,10 @@ class GameSpotIE(InfoExtractor):
     }
 
     def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
-        page_id = mobj.group('page_id')
+        page_id = self._match_id(url)
         webpage = self._download_webpage(url, page_id)
-        data_video_json = self._search_regex(r'data-video=["\'](.*?)["\']', webpage, 'data video')
+        data_video_json = self._search_regex(
+            r'data-video=["\'](.*?)["\']', webpage, 'data video')
         data_video = json.loads(unescapeHTML(data_video_json))
 
         # Transform the manifest url to a link to the mp4 files
@@ -41,7 +40,8 @@ class GameSpotIE(InfoExtractor):
         http_path = f4m_path[1:].split('/', 1)[1]
         http_template = re.sub(QUALITIES_RE, r'%s', http_path)
         http_template = http_template.replace('.csmil/manifest.f4m', '')
-        http_template = compat_urlparse.urljoin('http://video.gamespotcdn.com/', http_template)
+        http_template = compat_urlparse.urljoin(
+            'http://video.gamespotcdn.com/', http_template)
         formats = []
         for q in qualities:
             formats.append({
@@ -52,8 +52,9 @@ class GameSpotIE(InfoExtractor):
 
         return {
             'id': data_video['guid'],
+            'display_id': page_id,
             'title': compat_urllib_parse.unquote(data_video['title']),
             'formats': formats,
-            'description': get_meta_content('description', webpage),
+            'description': self._html_search_meta('description', webpage),
             'thumbnail': self._og_search_thumbnail(webpage),
         }

youtube_dl/extractor/generic.py

@@ -7,11 +7,12 @@ import re
 
 from .common import InfoExtractor
 from .youtube import YoutubeIE
-from ..utils import (
+from ..compat import (
     compat_urllib_parse,
     compat_urlparse,
     compat_xml_parse_error,
-
+)
+from ..utils import (
     determine_ext,
     ExtractorError,
     float_or_none,
@@ -99,6 +100,22 @@ class GenericIE(InfoExtractor):
                 'uploader': 'Championat',
             },
         },
+        {
+            # https://github.com/rg3/youtube-dl/issues/3541
+            'add_ie': ['Brightcove'],
+            'url': 'http://www.kijk.nl/sbs6/leermijvrouwenkennen/videos/jqMiXKAYan2S/aflevering-1',
+            'info_dict': {
+                'id': '3866516442001',
+                'ext': 'mp4',
+                'title': 'Leer mij vrouwen kennen: Aflevering 1',
+                'description': 'Leer mij vrouwen kennen: Aflevering 1',
+                'uploader': 'SBS Broadcasting',
+            },
+            'skip': 'Restricted to Netherlands',
+            'params': {
+                'skip_download': True,  # m3u8 download
+            },
+        },
         # Direct link to a video
         {
             'url': 'http://media.w3.org/2010/05/sintel/trailer.mp4',
@@ -417,7 +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):
@@ -559,6 +586,7 @@ class GenericIE(InfoExtractor):
             return {
                 'id': video_id,
                 'title': os.path.splitext(url_basename(url))[0],
+                'direct': True,
                 'formats': [{
                     'format_id': m.group('format_id'),
                     'url': url,
@@ -898,6 +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)
@@ -945,7 +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/globo.py

@@ -5,13 +5,15 @@ import random
 import math
 
 from .common import InfoExtractor
-from ..utils import (
-    ExtractorError,
-    float_or_none,
+from ..compat import (
     compat_str,
     compat_chr,
     compat_ord,
 )
+from ..utils import (
+    ExtractorError,
+    float_or_none,
+)
 
 
 class GloboIE(InfoExtractor):

youtube_dl/extractor/goldenmoustache.py

@@ -0,0 +1,57 @@
+from __future__ import unicode_literals
+
+from .common import InfoExtractor
+from ..utils import (
+    int_or_none,
+)
+
+
+class GoldenMoustacheIE(InfoExtractor):
+    _VALID_URL = r'https?://(?:www\.)?goldenmoustache\.com/(?P<display_id>[\w-]+)-(?P<id>\d+)'
+    _TESTS = [{
+        'url': 'http://www.goldenmoustache.com/suricate-le-poker-3700/',
+        'md5': '0f904432fa07da5054d6c8beb5efb51a',
+        'info_dict': {
+            'id': '3700',
+            'ext': 'mp4',
+            'title': 'Suricate - Le Poker',
+            'description': 'md5:3d1f242f44f8c8cb0a106f1fd08e5dc9',
+            '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)
+        webpage = self._download_webpage(url, video_id)
+
+        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')
+        thumbnail = self._og_search_thumbnail(webpage)
+        description = self._og_search_description(webpage)
+        view_count = int_or_none(self._html_search_regex(
+            r'<strong>([0-9]+)</strong>\s*VUES</span>',
+            webpage, 'view count', fatal=False))
+
+        return {
+            'id': video_id,
+            'url': video_url,
+            'ext': 'mp4',
+            'title': title,
+            'description': description,
+            'thumbnail': thumbnail,
+            'view_count': view_count,
+        }

youtube_dl/extractor/goshgay.py

@@ -1,15 +1,11 @@
 # -*- coding: utf-8 -*-
 from __future__ import unicode_literals
 
-import re
-
 from .common import InfoExtractor
 from ..utils import (
     compat_urlparse,
-    str_to_int,
     ExtractorError,
 )
-import json
 
 
 class GoshgayIE(InfoExtractor):
@@ -27,36 +23,27 @@ class GoshgayIE(InfoExtractor):
     }
 
     def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
-        video_id = mobj.group('id')
+        video_id = self._match_id(url)
 
         webpage = self._download_webpage(url, video_id)
-        title = self._search_regex(r'class="video-title"><h1>(.+?)<', webpage, 'title')
+        title = self._og_search_title(webpage)
+        thumbnail = self._og_search_thumbnail(webpage)
+        family_friendly = self._html_search_meta(
+            'isFamilyFriendly', webpage, default='false')
+        config_url = self._search_regex(
+            r"'config'\s*:\s*'([^']+)'", webpage, 'config URL')
 
-        player_config = self._search_regex(
-            r'(?s)jwplayer\("player"\)\.setup\(({.+?})\)', webpage, 'config settings')
-        player_vars = json.loads(player_config.replace("'", '"'))
-        width = str_to_int(player_vars.get('width'))
-        height = str_to_int(player_vars.get('height'))
-        config_uri = player_vars.get('config')
+        config = self._download_xml(
+            config_url, video_id, 'Downloading player config XML')
 
-        if config_uri is None:
-            raise ExtractorError('Missing config URI')
-        node = self._download_xml(config_uri, video_id, 'Downloading player config XML',
-                                  errnote='Unable to download XML')
-        if node is None:
+        if config is None:
             raise ExtractorError('Missing config XML')
-        if node.tag != 'config':
+        if config.tag != 'config':
             raise ExtractorError('Missing config attribute')
-        fns = node.findall('file')
-        imgs = node.findall('image')
-        if len(fns) != 1:
+        fns = config.findall('file')
+        if len(fns) < 1:
             raise ExtractorError('Missing media URI')
         video_url = fns[0].text
-        if len(imgs) < 1:
-            thumbnail = None
-        else:
-            thumbnail = imgs[0].text
 
         url_comp = compat_urlparse.urlparse(url)
         ref = "%s://%s%s" % (url_comp[0], url_comp[1], url_comp[2])
@@ -65,9 +52,7 @@ class GoshgayIE(InfoExtractor):
             'id': video_id,
             'url': video_url,
             'title': title,
-            'width': width,
-            'height': height,
             'thumbnail': thumbnail,
             'http_referer': ref,
-            'age_limit': 18,
+            'age_limit': 0 if family_friendly == 'true' else 18,
         }

youtube_dl/extractor/grooveshark.py

@@ -8,12 +8,13 @@ import re
 
 
 from .common import InfoExtractor
-from ..utils import ExtractorError, compat_urllib_request, compat_html_parser
-
-from ..utils import (
+from ..compat import (
+    compat_html_parser,
     compat_urllib_parse,
+    compat_urllib_request,
     compat_urlparse,
 )
+from ..utils import ExtractorError
 
 
 class GroovesharkHtmlParser(compat_html_parser.HTMLParser):

youtube_dl/extractor/heise.py

@@ -3,7 +3,7 @@ from __future__ import unicode_literals
 
 from .common import InfoExtractor
 from ..utils import (
-    get_meta_content,
+    determine_ext,
     int_or_none,
     parse_iso8601,
 )
@@ -25,11 +25,11 @@ class HeiseIE(InfoExtractor):
             'title': (
                 "Podcast: c't uplink 3.3 – Owncloud / Tastaturen / Peilsender Smartphone"
             ),
-            'format_id': 'mp4_720',
+            'format_id': 'mp4_720p',
             'timestamp': 1411812600,
             'upload_date': '20140927',
             'description': 'In uplink-Episode 3.3 geht es darum, wie man sich von Cloud-Anbietern emanzipieren kann, worauf man beim Kauf einer Tastatur achten sollte und was Smartphones über uns verraten.',
-            'thumbnail': 're:https?://.*\.jpg$',
+            'thumbnail': 're:^https?://.*\.jpe?g$',
         }
     }
 
