release 2015.02.02.3

'' might actually be passed in, so check for None.

AUTHORS

@@ -107,5 +107,4 @@ Yen Chi Hsuan
 Enam Mijbah Noor
 David Luhmer
 Shaya Goldberg
-Yen Chi Hsuan
 Paul Hartmann

README.md

@@ -368,11 +368,11 @@ which means you can modify it, redistribute it or use it however you like.
     --add-metadata                   write metadata to the video file
     --xattrs                         write metadata to the video file's xattrs
                                      (using dublin core and xdg standards)
-    --fixup POLICY                   (experimental) Automatically correct known
-                                     faults of the file. One of never (do
-                                     nothing), warn (only emit a warning),
-                                     detect_or_warn(check whether we can do
-                                     anything about it, warn otherwise
+    --fixup POLICY                   Automatically correct known faults of the
+                                     file. One of never (do nothing), warn (only
+                                     emit a warning), detect_or_warn(the
+                                     default; fix file if we can, warn
+                                     otherwise)
     --prefer-avconv                  Prefer avconv over ffmpeg for running the
                                      postprocessors (default)
     --prefer-ffmpeg                  Prefer ffmpeg over avconv for running the
@@ -525,9 +525,16 @@ From then on, after restarting your shell, you will be able to access both youtu
 
 Use the `-o` to specify an [output template](#output-template), for example `-o "/home/user/videos/%(title)s-%(id)s.%(ext)s"`. If you want this for all of your downloads, put the option into your [configuration file](#configuration).
 
+### How do I download a video starting with a `-` ?
+
+Either prepend `http://www.youtube.com/watch?v=` or separate the ID from the options with `--`:
+
+    youtube-dl -- -wNyEUrxzFU
+    youtube-dl "http://www.youtube.com/watch?v=-wNyEUrxzFU"
+
 ### How can I detect whether a given URL is supported by youtube-dl?
 
-For one, have a look at the [list of supported sites](docs/supportedsites). Note that it can sometimes happen that the site changes its URL scheme (say, from http://example.com/v/1234567 to http://example.com/v/1234567 ) and youtube-dl reports an URL of a service in that list as unsupported. In that case, simply report a bug.
+For one, have a look at the [list of supported sites](docs/supportedsites.md). Note that it can sometimes happen that the site changes its URL scheme (say, from http://example.com/v/1234567 to http://example.com/v/1234567 ) and youtube-dl reports an URL of a service in that list as unsupported. In that case, simply report a bug.
 
 It is *not* possible to detect whether a URL is supported or not. That's because youtube-dl contains a generic extractor which matches **all** URLs. You may be tempted to disable, exclude, or remove the generic extractor, but the generic extractor not only allows users to extract videos from lots of websites that embed a video from another service, but may also be used to extract video from a service that it's hosting itself. Therefore, we neither recommend nor support disabling, excluding, or removing the generic extractor.
 

test/helper.py

@@ -148,9 +148,15 @@ def expect_info_dict(self, got_dict, expected_dict):
                 return "'%s'" % v.replace('\\', '\\\\').replace("'", "\\'").replace('\n', '\\n')
             else:
                 return repr(v)
-        info_dict_str = ''.join(
-            '    %s: %s,\n' % (_repr(k), _repr(v))
-            for k, v in test_info_dict.items())
+        info_dict_str = ''
+        if len(missing_keys) != len(expected_dict):
+            info_dict_str += ''.join(
+                '    %s: %s,\n' % (_repr(k), _repr(v))
+                for k, v in test_info_dict.items() if k not in missing_keys)
+            info_dict_str += '\n'
+        info_dict_str += ''.join(
+            '    %s: %s,\n' % (_repr(k), _repr(test_info_dict[k]))
+            for k in missing_keys)
         write_string(
             '\n\'info_dict\': {\n' + info_dict_str + '}\n', out=sys.stderr)
         self.assertFalse(

test/test_download.py

@@ -89,7 +89,7 @@ def generator(test_case):
 
         for tc in test_cases:
             info_dict = tc.get('info_dict', {})
-            if not tc.get('file') and not (info_dict.get('id') and info_dict.get('ext')):
+            if not (info_dict.get('id') and info_dict.get('ext')):
                 raise Exception('Test definition incorrect. The output file cannot be known. Are both \'id\' and \'ext\' keys present?')
 
         if 'skip' in test_case:
@@ -116,7 +116,7 @@ def generator(test_case):
         expect_warnings(ydl, test_case.get('expected_warnings', []))
 
         def get_tc_filename(tc):
-            return tc.get('file') or ydl.prepare_filename(tc.get('info_dict', {}))
+            return ydl.prepare_filename(tc.get('info_dict', {}))
 
         res_dict = None
 

test/test_jsinterp.py

@@ -0,0 +1,95 @@
+#!/usr/bin/env python
+
+from __future__ import unicode_literals
+
+# Allow direct execution
+import os
+import sys
+import unittest
+sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
+
+from youtube_dl.jsinterp import JSInterpreter
+
+
+class TestJSInterpreter(unittest.TestCase):
+    def test_basic(self):
+        jsi = JSInterpreter('function x(){;}')
+        self.assertEqual(jsi.call_function('x'), None)
+
+        jsi = JSInterpreter('function x3(){return 42;}')
+        self.assertEqual(jsi.call_function('x3'), 42)
+
+    def test_calc(self):
+        jsi = JSInterpreter('function x4(a){return 2*a+1;}')
+        self.assertEqual(jsi.call_function('x4', 3), 7)
+
+    def test_empty_return(self):
+        jsi = JSInterpreter('function f(){return; y()}')
+        self.assertEqual(jsi.call_function('f'), None)
+
+    def test_morespace(self):
+        jsi = JSInterpreter('function x (a) { return 2 * a + 1 ; }')
+        self.assertEqual(jsi.call_function('x', 3), 7)
+
+        jsi = JSInterpreter('function f () { x =  2  ; return x; }')
+        self.assertEqual(jsi.call_function('f'), 2)
+
+    def test_strange_chars(self):
+        jsi = JSInterpreter('function $_xY1 ($_axY1) { var $_axY2 = $_axY1 + 1; return $_axY2; }')
+        self.assertEqual(jsi.call_function('$_xY1', 20), 21)
+
+    def test_operators(self):
+        jsi = JSInterpreter('function f(){return 1 << 5;}')
+        self.assertEqual(jsi.call_function('f'), 32)
+
+        jsi = JSInterpreter('function f(){return 19 & 21;}')
+        self.assertEqual(jsi.call_function('f'), 17)
+
+        jsi = JSInterpreter('function f(){return 11 >> 2;}')
+        self.assertEqual(jsi.call_function('f'), 2)
+
+    def test_array_access(self):
+        jsi = JSInterpreter('function f(){var x = [1,2,3]; x[0] = 4; x[0] = 5; x[2] = 7; return x;}')
+        self.assertEqual(jsi.call_function('f'), [5, 2, 7])
+
+    def test_parens(self):
+        jsi = JSInterpreter('function f(){return (1) + (2) * ((( (( (((((3)))))) )) ));}')
+        self.assertEqual(jsi.call_function('f'), 7)
+
+        jsi = JSInterpreter('function f(){return (1 + 2) * 3;}')
+        self.assertEqual(jsi.call_function('f'), 9)
+
+    def test_assignments(self):
+        jsi = JSInterpreter('function f(){var x = 20; x = 30 + 1; return x;}')
+        self.assertEqual(jsi.call_function('f'), 31)
+
+        jsi = JSInterpreter('function f(){var x = 20; x += 30 + 1; return x;}')
+        self.assertEqual(jsi.call_function('f'), 51)
+
+        jsi = JSInterpreter('function f(){var x = 20; x -= 30 + 1; return x;}')
+        self.assertEqual(jsi.call_function('f'), -11)
+
+    def test_comments(self):
+        jsi = JSInterpreter('''
+        function x() {
+            var x = /* 1 + */ 2;
+            var y = /* 30
+            * 40 */ 50;
+            return x + y;
+        }
+        ''')
+        self.assertEqual(jsi.call_function('x'), 52)
+
+    def test_precedence(self):
+        jsi = JSInterpreter('''
+        function x() {
+            var a = [10, 20, 30, 40, 50];
+            var b = 6;
+            a[0]=a[b%a.length];
+            return a;
+        }''')
+        self.assertEqual(jsi.call_function('x'), [20, 20, 30, 40, 50])
+
+
+if __name__ == '__main__':
+    unittest.main()

test/test_utils.py

@@ -238,6 +238,8 @@ class TestUtil(unittest.TestCase):
         self.assertEqual(parse_duration('5 s'), 5)
         self.assertEqual(parse_duration('3 min'), 180)
         self.assertEqual(parse_duration('2.5 hours'), 9000)
+        self.assertEqual(parse_duration('02:03:04'), 7384)
+        self.assertEqual(parse_duration('01:02:03:04'), 93784)
 
     def test_fix_xml_ampersands(self):
         self.assertEqual(
@@ -371,6 +373,16 @@ class TestUtil(unittest.TestCase):
         on = js_to_json('{"abc": true}')
         self.assertEqual(json.loads(on), {'abc': True})
 
+        # Ignore JavaScript code as well
+        on = js_to_json('''{
+            "x": 1,
+            y: "a",
+            z: some.code
+        }''')
+        d = json.loads(on)
+        self.assertEqual(d['x'], 1)
+        self.assertEqual(d['y'], 'a')
+
     def test_clean_html(self):
         self.assertEqual(clean_html('a:\nb'), 'a: b')
         self.assertEqual(clean_html('a:\n   "b"'), 'a:    "b"')

youtube_dl/YoutubeDL.py

@@ -25,6 +25,7 @@ if os.name == 'nt':
     import ctypes
 
 from .compat import (
+    compat_basestring,
     compat_cookiejar,
     compat_expanduser,
     compat_http_client,
@@ -1074,8 +1075,8 @@ class YoutubeDL(object):
                             selected_format = {
                                 'requested_formats': formats_info,
                                 'format': rf,
-                                'format_id': rf,
-                                'ext': formats_info[0]['ext'],
+                                'format_id': '%s+%s' % (formats_info[0].get('format_id'),
+                                                        formats_info[1].get('format_id')),
                                 'width': formats_info[0].get('width'),
                                 'height': formats_info[0].get('height'),
                                 'resolution': formats_info[0].get('resolution'),
@@ -1558,7 +1559,7 @@ class YoutubeDL(object):
         # urllib chokes on URLs with non-ASCII characters (see http://bugs.python.org/issue3991)
         # To work around aforementioned issue we will replace request's original URL with
         # percent-encoded one
-        req_is_string = isinstance(req, basestring if sys.version_info < (3, 0) else compat_str)
+        req_is_string = isinstance(req, compat_basestring)
         url = req if req_is_string else req.get_full_url()
         url_escaped = escape_url(url)
 

youtube_dl/compat.py

@@ -114,6 +114,26 @@ except ImportError:
             string += pct_sequence.decode(encoding, errors)
         return string
 
+try:
+    compat_str = unicode  # Python 2
+except NameError:
+    compat_str = str
+
+try:
+    compat_basestring = basestring  # Python 2
+except NameError:
+    compat_basestring = 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 urllib.parse import parse_qs as compat_parse_qs
@@ -123,7 +143,7 @@ except ImportError:  # Python 2
 
     def _parse_qsl(qs, keep_blank_values=False, strict_parsing=False,
                    encoding='utf-8', errors='replace'):
-        qs, _coerce_result = qs, unicode
+        qs, _coerce_result = qs, compat_str
         pairs = [s2 for s1 in qs.split('&') for s2 in s1.split(';')]
         r = []
         for name_value in pairs:
@@ -162,21 +182,6 @@ except ImportError:  # Python 2
                 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
@@ -362,6 +367,7 @@ def workaround_optparse_bug9161():
 
