release 2015.09.09

[xuite] fix test md5

.travis.yml

@@ -5,9 +5,7 @@ python:
   - "3.2"
   - "3.3"
   - "3.4"
-before_install:
-  - sudo apt-get update -qq
-  - sudo apt-get install -yqq rtmpdump
+sudo: false
 script: nosetests test --verbose
 notifications:
   email:

AUTHORS

@@ -140,3 +140,6 @@ Behrouz Abbasi
 ngld
 nyuszika7h
 Shaun Walbridge
+Lee Jenkins
+Anssi Hannula
+Lukáš Lalinský

CONTRIBUTING.md

@@ -125,7 +125,7 @@ If you want to add support for a new site, you can follow this quick list (assum
     ```
 5. Add an import in [`youtube_dl/extractor/__init__.py`](https://github.com/rg3/youtube-dl/blob/master/youtube_dl/extractor/__init__.py).
 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 and may return](https://github.com/rg3/youtube-dl/blob/master/youtube_dl/extractor/common.py#L62-L200). Add tests and code for as many as you want.
+7. Have a look at [`youtube_dl/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 and may return](https://github.com/rg3/youtube-dl/blob/master/youtube_dl/extractor/common.py#L62-L200). Add tests and code for as many as you want.
 8. If you can, check the code with [flake8](https://pypi.python.org/pypi/flake8).
 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:
 

README.md

@@ -34,7 +34,7 @@ You can also use pip:
 
     sudo pip install youtube-dl
 
-Alternatively, refer to the developer instructions below for how to check out and work with the git repository. For further options, including PGP signatures, see https://rg3.github.io/youtube-dl/download.html .
+Alternatively, refer to the [developer instructions](#developer-instructions) for how to check out and work with the git repository. For further options, including PGP signatures, see https://rg3.github.io/youtube-dl/download.html .
 
 # DESCRIPTION
 **youtube-dl** is a small command-line program to download videos from
@@ -207,7 +207,7 @@ which means you can modify it, redistribute it or use it however you like.
     -p, --password PASSWORD          Account password. If this option is left out, youtube-dl will ask interactively.
     -2, --twofactor TWOFACTOR        Two-factor auth code
     -n, --netrc                      Use .netrc authentication data
-    --video-password PASSWORD        Video password (vimeo, smotri)
+    --video-password PASSWORD        Video password (vimeo, smotri, youku)
 
 ## Post-processing Options:
     -x, --extract-audio              Convert video files to audio-only files (requires ffmpeg or avconv and ffprobe or avprobe)
@@ -552,7 +552,7 @@ If you want to add support for a new site, you can follow this quick list (assum
     ```
 5. Add an import in [`youtube_dl/extractor/__init__.py`](https://github.com/rg3/youtube-dl/blob/master/youtube_dl/extractor/__init__.py).
 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 and may return](https://github.com/rg3/youtube-dl/blob/master/youtube_dl/extractor/common.py#L62-L200). Add tests and code for as many as you want.
+7. Have a look at [`youtube_dl/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 and may return](https://github.com/rg3/youtube-dl/blob/master/youtube_dl/extractor/common.py#L62-L200). Add tests and code for as many as you want.
 8. If you can, check the code with [flake8](https://pypi.python.org/pypi/flake8).
 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:
 

docs/supportedsites.md

@@ -166,7 +166,7 @@
  - **Folketinget**: Folketinget (ft.dk; Danish parliament)
  - **FootyRoom**
  - **Foxgay**
- - **FoxNews**
+ - **FoxNews**: Fox News and Fox Business Video
  - **FoxSports**
  - **france2.fr:generation-quoi**
  - **FranceCulture**
@@ -195,7 +195,7 @@
  - **GodTube**
  - **GoldenMoustache**
  - **Golem**
- - **GorillaVid**: GorillaVid.in, daclips.in, movpod.in, fastvideo.in and realvid.net
+ - **GorillaVid**: GorillaVid.in, daclips.in, movpod.in, fastvideo.in, realvid.net and filehoot.com
  - **Goshgay**
  - **Groupon**
  - **Hark**
@@ -220,6 +220,7 @@
  - **imdb**: Internet Movie Database trailers
  - **imdb:list**: Internet Movie Database lists
  - **Imgur**
+ - **ImgurAlbum**
  - **Ina**
  - **Indavideo**
  - **IndavideoEmbed**
@@ -301,13 +302,15 @@
  - **Moviezine**
  - **movshare**: MovShare
  - **MPORA**
+ - **MSNBC**
  - **MTV**
+ - **mtv.de**
  - **mtviggy.com**
  - **mtvservices:embedded**
  - **MuenchenTV**: münchen.tv
  - **MusicPlayOn**
- - **MusicVault**
  - **muzu.tv**
+ - **Mwave**
  - **MySpace**
  - **MySpace:album**
  - **MySpass**
@@ -392,6 +395,8 @@
  - **Playtvak**: Playtvak.cz, iDNES.cz and Lidovky.cz
  - **Playvid**
  - **Playwire**
+ - **pluralsight**
+ - **pluralsight:course**
  - **plus.google**: Google Plus
  - **pluzz.francetv.fr**
  - **podomatic**
@@ -461,7 +466,7 @@
  - **Sexu**
  - **SexyKarma**: Sexy Karma and Watch Indian Porn
  - **Shahid**
- - **Shared**
+ - **Shared**: shared.sx and vivo.sx
  - **ShareSix**
  - **Sina**
  - **Slideshare**
@@ -534,6 +539,7 @@
  - **TF1**
  - **TheOnion**
  - **ThePlatform**
+ - **ThePlatformFeed**
  - **TheSixtyOne**
  - **ThisAmericanLife**
  - **ThisAV**
@@ -599,7 +605,6 @@
  - **Viddler**
  - **video.google:search**: Google Video search
  - **video.mit.edu**
- - **VideoBam**
  - **VideoDetective**
  - **videofy.me**
  - **videolectures.net**
@@ -627,6 +632,7 @@
  - **vine:user**
  - **vk**: VK
  - **vk:uservideos**: VK - User's Videos
+ - **vlive**
  - **Vodlocker**
  - **VoiceRepublic**
  - **Vporn**