@@ -49,11 +49,12 @@ class HeiseIE(InfoExtractor):
         info = {
             'id': video_id,
             'thumbnail': self._og_search_thumbnail(webpage),
-            'timestamp': parse_iso8601(get_meta_content('date', webpage)),
+            'timestamp': parse_iso8601(
+                self._html_search_meta('date', webpage)),
             'description': self._og_search_description(webpage),
         }
 
-        title = get_meta_content('fulltitle', webpage)
+        title = self._html_search_meta('fulltitle', webpage)
         if title:
             info['title'] = title
         else:
@@ -64,9 +65,12 @@ class HeiseIE(InfoExtractor):
             label = source_node.attrib['label']
             height = int_or_none(self._search_regex(
                 r'^(.*?_)?([0-9]+)p$', label, 'height', default=None))
+            video_url = source_node.attrib['file']
+            ext = determine_ext(video_url, '')
             formats.append({
-                'url': source_node.attrib['file'],
+                'url': video_url,
                 'format_note': label,
+                'format_id': '%s_%s' % (ext, label),
                 'height': height,
             })
         self._sort_formats(formats)

youtube_dl/extractor/imdb.py

@@ -6,7 +6,6 @@ import json
 from .common import InfoExtractor
 from ..utils import (
     compat_urlparse,
-    get_element_by_attribute,
 )
 
 
@@ -27,10 +26,11 @@ class ImdbIE(InfoExtractor):
     }
 
     def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
-        video_id = mobj.group('id')
+        video_id = self._match_id(url)
         webpage = self._download_webpage('http://www.imdb.com/video/imdb/vi%s' % video_id, video_id)
-        descr = get_element_by_attribute('itemprop', 'description', webpage)
+        descr = self._html_search_regex(
+            r'(?s)<span itemprop="description">(.*?)</span>',
+            webpage, 'description', fatal=False)
         available_formats = re.findall(
             r'case \'(?P<f_id>.*?)\' :$\s+url = \'(?P<path>.*?)\'', webpage,
             flags=re.MULTILINE)
@@ -73,9 +73,7 @@ class ImdbListIE(InfoExtractor):
     }
     
     def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
-        list_id = mobj.group('id')
-
+        list_id = self._match_id(url)
         webpage = self._download_webpage(url, list_id)
         entries = [
             self.url_result('http://www.imdb.com' + m, 'Imdb')

youtube_dl/extractor/izlesene.py

@@ -5,11 +5,11 @@ import re
 
 from .common import InfoExtractor
 from ..utils import (
-    get_element_by_id,
-    parse_iso8601,
     determine_ext,
-    int_or_none,
     float_or_none,
+    get_element_by_id,
+    int_or_none,
+    parse_iso8601,
     str_to_int,
 )
 
@@ -30,7 +30,7 @@ class IzleseneIE(InfoExtractor):
                 'description': 'md5:253753e2655dde93f59f74b572454f6d',
                 'thumbnail': 're:^http://.*\.jpg',
                 'uploader_id': 'pelikzzle',
-                'timestamp': 1404298698,
+                'timestamp': 1404302298,
                 'upload_date': '20140702',
                 'duration': 95.395,
                 'age_limit': 0,
@@ -46,7 +46,7 @@ class IzleseneIE(InfoExtractor):
                 'description': 'Tarkan Dortmund 2006 Konseri',
                 'thumbnail': 're:^http://.*\.jpg',
                 'uploader_id': 'parlayankiz',
-                'timestamp': 1163318593,
+                'timestamp': 1163322193,
                 'upload_date': '20061112',
                 'duration': 253.666,
                 'age_limit': 0,
@@ -55,10 +55,9 @@ class IzleseneIE(InfoExtractor):
     ]
 
     def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
-        video_id = mobj.group('id')
-        url = 'http://www.izlesene.com/video/%s' % video_id
+        video_id = self._match_id(url)
 
+        url = 'http://www.izlesene.com/video/%s' % video_id
         webpage = self._download_webpage(url, video_id)
 
         title = self._og_search_title(webpage)

youtube_dl/extractor/laola1tv.py

@@ -4,6 +4,7 @@ import random
 import re
 
 from .common import InfoExtractor
+from ..utils import ExtractorError
 
 
 class Laola1TvIE(InfoExtractor):

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']
-        content_id = str(movie['contentId'])
-        title = movie['title']
+        meta_data = video_data['meta']
+        content_id = '%s_%s' % (
+            meta_data.get('accId', ''), meta_data['itemId'])
+        title = meta_data['title']
         if title.endswith('.mp4'):
             title = title[:-4]
-        thumbnail = movie['poster']
-        duration = movie['duration']
-
-        view_count = video_data['views_count']
+        thumbnail = meta_data['poster']
+        duration = meta_data['duration']
+        timestamp = meta_data['timestamp']
 
         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

@@ -33,7 +33,7 @@ class MTVServicesInfoExtractor(InfoExtractor):
         m = re.match(r'^rtmpe?://.*?/(?P<finalid>gsp\..+?/.*)$', rtmp_video_url)
         if not m:
             return rtmp_video_url
-        base = 'http://mtvnmobile.vo.llnwd.net/kip0/_pxn=1+_pxI0=Ripod-h264+_pxL0=undefined+_pxM0=+_pxK=18639+_pxE=mp4/44620/mtvnorigin/'
+        base = 'http://viacommtvstrmfs.fplive.net/'
         return base + m.group('finalid')
 
     def _get_feed_url(self, uri):
@@ -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/myspass.py

@@ -13,9 +13,10 @@ class MySpassIE(InfoExtractor):
     _VALID_URL = r'http://www\.myspass\.de/.*'
     _TEST = {
         'url': 'http://www.myspass.de/myspass/shows/tvshows/absolute-mehrheit/Absolute-Mehrheit-vom-17022013-Die-Highlights-Teil-2--/11741/',
-        'file': '11741.mp4',
         'md5': '0b49f4844a068f8b33f4b7c88405862b',
         'info_dict': {
+            'id': '11741',
+            'ext': 'mp4',
             "description": "Wer kann in die Fu\u00dfstapfen von Wolfgang Kubicki treten und die Mehrheit der Zuschauer hinter sich versammeln? Wird vielleicht sogar die Absolute Mehrheit geknackt und der Jackpot von 200.000 Euro mit nach Hause genommen?",
             "title": "Absolute Mehrheit vom 17.02.2013 - Die Highlights, Teil 2",
         },

youtube_dl/extractor/myvideo.py

@@ -7,11 +7,12 @@ import re
 import json
 
 from .common import InfoExtractor
-from ..utils import (
+from ..compat import (
     compat_ord,
     compat_urllib_parse,
     compat_urllib_request,
-
+)
+from ..utils import (
     ExtractorError,
 )
 

youtube_dl/extractor/ndr.py

@@ -67,7 +67,7 @@ class NDRIE(InfoExtractor):
 
         thumbnail = None
 
-        video_url = re.search(r'''3: \{src:'(?P<video>.+?)\.hi\.mp4', type:"video/mp4"},''', page)
+        video_url = re.search(r'''3: \{src:'(?P<video>.+?)\.(lo|hi|hq)\.mp4', type:"video/mp4"},''', page)
         if video_url:
             thumbnails = re.findall(r'''\d+: \{src: "([^"]+)"(?: \|\| '[^']+')?, quality: '([^']+)'}''', page)
             if thumbnails:

youtube_dl/extractor/niconico.py

@@ -12,6 +12,7 @@ from ..utils import (
     unified_strdate,
     parse_duration,
     int_or_none,
+    ExtractorError,
 )
 
 
@@ -108,6 +109,9 @@ class NiconicoIE(InfoExtractor):
                 flv_info_request, video_id,
                 note='Downloading flv info', errnote='Unable to download flv info')
 
+        if 'deleted=' in flv_info_webpage:
+            raise ExtractorError('The video has been deleted.',
+                expected=True)
         video_real_url = compat_urlparse.parse_qs(flv_info_webpage)['url'][0]
 