 __all__ = [
     'compat_HTTPError',
+    'compat_basestring',
     'compat_chr',
     'compat_cookiejar',
     'compat_expanduser',

youtube_dl/downloader/external.py

@@ -45,6 +45,12 @@ class ExternalFD(FileDownloader):
     def supports(cls, info_dict):
         return info_dict['protocol'] in ('http', 'https', 'ftp', 'ftps')
 
+    def _source_address(self, command_option):
+        source_address = self.params.get('source_address')
+        if source_address is None:
+            return []
+        return [command_option, source_address]
+
     def _call_downloader(self, tmpfilename, info_dict):
         """ Either overwrite this or implement _make_cmd """
         cmd = self._make_cmd(tmpfilename, info_dict)
@@ -72,6 +78,7 @@ class CurlFD(ExternalFD):
         cmd = [self.exe, '-o', tmpfilename]
         for key, val in info_dict['http_headers'].items():
             cmd += ['--header', '%s: %s' % (key, val)]
+        cmd += self._source_address('--interface')
         cmd += ['--', info_dict['url']]
         return cmd
 
@@ -81,6 +88,7 @@ class WgetFD(ExternalFD):
         cmd = [self.exe, '-O', tmpfilename, '-nv', '--no-cookies']
         for key, val in info_dict['http_headers'].items():
             cmd += ['--header', '%s: %s' % (key, val)]
+        cmd += self._source_address('--bind-address')
         cmd += ['--', info_dict['url']]
         return cmd
 
@@ -96,6 +104,7 @@ class Aria2cFD(ExternalFD):
         cmd += ['--out', os.path.basename(tmpfilename)]
         for key, val in info_dict['http_headers'].items():
             cmd += ['--header', '%s: %s' % (key, val)]
+        cmd += self._source_address('--interface')
         cmd += ['--', info_dict['url']]
         return cmd
 

youtube_dl/downloader/f4m.py

@@ -230,6 +230,23 @@ class F4mFD(FileDownloader):
     A downloader for f4m manifests or AdobeHDS.
     """
 
+    def _get_unencrypted_media(self, doc):
+        media = doc.findall(_add_ns('media'))
+        if not media:
+            self.report_error('No media found')
+        for e in (doc.findall(_add_ns('drmAdditionalHeader')) +
+                  doc.findall(_add_ns('drmAdditionalHeaderSet'))):
+            # If id attribute is missing it's valid for all media nodes
+            # without drmAdditionalHeaderId or drmAdditionalHeaderSetId attribute
+            if 'id' not in e.attrib:
+                self.report_error('Missing ID in f4m DRM')
+        media = list(filter(lambda e: 'drmAdditionalHeaderId' not in e.attrib and
+                                      'drmAdditionalHeaderSetId' not in e.attrib,
+                            media))
+        if not media:
+            self.report_error('Unsupported DRM')
+        return media
+
     def real_download(self, filename, info_dict):
         man_url = info_dict['url']
         requested_bitrate = info_dict.get('tbr')
@@ -248,7 +265,8 @@ class F4mFD(FileDownloader):
         )
 
         doc = etree.fromstring(manifest)
-        formats = [(int(f.attrib.get('bitrate', -1)), f) for f in doc.findall(_add_ns('media'))]
+        formats = [(int(f.attrib.get('bitrate', -1)), f)
+                   for f in self._get_unencrypted_media(doc)]
         if requested_bitrate is None:
             # get the best format
             formats = sorted(formats, key=lambda f: f[0])

youtube_dl/downloader/hls.py

@@ -11,6 +11,7 @@ from ..compat import (
     compat_urllib_request,
 )
 from ..utils import (
+    encodeArgument,
     encodeFilename,
 )
 
@@ -21,23 +22,22 @@ class HlsFD(FileDownloader):
         self.report_destination(filename)
         tmpfilename = self.temp_name(filename)
 
-        args = [
-            '-y', '-i', url, '-f', 'mp4', '-c', 'copy',
-            '-bsf:a', 'aac_adtstoasc',
-            encodeFilename(tmpfilename, for_subprocess=True)]
-
         ffpp = FFmpegPostProcessor(downloader=self)
         program = ffpp._executable
         if program is None:
             self.report_error('m3u8 download detected but ffmpeg or avconv could not be found. Please install one.')
             return False
         ffpp.check_version()
-        cmd = [program] + args
 
-        retval = subprocess.call(cmd)
+        args = [
+            encodeArgument(opt)
+            for opt in (program, '-y', '-i', url, '-f', 'mp4', '-c', 'copy', '-bsf:a', 'aac_adtstoasc')]
+        args.append(encodeFilename(tmpfilename, True))
+
+        retval = subprocess.call(args)
         if retval == 0:
             fsize = os.path.getsize(encodeFilename(tmpfilename))
-            self.to_screen('\r[%s] %s bytes' % (cmd[0], fsize))
+            self.to_screen('\r[%s] %s bytes' % (args[0], fsize))
             self.try_rename(tmpfilename, filename)
             self._hook_progress({
                 'downloaded_bytes': fsize,

youtube_dl/downloader/http.py

@@ -3,6 +3,9 @@ from __future__ import unicode_literals
 import os
 import time
 
+from socket import error as SocketError
+import errno
+
 from .common import FileDownloader
 from ..compat import (
     compat_urllib_request,
@@ -99,6 +102,11 @@ class HttpFD(FileDownloader):
                             resume_len = 0
                             open_mode = 'wb'
                             break
+            except SocketError as e:
+                if e.errno != errno.ECONNRESET:
+                    # Connection reset is no problem, just retry
+                    raise
+
             # Retry
             count += 1
             if count <= retries:

youtube_dl/extractor/__init__.py

@@ -182,6 +182,7 @@ from .heise import HeiseIE
 from .hellporno import HellPornoIE
 from .helsinki import HelsinkiIE
 from .hentaistigma import HentaiStigmaIE
+from .historicfilms import HistoricFilmsIE
 from .hitbox import HitboxIE, HitboxLiveIE
 from .hornbunny import HornBunnyIE
 from .hostingbulk import HostingBulkIE
@@ -294,7 +295,11 @@ from .nextmedia import (
 )
 from .nfb import NFBIE
 from .nfl import NFLIE
-from .nhl import NHLIE, NHLVideocenterIE
+from .nhl import (
+    NHLIE,
+    NHLNewsIE,
+    NHLVideocenterIE,
+)
 from .niconico import NiconicoIE, NiconicoPlaylistIE
 from .ninegag import NineGagIE
 from .noco import NocoIE
@@ -312,7 +317,8 @@ from .nrk import (
     NRKIE,
     NRKTVIE,
 )
-from .ntv import NTVIE
+from .ntvde import NTVDeIE
+from .ntvru import NTVRuIE
 from .nytimes import NYTimesIE
 from .nuvid import NuvidIE
 from .oktoberfesttv import OktoberfestTVIE

youtube_dl/extractor/appletrailers.py

@@ -122,7 +122,6 @@ class AppleTrailersIE(InfoExtractor):
             playlist.append({
                 '_type': 'video',
                 'id': video_id,
-                'title': title,
                 'formats': formats,
                 'title': title,
                 'duration': duration,

youtube_dl/extractor/ard.py

@@ -23,13 +23,7 @@ class ARDMediathekIE(InfoExtractor):
 
     _TESTS = [{
         'url': 'http://mediathek.daserste.de/sendungen_a-z/328454_anne-will/22429276_vertrauen-ist-gut-spionieren-ist-besser-geht',
-        'file': '22429276.mp4',
-        'md5': '469751912f1de0816a9fc9df8336476c',
-        'info_dict': {
-            'title': 'Vertrauen ist gut, Spionieren ist besser - Geht so deutsch-amerikanische Freundschaft?',
-            'description': 'Das Erste Mediathek [ARD]: Vertrauen ist gut, Spionieren ist besser - Geht so deutsch-amerikanische Freundschaft?, Anne Will, Über die Spionage-Affäre diskutieren Clemens Binninger, Katrin Göring-Eckardt, Georg Mascolo, Andrew B. Denison und Constanze Kurz.. Das Video zur Sendung Anne Will am Mittwoch, 16.07.2014',
-        },
-        'skip': 'Blocked outside of Germany',
+        'only_matching': True,
     }, {
         'url': 'http://www.ardmediathek.de/tv/Tatort/Das-Wunder-von-Wolbeck-Video-tgl-ab-20/Das-Erste/Video?documentId=22490580&bcastId=602916',
         'info_dict': {

youtube_dl/extractor/bbccouk.py

@@ -10,7 +10,7 @@ from ..compat import compat_HTTPError
 class BBCCoUkIE(SubtitlesInfoExtractor):
     IE_NAME = 'bbc.co.uk'
     IE_DESC = 'BBC iPlayer'
-    _VALID_URL = r'https?://(?:www\.)?bbc\.co\.uk/(?:(?:(?:programmes|iplayer/(?:episode|playlist))/)|music/clips[/#])(?P<id>[\da-z]{8})'
+    _VALID_URL = r'https?://(?:www\.)?bbc\.co\.uk/(?:(?:(?:programmes|iplayer(?:/[^/]+)?/(?:episode|playlist))/)|music/clips[/#])(?P<id>[\da-z]{8})'
 
     _TESTS = [
         {
@@ -118,6 +118,9 @@ class BBCCoUkIE(SubtitlesInfoExtractor):
         }, {
             'url': 'http://www.bbc.co.uk/music/clips#p02frcc3',
             'only_matching': True,
+        }, {
+            'url': 'http://www.bbc.co.uk/iplayer/cbeebies/episode/b0480276/bing-14-atchoo',
+            'only_matching': True,
         }
     ]
 

youtube_dl/extractor/cinchcast.py

@@ -28,12 +28,10 @@ class CinchcastIE(InfoExtractor):
             item, './{http://developer.longtailvideo.com/trac/}date')
         upload_date = unified_strdate(date_str, day_first=False)
         # duration is present but wrong
-        formats = []
-        formats.append({
+        formats = [{
             'format_id': 'main',
-            'url': item.find(
-                './{http://search.yahoo.com/mrss/}content').attrib['url'],
-        })
+            'url': item.find('./{http://search.yahoo.com/mrss/}content').attrib['url'],
+        }]
         backup_url = xpath_text(
             item, './{http://developer.longtailvideo.com/trac/}backupContent')
         if backup_url:

youtube_dl/extractor/comedycentral.py

@@ -49,7 +49,9 @@ class ComedyCentralShowsIE(MTVServicesInfoExtractor):
                               |(watch/(?P<date>[^/]*)/(?P<tdstitle>.*))
                           )|
                           (?P<interview>
-                              extended-interviews/(?P<interID>[0-9a-z]+)/(?:playlist_tds_extended_)?(?P<interview_title>.*?)(/.*?)?)))
+                              extended-interviews/(?P<interID>[0-9a-z]+)/
+                              (?:playlist_tds_extended_)?(?P<interview_title>[^/?#]*?)
+                              (?:/[^/?#]?|[?#]|$))))
                      '''
     _TESTS = [{
         'url': 'http://thedailyshow.cc.com/watch/thu-december-13-2012/kristen-stewart',
@@ -62,6 +64,38 @@ class ComedyCentralShowsIE(MTVServicesInfoExtractor):
             'uploader': 'thedailyshow',
             'title': 'thedailyshow kristen-stewart part 1',
         }
+    }, {
+        'url': 'http://thedailyshow.cc.com/extended-interviews/b6364d/sarah-chayes-extended-interview',
+        'info_dict': {
+            'id': 'sarah-chayes-extended-interview',
+            'description': 'Carnegie Endowment Senior Associate Sarah Chayes discusses how corrupt institutions function throughout the world in her book "Thieves of State: Why Corruption Threatens Global Security."',
+            'title': 'thedailyshow Sarah Chayes Extended Interview',
+        },
+        'playlist': [
+            {
+                'info_dict': {
+                    'id': '0baad492-cbec-4ec1-9e50-ad91c291127f',
+                    'ext': 'mp4',
+                    'upload_date': '20150129',
+                    'description': 'Carnegie Endowment Senior Associate Sarah Chayes discusses how corrupt institutions function throughout the world in her book "Thieves of State: Why Corruption Threatens Global Security."',
+                    'uploader': 'thedailyshow',
+                    'title': 'thedailyshow sarah-chayes-extended-interview part 1',
+                },
+            },
+            {
+                'info_dict': {
+                    'id': '1e4fb91b-8ce7-4277-bd7c-98c9f1bbd283',
+                    'ext': 'mp4',
+                    'upload_date': '20150129',
+                    'description': 'Carnegie Endowment Senior Associate Sarah Chayes discusses how corrupt institutions function throughout the world in her book "Thieves of State: Why Corruption Threatens Global Security."',
+                    'uploader': 'thedailyshow',
+                    'title': 'thedailyshow sarah-chayes-extended-interview part 2',
+                },
+            },
+        ],
+        'params': {
+            'skip_download': True,
+        },
     }, {
         'url': 'http://thedailyshow.cc.com/extended-interviews/xm3fnq/andrew-napolitano-extended-interview',
         'only_matching': True,
@@ -230,6 +264,7 @@ class ComedyCentralShowsIE(MTVServicesInfoExtractor):
 
         return {
             '_type': 'playlist',
+            'id': epTitle,
             'entries': entries,
             'title': show_name + ' ' + title,
             'description': description,

youtube_dl/extractor/common.py

@@ -89,7 +89,8 @@ class InfoExtractor(object):
                     * player_url SWF Player URL (used for rtmpdump).
                     * protocol   The protocol that will be used for the actual
                                  download, lower-case.
-                                 "http", "https", "rtsp", "rtmp", "m3u8" or so.
+                                 "http", "https", "rtsp", "rtmp", "rtmpe",
+                                 "m3u8", or "m3u8_native".
                     * preference Order number of this format. If this field is
                                  present and not None, the formats get sorted
                                  by this field, regardless of all other values.

youtube_dl/extractor/defense.py

@@ -1,40 +1,38 @@
 from __future__ import unicode_literals
 
-import re
-import json
-
 from .common import InfoExtractor
 
 
 class DefenseGouvFrIE(InfoExtractor):
     IE_NAME = 'defense.gouv.fr'
-    _VALID_URL = (r'http://.*?\.defense\.gouv\.fr/layout/set/'
-                  r'ligthboxvideo/base-de-medias/webtv/(.*)')
+    _VALID_URL = r'http://.*?\.defense\.gouv\.fr/layout/set/ligthboxvideo/base-de-medias/webtv/(?P<id>[^/?#]*)'
 
     _TEST = {
         'url': 'http://www.defense.gouv.fr/layout/set/ligthboxvideo/base-de-medias/webtv/attaque-chimique-syrienne-du-21-aout-2013-1',
-        'file': '11213.mp4',
         'md5': '75bba6124da7e63d2d60b5244ec9430c',
-        "info_dict": {
-            "title": "attaque-chimique-syrienne-du-21-aout-2013-1"
+        'info_dict': {
+            'id': '11213',
+            'ext': 'mp4',
+            'title': 'attaque-chimique-syrienne-du-21-aout-2013-1'
         }
     }
 
     def _real_extract(self, url):
-        title = re.match(self._VALID_URL, url).group(1)
+        title = self._match_id(url)
         webpage = self._download_webpage(url, title)
+
         video_id = self._search_regex(
             r"flashvars.pvg_id=\"(\d+)\";",
             webpage, 'ID')
 
         json_url = ('http://static.videos.gouv.fr/brightcovehub/export/json/'
                     + video_id)
-        info = self._download_webpage(json_url, title,
-                                      'Downloading JSON config')
-        video_url = json.loads(info)['renditions'][0]['url']
+        info = self._download_json(json_url, title, 'Downloading JSON config')
+        video_url = info['renditions'][0]['url']
 
-        return {'id': video_id,
-                'ext': 'mp4',
-                'url': video_url,
-                'title': title,
-                }
+        return {
+            'id': video_id,
+            'ext': 'mp4',
+            'url': video_url,
+            'title': title,
+        }

youtube_dl/extractor/drtv.py

@@ -6,7 +6,7 @@ from ..utils import parse_iso8601
 
 
 class DRTVIE(SubtitlesInfoExtractor):
-    _VALID_URL = r'https?://(?:www\.)?dr\.dk/tv/se/(?:[^/]+/)+(?P<id>[\da-z-]+)(?:[/#?]|$)'
+    _VALID_URL = r'https?://(?:www\.)?dr\.dk/tv/se/(?:[^/]+/)*(?P<id>[\da-z-]+)(?:[/#?]|$)'
 
     _TEST = {
         'url': 'http://www.dr.dk/tv/se/partiets-mand/partiets-mand-7-8',
@@ -25,9 +25,15 @@ class DRTVIE(SubtitlesInfoExtractor):
     def _real_extract(self, url):
         video_id = self._match_id(url)
 
-        programcard = self._download_json(
-            'http://www.dr.dk/mu/programcard/expanded/%s' % video_id, video_id, 'Downloading video JSON')
+        webpage = self._download_webpage(url, video_id)
 
+        video_id = self._search_regex(
+            r'data-(?:material-identifier|episode-slug)="([^"]+)"',
+            webpage, 'video id')
+
+        programcard = self._download_json(
+            'http://www.dr.dk/mu/programcard/expanded/%s' % video_id,
+            video_id, 'Downloading video JSON')
         data = programcard['Data'][0]
 
         title = data['Title']

youtube_dl/extractor/francetv.py

@@ -230,12 +230,13 @@ class FranceTVIE(FranceTVBaseInfoExtractor):
 
 class GenerationQuoiIE(InfoExtractor):
     IE_NAME = 'france2.fr:generation-quoi'
-    _VALID_URL = r'https?://generation-quoi\.france2\.fr/portrait/(?P<name>.*)(\?|$)'
+    _VALID_URL = r'https?://generation-quoi\.france2\.fr/portrait/(?P<id>[^/?#]+)'
 
     _TEST = {
         'url': 'http://generation-quoi.france2.fr/portrait/garde-a-vous',
-        'file': 'k7FJX8VBcvvLmX4wA5Q.mp4',
         'info_dict': {
+            'id': 'k7FJX8VBcvvLmX4wA5Q',
+            'ext': 'mp4',
             'title': 'Génération Quoi - Garde à Vous',
             'uploader': 'Génération Quoi',
         },
@@ -243,14 +244,12 @@ class GenerationQuoiIE(InfoExtractor):
             # It uses Dailymotion
             'skip_download': True,
         },
-        'skip': 'Only available from France',
     }
 
     def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
-        name = mobj.group('name')
-        info_url = compat_urlparse.urljoin(url, '/medias/video/%s.json' % name)
-        info_json = self._download_webpage(info_url, name)
+        display_id = self._match_id(url)
+        info_url = compat_urlparse.urljoin(url, '/medias/video/%s.json' % display_id)
+        info_json = self._download_webpage(info_url, display_id)
         info = json.loads(info_json)
         return self.url_result('http://www.dailymotion.com/video/%s' % info['id'],
                                ie='Dailymotion')

youtube_dl/extractor/generic.py

@@ -1073,7 +1073,7 @@ class GenericIE(InfoExtractor):
             found = filter_video(re.findall(r'''(?xs)
                 flowplayer\("[^"]+",\s*
                     \{[^}]+?\}\s*,
-                    \s*{[^}]+? ["']?clip["']?\s*:\s*\{\s*
+                    \s*\{[^}]+? ["']?clip["']?\s*:\s*\{\s*
                         ["']?url["']?\s*:\s*["']([^"']+)["']
             ''', webpage))
         if not found:

youtube_dl/extractor/globo.py

@@ -70,6 +70,19 @@ class GloboIE(InfoExtractor):
                 'like_count': int,
             }
         },
+        {
+            'url': 'http://globotv.globo.com/canal-brasil/sangue-latino/t/todos-os-videos/v/ator-e-diretor-argentino-ricado-darin-fala-sobre-utopias-e-suas-perdas/3928201/',
+            'md5': 'c1defca721ce25b2354e927d3e4b3dec',
+            'info_dict': {
+                'id': '3928201',
+                'ext': 'mp4',
+                'title': 'Ator e diretor argentino, Ricado Darín fala sobre utopias e suas perdas',
+                'duration': 1472.906,
+                'uploader': 'Canal Brasil',
+                'uploader_id': 705,
+                'like_count': int,
+            }
+        },
     ]
 
     class MD5():
@@ -381,11 +394,16 @@ class GloboIE(InfoExtractor):
             signed_md5 = self.MD5.b64_md5(received_md5 + compat_str(sign_time) + padding)
             signed_hash = hash_code + compat_str(received_time) + received_random + compat_str(sign_time) + padding + signed_md5
 
-            formats.append({
-                'url': '%s?h=%s&k=%s' % (resource['url'], signed_hash, 'flash'),
-                'format_id': resource_id,
-                'height': resource['height']
-            })
+            resource_url = resource['url']
+            signed_url = '%s?h=%s&k=%s' % (resource_url, signed_hash, 'flash')
+            if resource_id.endswith('m3u8') or resource_url.endswith('.m3u8'):
+                formats.extend(self._extract_m3u8_formats(signed_url, resource_id, 'mp4'))
+            else:
+                formats.append({
+                    'url': signed_url,
+                    'format_id': resource_id,
+                    'height': resource.get('height'),
+                })
 
         self._sort_formats(formats)
 

youtube_dl/extractor/grooveshark.py

@@ -83,7 +83,7 @@ class GroovesharkIE(InfoExtractor):
         return compat_urlparse.urlunparse((uri.scheme, uri.netloc, obj['attrs']['data'], None, None, None))
 
     def _transform_bootstrap(self, js):
-        return re.split('(?m)^\s*try\s*{', js)[0] \
+        return re.split('(?m)^\s*try\s*\{', js)[0] \
                  .split(' = ', 1)[1].strip().rstrip(';')
 
     def _transform_meta(self, js):

youtube_dl/extractor/historicfilms.py

@@ -0,0 +1,46 @@
+from __future__ import unicode_literals
+
+from .common import InfoExtractor
+from ..utils import parse_duration
+
+
+class HistoricFilmsIE(InfoExtractor):
+    _VALID_URL = r'https?://(?:www\.)?historicfilms\.com/(?:tapes/|play)(?P<id>\d+)'
+    _TEST = {
+        'url': 'http://www.historicfilms.com/tapes/4728',
+        'md5': 'd4a437aec45d8d796a38a215db064e9a',
+        'info_dict': {
+            'id': '4728',
+            'ext': 'mov',
+            'title': 'Historic Films: GP-7',
+            'description': 'md5:1a86a0f3ac54024e419aba97210d959a',
+            'thumbnail': 're:^https?://.*\.jpg$',
+            'duration': 2096,
+        },
+    }
+
+    def _real_extract(self, url):
+        video_id = self._match_id(url)
+
+        webpage = self._download_webpage(url, video_id)
+
+        tape_id = self._search_regex(
+            r'class="tapeId">([^<]+)<', webpage, 'tape id')
+
+        title = self._og_search_title(webpage)
+        description = self._og_search_description(webpage)
+        thumbnail = self._html_search_meta(
+            'thumbnailUrl', webpage, 'thumbnails') or self._og_search_thumbnail(webpage)
+        duration = parse_duration(self._html_search_meta(
+            'duration', webpage, 'duration'))
+
+        video_url = 'http://www.historicfilms.com/video/%s_%s_web.mov' % (tape_id, video_id)
+
+        return {
+            'id': video_id,
+            'url': video_url,
+            'title': title,
+            'description': description,
+            'thumbnail': thumbnail,
+            'duration': duration,
+        }

youtube_dl/extractor/kankan.py

@@ -13,17 +13,17 @@ class KankanIE(InfoExtractor):
 