test/test_compat.py

@@ -14,6 +14,7 @@ from youtube_dl.utils import get_filesystem_encoding
 from youtube_dl.compat import (
     compat_getenv,
     compat_expanduser,
+    compat_shlex_split,
     compat_urllib_parse_unquote,
     compat_urllib_parse_unquote_plus,
 )
@@ -67,5 +68,8 @@ class TestCompat(unittest.TestCase):
         self.assertEqual(compat_urllib_parse_unquote_plus('abc%20def'), 'abc def')
         self.assertEqual(compat_urllib_parse_unquote_plus('%7e/abc+def'), '~/abc def')
 
+    def test_compat_shlex_split(self):
+        self.assertEqual(compat_shlex_split('-option "one two"'), ['-option', 'one two'])
+
 if __name__ == '__main__':
     unittest.main()

test/test_subtitles.py

@@ -25,6 +25,7 @@ from youtube_dl.extractor import (
     RaiIE,
     VikiIE,
     ThePlatformIE,
+    ThePlatformFeedIE,
     RTVEALaCartaIE,
     FunnyOrDieIE,
 )
@@ -307,6 +308,18 @@ class TestThePlatformSubtitles(BaseTestSubtitles):
         self.assertEqual(md5(subtitles['en']), '97e7670cbae3c4d26ae8bcc7fdd78d4b')
 
 
+class TestThePlatformFeedSubtitles(BaseTestSubtitles):
+    url = 'http://feed.theplatform.com/f/7wvmTC/msnbc_video-p-test?form=json&pretty=true&range=-40&byGuid=n_hardball_5biden_140207'
+    IE = ThePlatformFeedIE
+
+    def test_allsubtitles(self):
+        self.DL.params['writesubtitles'] = True
+        self.DL.params['allsubtitles'] = True
+        subtitles = self.getSubtitles()
+        self.assertEqual(set(subtitles.keys()), set(['en']))
+        self.assertEqual(md5(subtitles['en']), '48649a22e82b2da21c9a67a395eedade')
+
+
 class TestRtveSubtitles(BaseTestSubtitles):
     url = 'http://www.rtve.es/alacarta/videos/los-misterios-de-laura/misterios-laura-capitulo-32-misterio-del-numero-17-2-parte/2428621/'
     IE = RTVEALaCartaIE

test/test_utils.py

@@ -57,11 +57,16 @@ from youtube_dl.utils import (
     urlencode_postdata,
     version_tuple,
     xpath_with_ns,
+    xpath_element,
     xpath_text,
+    xpath_attr,
     render_table,
     match_str,
     parse_dfxp_time_expr,
     dfxp2srt,
+    cli_option,
+    cli_valueless_option,
+    cli_bool_option,
 )
 
 
@@ -264,6 +269,16 @@ class TestUtil(unittest.TestCase):
         self.assertEqual(find('media:song/media:author').text, 'The Author')
         self.assertEqual(find('media:song/url').text, 'http://server.com/download.mp3')
 
+    def test_xpath_element(self):
+        doc = xml.etree.ElementTree.Element('root')
+        div = xml.etree.ElementTree.SubElement(doc, 'div')
+        p = xml.etree.ElementTree.SubElement(div, 'p')
+        p.text = 'Foo'
+        self.assertEqual(xpath_element(doc, 'div/p'), p)
+        self.assertEqual(xpath_element(doc, 'div/bar', default='default'), 'default')
+        self.assertTrue(xpath_element(doc, 'div/bar') is None)
+        self.assertRaises(ExtractorError, xpath_element, doc, 'div/bar', fatal=True)
+
     def test_xpath_text(self):
         testxml = '''<root>
             <div>
@@ -272,9 +287,25 @@ class TestUtil(unittest.TestCase):
         </root>'''
         doc = xml.etree.ElementTree.fromstring(testxml)
         self.assertEqual(xpath_text(doc, 'div/p'), 'Foo')
+        self.assertEqual(xpath_text(doc, 'div/bar', default='default'), 'default')
         self.assertTrue(xpath_text(doc, 'div/bar') is None)
         self.assertRaises(ExtractorError, xpath_text, doc, 'div/bar', fatal=True)
 