         # Start extracting information
@@ -171,7 +175,8 @@ class NiconicoPlaylistIE(InfoExtractor):
         entries = [{
             '_type': 'url',
             'ie_key': NiconicoIE.ie_key(),
-            'url': 'http://www.nicovideo.jp/watch/%s' % entry['item_id'],
+            'url': ('http://www.nicovideo.jp/watch/%s' %
+                entry['item_data']['video_id']),
         } for entry in entries]
 
         return {

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/played.py

@@ -6,6 +6,7 @@ import os.path
 
 from .common import InfoExtractor
 from ..utils import (
+    ExtractorError,
     compat_urllib_parse,
     compat_urllib_request,
 )
@@ -29,6 +30,12 @@ class PlayedIE(InfoExtractor):
         video_id = self._match_id(url)
 
         orig_webpage = self._download_webpage(url, video_id)
+
+        m_error = re.search(
+            r'(?s)Reason for deletion:.*?<b class="err"[^>]*>(?P<msg>[^<]+)</b>', orig_webpage)
+        if m_error:
+            raise ExtractorError(m_error.group('msg'), expected=True)
+
         fields = re.findall(
             r'type="hidden" name="([^"]+)"\s+value="([^"]+)">', orig_webpage)
         data = dict(fields)

youtube_dl/extractor/ro220.py

@@ -1,7 +1,7 @@
 from __future__ import unicode_literals
 
 from .common import InfoExtractor
-from ..utils import compat_urllib_parse_unquote
+from ..compat import compat_urllib_parse_unquote
 
 
 class Ro220IE(InfoExtractor):

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']
-        f4m_url = 'http://manifest.us.rtl.nl' + videopath
+        # Use unencrypted m3u8 streams (See https://github.com/rg3/youtube-dl/issues/4118)
+        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/sexu.py

@@ -0,0 +1,61 @@
+from __future__ import unicode_literals
+
+import re
+
+from .common import InfoExtractor
+
+
+class SexuIE(InfoExtractor):
+    _VALID_URL = r'https?://(?:www\.)?sexu\.com/(?P<id>\d+)'
+    _TEST = {
+        'url': 'http://sexu.com/961791/',
+        'md5': 'ff615aca9691053c94f8f10d96cd7884',
+        'info_dict': {
+            'id': '961791',
+            'ext': 'mp4',
+            'title': 'md5:4d05a19a5fc049a63dbbaf05fb71d91b',
+            'description': 'md5:c5ed8625eb386855d5a7967bd7b77a54',
+            'categories': list,  # NSFW
+            'thumbnail': 're:https?://.*\.jpg$',
+            'age_limit': 18,
+        }
+    }
+
+    def _real_extract(self, url):
+        video_id = self._match_id(url)
+        webpage = self._download_webpage(url, video_id)
+
+        quality_arr = self._search_regex(
+            r'sources:\s*\[([^\]]+)\]', webpage, 'forrmat string')
+        formats = [{
+            'url': fmt[0].replace('\\', ''),
+            'format_id': fmt[1],
+            'height': int(fmt[1][:3]),
+        } for fmt in re.findall(r'"file":"([^"]+)","label":"([^"]+)"', quality_arr)]
+        self._sort_formats(formats)
+
+        title = self._html_search_regex(
+            r'<title>([^<]+)\s*-\s*Sexu\.Com</title>', webpage, 'title')
+
+        description = self._html_search_meta(
+            'description', webpage, 'description')
+
+        thumbnail = self._html_search_regex(
+            r'image:\s*"([^"]+)"',
+            webpage, 'thumbnail', fatal=False)
+
+        categories_str = self._html_search_meta(
+            'keywords', webpage, 'categories')
+        categories = (
+            None if categories_str is None
+            else categories_str.split(','))
+
+        return {
+            'id': video_id,
+            'title': title,
+            'description': description,
+            'thumbnail': thumbnail,
+            'categories': categories,
+            'formats': formats,
+            'age_limit': 18,
+        }

youtube_dl/extractor/spiegel.py

@@ -4,11 +4,12 @@ from __future__ import unicode_literals
 import re
 
 from .common import InfoExtractor
-from ..utils import compat_urlparse
+from ..compat import compat_urlparse
+from .spiegeltv import SpiegeltvIE
 
 
 class SpiegelIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:www\.)?spiegel\.de/video/[^/]*-(?P<videoID>[0-9]+)(?:\.html)?(?:#.*)?$'
+    _VALID_URL = r'https?://(?:www\.)?spiegel\.de/video/[^/]*-(?P<id>[0-9]+)(?:-embed)?(?:\.html)?(?:#.*)?$'
     _TESTS = [{
         'url': 'http://www.spiegel.de/video/vulkan-tungurahua-in-ecuador-ist-wieder-aktiv-video-1259285.html',
         'md5': '2c2754212136f35fb4b19767d242f66e',
@@ -29,16 +30,28 @@ class SpiegelIE(InfoExtractor):
             'description': 'md5:c2322b65e58f385a820c10fa03b2d088',
             'duration': 983,
         },
+    }, {
+        'url': 'http://www.spiegel.de/video/astronaut-alexander-gerst-von-der-iss-station-beantwortet-fragen-video-1519126-embed.html',
+        'md5': 'd8eeca6bfc8f1cd6f490eb1f44695d51',
+        'info_dict': {
+            'id': '1519126',
+            'ext': 'mp4',
+            'description': 'SPIEGEL ONLINE-Nutzer durften den deutschen Astronauten Alexander Gerst über sein Leben auf der ISS-Station befragen. Hier kommen seine Antworten auf die besten sechs Fragen.',
+            'title': 'Fragen an Astronaut Alexander Gerst: "Bekommen Sie die Tageszeiten mit?"',
+        }
     }]
 
     def _real_extract(self, url):
-        m = re.match(self._VALID_URL, url)
-        video_id = m.group('videoID')
+        video_id = self._match_id(url)
+        webpage, handle = self._download_webpage_handle(url, video_id)
 
-        webpage = self._download_webpage(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 = self._html_search_regex(
-            r'<div class="module-title">(.*?)</div>', webpage, 'title')
+        title = re.sub(r'\s+', ' ', self._html_search_regex(
+            r'(?s)<(?:h1|div) class="module-title"[^>]*>(.*?)</(?:h1|div)>',
+            webpage, 'title'))
         description = self._html_search_meta('description', webpage, 'description')
 
         base_url = self._search_regex(
@@ -79,7 +92,7 @@ class SpiegelArticleIE(InfoExtractor):
     _VALID_URL = 'https?://www\.spiegel\.de/(?!video/)[^?#]*?-(?P<id>[0-9]+)\.html'
     IE_NAME = 'Spiegel:Article'
     IE_DESC = 'Articles on spiegel.de'
-    _TEST = {
+    _TESTS = [{
         'url': 'http://www.spiegel.de/sport/sonst/badminton-wm-die-randsportart-soll-populaerer-werden-a-987092.html',
         'info_dict': {
             'id': '1516455',
@@ -87,20 +100,34 @@ class SpiegelArticleIE(InfoExtractor):
             'title': 'Faszination Badminton: Nennt es bloß nicht Federball',
             'description': 're:^Patrick Kämnitz gehört.{100,}',
         },
-    }
+    }, {
+        'url': 'http://www.spiegel.de/wissenschaft/weltall/astronaut-alexander-gerst-antwortet-spiegel-online-lesern-a-989876.html',
+        'info_dict': {
+
+        },
+        'playlist_count': 6,
+    }]
 
     def _real_extract(self, url):
-        m = re.match(self._VALID_URL, url)
-        video_id = m.group('id')
-
+        video_id = self._match_id(url)
         webpage = self._download_webpage(url, video_id)
+
+        # Single video on top of the page
         video_link = self._search_regex(
             r'<a href="([^"]+)" onclick="return spOpenVideo\(this,', webpage,
-            'video page URL')
-        video_url = compat_urlparse.urljoin(
-            self.http_scheme() + '//spiegel.de/', video_link)
+            'video page URL', default=None)
+        if video_link:
+            video_url = compat_urlparse.urljoin(
+                self.http_scheme() + '//spiegel.de/', video_link)
+            return self.url_result(video_url)
 
-        return {
-            '_type': 'url',
-            'url': video_url,
-        }
+        # Multiple embedded videos
+        embeds = re.findall(
+            r'<div class="vid_holder[0-9]+.*?</div>\s*.*?url\s*=\s*"([^"]+)"',
+            webpage)
+        entries = [
+            self.url_result(compat_urlparse.urljoin(
+                self.http_scheme() + '//spiegel.de/', embed_path))
+            for embed_path in embeds
+        ]
+        return self.playlist_result(entries)