     _TEST = {
         'url': 'http://yinyue.kankan.com/vod/48/48863.shtml',
-        'file': '48863.flv',
         'md5': '29aca1e47ae68fc28804aca89f29507e',
         'info_dict': {
+            'id': '48863',
+            'ext': 'flv',
             'title': 'Ready To Go',
         },
         'skip': 'Only available from China',
     }
 
     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'(?:G_TITLE=|G_MOVIE_TITLE = )[\'"](.+?)[\'"]', webpage, 'video title')

youtube_dl/extractor/keezmovies.py

@@ -7,10 +7,6 @@ from .common import InfoExtractor
 from ..compat import (
     compat_urllib_parse_urlparse,
     compat_urllib_request,
-    compat_urllib_parse,
-)
-from ..aes import (
-    aes_decrypt_text
 )
 
 
@@ -18,9 +14,10 @@ class KeezMoviesIE(InfoExtractor):
     _VALID_URL = r'https?://(?:www\.)?keezmovies\.com/video/.+?(?P<id>[0-9]+)(?:[/?&]|$)'
     _TEST = {
         'url': 'http://www.keezmovies.com/video/petite-asian-lady-mai-playing-in-bathtub-1214711',
-        'file': '1214711.mp4',
         'md5': '6e297b7e789329923fcf83abb67c9289',
         'info_dict': {
+            'id': '1214711',
+            'ext': 'mp4',
             'title': 'Petite Asian Lady Mai Playing In Bathtub',
             'age_limit': 18,
         }
@@ -39,11 +36,10 @@ class KeezMoviesIE(InfoExtractor):
             embedded_url = mobj.group(1)
             return self.url_result(embedded_url)
 
-        video_title = self._html_search_regex(r'<h1 [^>]*>([^<]+)', webpage, 'title')
-        video_url = compat_urllib_parse.unquote(self._html_search_regex(r'video_url=(.+?)&amp;', webpage, 'video_url'))
-        if 'encrypted=true' in webpage:
-            password = self._html_search_regex(r'video_title=(.+?)&amp;', webpage, 'password')
-            video_url = aes_decrypt_text(video_url, password, 32).decode('utf-8')
+        video_title = self._html_search_regex(
+            r'<h1 [^>]*>([^<]+)', webpage, 'title')
+        video_url = self._html_search_regex(
+            r'(?s)html5VideoPlayer = .*?src="([^"]+)"', webpage, 'video URL')
         path = compat_urllib_parse_urlparse(video_url).path
         extension = os.path.splitext(path)[1][1:]
         format = path.split('/')[4].split('_')[:2]

youtube_dl/extractor/la7.py

@@ -1,7 +1,5 @@
 from __future__ import unicode_literals
 
-import re
-
 from .common import InfoExtractor
 from ..utils import (
     parse_duration,
@@ -20,9 +18,10 @@ class LA7IE(InfoExtractor):
 
     _TEST = {
         'url': 'http://www.la7.tv/richplayer/?assetid=50355319',
-        'file': '50355319.mp4',
         'md5': 'ec7d1f0224d20ba293ab56cf2259651f',
         'info_dict': {
+            'id': '50355319',
+            'ext': 'mp4',
             'title': 'IL DIVO',
             'description': 'Un film di Paolo Sorrentino con Toni Servillo, Anna Bonaiuto, Giulio Bosetti  e Flavio Bucci',
             'duration': 6254,
@@ -31,9 +30,7 @@ class LA7IE(InfoExtractor):
     }
 
     def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
-        video_id = mobj.group('id')
-
+        video_id = self._match_id(url)
         xml_url = 'http://www.la7.tv/repliche/content/index.php?contentId=%s' % video_id
         doc = self._download_xml(xml_url, video_id)
 

youtube_dl/extractor/macgamestore.py

@@ -1,7 +1,5 @@
 from __future__ import unicode_literals
 
-import re
-
 from .common import InfoExtractor
 from ..utils import ExtractorError
 
@@ -13,21 +11,22 @@ class MacGameStoreIE(InfoExtractor):
 
     _TEST = {
         'url': 'http://www.macgamestore.com/mediaviewer.php?trailer=2450',
-        'file': '2450.m4v',
         'md5': '8649b8ea684b6666b4c5be736ecddc61',
         'info_dict': {
+            'id': '2450',
+            'ext': 'm4v',
             'title': 'Crow',
         }
     }
 
     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, 'Downloading trailer page')
 
-        webpage = self._download_webpage(url, video_id, 'Downloading trailer page')
-
-        if re.search(r'>Missing Media<', webpage) is not None:
-            raise ExtractorError('Trailer %s does not exist' % video_id, expected=True)
+        if '>Missing Media<' in webpage:
+            raise ExtractorError(
+                'Trailer %s does not exist' % video_id, expected=True)
 
         video_title = self._html_search_regex(
             r'<title>MacGameStore: (.*?) Trailer</title>', webpage, 'title')

youtube_dl/extractor/mixcloud.py

@@ -9,7 +9,7 @@ from ..compat import (
 from ..utils import (
     ExtractorError,
     HEADRequest,
-    int_or_none,
+    str_to_int,
     parse_iso8601,
 )
 
@@ -85,15 +85,17 @@ class MixcloudIE(InfoExtractor):
         uploader_id = self._search_regex(
             r'\s+"profile": "([^"]+)",', webpage, 'uploader id', fatal=False)
         description = self._og_search_description(webpage)
-        like_count = int_or_none(self._search_regex(
-            r'<meta itemprop="interactionCount" content="UserLikes:([0-9]+)"',
+        like_count = str_to_int(self._search_regex(
+            [r'<meta itemprop="interactionCount" content="UserLikes:([0-9]+)"',
+             r'/favorites/?">([0-9]+)<'],
             webpage, 'like count', fatal=False))
-        view_count = int_or_none(self._search_regex(
-            r'<meta itemprop="interactionCount" content="UserPlays:([0-9]+)"',
+        view_count = str_to_int(self._search_regex(
+            [r'<meta itemprop="interactionCount" content="UserPlays:([0-9]+)"',
+             r'/listeners/?">([0-9,.]+)</a>'],
             webpage, 'play count', fatal=False))
         timestamp = parse_iso8601(self._search_regex(
             r'<time itemprop="dateCreated" datetime="([^"]+)">',
-            webpage, 'upload date'))
+            webpage, 'upload date', default=None))
 
         return {
             'id': track_id,

youtube_dl/extractor/mpora.py

@@ -1,21 +1,19 @@
 from __future__ import unicode_literals
 
-import json
-import re
-
 from .common import InfoExtractor
 from ..utils import int_or_none
 
 
 class MporaIE(InfoExtractor):
-    _VALID_URL = r'^https?://(www\.)?mpora\.(?:com|de)/videos/(?P<id>[^?#/]+)'
+    _VALID_URL = r'https?://(www\.)?mpora\.(?:com|de)/videos/(?P<id>[^?#/]+)'
     IE_NAME = 'MPORA'
 
     _TEST = {
         'url': 'http://mpora.de/videos/AAdo8okx4wiz/embed?locale=de',
-        'file': 'AAdo8okx4wiz.mp4',
         'md5': 'a7a228473eedd3be741397cf452932eb',
         'info_dict': {
+            'id': 'AAdo8okx4wiz',
+            'ext': 'mp4',
             'title': 'Katy Curd -  Winter in the Forest',
             'duration': 416,
             'uploader': 'Peter Newman Media',
@@ -23,14 +21,12 @@ class MporaIE(InfoExtractor):
     }
 
     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)
+
         data_json = self._search_regex(
             r"new FM\.Player\('[^']+',\s*(\{.*?)\).player;", webpage, 'json')
-
-        data = json.loads(data_json)
+        data = self._parse_json(data_json, video_id)
 
         uploader = data['info_overlay'].get('username')
         duration = data['video']['duration'] // 1000

youtube_dl/extractor/mtv.py

@@ -2,10 +2,11 @@ from __future__ import unicode_literals
 
 import re
 
-from .common import InfoExtractor
+from .subtitles import SubtitlesInfoExtractor
 from ..compat import (
     compat_urllib_parse,
     compat_urllib_request,
+    compat_str,
 )
 from ..utils import (
     ExtractorError,
@@ -22,7 +23,7 @@ def _media_xml_tag(tag):
     return '{http://search.yahoo.com/mrss/}%s' % tag
 
 
-class MTVServicesInfoExtractor(InfoExtractor):
+class MTVServicesInfoExtractor(SubtitlesInfoExtractor):
     _MOBILE_TEMPLATE = None
 
     @staticmethod
@@ -78,17 +79,42 @@ class MTVServicesInfoExtractor(InfoExtractor):
             try:
                 _, _, ext = rendition.attrib['type'].partition('/')
                 rtmp_video_url = rendition.find('./src').text
-                formats.append({'ext': ext,
-                                'url': self._transform_rtmp_url(rtmp_video_url),
-                                'format_id': rendition.get('bitrate'),
-                                'width': int(rendition.get('width')),
-                                'height': int(rendition.get('height')),
-                                })
+                if rtmp_video_url.endswith('siteunavail.png'):
+                    continue
+                formats.append({
+                    'ext': ext,
+                    'url': self._transform_rtmp_url(rtmp_video_url),
+                    'format_id': rendition.get('bitrate'),
+                    'width': int(rendition.get('width')),
+                    'height': int(rendition.get('height')),
+                })
             except (KeyError, TypeError):
                 raise ExtractorError('Invalid rendition field.')
         self._sort_formats(formats)
         return formats
 