+    def test_xpath_attr(self):
+        testxml = '''<root>
+            <div>
+                <p x="a">Foo</p>
+            </div>
+        </root>'''
+        doc = xml.etree.ElementTree.fromstring(testxml)
+        self.assertEqual(xpath_attr(doc, 'div/p', 'x'), 'a')
+        self.assertEqual(xpath_attr(doc, 'div/bar', 'x'), None)
+        self.assertEqual(xpath_attr(doc, 'div/p', 'y'), None)
+        self.assertEqual(xpath_attr(doc, 'div/bar', 'x', default='default'), 'default')
+        self.assertEqual(xpath_attr(doc, 'div/p', 'y', default='default'), 'default')
+        self.assertRaises(ExtractorError, xpath_attr, doc, 'div/bar', 'x', fatal=True)
+        self.assertRaises(ExtractorError, xpath_attr, doc, 'div/p', 'y', fatal=True)
+
     def test_smuggle_url(self):
         data = {"ö": "ö", "abc": [3]}
         url = 'https://foo.bar/baz?x=y#a'
@@ -646,6 +677,51 @@ The first line
 '''
         self.assertEqual(dfxp2srt(dfxp_data_no_default_namespace), srt_data)
 
+    def test_cli_option(self):
+        self.assertEqual(cli_option({'proxy': '127.0.0.1:3128'}, '--proxy', 'proxy'), ['--proxy', '127.0.0.1:3128'])
+        self.assertEqual(cli_option({'proxy': None}, '--proxy', 'proxy'), [])
+        self.assertEqual(cli_option({}, '--proxy', 'proxy'), [])
+
+    def test_cli_valueless_option(self):
+        self.assertEqual(cli_valueless_option(
+            {'downloader': 'external'}, '--external-downloader', 'downloader', 'external'), ['--external-downloader'])
+        self.assertEqual(cli_valueless_option(
+            {'downloader': 'internal'}, '--external-downloader', 'downloader', 'external'), [])
+        self.assertEqual(cli_valueless_option(
+            {'nocheckcertificate': True}, '--no-check-certificate', 'nocheckcertificate'), ['--no-check-certificate'])
+        self.assertEqual(cli_valueless_option(
+            {'nocheckcertificate': False}, '--no-check-certificate', 'nocheckcertificate'), [])
+        self.assertEqual(cli_valueless_option(
+            {'checkcertificate': True}, '--no-check-certificate', 'checkcertificate', False), [])
+        self.assertEqual(cli_valueless_option(
+            {'checkcertificate': False}, '--no-check-certificate', 'checkcertificate', False), ['--no-check-certificate'])
+
+    def test_cli_bool_option(self):
+        self.assertEqual(
+            cli_bool_option(
+                {'nocheckcertificate': True}, '--no-check-certificate', 'nocheckcertificate'),
+            ['--no-check-certificate', 'true'])
+        self.assertEqual(
+            cli_bool_option(
+                {'nocheckcertificate': True}, '--no-check-certificate', 'nocheckcertificate', separator='='),
+            ['--no-check-certificate=true'])
+        self.assertEqual(
+            cli_bool_option(
+                {'nocheckcertificate': True}, '--check-certificate', 'nocheckcertificate', 'false', 'true'),
+            ['--check-certificate', 'false'])
+        self.assertEqual(
+            cli_bool_option(
+                {'nocheckcertificate': True}, '--check-certificate', 'nocheckcertificate', 'false', 'true', '='),
+            ['--check-certificate=false'])
+        self.assertEqual(
+            cli_bool_option(
+                {'nocheckcertificate': False}, '--check-certificate', 'nocheckcertificate', 'false', 'true'),
+            ['--check-certificate', 'true'])
+        self.assertEqual(
+            cli_bool_option(
+                {'nocheckcertificate': False}, '--check-certificate', 'nocheckcertificate', 'false', 'true', '='),
+            ['--check-certificate=true'])
+
 
 if __name__ == '__main__':
     unittest.main()

test/test_write_annotations.py

@@ -33,7 +33,7 @@ params = get_params({
 
 
 TEST_ID = 'gr51aVj-mLg'
-ANNOTATIONS_FILE = TEST_ID + '.flv.annotations.xml'
+ANNOTATIONS_FILE = TEST_ID + '.annotations.xml'
 EXPECTED_ANNOTATIONS = ['Speech bubble', 'Note', 'Title', 'Spotlight', 'Label']
 
 

youtube_dl/YoutubeDL.py

@@ -69,6 +69,7 @@ from .utils import (
     version_tuple,
     write_json_file,
     write_string,
+    YoutubeDLCookieProcessor,
     YoutubeDLHandler,
     prepend_extension,
     replace_extension,
@@ -284,7 +285,11 @@ class YoutubeDL(object):
         self._num_downloads = 0
         self._screen_file = [sys.stdout, sys.stderr][params.get('logtostderr', False)]
         self._err_file = sys.stderr
-        self.params = params
+        self.params = {
+            # Default parameters
+            'nocheckcertificate': False,
+        }
+        self.params.update(params)
         self.cache = Cache(self)
 
         if params.get('bidi_workaround', False):
@@ -1939,8 +1944,7 @@ class YoutubeDL(object):
             if os.access(opts_cookiefile, os.R_OK):
                 self.cookiejar.load()
 