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]+)'
-    _TEST = {
+    _VALID_URL = r'https?://(?:www\.)?spiegel\.tv/(?:#/)?filme/(?P<id>[\-a-z0-9]+)'
+    _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)
-        video_id = mobj.group('id')
-
+        if '/#/' in url:
+            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.
-
-        if is_wide:
-            format = '16x9'
-        else:
-            format = '4x3'
+        duration = float_or_none(media_json.get('duration_in_ms'), scale=1000)
+        format = '16x9' if is_wide else '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_DESC = u'Stanford Open ClassRoom'
-    _VALID_URL = r'^(?:https?://)?openclassroom\.stanford\.edu(?P<path>/?|(/MainFolder/(?:HomePage|CoursePage|VideoPage)\.php([?]course=(?P<course>[^&]+)(&video=(?P<video>[^&]+))?(&.*)?)?))$'
+    IE_NAME = 'stanfordoc'
+    IE_DESC = 'Stanford Open ClassRoom'
+    _VALID_URL = r'https?://openclassroom\.stanford\.edu(?P<path>/?|(/MainFolder/(?:HomePage|CoursePage|VideoPage)\.php([?]course=(?P<course>[^&]+)(&video=(?P<video>[^&]+))?(&.*)?)?))$'
     _TEST = {
-        u'url': u'http://openclassroom.stanford.edu/MainFolder/VideoPage.php?course=PracticalUnix&video=intro-environment&speed=100',
-        u'file': u'PracticalUnix_intro-environment.mp4',
-        u'md5': u'544a9468546059d4e80d76265b0443b8',
-        u'info_dict': {
-            u"title": u"Intro Environment"
+        'url': 'http://openclassroom.stanford.edu/MainFolder/VideoPage.php?course=PracticalUnix&video=intro-environment&speed=100',
+        'md5': '544a9468546059d4e80d76265b0443b8',
+        'info_dict': {
+            'id': 'PracticalUnix_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')
-            info['ext'] = info['url'].rpartition('.')[2]
-            return [info]
-        elif mobj.group('course'): # A course page
+                raise ExtractorError('Invalid metadata XML file')
+            return info
+        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'],
-                                        note='Downloading course info page',
-                                        errnote='Unable to download course info page')
+            coursepage = self._download_webpage(
+                url, info['id'],
+                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>',
-                coursepage, u'description', fatal=False)
+            info['description'] = self._html_search_regex(
+                r'(?s)<description>([^<]+)</description>',
+                coursepage, 'description', fatal=False)
 
             links = orderedSet(re.findall('<a href="(VideoPage.php\?[^"]+)">', coursepage))
-            info['list'] = [
-                {
-                    'type': 'reference',
-                    'url': 'http://openclassroom.stanford.edu/MainFolder/' + unescapeHTML(vpage),
-                }
-                    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['entries'] = [self.url_result(
+                'http://openclassroom.stanford.edu/MainFolder/%s' % unescapeHTML(l)
+            ) for l in links]
+            return info
+        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')
-
-            info['title'] = info['id']
+                errnote='Unable to download course info page')
 
             links = orderedSet(re.findall('<a href="(CoursePage.php\?[^"]+)">', rootpage))
-            info['list'] = [
-                {
-                    'type': 'reference',
-                    'url': 'http://openclassroom.stanford.edu/MainFolder/' + unescapeHTML(cpage),
-                }
-                    for cpage in links]
-
-            results = []
-            for entry in info['list']:
-                assert entry['type'] == 'reference'
-                results += self.extract(entry['url'])
-            return results
+            info['entries'] = [self.url_result(
+                'http://openclassroom.stanford.edu/MainFolder/%s' % unescapeHTML(l)
+            ) for l in links]
+            return info

youtube_dl/extractor/streamcloud.py

@@ -13,7 +13,7 @@ from ..utils import (
 
 class StreamcloudIE(InfoExtractor):
     IE_NAME = 'streamcloud.eu'
-    _VALID_URL = r'https?://streamcloud\.eu/(?P<id>[a-zA-Z0-9_-]+)/(?P<fname>[^#?]*)\.html'
+    _VALID_URL = r'https?://streamcloud\.eu/(?P<id>[a-zA-Z0-9_-]+)(?:/(?P<fname>[^#?]*)\.html)?'
 
     _TEST = {
         'url': 'http://streamcloud.eu/skp9j99s4bpz/youtube-dl_test_video_____________-BaW_jenozKc.mp4.html',
@@ -27,8 +27,8 @@ class StreamcloudIE(InfoExtractor):
     }
 
     def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
-        video_id = mobj.group('id')
+        video_id = self._match_id(url)
+        url = 'http://streamcloud.eu/%s' % video_id
 
         orig_webpage = self._download_webpage(url, video_id)
 

youtube_dl/extractor/swrmediathek.py

@@ -80,7 +80,7 @@ class SWRMediathekIE(InfoExtractor):
 
             if media_type == 'Video':
                 fmt.update({
-                    'format_note': ['144p', '288p', '544p'][quality-1],
+                    'format_note': ['144p', '288p', '544p', '720p'][quality-1],
                     'vcodec': codec,
                 })
             elif media_type == 'Audio':

youtube_dl/extractor/sztvhu.py

@@ -1,27 +1,24 @@
 # -*- coding: utf-8 -*-
-
-import re
+from __future__ import unicode_literals
 
 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',
-        u'file': u'20130909.mp4',
-        u'md5': u'a6df607b11fb07d0e9f2ad94613375cb',
-        u'info_dict': {
-            u"title": u"Cserkészek népszerűsítették a környezettudatos életmódot a Savaria téren",
-            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...',
+        'url': 'http://sztv.hu/hirek/cserkeszek-nepszerusitettek-a-kornyezettudatos-eletmodot-a-savaria-teren-20130909',
+        'md5': 'a6df607b11fb07d0e9f2ad94613375cb',
+        'info_dict': {
+            'id': '20130909',
+            'ext': 'mp4',
+            '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 = mobj.group('id')
+        video_id = self._match_id(url)
         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/tapely.py

@@ -50,6 +50,7 @@ class TapelyIE(InfoExtractor):
         request = compat_urllib_request.Request(playlist_url)
         request.add_header('X-Requested-With', 'XMLHttpRequest')
         request.add_header('Accept', 'application/json')
+        request.add_header('Referer', url)
 
         playlist = self._download_json(request, display_id)
 

youtube_dl/extractor/ted.py

@@ -38,6 +38,7 @@ class TEDIE(SubtitlesInfoExtractor):
                 'actively fooling us.'),
             'uploader': 'Dan Dennett',
             'width': 854,
+            'duration': 1308,
         }
     }, {
         'url': 'http://www.ted.com/watch/ted-institute/ted-bcg/vishal-sikka-the-beauty-and-power-of-algorithms',
@@ -57,6 +58,7 @@ class TEDIE(SubtitlesInfoExtractor):
             'title': 'Be passionate. Be courageous. Be your best.',
             'uploader': 'Gabby Giffords and Mark Kelly',
             'description': 'md5:5174aed4d0f16021b704120360f72b92',
+            'duration': 1128,
         },
     }, {
         'url': 'http://www.ted.com/playlists/who_are_the_hackers',
@@ -178,6 +180,7 @@ class TEDIE(SubtitlesInfoExtractor):
             'description': self._og_search_description(webpage),
             'subtitles': video_subtitles,
             'formats': formats,
+            'duration': talk_info.get('duration'),
         }
 
     def _get_available_subtitles(self, video_id, talk_info):

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/trutube.py

@@ -29,7 +29,7 @@ class TruTubeIE(InfoExtractor):
 
         # filehd is always 404
         video_url = xpath_text(config, './file', 'video URL', fatal=True)
-        title = xpath_text(config, './title', 'title')
+        title = xpath_text(config, './title', 'title').strip()
         thumbnail = xpath_text(config, './image', ' thumbnail')
 
         return {

youtube_dl/extractor/tvplay.py

@@ -4,9 +4,9 @@ from __future__ import unicode_literals
 import re
 
 from .common import InfoExtractor
+from ..compat import compat_str
 from ..utils import (
     ExtractorError,
-    compat_str,
     parse_iso8601,
     qualities,
 )
@@ -176,8 +176,7 @@ class TVPlayIE(InfoExtractor):
     ]
 
     def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
-        video_id = mobj.group('id')
+        video_id = self._match_id(url)
 
         video = self._download_json(
             'http://playapi.mtgx.tv/v1/videos/%s' % video_id, video_id, 'Downloading video JSON')
@@ -208,6 +207,10 @@ class TVPlayIE(InfoExtractor):
                     'app': m.group('app'),
                     'play_path': m.group('playpath'),
                 })
+            elif video_url.endswith('.f4m'):
+                formats.extend(self._extract_f4m_formats(
+                    video_url + '?hdcore=3.5.0&plugin=aasp-3.5.0.151.81', video_id))
+                continue
             else:
                 fmt.update({
                     'url': video_url,

youtube_dl/extractor/ustream.py

@@ -5,7 +5,6 @@ import re
 from .common import InfoExtractor
 from ..utils import (
     compat_urlparse,
-    get_meta_content,
 )
 