+    def _extract_subtitles(self, mdoc, mtvn_id):
+        subtitles = {}
+        FORMATS = {
+            'scc': 'cea-608',
+            'eia-608': 'cea-608',
+            'xml': 'ttml',
+        }
+        subtitles_format = FORMATS.get(
+            self._downloader.params.get('subtitlesformat'), 'ttml')
+        for transcript in mdoc.findall('.//transcript'):
+            if transcript.get('kind') != 'captions':
+                continue
+            lang = transcript.get('srclang')
+            for typographic in transcript.findall('./typographic'):
+                captions_format = typographic.get('format')
+                if captions_format == subtitles_format:
+                    subtitles[lang] = compat_str(typographic.get('src'))
+                    break
+        if self._downloader.params.get('listsubtitles', False):
+            self._list_available_subtitles(mtvn_id, subtitles)
+        return self.extract_subtitles(mtvn_id, subtitles)
+
     def _get_video_info(self, itemdoc):
         uri = itemdoc.find('guid').text
         video_id = self._id_from_uri(uri)
@@ -135,6 +161,7 @@ class MTVServicesInfoExtractor(InfoExtractor):
         return {
             'title': title,
             'formats': self._extract_video_formats(mediagen_doc, mtvn_id),
+            'subtitles': self._extract_subtitles(mediagen_doc, mtvn_id),
             'id': video_id,
             'thumbnail': self._get_thumbnail_url(uri, itemdoc),
             'description': description,
@@ -167,7 +194,11 @@ class MTVServicesInfoExtractor(InfoExtractor):
             mgid = self._search_regex(
                 [r'data-mgid="(.*?)"', r'swfobject.embedSWF\(".*?(mgid:.*?)"'],
                 webpage, 'mgid')
-        return self._get_videos_info(mgid)
+
+        videos_info = self._get_videos_info(mgid)
+        if self._downloader.params.get('listsubtitles', False):
+            return
+        return videos_info
 
 
 class MTVServicesEmbeddedIE(MTVServicesInfoExtractor):
@@ -212,25 +243,14 @@ class MTVIE(MTVServicesInfoExtractor):
     _TESTS = [
         {
             'url': 'http://www.mtv.com/videos/misc/853555/ours-vh1-storytellers.jhtml',
-            'file': '853555.mp4',
             'md5': '850f3f143316b1e71fa56a4edfd6e0f8',
             'info_dict': {
+                'id': '853555',
+                'ext': 'mp4',
                 'title': 'Taylor Swift - "Ours (VH1 Storytellers)"',
                 'description': 'Album: Taylor Swift performs "Ours" for VH1 Storytellers at Harvey Mudd College.',
             },
         },
-        {
-            'add_ie': ['Vevo'],
-            'url': 'http://www.mtv.com/videos/taylor-swift/916187/everything-has-changed-ft-ed-sheeran.jhtml',
-            'file': 'USCJY1331283.mp4',
-            'md5': '73b4e7fcadd88929292fe52c3ced8caf',
-            'info_dict': {
-                'title': 'Everything Has Changed',
-                'upload_date': '20130606',
-                'uploader': 'Taylor Swift',
-            },
-            'skip': 'VEVO is only available in some countries',
-        },
     ]
 
     def _get_thumbnail_url(self, uri, itemdoc):
@@ -244,8 +264,8 @@ class MTVIE(MTVServicesInfoExtractor):
             webpage = self._download_webpage(url, video_id)
 
             # Some videos come from Vevo.com
-            m_vevo = re.search(r'isVevoVideo = true;.*?vevoVideoId = "(.*?)";',
-                               webpage, re.DOTALL)
+            m_vevo = re.search(
+                r'(?s)isVevoVideo = true;.*?vevoVideoId = "(.*?)";', webpage)
             if m_vevo:
                 vevo_id = m_vevo.group(1)
                 self.to_screen('Vevo video detected: %s' % vevo_id)

youtube_dl/extractor/nerdcubed.py

@@ -11,6 +11,7 @@ class NerdCubedFeedIE(InfoExtractor):
     _TEST = {
         'url': 'http://www.nerdcubed.co.uk/feed.json',
         'info_dict': {
+            'id': 'nerdcubed-feed',
             'title': 'nerdcubed.co.uk feed',
         },
         'playlist_mincount': 1300,

youtube_dl/extractor/nhl.py

@@ -20,6 +20,12 @@ class NHLBaseInfoExtractor(InfoExtractor):
     def _fix_json(json_string):
         return json_string.replace('\\\'', '\'')
 
+    def _real_extract_video(self, video_id):
+        json_url = 'http://video.nhl.com/videocenter/servlets/playlist?ids=%s&format=json' % video_id
+        data = self._download_json(
+            json_url, video_id, transform_source=self._fix_json)
+        return self._extract_video(data[0])
+
     def _extract_video(self, info):
         video_id = info['id']
         self.report_extraction(video_id)
@@ -54,7 +60,7 @@ class NHLBaseInfoExtractor(InfoExtractor):
 
 class NHLIE(NHLBaseInfoExtractor):
     IE_NAME = 'nhl.com'
-    _VALID_URL = r'https?://video(?P<team>\.[^.]*)?\.nhl\.com/videocenter/console(?:\?(?:.*?[?&])?)id=(?P<id>[-0-9a-zA-Z]+)'
+    _VALID_URL = r'https?://video(?P<team>\.[^.]*)?\.nhl\.com/videocenter/(?:console)?(?:\?(?:.*?[?&])?)id=(?P<id>[-0-9a-zA-Z]+)'
 
     _TESTS = [{
         'url': 'http://video.canucks.nhl.com/videocenter/console?catid=6?id=453614',
@@ -92,15 +98,41 @@ class NHLIE(NHLBaseInfoExtractor):
     }, {
         'url': 'http://video.flames.nhl.com/videocenter/console?id=630616',
         'only_matching': True,
+    }, {
+        'url': 'http://video.nhl.com/videocenter/?id=736722',
+        'only_matching': True,
     }]
 
     def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
-        video_id = mobj.group('id')
-        json_url = 'http://video.nhl.com/videocenter/servlets/playlist?ids=%s&format=json' % video_id
-        data = self._download_json(
-            json_url, video_id, transform_source=self._fix_json)
-        return self._extract_video(data[0])
+        video_id = self._match_id(url)
+        return self._real_extract_video(video_id)
+
+
+class NHLNewsIE(NHLBaseInfoExtractor):
+    IE_NAME = 'nhl.com:news'
+    IE_DESC = 'NHL news'
+    _VALID_URL = r'https?://(?:www\.)?nhl\.com/ice/news\.html?(?:\?(?:.*?[?&])?)id=(?P<id>[-0-9a-zA-Z]+)'
+
+    _TEST = {
+        'url': 'http://www.nhl.com/ice/news.htm?id=750727',
+        'md5': '4b3d1262e177687a3009937bd9ec0be8',
+        'info_dict': {
+            'id': '736722',
+            'ext': 'mp4',
+            'title': 'Cal Clutterbuck has been fined $2,000',
+            'description': 'md5:45fe547d30edab88b23e0dd0ab1ed9e6',
+            'duration': 37,
+            'upload_date': '20150128',
+        },
+    }
+
+    def _real_extract(self, url):
+        news_id = self._match_id(url)
+        webpage = self._download_webpage(url, news_id)
+        video_id = self._search_regex(
+            [r'pVid(\d+)', r"nlid\s*:\s*'(\d+)'"],
+            webpage, 'video id')
+        return self._real_extract_video(video_id)
 
 
 class NHLVideocenterIE(NHLBaseInfoExtractor):

youtube_dl/extractor/ntvde.py

@@ -0,0 +1,68 @@
+# encoding: utf-8
+from __future__ import unicode_literals
+
+from .common import InfoExtractor
+from ..utils import (
+    int_or_none,
+    js_to_json,
+    parse_duration,
+)
+
+
+class NTVDeIE(InfoExtractor):
+    IE_NAME = 'n-tv.de'
+    _VALID_URL = r'https?://(?:www\.)?n-tv\.de/mediathek/videos/[^/?#]+/[^/?#]+-article(?P<id>.+)\.html'
+
+    _TESTS = [{
+        'url': 'http://www.n-tv.de/mediathek/videos/panorama/Schnee-und-Glaette-fuehren-zu-zahlreichen-Unfaellen-und-Staus-article14438086.html',
+        'md5': '6ef2514d4b1e8e03ca24b49e2f167153',
+        'info_dict': {
+            'id': '14438086',
+            'ext': 'mp4',
+            'thumbnail': 're:^https?://.*\.jpg$',
+            'title': 'Schnee und Glätte führen zu zahlreichen Unfällen und Staus',
+            'alt_title': 'Winterchaos auf deutschen Straßen',
+            'description': 'Schnee und Glätte sorgen deutschlandweit für einen chaotischen Start in die Woche: Auf den Straßen kommt es zu kilometerlangen Staus und Dutzenden Glätteunfällen. In Düsseldorf und München wirbelt der Schnee zudem den Flugplan durcheinander. Dutzende Flüge landen zu spät, einige fallen ganz aus.',
+            'duration': 4020,
+            'timestamp': 1422892797,
+            'upload_date': '20150202',
+        },
+    }]
+
+    def _real_extract(self, url):
+        video_id = self._match_id(url)
+        webpage = self._download_webpage(url, video_id)
+
+        info = self._parse_json(self._search_regex(
+            r'(?s)ntv.pageInfo.article =\s(\{.*?\});', webpage, 'info'),
+            video_id, transform_source=js_to_json)
+        timestamp = int_or_none(info.get('publishedDateAsUnixTimeStamp'))
+        vdata = self._parse_json(self._search_regex(
+            r'(?s)\$\(\s*"\#player"\s*\)\s*\.data\(\s*"player",\s*(\{.*?\})\);',
+            webpage, 'player data'),
+            video_id, transform_source=js_to_json)
+        duration = parse_duration(vdata.get('duration'))
+        formats = [{
+            'format_id': 'flash',
+            'url': 'rtmp://fms.n-tv.de/' + vdata['video'],
+        }, {
+            'format_id': 'mobile',
+            'url': 'http://video.n-tv.de' + vdata['videoMp4'],
+            'tbr': 400,  # estimation
+        }]
+        m3u8_url = 'http://video.n-tv.de' + vdata['videoM3u8']
+        formats.extend(self._extract_m3u8_formats(
+            m3u8_url, video_id, ext='mp4',
+            entry_protocol='m3u8_native', preference=0))
+        self._sort_formats(formats)
+
+        return {
+            'id': video_id,
+            'title': info['headline'],
+            'description': info.get('intro'),
+            'alt_title': info.get('kicker'),
+            'timestamp': timestamp,
+            'thumbnail': vdata.get('html5VideoPoster'),
+            'duration': duration,
+            'formats': formats,
+        }

youtube_dl/extractor/ntvru.py

@@ -1,15 +1,14 @@
 # encoding: utf-8
 from __future__ import unicode_literals
 
-import re
-
 from .common import InfoExtractor
 from ..utils import (
     unescapeHTML
 )
 
 
-class NTVIE(InfoExtractor):
+class NTVRuIE(InfoExtractor):
+    IE_NAME = 'ntv.ru'
     _VALID_URL = r'http://(?:www\.)?ntv\.ru/(?P<id>.+)'
 
     _TESTS = [
@@ -92,9 +91,7 @@ class NTVIE(InfoExtractor):
     ]
 
     def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
-        video_id = mobj.group('id')
-
+        video_id = self._match_id(url)
         page = self._download_webpage(url, video_id)
 
         video_id = self._html_search_regex(self._VIDEO_ID_REGEXES, page, 'video id')

youtube_dl/extractor/ringtv.py

@@ -6,12 +6,13 @@ from .common import InfoExtractor
 
 
 class RingTVIE(InfoExtractor):
-    _VALID_URL = r'(?:http://)?(?:www\.)?ringtv\.craveonline\.com/(?P<type>news|videos/video)/(?P<id>[^/?#]+)'
+    _VALID_URL = r'http://(?:www\.)?ringtv\.craveonline\.com/(?P<type>news|videos/video)/(?P<id>[^/?#]+)'
     _TEST = {
         "url": "http://ringtv.craveonline.com/news/310833-luis-collazo-says-victor-ortiz-better-not-quit-on-jan-30",
-        "file": "857645.mp4",
         "md5": "d25945f5df41cdca2d2587165ac28720",
         "info_dict": {
+            'id': '857645',
+            'ext': 'mp4',
             "title": 'Video: Luis Collazo says Victor Ortiz "better not quit on Jan. 30" - Ring TV',
             "description": 'Luis Collazo is excited about his Jan. 30 showdown with fellow former welterweight titleholder Victor Ortiz at Barclays Center in his hometown of Brooklyn. The SuperBowl week fight headlines a Golden Boy Live! card on Fox Sports 1.',
         }

youtube_dl/extractor/rottentomatoes.py

@@ -10,8 +10,9 @@ class RottenTomatoesIE(VideoDetectiveIE):
 