-        cookie_processor = compat_urllib_request.HTTPCookieProcessor(
-            self.cookiejar)
+        cookie_processor = YoutubeDLCookieProcessor(self.cookiejar)
         if opts_proxy is not None:
             if opts_proxy == '':
                 proxies = {}
@@ -2009,7 +2013,7 @@ class YoutubeDL(object):
                                (info_dict['extractor'], info_dict['id'], thumb_display_id))
                 try:
                     uf = self.urlopen(t['url'])
-                    with open(thumb_filename, 'wb') as thumbf:
+                    with open(encodeFilename(thumb_filename), 'wb') as thumbf:
                         shutil.copyfileobj(uf, thumbf)
                     self.to_screen('[%s] %s: Writing thumbnail %sto: %s' %
                                    (info_dict['extractor'], info_dict['id'], thumb_display_id, thumb_filename))

youtube_dl/__init__.py

@@ -9,7 +9,6 @@ import codecs
 import io
 import os
 import random
-import shlex
 import sys
 
 
@@ -20,6 +19,7 @@ from .compat import (
     compat_expanduser,
     compat_getpass,
     compat_print,
+    compat_shlex_split,
     workaround_optparse_bug9161,
 )
 from .utils import (
@@ -262,10 +262,10 @@ def _real_main(argv=None):
             parser.error('setting filesize xattr requested but python-xattr is not available')
     external_downloader_args = None
     if opts.external_downloader_args:
-        external_downloader_args = shlex.split(opts.external_downloader_args)
+        external_downloader_args = compat_shlex_split(opts.external_downloader_args)
     postprocessor_args = None
     if opts.postprocessor_args:
-        postprocessor_args = shlex.split(opts.postprocessor_args)
+        postprocessor_args = compat_shlex_split(opts.postprocessor_args)
     match_filter = (
         None if opts.match_filter is None
         else match_filter_func(opts.match_filter))

youtube_dl/compat.py

@@ -5,6 +5,7 @@ import getpass
 import optparse
 import os
 import re
+import shlex
 import shutil
 import socket
 import subprocess
@@ -227,6 +228,17 @@ except ImportError:  # Python < 3.3
             return "'" + s.replace("'", "'\"'\"'") + "'"
 
 
+if sys.version_info >= (2, 7, 3):
+    compat_shlex_split = shlex.split
+else:
+    # Working around shlex issue with unicode strings on some python 2
+    # versions (see http://bugs.python.org/issue1548891)
+    def compat_shlex_split(s, comments=False, posix=True):
+        if isinstance(s, unicode):
+            s = s.encode('utf-8')
+        return shlex.split(s, comments, posix)
+
+
 def compat_ord(c):
     if type(c) is int:
         return c
@@ -459,6 +471,7 @@ __all__ = [
     'compat_ord',
     'compat_parse_qs',
     'compat_print',
+    'compat_shlex_split',
     'compat_socket_create_connection',
     'compat_str',
     'compat_subprocess_get_DEVNULL',

youtube_dl/downloader/external.py

@@ -5,6 +5,10 @@ import subprocess
 
 from .common import FileDownloader
 from ..utils import (
+    cli_option,
+    cli_valueless_option,
+    cli_bool_option,
+    cli_configuration_args,
     encodeFilename,
     encodeArgument,
 )
@@ -46,19 +50,16 @@ class ExternalFD(FileDownloader):
         return info_dict['protocol'] in ('http', 'https', 'ftp', 'ftps')
 
     def _option(self, command_option, param):
-        param = self.params.get(param)
-        if param is None:
-            return []
-        if isinstance(param, bool):
-            return [command_option]
-        return [command_option, param]
+        return cli_option(self.params, command_option, param)
+
+    def _bool_option(self, command_option, param, true_value='true', false_value='false', separator=None):
+        return cli_bool_option(self.params, command_option, param, true_value, false_value, separator)
+
+    def _valueless_option(self, command_option, param, expected_value=True):
+        return cli_valueless_option(self.params, command_option, param, expected_value)
 
     def _configuration_args(self, default=[]):
-        ex_args = self.params.get('external_downloader_args')
-        if ex_args is None:
-            return default
-        assert isinstance(ex_args, list)
-        return ex_args
+        return cli_configuration_args(self.params, 'external_downloader_args', default)
 
     def _call_downloader(self, tmpfilename, info_dict):
         """ Either overwrite this or implement _make_cmd """
@@ -80,6 +81,8 @@ class CurlFD(ExternalFD):
         for key, val in info_dict['http_headers'].items():
             cmd += ['--header', '%s: %s' % (key, val)]
         cmd += self._option('--interface', 'source_address')
+        cmd += self._option('--proxy', 'proxy')
+        cmd += self._valueless_option('--insecure', 'nocheckcertificate')
         cmd += self._configuration_args()
         cmd += ['--', info_dict['url']]
         return cmd
@@ -102,7 +105,7 @@ class WgetFD(ExternalFD):
             cmd += ['--header', '%s: %s' % (key, val)]
         cmd += self._option('--bind-address', 'source_address')
         cmd += self._option('--proxy', 'proxy')
-        cmd += self._option('--no-check-certificate', 'nocheckcertificate')
+        cmd += self._valueless_option('--no-check-certificate', 'nocheckcertificate')
         cmd += self._configuration_args()
         cmd += ['--', info_dict['url']]
         return cmd
@@ -121,6 +124,7 @@ class Aria2cFD(ExternalFD):
             cmd += ['--header', '%s: %s' % (key, val)]
         cmd += self._option('--interface', 'source_address')
         cmd += self._option('--all-proxy', 'proxy')
+        cmd += self._bool_option('--check-certificate', 'nocheckcertificate', 'false', 'true', '=')
         cmd += ['--', info_dict['url']]
         return cmd
 