 
@@ -79,7 +78,7 @@ class UstreamChannelIE(InfoExtractor):
         m = re.match(self._VALID_URL, url)
         display_id = m.group('slug')
         webpage = self._download_webpage(url, display_id)
-        channel_id = get_meta_content('ustream:channel_id', webpage)
+        channel_id = self._html_search_meta('ustream:channel_id', webpage)
 
         BASE = 'http://www.ustream.tv'
         next_url = '/ajax/socialstream/videos/%s/1.json' % channel_id

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/vice.py

@@ -0,0 +1,38 @@
+from __future__ import unicode_literals
+import re
+
+from .common import InfoExtractor
+from .ooyala import OoyalaIE
+from ..utils import ExtractorError
+
+
+class ViceIE(InfoExtractor):
+    _VALID_URL = r'http://www\.vice\.com/.*?/(?P<name>.+)'
+
+    _TEST = {
+        'url': 'http://www.vice.com/Fringes/cowboy-capitalists-part-1',
+        'info_dict': {
+            'id': '43cW1mYzpia9IlestBjVpd23Yu3afAfp',
+            'ext': 'mp4',
+            'title': 'VICE_COWBOYCAPITALISTS_PART01_v1_VICE_WM_1080p.mov',
+        },
+        'params': {
+            # Requires ffmpeg (m3u8 manifest)
+            'skip_download': True,
+        },
+    }
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        name = mobj.group('name')
+        webpage = self._download_webpage(url, name)
+        try:
+            embed_code = self._search_regex(
+                r'embedCode=([^&\'"]+)', webpage,
+                'ooyala embed code')
+            ooyala_url = OoyalaIE._url_for_embed_code(embed_code)
+            print(ooyala_url)
+        except ExtractorError:
+            raise ExtractorError('The page doesn\'t contain a video', expected=True)
+        return self.url_result(ooyala_url, ie='Ooyala')
+

youtube_dl/extractor/vimeo.py

@@ -7,11 +7,13 @@ import itertools
 
 from .common import InfoExtractor
 from .subtitles import SubtitlesInfoExtractor
-from ..utils import (
+from ..compat import (
     compat_HTTPError,
     compat_urllib_parse,
     compat_urllib_request,
     compat_urlparse,
+)
+from ..utils import (
     ExtractorError,
     InAdvancePagedList,
     int_or_none,

youtube_dl/extractor/wimp.py

@@ -37,7 +37,7 @@ class WimpIE(InfoExtractor):
         video_id = mobj.group(1)
         webpage = self._download_webpage(url, video_id)
         video_url = self._search_regex(
-            r"'file'\s*:\s*'([^']+)'", webpage, 'video URL')
+            r"[\"']file[\"']\s*[:,]\s*[\"'](.+?)[\"']", webpage, 'video URL')
         if YoutubeIE.suitable(video_url):
             self.to_screen('Found YouTube video')
             return {

youtube_dl/extractor/wrzuta.py

@@ -27,15 +27,15 @@ class WrzutaIE(InfoExtractor):
             'description': 'md5:7fb5ef3c21c5893375fda51d9b15d9cd',
         },
     }, {
-        'url': 'http://w729.wrzuta.pl/audio/9oXJqdcndqv/david_guetta_amp_showtek_ft._vassy_-_bad',
-        'md5': '1e546a18e1c22ac6e9adce17b8961ff5',
+        'url': 'http://jolka85.wrzuta.pl/audio/063jOPX5ue2/liber_natalia_szroeder_-_teraz_ty',
+        'md5': 'bc78077859bea7bcfe4295d7d7fc9025',
         'info_dict': {
-            'id': '9oXJqdcndqv',
+            'id': '063jOPX5ue2',
             'ext': 'ogg',
-            'title': 'David Guetta & Showtek ft. Vassy - Bad',
-            'duration': 270,
-            'uploader_id': 'w729',
-            'description': 'md5:4628f01c666bbaaecefa83476cfa794a',
+            'title': 'Liber & Natalia Szroeder - Teraz Ty',
+            'duration': 203,
+            'uploader_id': 'jolka85',
+            'description': 'md5:2d2b6340f9188c8c4cd891580e481096',
         },
     }]
 
@@ -49,16 +49,17 @@ class WrzutaIE(InfoExtractor):
 
         quality = qualities(['SD', 'MQ', 'HQ', 'HD'])
 
-        audio_table = {'flv': 'mp3', 'webm': 'ogg'}
+        audio_table = {'flv': 'mp3', 'webm': 'ogg', '???': 'mp3'}
 
         embedpage = self._download_json('http://www.wrzuta.pl/npp/embed/%s/%s' % (uploader, video_id), video_id)
 
         formats = []
         for media in embedpage['url']:
+            fmt = media['type'].split('@')[0]
             if typ == 'audio':
-                ext = audio_table[media['type'].split('@')[0]]
+                ext = audio_table.get(fmt, fmt)
             else:
-                ext = media['type'].split('@')[0]
+                ext = fmt
 
             formats.append({
                 'format_id': '%s_%s' % (ext, media['quality'].lower()),

youtube_dl/extractor/youjizz.py

@@ -9,40 +9,30 @@ from ..utils import (
 
 
 class YouJizzIE(InfoExtractor):
-    _VALID_URL = r'^https?://(?:\w+\.)?youjizz\.com/videos/(?P<videoid>[^.]+)\.html$'
+    _VALID_URL = r'https?://(?:\w+\.)?youjizz\.com/videos/[^/#?]+-(?P<id>[0-9]+)\.html(?:$|[?#])'
     _TEST = {
         'url': 'http://www.youjizz.com/videos/zeichentrick-1-2189178.html',
-        'file': '2189178.flv',
         'md5': '07e15fa469ba384c7693fd246905547c',
         'info_dict': {
+            'id': '2189178',
+            'ext': 'flv',
             "title": "Zeichentrick 1",
             "age_limit": 18,
         }
     }
 
     def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
-
-        video_id = mobj.group('videoid')
-
-        # Get webpage content
+        video_id = self._match_id(url)
         webpage = self._download_webpage(url, video_id)
-
         age_limit = self._rta_search(webpage)
+        video_title = self._html_search_regex(
+            r'<title>\s*(.*)\s*</title>', webpage, 'title')
 
-        # Get the video title
-        video_title = self._html_search_regex(r'<title>(?P<title>.*)</title>',
-            webpage, 'title').strip()
-
-        # Get the embed page
-        result = re.search(r'https?://www.youjizz.com/videos/embed/(?P<videoid>[0-9]+)', webpage)
-        if result is None:
-            raise ExtractorError('ERROR: unable to extract embed page')
-
-        embed_page_url = result.group(0).strip()
-        video_id = result.group('videoid')
-
-        webpage = self._download_webpage(embed_page_url, video_id)
+        embed_page_url = self._search_regex(
+            r'(https?://www.youjizz.com/videos/embed/[0-9]+)',
+            webpage, 'embed page')
+        webpage = self._download_webpage(
+            embed_page_url, video_id, note='downloading embed page')
 
         # Get the video URL
         m_playlist = re.search(r'so.addVariable\("playlist", ?"(?P<playlist>.+?)"\);', webpage)

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):
@@ -510,7 +528,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor):
 
     def _parse_sig_js(self, jscode):
         funcname = self._search_regex(
-            r'signature=([$a-zA-Z]+)', jscode,
+            r'\.sig\|\|([a-zA-Z0-9]+)\(', jscode,
              'Initial JS player signature function name')
 
         jsi = JSInterpreter(jscode)
@@ -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']
@@ -684,7 +702,6 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor):
         # Get video info
         self.report_video_info_webpage_download(video_id)
         if re.search(r'player-age-gate-content">', video_webpage) is not None:
-            self.report_age_confirmation()
             age_gate = True
             # We simulate the access to the video from www.youtube.com/v/{video_id}
             # this can be viewed without login into Youtube
@@ -692,12 +709,13 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor):
                 'video_id': video_id,
                 'eurl': 'https://youtube.googleapis.com/v/' + video_id,
                 'sts': self._search_regex(
-                    r'"sts"\s*:\s*(\d+)', video_webpage, 'sts'),
+                    r'"sts"\s*:\s*(\d+)', video_webpage, 'sts', default=''),
             })
             video_info_url = proto + '://www.youtube.com/get_video_info?' + data
-            video_info_webpage = self._download_webpage(video_info_url, video_id,
-                                    note=False,
-                                    errnote='unable to download video info webpage')
+            video_info_webpage = self._download_webpage(
+                video_info_url, video_id,
+                note='Refetching age-gated info webpage',
+                errnote='unable to download video info webpage')
             video_info = compat_parse_qs(video_info_webpage)
         else:
             age_gate = False
@@ -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)
 