     _TEST = {
         'url': 'http://www.rottentomatoes.com/m/toy_story_3/trailers/11028566/',
-        'file': '613340.mp4',
         'info_dict': {
+            'id': '613340',
+            'ext': 'mp4',
             'title': 'TOY STORY 3',
             'description': 'From the creators of the beloved TOY STORY films, comes a story that will reunite the gang in a whole new way.',
         },

youtube_dl/extractor/rtve.py

@@ -57,7 +57,7 @@ def _decrypt_url(png):
 class RTVEALaCartaIE(InfoExtractor):
     IE_NAME = 'rtve.es:alacarta'
     IE_DESC = 'RTVE a la carta'
-    _VALID_URL = r'http://www\.rtve\.es/alacarta/videos/[^/]+/[^/]+/(?P<id>\d+)'
+    _VALID_URL = r'http://www\.rtve\.es/(m/)?alacarta/videos/[^/]+/[^/]+/(?P<id>\d+)'
 
     _TESTS = [{
         'url': 'http://www.rtve.es/alacarta/videos/balonmano/o-swiss-cup-masculina-final-espana-suecia/2491869/',
@@ -74,7 +74,11 @@ class RTVEALaCartaIE(InfoExtractor):
             'id': '1694255',
             'ext': 'flv',
             'title': 'TODO',
-        }
+        },
+        'skip': 'The f4m manifest can\'t be used yet',
+    }, {
+        'url': 'http://www.rtve.es/m/alacarta/videos/cuentame-como-paso/cuentame-como-paso-t16-ultimo-minuto-nuestra-vida-capitulo-276/2969138/?media=tve',
+        'only_matching': True,
     }]
 
     def _real_extract(self, url):
@@ -86,6 +90,18 @@ class RTVEALaCartaIE(InfoExtractor):
         png_url = 'http://www.rtve.es/ztnr/movil/thumbnail/default/videos/%s.png' % video_id
         png = self._download_webpage(png_url, video_id, 'Downloading url information')
         video_url = _decrypt_url(png)
+        if not video_url.endswith('.f4m'):
+            auth_url = video_url.replace(
+                'resources/', 'auth/resources/'
+            ).replace('.net.rtve', '.multimedia.cdn.rtve')
+            video_path = self._download_webpage(
+                auth_url, video_id, 'Getting video url')
+            # Use mvod.akcdn instead of flash.akamaihd.multimedia.cdn to get
+            # the right Content-Length header and the mp4 format
+            video_url = (
+                'http://mvod.akcdn.rtve.es/{0}&v=2.6.8'
+                '&fp=MAC%2016,0,0,296&r=MRUGG&g=OEOJWFXNFGCP'.format(video_path)
+            )
 
         return {
             'id': video_id,

youtube_dl/extractor/rutv.py

@@ -162,10 +162,8 @@ class RUTVIE(InfoExtractor):
                         'vbr': int(quality),
                     }
                 elif transport == 'm3u8':
-                    fmt = {
-                        'url': url,
-                        'ext': 'mp4',
-                    }
+                    formats.extend(self._extract_m3u8_formats(url, video_id, 'mp4'))
+                    continue
                 else:
                     fmt = {
                         'url': url

youtube_dl/extractor/servingsys.py

@@ -1,7 +1,5 @@
 from __future__ import unicode_literals
 
-import re
-
 from .common import InfoExtractor
 from ..utils import (
     int_or_none,
@@ -13,10 +11,15 @@ class ServingSysIE(InfoExtractor):
 
     _TEST = {
         'url': 'http://bs.serving-sys.com/BurstingPipe/adServer.bs?cn=is&c=23&pl=VAST&pli=5349193&PluID=0&pos=7135&ord=[timestamp]&cim=1?',
+        'info_dict': {
+            'id': '5349193',
+            'title': 'AdAPPter_Hyundai_demo',
+        },
         'playlist': [{
-            'file': '29955898.flv',
             'md5': 'baed851342df6846eb8677a60a011a0f',
             'info_dict': {
+                'id': '29955898',
+                'ext': 'flv',
                 'title': 'AdAPPter_Hyundai_demo (1)',
                 'duration': 74,
                 'tbr': 1378,
@@ -24,9 +27,10 @@ class ServingSysIE(InfoExtractor):
                 'height': 400,
             },
         }, {
-            'file': '29907998.flv',
             'md5': '979b4da2655c4bc2d81aeb915a8c5014',
             'info_dict': {
+                'id': '29907998',
+                'ext': 'flv',
                 'title': 'AdAPPter_Hyundai_demo (2)',
                 'duration': 34,
                 'width': 854,
@@ -37,14 +41,13 @@ class ServingSysIE(InfoExtractor):
         'params': {
             'playlistend': 2,
         },
-        'skip': 'Blocked in the US [sic]',
+        '_skip': 'Blocked in the US [sic]',
     }
 
     def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
-        pl_id = mobj.group('id')
-
+        pl_id = self._match_id(url)
         vast_doc = self._download_xml(url, pl_id)
+
         title = vast_doc.find('.//AdTitle').text
         media = vast_doc.find('.//MediaFile').text
         info_url = self._search_regex(r'&adData=([^&]+)&', media, 'info URL')

youtube_dl/extractor/sina.py

@@ -11,7 +11,7 @@ from ..compat import (
 
 
 class SinaIE(InfoExtractor):
-    _VALID_URL = r'''https?://(.*?\.)?video\.sina\.com\.cn/
+    _VALID_URL = r'''(?x)https?://(.*?\.)?video\.sina\.com\.cn/
                         (
                             (.+?/(((?P<pseudo_id>\d+).html)|(.*?(\#|(vid=)|b/)(?P<id>\d+?)($|&|\-))))
                             |
@@ -23,9 +23,10 @@ class SinaIE(InfoExtractor):
     _TESTS = [
         {
             'url': 'http://video.sina.com.cn/news/vlist/zt/chczlj2013/?opsubject_id=top12#110028898',
-            'file': '110028898.flv',
             'md5': 'd65dd22ddcf44e38ce2bf58a10c3e71f',
             'info_dict': {
+                'id': '110028898',
+                'ext': 'flv',
                 'title': '《中国新闻》 朝鲜要求巴拿马立即释放被扣船员',
             }
         },
@@ -39,10 +40,6 @@ class SinaIE(InfoExtractor):
         },
     ]
 
-    @classmethod
-    def suitable(cls, url):
-        return re.match(cls._VALID_URL, url, flags=re.VERBOSE) is not None
-
     def _extract_video(self, video_id):
         data = compat_urllib_parse.urlencode({'vid': video_id})
         url_doc = self._download_xml('http://v.iask.com/v_play.php?%s' % data,
@@ -59,7 +56,7 @@ class SinaIE(InfoExtractor):
                 }
 
     def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url, flags=re.VERBOSE)
+        mobj = re.match(self._VALID_URL, url)
         video_id = mobj.group('id')
         if mobj.group('token') is not None:
             # The video id is in the redirected url

youtube_dl/extractor/smotri.py

@@ -108,7 +108,7 @@ class SmotriIE(InfoExtractor):
         # swf player
         {
             'url': 'http://pics.smotri.com/scrubber_custom8.swf?file=v9188090500',
-            'md5': '4d47034979d9390d14acdf59c4935bc2',
+            'md5': '31099eeb4bc906712c5f40092045108d',
             'info_dict': {
                 'id': 'v9188090500',
                 'ext': 'mp4',
@@ -139,9 +139,6 @@ class SmotriIE(InfoExtractor):
     def _search_meta(self, name, html, display_name=None):
         if display_name is None:
             display_name = name
-        return self._html_search_regex(
-            r'<meta itemprop="%s" content="([^"]+)" />' % re.escape(name),
-            html, display_name, fatal=False)
         return self._html_search_meta(name, html, display_name)
 
     def _real_extract(self, url):

youtube_dl/extractor/soundcloud.py

@@ -246,6 +246,7 @@ class SoundcloudSetIE(SoundcloudIE):
     _TESTS = [{
         'url': 'https://soundcloud.com/the-concept-band/sets/the-royal-concept-ep',
         'info_dict': {
+            'id': '2284613',
             'title': 'The Royal Concept EP',
         },
         'playlist_mincount': 6,
@@ -279,7 +280,7 @@ class SoundcloudSetIE(SoundcloudIE):
         return {
             '_type': 'playlist',
             'entries': [self._extract_info_dict(track, secret_token=token) for track in info['tracks']],
-            'id': info['id'],
+            'id': '%s' % info['id'],
             'title': info['title'],
         }
 

youtube_dl/extractor/spike.py

@@ -1,14 +1,12 @@
 from __future__ import unicode_literals
 
-import re
-
 from .mtv import MTVServicesInfoExtractor
 
 
 class SpikeIE(MTVServicesInfoExtractor):
     _VALID_URL = r'''(?x)https?://
-        (www\.spike\.com/(video-clips|episodes)/.+|
-         m\.spike\.com/videos/video.rbml\?id=(?P<mobile_id>[^&]+))
+        (?:www\.spike\.com/(?:video-clips|(?:full-)?episodes)/.+|
+         m\.spike\.com/videos/video\.rbml\?id=(?P<id>[^&]+))
         '''
     _TEST = {
         'url': 'http://www.spike.com/video-clips/lhtu8m/auction-hunters-can-allen-ride-a-hundred-year-old-motorcycle',
@@ -25,8 +23,7 @@ class SpikeIE(MTVServicesInfoExtractor):
     _MOBILE_TEMPLATE = 'http://m.spike.com/videos/video.rbml?id=%s'
 
     def _real_extract(self, url):
-        mobj = re.search(self._VALID_URL, url)
-        mobile_id = mobj.group('mobile_id')
-        if mobile_id is not None:
+        mobile_id = self._match_id(url)
+        if mobile_id:
             url = 'http://www.spike.com/video-clips/%s' % mobile_id
         return super(SpikeIE, self)._real_extract(url)

youtube_dl/extractor/teamcoco.py

@@ -10,17 +10,19 @@ class TeamcocoIE(InfoExtractor):
     _TESTS = [
         {
             'url': 'http://teamcoco.com/video/80187/conan-becomes-a-mary-kay-beauty-consultant',
-            'file': '80187.mp4',
             'md5': '3f7746aa0dc86de18df7539903d399ea',
             'info_dict': {
+                'id': '80187',
+                'ext': 'mp4',
                 'title': 'Conan Becomes A Mary Kay Beauty Consultant',
                 'description': 'Mary Kay is perhaps the most trusted name in female beauty, so of course Conan is a natural choice to sell their products.'
             }
         }, {
             'url': 'http://teamcoco.com/video/louis-ck-interview-george-w-bush',
-            'file': '19705.mp4',
             'md5': 'cde9ba0fa3506f5f017ce11ead928f9a',
             'info_dict': {
+                'id': '19705',
+                'ext': 'mp4',
                 "description": "Louis C.K. got starstruck by George W. Bush, so what? Part one.",
                 "title": "Louis C.K. Interview Pt. 1 11/3/11"
             }
@@ -36,7 +38,7 @@ class TeamcocoIE(InfoExtractor):
         video_id = mobj.group("video_id")
         if not video_id:
             video_id = self._html_search_regex(
-                r'data-node-id="(\d+?)"',
+                r'<div\s+class="player".*?data-id="(\d+?)"',
                 webpage, 'video id')
 
         data_url = 'http://teamcoco.com/cvp/2.0/%s.xml' % video_id

youtube_dl/extractor/teletask.py

@@ -11,6 +11,7 @@ class TeleTaskIE(InfoExtractor):
     _TEST = {
         'url': 'http://www.tele-task.de/archive/video/html5/26168/',
         'info_dict': {
+            'id': '26168',
             'title': 'Duplicate Detection',
         },
         'playlist': [{
@@ -34,7 +35,6 @@ class TeleTaskIE(InfoExtractor):
 
     def _real_extract(self, url):
         lecture_id = self._match_id(url)
-
         webpage = self._download_webpage(url, lecture_id)
 
         title = self._html_search_regex(

youtube_dl/extractor/toutv.py

@@ -16,8 +16,9 @@ class TouTvIE(InfoExtractor):
 