youtube_dl/downloader/f4m.py

@@ -13,6 +13,8 @@ from ..compat import (
     compat_urllib_error,
 )
 from ..utils import (
+    encodeFilename,
+    sanitize_open,
     struct_pack,
     struct_unpack,
     xpath_text,
@@ -343,18 +345,19 @@ class F4mFD(FragmentFD):
                 success = ctx['dl'].download(frag_filename, {'url': url})
                 if not success:
                     return False
-                with open(frag_filename, 'rb') as down:
-                    down_data = down.read()
-                    reader = FlvReader(down_data)
-                    while True:
-                        _, box_type, box_data = reader.read_box_info()
-                        if box_type == b'mdat':
-                            dest_stream.write(box_data)
-                            break
+                (down, frag_sanitized) = sanitize_open(frag_filename, 'rb')
+                down_data = down.read()
+                down.close()
+                reader = FlvReader(down_data)
+                while True:
+                    _, box_type, box_data = reader.read_box_info()
+                    if box_type == b'mdat':
+                        dest_stream.write(box_data)
+                        break
                 if live:
-                    os.remove(frag_filename)
+                    os.remove(encodeFilename(frag_sanitized))
                 else:
-                    frags_filenames.append(frag_filename)
+                    frags_filenames.append(frag_sanitized)
             except (compat_urllib_error.HTTPError, ) as err:
                 if live and (err.code == 404 or err.code == 410):
                     # We didn't keep up with the live window. Continue
@@ -375,6 +378,6 @@ class F4mFD(FragmentFD):
         self._finish_frag_download(ctx)
 
         for frag_file in frags_filenames:
-            os.remove(frag_file)
+            os.remove(encodeFilename(frag_file))
 
         return True

youtube_dl/downloader/hls.py

@@ -12,6 +12,7 @@ from ..postprocessor.ffmpeg import FFmpegPostProcessor
 from ..utils import (
     encodeArgument,
     encodeFilename,
+    sanitize_open,
 )
 
 
@@ -89,13 +90,13 @@ class NativeHlsFD(FragmentFD):
             success = ctx['dl'].download(frag_filename, {'url': frag_url})
             if not success:
                 return False
-            with open(frag_filename, 'rb') as down:
-                ctx['dest_stream'].write(down.read())
-            frags_filenames.append(frag_filename)
+            down, frag_sanitized = sanitize_open(frag_filename, 'rb')
+            ctx['dest_stream'].write(down.read())
+            frags_filenames.append(frag_sanitized)
 
         self._finish_frag_download(ctx)
 
         for frag_file in frags_filenames:
-            os.remove(frag_file)
+            os.remove(encodeFilename(frag_file))
 
         return True

youtube_dl/extractor/__init__.py

@@ -241,7 +241,10 @@ from .imdb import (
     ImdbIE,
     ImdbListIE
 )
-from .imgur import ImgurIE
+from .imgur import (
+    ImgurIE,
+    ImgurAlbumIE,
+)
 from .ina import InaIE
 from .indavideo import (
     IndavideoIE,
@@ -340,11 +343,12 @@ from .mtv import (
     MTVIE,
     MTVServicesEmbeddedIE,
     MTVIggyIE,
+    MTVDEIE,
 )
 from .muenchentv import MuenchenTVIE
 from .musicplayon import MusicPlayOnIE
-from .musicvault import MusicVaultIE
 from .muzu import MuzuTVIE
+from .mwave import MwaveIE
 from .myspace import MySpaceIE, MySpaceAlbumIE
 from .myspass import MySpassIE
 from .myvi import MyviIE
@@ -358,6 +362,7 @@ from .nbc import (
     NBCNewsIE,
     NBCSportsIE,
     NBCSportsVPlayerIE,
+    MSNBCIE,
 )
 from .ndr import (
     NDRIE,
@@ -452,6 +457,10 @@ from .playfm import PlayFMIE
 from .playtvak import PlaytvakIE
 from .playvid import PlayvidIE
 from .playwire import PlaywireIE
+from .pluralsight import (
+    PluralsightIE,
+    PluralsightCourseIE,
+)
 from .podomatic import PodomaticIE
 from .porn91 import Porn91IE
 from .pornhd import PornHdIE
@@ -612,7 +621,10 @@ from .testurl import TestURLIE
 from .testtube import TestTubeIE
 from .tf1 import TF1IE
 from .theonion import TheOnionIE
-from .theplatform import ThePlatformIE
+from .theplatform import (
+    ThePlatformIE,
+    ThePlatformFeedIE,
+)
 from .thesixtyone import TheSixtyOneIE
 from .thisamericanlife import ThisAmericanLifeIE
 from .thisav import ThisAVIE
@@ -696,7 +708,6 @@ from .vgtv import (
 from .vh1 import VH1IE
 from .vice import ViceIE
 from .viddler import ViddlerIE
-from .videobam import VideoBamIE
 from .videodetective import VideoDetectiveIE
 from .videolecturesnet import VideoLecturesNetIE
 from .videofyme import VideofyMeIE
@@ -731,6 +742,7 @@ from .vk import (
     VKIE,
     VKUserVideosIE,
 )
+from .vlive import VLiveIE
 from .vodlocker import VodlockerIE
 from .voicerepublic import VoiceRepublicIE
 from .vporn import VpornIE

youtube_dl/extractor/abc.py

@@ -1,16 +1,20 @@
 from __future__ import unicode_literals
 
 import re
-import json
 
 from .common import InfoExtractor
+from ..utils import (
+    ExtractorError,
+    js_to_json,
+    int_or_none,
+)
 
 
 class ABCIE(InfoExtractor):
     IE_NAME = 'abc.net.au'
     _VALID_URL = r'http://www\.abc\.net\.au/news/[^/]+/[^/]+/(?P<id>\d+)'
 