@@ -1043,6 +1061,7 @@ class YoutubePlaylistIE(YoutubeBaseInfoExtractor):
         'url': 'https://www.youtube.com/playlist?list=PLwiyx1dc3P2JR9N8gQaQN_BCvlSlap7re',
         'info_dict': {
             'title': 'ytdl test PL',
+            'id': 'PLwiyx1dc3P2JR9N8gQaQN_BCvlSlap7re',
         },
         'playlist_count': 3,
     }, {

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

@@ -5,9 +5,12 @@ import optparse
 import shlex
 import sys
 
-from .utils import (
+from .compat import (
     compat_expanduser,
     compat_getenv,
+    compat_kwargs,
+)
+from .utils import (
     get_term_width,
     write_string,
 )
@@ -110,7 +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(
@@ -259,7 +262,16 @@ def parseOpts(overrideArguments=None):
     video_format.add_option(
         '-f', '--format',
         action='store', dest='format', metavar='FORMAT', default=None,
-        help='video format code, specify the order of preference using slashes: -f 22/17/18 .  -f mp4 , -f m4a and  -f flv  are also supported. You can also use the special names "best", "bestvideo", "bestaudio", "worst", "worstvideo" and "worstaudio". By default, youtube-dl will pick the best quality. Use commas to download multiple audio formats, such as  -f  136/137/mp4/bestvideo,140/m4a/bestaudio')
+        help='video format code, specify the order of preference using'
+            ' slashes: -f 22/17/18 .  -f mp4 , -f m4a and  -f flv  are also'
+            ' supported. You can also use the special names "best",'
+            ' "bestvideo", "bestaudio", "worst", "worstvideo" and'
+            ' "worstaudio". By default, youtube-dl will pick the best quality.'
+            ' Use commas to download multiple audio formats, such as'
+            ' -f  136/137/mp4/bestvideo,140/m4a/bestaudio.'
+            ' You can merge the video and audio of two formats into a single'
+            ' file using -f <video-format>+<audio-format> (requires ffmpeg or'
+            ' avconv), for example -f bestvideo+bestaudio.')
     video_format.add_option(
         '--all-formats',
         action='store_const', dest='format', const='all',
@@ -479,10 +491,12 @@ def parseOpts(overrideArguments=None):
               '%(format_id)s for the unique id of the format (like Youtube\'s itags: "137"), '
               '%(upload_date)s for the upload date (YYYYMMDD), '
               '%(extractor)s for the provider (youtube, metacafe, etc), '
-              '%(id)s for the video id, %(playlist)s for the playlist the video is in, '
-              '%(playlist_index)s for the position in the playlist and %% for a literal percent. '
+              '%(id)s for the video id, '
+              '%(playlist_title)s, %(playlist_id)s, or %(playlist)s (=title if present, ID otherwise) for the playlist the video is in, '
+              '%(playlist_index)s for the position in the playlist. '
               '%(height)s and %(width)s for the width and height of the video format. '
               '%(resolution)s for a textual description of the resolution of the video format. '
+              '%% for a literal percent. '
               'Use - to output to stdout. Can also be used to download to a different directory, '
               'for example with -o \'/my/downloads/%(uploader)s/%(title)s-%(id)s.%(ext)s\' .'))
     filesystem.add_option(

youtube_dl/postprocessor/atomicparsley.py

@@ -6,10 +6,11 @@ import os
 import subprocess
 
 from .common import PostProcessor
-
+from ..compat import (
+    compat_urlretrieve,
+)
 from ..utils import (
     check_executable,
-    compat_urlretrieve,
     encodeFilename,
     PostProcessingError,
     prepend_extension,

youtube_dl/postprocessor/execafterdownload.py

@@ -3,10 +3,8 @@ from __future__ import unicode_literals
 import subprocess
 
 from .common import PostProcessor
-from ..utils import (
-    shlex_quote,
-    PostProcessingError,
-)
+from ..compat import shlex_quote
+from ..utils import PostProcessingError
 
 
 class ExecAfterDownloadPP(PostProcessor):

youtube_dl/postprocessor/ffmpeg.py

@@ -1,5 +1,4 @@
 import os
-import re
 import subprocess
 import sys
 import time
@@ -7,10 +6,13 @@ import time
 
 from .common import AudioConversionError, PostProcessor
 
-from ..utils import (
+from ..compat import (
     compat_subprocess_get_DEVNULL,
+)
+from ..utils import (
     encodeArgument,
     encodeFilename,
+    get_exe_version,
     is_outdated_version,
     PostProcessingError,
     prepend_extension,
@@ -19,23 +21,6 @@ from ..utils import (
 )
 
 
-def get_version(executable):
-    """ Returns the version of the specified executable,
-    or False if the executable is not present """
-    try:
-        out, err = subprocess.Popen(
-            [executable, '-version'],
-            stdout=subprocess.PIPE, stderr=subprocess.STDOUT).communicate()
-    except OSError:
-        return False
-    firstline = out.partition(b'\n')[0].decode('ascii', 'ignore')
-    m = re.search(r'version\s+([0-9._-a-zA-Z]+)', firstline)
-    if not m:
-        return u'present'
-    else:
-        return m.group(1)
-
-
 class FFmpegPostProcessorError(PostProcessingError):
     pass
 
@@ -61,7 +46,7 @@ class FFmpegPostProcessor(PostProcessor):
     @staticmethod
     def get_versions():
         programs = ['avprobe', 'avconv', 'ffmpeg', 'ffprobe']
-        return dict((program, get_version(program)) for program in programs)
+        return dict((p, get_exe_version(p, args=['-version'])) for p in programs)
 
     @property
     def _executable(self):

youtube_dl/postprocessor/xattrpp.py

@@ -3,10 +3,12 @@ import subprocess
 import sys
 
 from .common import PostProcessor
+from ..compat import (
+    subprocess_check_output
+)
 from ..utils import (
     check_executable,
     hyphenate_date,
-    subprocess_check_output
 )
 
 

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
-                    res = scope.get(mname)
+
+                    if mname in scope:
+                        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

@@ -1,6 +1,8 @@
 #!/usr/bin/env python
 # -*- coding: utf-8 -*-
 
+from __future__ import unicode_literals
+
 import calendar
 import codecs
 import contextlib
@@ -8,7 +10,6 @@ import ctypes
 import datetime
 import email.utils
 import errno
-import getpass
 import gzip
 import itertools
 import io
@@ -29,254 +30,18 @@ import traceback
 import xml.etree.ElementTree
 import zlib
 