     _TEST = {
         'url': 'http://www.tou.tv/30-vies/S04E41',
-        'file': '30-vies_S04E41.mp4',
         'info_dict': {
+            'id': '30-vies_S04E41',
+            'ext': 'mp4',
             'title': '30 vies Saison 4 / Épisode 41',
             'description': 'md5:da363002db82ccbe4dafeb9cab039b09',
             'age_limit': 8,

youtube_dl/extractor/vevo.py

@@ -9,6 +9,7 @@ from ..compat import (
 )
 from ..utils import (
     ExtractorError,
+    int_or_none,
 )
 
 
@@ -192,9 +193,29 @@ class VevoIE(InfoExtractor):
         # Download via HLS API
         formats.extend(self._download_api_formats(video_id))
 
+        # Download SMIL
+        smil_blocks = sorted((
+            f for f in video_info['videoVersions']
+            if f['sourceType'] == 13),
+            key=lambda f: f['version'])
+        smil_url = '%s/Video/V2/VFILE/%s/%sr.smil' % (
+            self._SMIL_BASE_URL, video_id, video_id.lower())
+        if smil_blocks:
+            smil_url_m = self._search_regex(
+                r'url="([^"]+)"', smil_blocks[-1]['data'], 'SMIL URL',
+                default=None)
+            if smil_url_m is not None:
+                smil_url = smil_url_m
+        if smil_url:
+            smil_xml = self._download_webpage(
+                smil_url, video_id, 'Downloading SMIL info', fatal=False)
+            if smil_xml:
+                formats.extend(self._formats_from_smil(smil_xml))
+
         self._sort_formats(formats)
-        timestamp_ms = int(self._search_regex(
-            r'/Date\((\d+)\)/', video_info['launchDate'], 'launch date'))
+        timestamp_ms = int_or_none(self._search_regex(
+            r'/Date\((\d+)\)/',
+            video_info['launchDate'], 'launch date', fatal=False))
 
         return {
             'id': video_id,

youtube_dl/extractor/vimeo.py

@@ -501,9 +501,10 @@ class VimeoReviewIE(InfoExtractor):
     _VALID_URL = r'https?://vimeo\.com/[^/]+/review/(?P<id>[^/]+)'
     _TESTS = [{
         'url': 'https://vimeo.com/user21297594/review/75524534/3c257a1b5d',
-        'file': '75524534.mp4',
         'md5': 'c507a72f780cacc12b2248bb4006d253',
         'info_dict': {
+            'id': '75524534',
+            'ext': 'mp4',
             'title': "DICK HARDWICK 'Comedian'",
             'uploader': 'Richard Hardwick',
         }

youtube_dl/extractor/washingtonpost.py

@@ -1,3 +1,4 @@
+# coding: utf-8
 from __future__ import unicode_literals
 
 import re
@@ -11,9 +12,10 @@ from ..utils import (
 
 class WashingtonPostIE(InfoExtractor):
     _VALID_URL = r'https?://(?:www\.)?washingtonpost\.com/.*?/(?P<id>[^/]+)/(?:$|[?#])'
-    _TEST = {
+    _TESTS = [{
         'url': 'http://www.washingtonpost.com/sf/national/2014/03/22/sinkhole-of-bureaucracy/',
         'info_dict': {
+            'id': 'sinkhole-of-bureaucracy',
             'title': 'Sinkhole of bureaucracy',
         },
         'playlist': [{
@@ -40,15 +42,38 @@ class WashingtonPostIE(InfoExtractor):
                 'upload_date': '20140322',
                 'uploader': 'The Washington Post',
             },
+        }],
+    }, {
+        'url': 'http://www.washingtonpost.com/blogs/wonkblog/wp/2014/12/31/one-airline-figured-out-how-to-make-sure-its-airplanes-never-disappear/',
+        'info_dict': {
+            'id': 'one-airline-figured-out-how-to-make-sure-its-airplanes-never-disappear',
+            'title': 'One airline figured out how to make sure its airplanes never disappear',
+        },
+        'playlist': [{
+            'md5': 'a7c1b5634ba5e57a6a82cdffa5b1e0d0',
+            'info_dict': {
+                'id': '0e4bb54c-9065-11e4-a66f-0ca5037a597d',
+                'ext': 'mp4',
+                'description': 'Washington Post transportation reporter Ashley Halsey III explains why a plane\'s black box needs to be recovered from a crash site instead of having its information streamed in real time throughout the flight.',
+                'upload_date': '20141230',
+                'uploader': 'The Washington Post',
+                'timestamp': 1419974765,
+                'title': 'Why black boxes don’t transmit data in real time',
+            }
         }]
-    }
+    }]
 
     def _real_extract(self, url):
         page_id = self._match_id(url)
         webpage = self._download_webpage(url, page_id)
 
         title = self._og_search_title(webpage)
-        uuids = re.findall(r'data-video-uuid="([^"]+)"', webpage)
+
+        uuids = re.findall(r'''(?x)
+            (?:
+                <div\s+class="posttv-video-embed[^>]*?data-uuid=|
+                data-video-uuid=
+            )"([^"]+)"''', webpage)
         entries = []
         for i, uuid in enumerate(uuids, start=1):
             vinfo_all = self._download_json(
@@ -75,10 +100,11 @@ class WashingtonPostIE(InfoExtractor):
                 'filesize': s.get('fileSize'),
                 'url': s.get('url'),
                 'ext': 'mp4',
+                'preference': -100 if s.get('type') == 'smil' else None,
                 'protocol': {
                     'MP4': 'http',
                     'F4F': 'f4m',
-                }.get(s.get('type'))
+                }.get(s.get('type')),
             } for s in vinfo.get('streams', [])]
             source_media_url = vinfo.get('sourceMediaURL')
             if source_media_url:

youtube_dl/extractor/wdr.py

@@ -71,6 +71,9 @@ class WDRIE(InfoExtractor):
         {
             'url': 'http://www1.wdr.de/mediathek/video/sendungen/quarks_und_co/filterseite-quarks-und-co100.html',
             'playlist_mincount': 146,
+            'info_dict': {
+                'id': 'mediathek/video/sendungen/quarks_und_co/filterseite-quarks-und-co100',
+            }
         }
     ]
 

youtube_dl/extractor/youtube.py

@@ -809,6 +809,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor):
             player_url = None
 
         # Get video info
+        embed_webpage = None
         if re.search(r'player-age-gate-content">', video_webpage) is not None:
             age_gate = True
             # We simulate the access to the video from www.youtube.com/v/{video_id}
@@ -1016,10 +1017,21 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor):
                     url += '&signature=' + url_data['sig'][0]
                 elif 's' in url_data:
                     encrypted_sig = url_data['s'][0]
+                    ASSETS_RE = r'"assets":.+?"js":\s*("[^"]+")'
 
                     jsplayer_url_json = self._search_regex(
-                        r'"assets":.+?"js":\s*("[^"]+")',
-                        embed_webpage if age_gate else video_webpage, 'JS player URL')
+                        ASSETS_RE,
+                        embed_webpage if age_gate else video_webpage,
+                        'JS player URL (1)', default=None)
+                    if not jsplayer_url_json and not age_gate:
+                        # We need the embed website after all
+                        if embed_webpage is None:
+                            embed_url = proto + '://www.youtube.com/embed/%s' % video_id
+                            embed_webpage = self._download_webpage(
+                                embed_url, video_id, 'Downloading embed webpage')
+                        jsplayer_url_json = self._search_regex(
+                            ASSETS_RE, embed_webpage, 'JS player URL')
+
                     player_url = json.loads(jsplayer_url_json)
                     if player_url is None:
                         player_url_json = self._search_regex(
@@ -1148,6 +1160,7 @@ class YoutubePlaylistIE(YoutubeBaseInfoExtractor):
     }, {
         'url': 'https://www.youtube.com/playlist?list=PLtPgu7CB4gbZDA7i_euNxn75ISqxwZPYx',
         'info_dict': {
+            'id': 'PLtPgu7CB4gbZDA7i_euNxn75ISqxwZPYx',
             'title': 'YDL_Empty_List',
         },
         'playlist_count': 0,
@@ -1156,6 +1169,7 @@ class YoutubePlaylistIE(YoutubeBaseInfoExtractor):
         'url': 'https://www.youtube.com/playlist?list=PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
         'info_dict': {
             'title': '29C3: Not my department',
+            'id': 'PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
         },
         'playlist_count': 95,
     }, {
@@ -1163,6 +1177,7 @@ class YoutubePlaylistIE(YoutubeBaseInfoExtractor):
         'url': 'PLBB231211A4F62143',
         'info_dict': {
             'title': '[OLD]Team Fortress 2 (Class-based LP)',
+            'id': 'PLBB231211A4F62143',
         },
         'playlist_mincount': 26,
     }, {
@@ -1170,12 +1185,14 @@ class YoutubePlaylistIE(YoutubeBaseInfoExtractor):
         'url': 'https://www.youtube.com/playlist?list=UUBABnxM4Ar9ten8Mdjj1j0Q',
         'info_dict': {
             'title': 'Uploads from Cauchemar',
+            'id': 'UUBABnxM4Ar9ten8Mdjj1j0Q',
         },
         'playlist_mincount': 799,
     }, {
         'url': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
         'info_dict': {
             'title': 'YDL_safe_search',
+            'id': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
         },
         'playlist_count': 2,
     }, {
@@ -1184,6 +1201,7 @@ class YoutubePlaylistIE(YoutubeBaseInfoExtractor):
         'playlist_count': 4,
         'info_dict': {
             'title': 'JODA15',
+            'id': 'PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
         }
     }, {
         'note': 'Embedded SWF player',
@@ -1191,12 +1209,14 @@ class YoutubePlaylistIE(YoutubeBaseInfoExtractor):
         'playlist_count': 4,
         'info_dict': {
             'title': 'JODA7',
+            'id': 'YN5VISEtHet5D4NEvfTd0zcgFk84NqFZ',
         }
     }, {
         'note': 'Buggy playlist: the webpage has a "Load more" button but it doesn\'t have more videos',
         'url': 'https://www.youtube.com/playlist?list=UUXw-G3eDE9trcvY2sBMM_aA',
         'info_dict': {
-                'title': 'Uploads from Interstellar Movie',
+            'title': 'Uploads from Interstellar Movie',
+            'id': 'UUXw-G3eDE9trcvY2sBMM_aA',
         },
         'playlist_mincout': 21,
     }]
@@ -1302,6 +1322,9 @@ class YoutubeChannelIE(InfoExtractor):
         'note': 'paginated channel',
         'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
         'playlist_mincount': 91,
+        'info_dict': {
+            'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
+        }
     }]
 
     def extract_videos_from_page(self, page):

youtube_dl/jsinterp.py

@@ -1,59 +1,122 @@
 from __future__ import unicode_literals
 
 import json
+import operator
 import re
 
 from .utils import (
     ExtractorError,
 )
 
+_OPERATORS = [
+    ('|', operator.or_),
+    ('^', operator.xor),
+    ('&', operator.and_),
+    ('>>', operator.rshift),
+    ('<<', operator.lshift),
+    ('-', operator.sub),
+    ('+', operator.add),
+    ('%', operator.mod),
+    ('/', operator.truediv),
+    ('*', operator.mul),
+]
+_ASSIGN_OPERATORS = [(op + '=', opfunc) for op, opfunc in _OPERATORS]
+_ASSIGN_OPERATORS.append(('=', lambda cur, right: right))
+
+_NAME_RE = r'[a-zA-Z_$][a-zA-Z_$0-9]*'
+
 
 class JSInterpreter(object):
-    def __init__(self, code):
-        self.code = code
+    def __init__(self, code, objects=None):
+        if objects is None:
+            objects = {}
+        self.code = self._remove_comments(code)
         self._functions = {}
-        self._objects = {}
+        self._objects = objects
 