-    _TEST = {
+    _TESTS = [{
         'url': 'http://www.abc.net.au/news/2014-11-05/australia-to-staff-ebola-treatment-centre-in-sierra-leone/5868334',
         'md5': 'cb3dd03b18455a661071ee1e28344d9f',
         'info_dict': {
@@ -19,22 +23,47 @@ class ABCIE(InfoExtractor):
             'title': 'Australia to help staff Ebola treatment centre in Sierra Leone',
             'description': 'md5:809ad29c67a05f54eb41f2a105693a67',
         },
-    }
+    }, {
+        'url': 'http://www.abc.net.au/news/2015-08-17/warren-entsch-introduces-same-sex-marriage-bill/6702326',
+        'md5': 'db2a5369238b51f9811ad815b69dc086',
+        'info_dict': {
+            'id': 'NvqvPeNZsHU',
+            'ext': 'mp4',
+            'upload_date': '20150816',
+            'uploader': 'ABC News (Australia)',
+            'description': 'Government backbencher Warren Entsch introduces a cross-party sponsored bill to legalise same-sex marriage, saying the bill is designed to promote "an inclusive Australia, not a divided one.". Read more here: http://ab.co/1Mwc6ef',
+            'uploader_id': 'NewsOnABC',
+            'title': 'Marriage Equality: Warren Entsch introduces same sex marriage bill',
+        },
+        'add_ie': ['Youtube'],
+    }]
 
     def _real_extract(self, url):
         video_id = self._match_id(url)
         webpage = self._download_webpage(url, video_id)
 
-        urls_info_json = self._search_regex(
-            r'inlineVideoData\.push\((.*?)\);', webpage, 'video urls',
-            flags=re.DOTALL)
-        urls_info = json.loads(urls_info_json.replace('\'', '"'))
+        mobj = re.search(
+            r'inline(?P<type>Video|YouTube)Data\.push\((?P<json_data>[^)]+)\);',
+            webpage)
+        if mobj is None:
+            raise ExtractorError('Unable to extract video urls')
+
+        urls_info = self._parse_json(
+            mobj.group('json_data'), video_id, transform_source=js_to_json)
+
+        if not isinstance(urls_info, list):
+            urls_info = [urls_info]
+
+        if mobj.group('type') == 'YouTube':
+            return self.playlist_result([
+                self.url_result(url_info['url']) for url_info in urls_info])
+
         formats = [{
             'url': url_info['url'],
-            'width': int(url_info['width']),
-            'height': int(url_info['height']),
-            'tbr': int(url_info['bitrate']),
-            'filesize': int(url_info['filesize']),
+            'width': int_or_none(url_info.get('width')),
+            'height': int_or_none(url_info.get('height')),
+            'tbr': int_or_none(url_info.get('bitrate')),
+            'filesize': int_or_none(url_info.get('filesize')),
         } for url_info in urls_info]
         self._sort_formats(formats)
 

youtube_dl/extractor/academicearth.py

@@ -15,7 +15,7 @@ class AcademicEarthCourseIE(InfoExtractor):
             'title': 'Laws of Nature',
             'description': 'Introduce yourself to the laws of nature with these free online college lectures from Yale, Harvard, and MIT.',
         },
-        'playlist_count': 4,
+        'playlist_count': 3,
     }
 
     def _real_extract(self, url):

youtube_dl/extractor/adultswim.py

@@ -156,7 +156,7 @@ class AdultSwimIE(InfoExtractor):
                 xpath_text(idoc, './/trt', 'segment duration').strip())
 
             formats = []
-            file_els = idoc.findall('.//files/file')
+            file_els = idoc.findall('.//files/file') or idoc.findall('./files/file')
 
             for file_el in file_els:
                 bitrate = file_el.attrib.get('bitrate')

youtube_dl/extractor/airmozilla.py

@@ -20,14 +20,14 @@ class AirMozillaIE(InfoExtractor):
             'id': '6x4q2w',
             'ext': 'mp4',
             'title': 'Privacy Lab - a meetup for privacy minded people in San Francisco',
-            'thumbnail': 're:https://\w+\.cloudfront\.net/6x4q2w/poster\.jpg\?t=\d+',
+            'thumbnail': 're:https?://vid\.ly/(?P<id>[0-9a-z-]+)/poster',
             'description': 'Brings together privacy professionals and others interested in privacy at for-profits, non-profits, and NGOs in an effort to contribute to the state of the ecosystem...',
             'timestamp': 1422487800,
             'upload_date': '20150128',
             'location': 'SFO Commons',
             'duration': 3780,
             'view_count': int,
-            'categories': ['Main'],
+            'categories': ['Main', 'Privacy'],
         }
     }
 