-try:
-    import urllib.request as compat_urllib_request
-except ImportError: # Python 2
-    import urllib2 as compat_urllib_request
-
-try:
-    import urllib.error as compat_urllib_error
-except ImportError: # Python 2
-    import urllib2 as compat_urllib_error
-
-try:
-    import urllib.parse as compat_urllib_parse
-except ImportError: # Python 2
-    import urllib as compat_urllib_parse
-
-try:
-    from urllib.parse import urlparse as compat_urllib_parse_urlparse
-except ImportError: # Python 2
-    from urlparse import urlparse as compat_urllib_parse_urlparse
-
-try:
-    import urllib.parse as compat_urlparse
-except ImportError: # Python 2
-    import urlparse as compat_urlparse
-
-try:
-    import http.cookiejar as compat_cookiejar
-except ImportError: # Python 2
-    import cookielib as compat_cookiejar
-
-try:
-    import html.entities as compat_html_entities
-except ImportError: # Python 2
-    import htmlentitydefs as compat_html_entities
-
-try:
-    import html.parser as compat_html_parser
-except ImportError: # Python 2
-    import HTMLParser as compat_html_parser
-
-try:
-    import http.client as compat_http_client
-except ImportError: # Python 2
-    import httplib as compat_http_client
-
-try:
-    from urllib.error import HTTPError as compat_HTTPError
-except ImportError:  # Python 2
-    from urllib2 import HTTPError as compat_HTTPError
-
-try:
-    from urllib.request import urlretrieve as compat_urlretrieve
-except ImportError:  # Python 2
-    from urllib import urlretrieve as compat_urlretrieve
-
-
-try:
-    from subprocess import DEVNULL
-    compat_subprocess_get_DEVNULL = lambda: DEVNULL
-except ImportError:
-    compat_subprocess_get_DEVNULL = lambda: open(os.path.devnull, 'w')
-
-try:
-    from urllib.parse import unquote as compat_urllib_parse_unquote
-except ImportError:
-    def compat_urllib_parse_unquote(string, encoding='utf-8', errors='replace'):
-        if string == '':
-            return string
-        res = string.split('%')
-        if len(res) == 1:
-            return string
-        if encoding is None:
-            encoding = 'utf-8'
-        if errors is None:
-            errors = 'replace'
-        # pct_sequence: contiguous sequence of percent-encoded bytes, decoded
-        pct_sequence = b''
-        string = res[0]
-        for item in res[1:]:
-            try:
-                if not item:
-                    raise ValueError
-                pct_sequence += item[:2].decode('hex')
-                rest = item[2:]
-                if not rest:
-                    # This segment was just a single percent-encoded character.
-                    # May be part of a sequence of code units, so delay decoding.
-                    # (Stored in pct_sequence).
-                    continue
-            except ValueError:
-                rest = '%' + item
-            # Encountered non-percent-encoded characters. Flush the current
-            # pct_sequence.
-            string += pct_sequence.decode(encoding, errors) + rest
-            pct_sequence = b''
-        if pct_sequence:
-            # Flush the final pct_sequence
-            string += pct_sequence.decode(encoding, errors)
-        return string
-
-
-try:
-    from urllib.parse import parse_qs as compat_parse_qs
-except ImportError: # Python 2
-    # HACK: The following is the correct parse_qs implementation from cpython 3's stdlib.
-    # Python 2's version is apparently totally broken
-
-    def _parse_qsl(qs, keep_blank_values=False, strict_parsing=False,
-                encoding='utf-8', errors='replace'):
-        qs, _coerce_result = qs, unicode
-        pairs = [s2 for s1 in qs.split('&') for s2 in s1.split(';')]
-        r = []
-        for name_value in pairs:
-            if not name_value and not strict_parsing:
-                continue
-            nv = name_value.split('=', 1)
-            if len(nv) != 2:
-                if strict_parsing:
-                    raise ValueError("bad query field: %r" % (name_value,))
-                # Handle case of a control-name with no equal sign
-                if keep_blank_values:
-                    nv.append('')
-                else:
-                    continue
-            if len(nv[1]) or keep_blank_values:
-                name = nv[0].replace('+', ' ')
-                name = compat_urllib_parse_unquote(
-                    name, encoding=encoding, errors=errors)
-                name = _coerce_result(name)
-                value = nv[1].replace('+', ' ')
-                value = compat_urllib_parse_unquote(
-                    value, encoding=encoding, errors=errors)
-                value = _coerce_result(value)
-                r.append((name, value))
-        return r
-
-    def compat_parse_qs(qs, keep_blank_values=False, strict_parsing=False,
-                encoding='utf-8', errors='replace'):
-        parsed_result = {}
-        pairs = _parse_qsl(qs, keep_blank_values, strict_parsing,
-                        encoding=encoding, errors=errors)
-        for name, value in pairs:
-            if name in parsed_result:
-                parsed_result[name].append(value)
-            else:
-                parsed_result[name] = [value]
-        return parsed_result
-
-try:
-    compat_str = unicode # Python 2
-except NameError:
-    compat_str = str
-
-try:
-    compat_chr = unichr # Python 2
-except NameError:
-    compat_chr = chr
-
-try:
-    from xml.etree.ElementTree import ParseError as compat_xml_parse_error
-except ImportError:  # Python 2.6
-    from xml.parsers.expat import ExpatError as compat_xml_parse_error
-
-try:
-    from shlex import quote as shlex_quote
-except ImportError:  # Python < 3.3
-    def shlex_quote(s):
-        return "'" + s.replace("'", "'\"'\"'") + "'"
-
-
-def compat_ord(c):
-    if type(c) is int: return c
-    else: return ord(c)
-
-
-if sys.version_info >= (3, 0):
-    compat_getenv = os.getenv
-    compat_expanduser = os.path.expanduser
-else:
-    # Environment variables should be decoded with filesystem encoding.
-    # Otherwise it will fail if any non-ASCII characters present (see #3854 #3217 #2918)
-
-    def compat_getenv(key, default=None):
-        env = os.getenv(key, default)
-        if env:
-            env = env.decode(get_filesystem_encoding())
-        return env
-
-    # HACK: The default implementations of os.path.expanduser from cpython do not decode
-    # environment variables with filesystem encoding. We will work around this by
-    # providing adjusted implementations.
-    # The following are os.path.expanduser implementations from cpython 2.7.8 stdlib
-    # for different platforms with correct environment variables decoding.
-
-    if os.name == 'posix':
-        def compat_expanduser(path):
-            """Expand ~ and ~user constructions.  If user or $HOME is unknown,
-            do nothing."""
-            if not path.startswith('~'):
-                return path
-            i = path.find('/', 1)
-            if i < 0:
-                i = len(path)
-            if i == 1:
-                if 'HOME' not in os.environ:
-                    import pwd
-                    userhome = pwd.getpwuid(os.getuid()).pw_dir
-                else:
-                    userhome = compat_getenv('HOME')
-            else:
-                import pwd
-                try:
-                    pwent = pwd.getpwnam(path[1:i])
-                except KeyError:
-                    return path
-                userhome = pwent.pw_dir
-            userhome = userhome.rstrip('/')
-            return (userhome + path[i:]) or '/'
-    elif os.name == 'nt' or os.name == 'ce':
-        def compat_expanduser(path):
-            """Expand ~ and ~user constructs.
-
-            If user or $HOME is unknown, do nothing."""
-            if path[:1] != '~':
-                return path
-            i, n = 1, len(path)
-            while i < n and path[i] not in '/\\':
-                i = i + 1
-
-            if 'HOME' in os.environ:
-                userhome = compat_getenv('HOME')
-            elif 'USERPROFILE' in os.environ:
-                userhome = compat_getenv('USERPROFILE')
-            elif not 'HOMEPATH' in os.environ:
-                return path
-            else:
-                try:
-                    drive = compat_getenv('HOMEDRIVE')
-                except KeyError:
-                    drive = ''
-                userhome = os.path.join(drive, compat_getenv('HOMEPATH'))
-
-            if i != 1: #~user
-                userhome = os.path.join(os.path.dirname(userhome), path[1:i])
-
-            return userhome + path[i:]
-    else:
-        compat_expanduser = os.path.expanduser
+from .compat import (
+    compat_chr,
+    compat_getenv,
+    compat_html_entities,
+    compat_parse_qs,
+    compat_str,
+    compat_urllib_error,
+    compat_urllib_parse,
+    compat_urllib_parse_urlparse,
+    compat_urllib_request,
+    compat_urlparse,
+)
 
 
 # This is not clearly defined otherwise
@@ -298,28 +63,33 @@ def preferredencoding():
     """
     try:
         pref = locale.getpreferredencoding()
-        u'TEST'.encode(pref)
+        'TEST'.encode(pref)
     except:
         pref = 'UTF-8'
 
     return pref
 
-if sys.version_info < (3,0):
-    def compat_print(s):
-        print(s.encode(preferredencoding(), 'xmlcharrefreplace'))
-else:
-    def compat_print(s):
-        assert type(s) == type(u'')
-        print(s)
-
 
 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) + '.',
-        'dir': os.path.dirname(fn),
+        'prefix': path_basename(fn) + '.',
+        'dir': path_dirname(fn),
         'delete': False,
     }
 
@@ -338,6 +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:
@@ -394,127 +171,32 @@ def xpath_text(node, xpath, name=None, fatal=False):
     return n.text
 