-    def interpret_statement(self, stmt, local_vars, allow_recursion=20):
+    def _remove_comments(self, code):
+        return re.sub(r'(?s)/\*.*?\*/', '', code)
+
+    def interpret_statement(self, stmt, local_vars, allow_recursion=100):
         if allow_recursion < 0:
             raise ExtractorError('Recursion limit reached')
 
-        if stmt.startswith('var '):
-            stmt = stmt[len('var '):]
-        ass_m = re.match(r'^(?P<out>[a-z]+)(?:\[(?P<index>[^\]]+)\])?' +
-                         r'=(?P<expr>.*)$', stmt)
-        if ass_m:
-            if ass_m.groupdict().get('index'):
-                def assign(val):
-                    lvar = local_vars[ass_m.group('out')]
-                    idx = self.interpret_expression(
-                        ass_m.group('index'), local_vars, allow_recursion)
-                    assert isinstance(idx, int)
-                    lvar[idx] = val
-                    return val
-                expr = ass_m.group('expr')
-            else:
-                def assign(val):
-                    local_vars[ass_m.group('out')] = val
-                    return val
-                expr = ass_m.group('expr')
-        elif stmt.startswith('return '):
-            assign = lambda v: v
-            expr = stmt[len('return '):]
+        should_abort = False
+        stmt = stmt.lstrip()
+        stmt_m = re.match(r'var\s', stmt)
+        if stmt_m:
+            expr = stmt[len(stmt_m.group(0)):]
         else:
-            # Try interpreting it as an expression
-            expr = stmt
-            assign = lambda v: v
+            return_m = re.match(r'return(?:\s+|$)', stmt)
+            if return_m:
+                expr = stmt[len(return_m.group(0)):]
+                should_abort = True
+            else:
+                # Try interpreting it as an expression
+                expr = stmt
 
         v = self.interpret_expression(expr, local_vars, allow_recursion)
-        return assign(v)
+        return v, should_abort
 
     def interpret_expression(self, expr, local_vars, allow_recursion):
+        expr = expr.strip()
+
+        if expr == '':  # Empty expression
+            return None
+
+        if expr.startswith('('):
+            parens_count = 0
+            for m in re.finditer(r'[()]', expr):
+                if m.group(0) == '(':
+                    parens_count += 1
+                else:
+                    parens_count -= 1
+                    if parens_count == 0:
+                        sub_expr = expr[1:m.start()]
+                        sub_result = self.interpret_expression(
+                            sub_expr, local_vars, allow_recursion)
+                        remaining_expr = expr[m.end():].strip()
+                        if not remaining_expr:
+                            return sub_result
+                        else:
+                            expr = json.dumps(sub_result) + remaining_expr
+                        break
+            else:
+                raise ExtractorError('Premature end of parens in %r' % expr)
+
+        for op, opfunc in _ASSIGN_OPERATORS:
+            m = re.match(r'''(?x)
+                (?P<out>%s)(?:\[(?P<index>[^\]]+?)\])?
+                \s*%s
+                (?P<expr>.*)$''' % (_NAME_RE, re.escape(op)), expr)
+            if not m:
+                continue
+            right_val = self.interpret_expression(
+                m.group('expr'), local_vars, allow_recursion - 1)
+
+            if m.groupdict().get('index'):
+                lvar = local_vars[m.group('out')]
+                idx = self.interpret_expression(
+                    m.group('index'), local_vars, allow_recursion)
+                assert isinstance(idx, int)
+                cur = lvar[idx]
+                val = opfunc(cur, right_val)
+                lvar[idx] = val
+                return val
+            else:
+                cur = local_vars.get(m.group('out'))
+                val = opfunc(cur, right_val)
+                local_vars[m.group('out')] = val
+                return val
+
         if expr.isdigit():
             return int(expr)
 
-        if expr.isalpha():
-            return local_vars[expr]
+        var_m = re.match(
+            r'(?!if|return|true|false)(?P<name>%s)$' % _NAME_RE,
+            expr)
+        if var_m:
+            return local_vars[var_m.group('name')]
 
         try:
             return json.loads(expr)
@@ -61,7 +124,7 @@ class JSInterpreter(object):
             pass
 
         m = re.match(
-            r'^(?P<var>[$a-zA-Z0-9_]+)\.(?P<member>[^(]+)(?:\(+(?P<args>[^()]*)\))?$',
+            r'(?P<var>%s)\.(?P<member>[^(]+)(?:\(+(?P<args>[^()]*)\))?$' % _NAME_RE,
             expr)
         if m:
             variable = m.group('var')
@@ -114,23 +177,31 @@ class JSInterpreter(object):
             return obj[member](argvals)
 
         m = re.match(
-            r'^(?P<in>[a-z]+)\[(?P<idx>.+)\]$', expr)
+            r'(?P<in>%s)\[(?P<idx>.+)\]$' % _NAME_RE, expr)
         if m:
             val = local_vars[m.group('in')]
             idx = self.interpret_expression(
                 m.group('idx'), local_vars, allow_recursion - 1)
             return val[idx]
 
-        m = re.match(r'^(?P<a>.+?)(?P<op>[%])(?P<b>.+?)$', expr)
-        if m:
-            a = self.interpret_expression(
-                m.group('a'), local_vars, allow_recursion)
-            b = self.interpret_expression(
-                m.group('b'), local_vars, allow_recursion)
-            return a % b
+        for op, opfunc in _OPERATORS:
+            m = re.match(r'(?P<x>.+?)%s(?P<y>.+)' % re.escape(op), expr)
+            if not m:
+                continue
+            x, abort = self.interpret_statement(
+                m.group('x'), local_vars, allow_recursion - 1)
+            if abort:
+                raise ExtractorError(
+                    'Premature left-side return of %s in %r' % (op, expr))
+            y, abort = self.interpret_statement(
+                m.group('y'), local_vars, allow_recursion - 1)
+            if abort:
+                raise ExtractorError(
+                    'Premature right-side return of %s in %r' % (op, expr))
+            return opfunc(x, y)
 
         m = re.match(
-            r'^(?P<func>[a-zA-Z$]+)\((?P<args>[a-z0-9,]+)\)$', expr)
+            r'^(?P<func>%s)\((?P<args>[a-zA-Z0-9_$,]+)\)$' % _NAME_RE, expr)
         if m:
             fname = m.group('func')
             argvals = tuple([
@@ -139,6 +210,7 @@ class JSInterpreter(object):
             if fname not in self._functions:
                 self._functions[fname] = self.extract_function(fname)
             return self._functions[fname](argvals)
+
         raise ExtractorError('Unsupported JS expression %r' % expr)
 
     def extract_object(self, objname):
@@ -162,9 +234,11 @@ class JSInterpreter(object):
 
     def extract_function(self, funcname):
         func_m = re.search(
-            (r'(?:function %s|[{;]%s\s*=\s*function)' % (
-                re.escape(funcname), re.escape(funcname))) +
-            r'\((?P<args>[a-z,]+)\){(?P<code>[^}]+)}',
+            r'''(?x)
+                (?:function\s+%s|[{;]%s\s*=\s*function)\s*
+                \((?P<args>[^)]*)\)\s*
+                \{(?P<code>[^}]+)\}''' % (
+                re.escape(funcname), re.escape(funcname)),
             self.code)
         if func_m is None:
             raise ExtractorError('Could not find JS function %r' % funcname)
@@ -172,10 +246,16 @@ class JSInterpreter(object):
 
         return self.build_function(argnames, func_m.group('code'))
 
+    def call_function(self, funcname, *args):
+        f = self.extract_function(funcname)
+        return f(args)
+
     def build_function(self, argnames, code):
         def resf(args):
             local_vars = dict(zip(argnames, args))
             for stmt in code.split(';'):
-                res = self.interpret_statement(stmt, local_vars)
+                res, abort = self.interpret_statement(stmt, local_vars)
+                if abort:
+                    break
             return res
         return resf

youtube_dl/options.py

@@ -698,10 +698,9 @@ def parseOpts(overrideArguments=None):
     postproc.add_option(
         '--fixup',
         metavar='POLICY', dest='fixup', default='detect_or_warn',
-        help='(experimental) Automatically correct known faults of the file. '
+        help='Automatically correct known faults of the file. '
              'One of never (do nothing), warn (only emit a warning), '
-             'detect_or_warn(check whether we can do anything about it, warn '
-             'otherwise')
+             'detect_or_warn(the default; fix file if we can, warn otherwise)')
     postproc.add_option(
         '--prefer-avconv',
         action='store_false', dest='prefer_ffmpeg',

youtube_dl/utils.py

@@ -32,6 +32,7 @@ import xml.etree.ElementTree
 import zlib
 
 from .compat import (
+    compat_basestring,
     compat_chr,
     compat_getenv,
     compat_html_entities,
@@ -140,7 +141,7 @@ else:
     def find_xpath_attr(node, xpath, key, val):
         # Here comes the crazy part: In 2.6, if the xpath is a unicode,
         # .//node does not match if a node is a direct child of . !
-        if isinstance(xpath, unicode):
+        if isinstance(xpath, compat_str):
             xpath = xpath.encode('ascii')
 
         for f in node.findall(xpath):
@@ -1262,7 +1263,7 @@ def float_or_none(v, scale=1, invscale=1, default=None):
 
 
 def parse_duration(s):
-    if not isinstance(s, basestring if sys.version_info < (3, 0) else compat_str):
+    if not isinstance(s, compat_basestring):
         return None
 
     s = s.strip()
@@ -1274,7 +1275,10 @@ def parse_duration(s):
             (?P<only_hours>[0-9.]+)\s*(?:hours?)|
 
             (?:
-                (?:(?P<hours>[0-9]+)\s*(?:[:h]|hours?)\s*)?
+                (?:
+                    (?:(?P<days>[0-9]+)\s*(?:[:d]|days?)\s*)?
+                    (?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?)?
@@ -1292,6 +1296,8 @@ def parse_duration(s):
         res += int(m.group('mins')) * 60
     if m.group('hours'):
         res += int(m.group('hours')) * 60 * 60
+    if m.group('days'):
+        res += int(m.group('days')) * 24 * 60 * 60
     if m.group('ms'):
         res += float(m.group('ms'))
     return res
@@ -1426,7 +1432,7 @@ def uppercase_escape(s):
 
 def escape_rfc3986(s):
     """Escape non-ASCII characters as suggested by RFC 3986"""
-    if sys.version_info < (3, 0) and isinstance(s, unicode):
+    if sys.version_info < (3, 0) and isinstance(s, compat_str):
         s = s.encode('utf-8')
     return compat_urllib_parse.quote(s, b"%/;:@&=+$,!~*'()?#[]")
 
@@ -1542,7 +1548,7 @@ def js_to_json(code):
     res = re.sub(r'''(?x)
         "(?:[^"\\]*(?:\\\\|\\")?)*"|
         '(?:[^'\\]*(?:\\\\|\\')?)*'|
-        [a-zA-Z_][a-zA-Z_0-9]*
+        [a-zA-Z_][.a-zA-Z_0-9]*
         ''', fix_kv, code)
     res = re.sub(r',(\s*\])', lambda m: m.group(1), res)
     return res

youtube_dl/version.py

@@ -1,3 +1,3 @@
 from __future__ import unicode_literals
 
-__version__ = '2015.01.30.1'
+__version__ = '2015.02.02.3'