youtube_dl/extractor/aljazeera.py

@@ -16,6 +16,7 @@ class AlJazeeraIE(InfoExtractor):
             'uploader': 'Al Jazeera English',
         },
         'add_ie': ['Brightcove'],
+        'skip': 'Not accessible from Travis CI server',
     }
 
     def _real_extract(self, url):

youtube_dl/extractor/ceskatelevize.py

@@ -17,55 +17,81 @@ from ..utils import (
 
 
 class CeskaTelevizeIE(InfoExtractor):
-    _VALID_URL = r'https?://www\.ceskatelevize\.cz/(porady|ivysilani)/(.+/)?(?P<id>[^?#]+)'
-
-    _TESTS = [
-        {
-            'url': 'http://www.ceskatelevize.cz/ivysilani/ivysilani/10441294653-hyde-park-civilizace/214411058091220',
-            'info_dict': {
-                'id': '214411058091220',
-                'ext': 'mp4',
-                'title': 'Hyde Park Civilizace',
-                'description': 'Věda a současná civilizace. Interaktivní pořad - prostor pro vaše otázky a komentáře',
-                'thumbnail': 're:^https?://.*\.jpg',
-                'duration': 3350,
-            },
-            'params': {
-                # m3u8 download
-                'skip_download': True,
-            },
+    _VALID_URL = r'https?://www\.ceskatelevize\.cz/(porady|ivysilani)/(?:[^/]+/)*(?P<id>[^/#?]+)/*(?:[#?].*)?$'
+    _TESTS = [{
+        'url': 'http://www.ceskatelevize.cz/ivysilani/ivysilani/10441294653-hyde-park-civilizace/214411058091220',
+        'info_dict': {
+            'id': '61924494876951776',
+            'ext': 'mp4',
+            'title': 'Hyde Park Civilizace',
+            'description': 'md5:fe93f6eda372d150759d11644ebbfb4a',
+            'thumbnail': 're:^https?://.*\.jpg',
+            'duration': 3350,
         },
-        {
-            'url': 'http://www.ceskatelevize.cz/ivysilani/10532695142-prvni-republika/bonus/14716-zpevacka-z-duparny-bobina',
-            'info_dict': {
-                'id': '14716',
-                'ext': 'mp4',
-                'title': 'První republika: Zpěvačka z Dupárny Bobina',
-                'description': 'Sága mapující atmosféru první republiky od r. 1918 do r. 1945.',
-                'thumbnail': 're:^https?://.*\.jpg',
-                'duration': 88.4,
-            },
-            'params': {
-                # m3u8 download
-                'skip_download': True,
-            },
+        'params': {
+            # m3u8 download
+            'skip_download': True,
         },
-    ]
+    }, {
+        'url': 'http://www.ceskatelevize.cz/ivysilani/10532695142-prvni-republika/bonus/14716-zpevacka-z-duparny-bobina',
+        'info_dict': {
+            'id': '61924494876844374',
+            'ext': 'mp4',
+            'title': 'První republika: Zpěvačka z Dupárny Bobina',
+            'description': 'Sága mapující atmosféru první republiky od r. 1918 do r. 1945.',
+            'thumbnail': 're:^https?://.*\.jpg',
+            'duration': 88.4,
+        },
+        'params': {
+            # m3u8 download
+            'skip_download': True,
+        },
+    }, {
+        # video with 18+ caution trailer
+        'url': 'http://www.ceskatelevize.cz/porady/10520528904-queer/215562210900007-bogotart/',
+        'info_dict': {
+            'id': '215562210900007-bogotart',
+            'title': 'Queer: Bogotart',
+            'description': 'Alternativní průvodce současným queer světem',
+        },
+        'playlist': [{
+            'info_dict': {
+                'id': '61924494876844842',
+                'ext': 'mp4',
+                'title': 'Queer: Bogotart (Varování 18+)',
+                'duration': 10.2,
+            },
+        }, {
+            'info_dict': {
+                'id': '61924494877068022',
+                'ext': 'mp4',
+                'title': 'Queer: Bogotart (Queer)',
+                'thumbnail': 're:^https?://.*\.jpg',
+                'duration': 1558.3,
+            },
+        }],
+        'params': {
+            # m3u8 download
+            'skip_download': True,
+        },
+    }]
 
     def _real_extract(self, url):
         url = url.replace('/porady/', '/ivysilani/').replace('/video/', '')
 
         mobj = re.match(self._VALID_URL, url)
-        video_id = mobj.group('id')
+        playlist_id = mobj.group('id')
 
-        webpage = self._download_webpage(url, video_id)
+        webpage = self._download_webpage(url, playlist_id)
 
         NOT_AVAILABLE_STRING = 'This content is not available at your territory due to limited copyright.'
         if '%s</p>' % NOT_AVAILABLE_STRING in webpage:
             raise ExtractorError(NOT_AVAILABLE_STRING, expected=True)
 
-        typ = self._html_search_regex(r'getPlaylistUrl\(\[\{"type":"(.+?)","id":".+?"\}\],', webpage, 'type')
-        episode_id = self._html_search_regex(r'getPlaylistUrl\(\[\{"type":".+?","id":"(.+?)"\}\],', webpage, 'episode_id')
+        typ = self._html_search_regex(
+            r'getPlaylistUrl\(\[\{"type":"(.+?)","id":".+?"\}\],', webpage, 'type')
+        episode_id = self._html_search_regex(
+            r'getPlaylistUrl\(\[\{"type":".+?","id":"(.+?)"\}\],', webpage, 'episode_id')
 
         data = {
             'playlist[0][type]': typ,
@@ -83,7 +109,7 @@ class CeskaTelevizeIE(InfoExtractor):
         req.add_header('X-Requested-With', 'XMLHttpRequest')
         req.add_header('Referer', url)
 