 
-compat_html_parser.locatestarttagend = re.compile(r"""<[a-zA-Z][-.a-zA-Z0-9:_]*(?:\s+(?:(?<=['"\s])[^\s/>][^\s/=>]*(?:\s*=+\s*(?:'[^']*'|"[^"]*"|(?!['"])[^>\s]*))?\s*)*)?\s*""", re.VERBOSE) # backport bugfix
-class BaseHTMLParser(compat_html_parser.HTMLParser):
-    def __init(self):
-        compat_html_parser.HTMLParser.__init__(self)
-        self.html = None
-
-    def loads(self, html):
-        self.html = html
-        self.feed(html)
-        self.close()
-
-class AttrParser(BaseHTMLParser):
-    """Modified HTMLParser that isolates a tag with the specified attribute"""
-    def __init__(self, attribute, value):
-        self.attribute = attribute
-        self.value = value
-        self.result = None
-        self.started = False
-        self.depth = {}
-        self.watch_startpos = False
-        self.error_count = 0
-        BaseHTMLParser.__init__(self)
-
-    def error(self, message):
-        if self.error_count > 10 or self.started:
-            raise compat_html_parser.HTMLParseError(message, self.getpos())
-        self.rawdata = '\n'.join(self.html.split('\n')[self.getpos()[0]:]) # skip one line
-        self.error_count += 1
-        self.goahead(1)
-
-    def handle_starttag(self, tag, attrs):
-        attrs = dict(attrs)
-        if self.started:
-            self.find_startpos(None)
-        if self.attribute in attrs and attrs[self.attribute] == self.value:
-            self.result = [tag]
-            self.started = True
-            self.watch_startpos = True
-        if self.started:
-            if not tag in self.depth: self.depth[tag] = 0
-            self.depth[tag] += 1
-
-    def handle_endtag(self, tag):
-        if self.started:
-            if tag in self.depth: self.depth[tag] -= 1
-            if self.depth[self.result[0]] == 0:
-                self.started = False
-                self.result.append(self.getpos())
-
-    def find_startpos(self, x):
-        """Needed to put the start position of the result (self.result[1])
-        after the opening tag with the requested id"""
-        if self.watch_startpos:
-            self.watch_startpos = False
-            self.result.append(self.getpos())
-    handle_entityref = handle_charref = handle_data = handle_comment = \
-    handle_decl = handle_pi = unknown_decl = find_startpos
-
-    def get_result(self):
-        if self.result is None:
-            return None
-        if len(self.result) != 3:
-            return None
-        lines = self.html.split('\n')
-        lines = lines[self.result[1][0]-1:self.result[2][0]]
-        lines[0] = lines[0][self.result[1][1]:]
-        if len(lines) == 1:
-            lines[-1] = lines[-1][:self.result[2][1]-self.result[1][1]]
-        lines[-1] = lines[-1][:self.result[2][1]]
-        return '\n'.join(lines).strip()
-# Hack for https://github.com/rg3/youtube-dl/issues/662
-if sys.version_info < (2, 7, 3):
-    AttrParser.parse_endtag = (lambda self, i:
-        i + len("</scr'+'ipt>")
-        if self.rawdata[i:].startswith("</scr'+'ipt>")
-        else compat_html_parser.HTMLParser.parse_endtag(self, i))
-
 def get_element_by_id(id, html):
     """Return the content of the tag with the specified ID in the passed HTML document"""
     return get_element_by_attribute("id", id, html)
 
+
 def get_element_by_attribute(attribute, value, html):
     """Return the content of the tag with the specified attribute in the passed HTML document"""
-    parser = AttrParser(attribute, value)
-    try:
-        parser.loads(html)
-    except compat_html_parser.HTMLParseError:
-        pass
-    return parser.get_result()
 
-class MetaParser(BaseHTMLParser):
-    """
-    Modified HTMLParser that isolates a meta tag with the specified name 
-    attribute.
-    """
-    def __init__(self, name):
-        BaseHTMLParser.__init__(self)
-        self.name = name
-        self.content = None
-        self.result = None
+    m = re.search(r'''(?xs)
+        <([a-zA-Z0-9:._-]+)
+         (?:\s+[a-zA-Z0-9:._-]+(?:=[a-zA-Z0-9:._-]+|="[^"]+"|='[^']+'))*?
+         \s+%s=['"]?%s['"]?
+         (?:\s+[a-zA-Z0-9:._-]+(?:=[a-zA-Z0-9:._-]+|="[^"]+"|='[^']+'))*?
+        \s*>
+        (?P<content>.*?)
+        </\1>
+    ''' % (re.escape(attribute), re.escape(value)), html)
 
-    def handle_starttag(self, tag, attrs):
-        if tag != 'meta':
-            return
-        attrs = dict(attrs)
-        if attrs.get('name') == self.name:
-            self.result = attrs.get('content')
+    if not m:
+        return None
+    res = m.group('content')
 
-    def get_result(self):
-        return self.result
+    if res.startswith('"') or res.startswith("'"):
+        res = res[1:-1]
 
-def get_meta_content(name, html):
-    """
-    Return the content attribute from the meta tag with the given name attribute.
-    """
-    parser = MetaParser(name)
-    try:
-        parser.loads(html)
-    except compat_html_parser.HTMLParseError:
-        pass
-    return parser.get_result()
+    return unescapeHTML(res)
 
 
 def clean_html(html):
@@ -541,7 +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)
@@ -554,7 +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:
@@ -593,7 +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('__', '_')
@@ -623,15 +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):
@@ -655,7 +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:
@@ -738,6 +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):
@@ -750,9 +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
@@ -763,7 +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):
@@ -991,17 +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):
     """
@@ -1181,10 +870,7 @@ def bytes_to_intlist(bs):
 def intlist_to_bytes(xs):
     if not xs:
         return b''
-    if isinstance(chr(0), bytes):  # Python 2
-        return ''.join([chr(x) for x in xs])
-    else:
-        return bytes(xs)
+    return struct_pack('%dB' % len(xs), *xs)
 
 
 # Cross-platform file locking
@@ -1296,7 +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):
@@ -1312,31 +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)})
-    return url + u'#' + sdata
+        {'__youtubedl_smuggle': json.dumps(data)})
+    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'#')
-    jsond = compat_parse_qs(sdata)[u'__youtubedl_smuggle'][0]
+    url, _, sdata = smug_url.rpartition('#')
+    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():
@@ -1359,8 +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',
-        u'July', u'August', u'September', u'October', u'November', u'December']
+        'January', 'February', 'March', 'April', 'May', 'June',
+        'July', 'August', 'September', 'October', 'November', 'December']
     try:
         return ENGLISH_NAMES.index(name) + 1
     except ValueError:
@@ -1371,7 +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)
 
 
@@ -1404,7 +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):
@@ -1429,7 +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)
 
 
@@ -1444,7 +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'))
@@ -1459,7 +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=[]):
@@ -1472,6 +1163,25 @@ def check_executable(exe, args=[]):
     return exe
 
 
+def get_exe_version(exe, args=['--version'],
+                    version_re=r'version\s+([0-9._-a-zA-Z]+)',
+                    unrecognized='present'):
+    """ Returns the version of the specified executable,
+    or False if the executable is not present """
+    try:
+        out, err = subprocess.Popen(
+            [exe] + args,
+            stdout=subprocess.PIPE, stderr=subprocess.STDOUT).communicate()
+    except OSError:
+        return False
+    firstline = out.partition(b'\n')[0].decode('ascii', 'ignore')
+    m = re.search(version_re, firstline)
+    if m:
+        return m.group(1)
+    else:
+        return unrecognized
+
+
 class PagedList(object):
     def __len__(self):
         # This is only useful for tests
@@ -1562,7 +1272,7 @@ def escape_rfc3986(s):
     """Escape non-ASCII characters as suggested by RFC 3986"""
     if sys.version_info < (3, 0) and isinstance(s, unicode):
         s = s.encode('utf-8')
-    return compat_urllib_parse.quote(s, "%/;:@&=+$,!~*'()?#[]")
+    return compat_urllib_parse.quote(s, b"%/;:@&=+$,!~*'()?#[]")
 
 
 def escape_url(url):
@@ -1576,7 +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):
@@ -1597,7 +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()
@@ -1636,15 +1346,6 @@ def parse_xml(s):
     return tree
 
 
-if sys.version_info < (3, 0) and sys.platform == 'win32':
-    def compat_getpass(prompt, *args, **kwargs):
-        if isinstance(prompt, compat_str):
-            prompt = prompt.encode(preferredencoding())
-        return getpass.getpass(prompt, *args, **kwargs)
-else:
-    compat_getpass = getpass.getpass
-
-
 US_RATINGS = {
     'G': 0,
     'PG': 10,
@@ -1662,7 +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):
@@ -1702,18 +1404,6 @@ def qualities(quality_ids):
 
 DEFAULT_OUTTMPL = '%(title)s-%(id)s.%(ext)s'
 
-try:
-    subprocess_check_output = subprocess.check_output
-except AttributeError:
-    def subprocess_check_output(*args, **kwargs):
-        assert 'input' not in kwargs
-        p = subprocess.Popen(*args, stdout=subprocess.PIPE, **kwargs)
-        output, _ = p.communicate()
-        ret = p.poll()
-        if ret:
-            raise subprocess.CalledProcessError(ret, p.args, output=output)
-        return output
-
 
 def limit_length(s, length):
     """ Add ellipses to overly long strings """
@@ -1736,3 +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__ = '2014.11.02.1'
+__version__ = '2014.11.23'