-        playlistpage = self._download_json(req, video_id)
+        playlistpage = self._download_json(req, playlist_id)
 
         playlist_url = playlistpage['url']
         if playlist_url == 'error_region':
@@ -92,33 +118,43 @@ class CeskaTelevizeIE(InfoExtractor):
         req = compat_urllib_request.Request(compat_urllib_parse_unquote(playlist_url))
         req.add_header('Referer', url)
 
-        playlist = self._download_json(req, video_id)
+        playlist_title = self._og_search_title(webpage)
+        playlist_description = self._og_search_description(webpage)
 
-        item = playlist['playlist'][0]
-        formats = []
-        for format_id, stream_url in item['streamUrls'].items():
-            formats.extend(self._extract_m3u8_formats(stream_url, video_id, 'mp4'))
-        self._sort_formats(formats)
+        playlist = self._download_json(req, playlist_id)['playlist']
+        playlist_len = len(playlist)
 
-        title = self._og_search_title(webpage)
-        description = self._og_search_description(webpage)
-        duration = float_or_none(item.get('duration'))
-        thumbnail = item.get('previewImageUrl')
+        entries = []
+        for item in playlist:
+            formats = []
+            for format_id, stream_url in item['streamUrls'].items():
+                formats.extend(self._extract_m3u8_formats(
+                    stream_url, playlist_id, 'mp4', entry_protocol='m3u8_native'))
+            self._sort_formats(formats)
 
-        subtitles = {}
-        subs = item.get('subtitles')
-        if subs:
-            subtitles = self.extract_subtitles(episode_id, subs)
+            item_id = item.get('id') or item['assetId']
+            title = item['title']
 
-        return {
-            'id': episode_id,
-            'title': title,
-            'description': description,
-            'thumbnail': thumbnail,
-            'duration': duration,
-            'formats': formats,
-            'subtitles': subtitles,
-        }
+            duration = float_or_none(item.get('duration'))
+            thumbnail = item.get('previewImageUrl')
+
+            subtitles = {}
+            if item.get('type') == 'VOD':
+                subs = item.get('subtitles')
+                if subs:
+                    subtitles = self.extract_subtitles(episode_id, subs)
+
+            entries.append({
+                'id': item_id,
+                'title': playlist_title if playlist_len == 1 else '%s (%s)' % (playlist_title, title),
+                'description': playlist_description if playlist_len == 1 else None,
+                'thumbnail': thumbnail,
+                'duration': duration,
+                'formats': formats,
+                'subtitles': subtitles,
+            })
+
+        return self.playlist_result(entries, playlist_id, playlist_title, playlist_description)
 
     def _get_subtitles(self, episode_id, subs):
         original_subtitles = self._download_webpage(

youtube_dl/extractor/cinemassacre.py

@@ -6,6 +6,7 @@ import re
 from .common import InfoExtractor
 from ..utils import ExtractorError
 from .bliptv import BlipTVIE
+from .screenwavemedia import ScreenwaveMediaIE
 
 
 class CinemassacreIE(InfoExtractor):
@@ -83,10 +84,10 @@ class CinemassacreIE(InfoExtractor):
 
         playerdata_url = self._search_regex(
             [
-                r'src="(http://(?:player2\.screenwavemedia\.com|player\.screenwavemedia\.com/play)/[a-zA-Z]+\.php\?[^"]*\bid=.+?)"',
-                r'<iframe[^>]+src="((?:https?:)?//(?:[^.]+\.)?youtube\.com/.+?)"',
+                ScreenwaveMediaIE.EMBED_PATTERN,
+                r'<iframe[^>]+src="(?P<url>(?:https?:)?//(?:[^.]+\.)?youtube\.com/.+?)"',
             ],
-            webpage, 'player data URL', default=None)
+            webpage, 'player data URL', default=None, group='url')
         if not playerdata_url:
             playerdata_url = BlipTVIE._extract_url(webpage)
         if not playerdata_url:

youtube_dl/extractor/common.py

@@ -510,6 +510,12 @@ class InfoExtractor(object):
         """Report attempt to log in."""
         self.to_screen('Logging in')
 
+    @staticmethod
+    def raise_login_required(msg='This video is only available for registered users'):
+        raise ExtractorError(
+            '%s. Use --username and --password or --netrc to provide account credentials.' % msg,
+            expected=True)
+
     # Methods for following #608
     @staticmethod
     def url_result(url, ie=None, video_id=None, video_title=None):
@@ -727,7 +733,7 @@ class InfoExtractor(object):
     def _hidden_inputs(html):
         hidden_inputs = {}
         for input in re.findall(r'<input([^>]+)>', html):
-            if not re.search(r'type=(["\'])hidden\1', input):
+            if not re.search(r'type=(["\'])(?:hidden|submit)\1', input):
                 continue
             name = re.search(r'name=(["\'])(?P<value>.+?)\1', input)
             if not name:
@@ -1052,7 +1058,7 @@ class InfoExtractor(object):
         return self._search_regex(
             r'(?i)^{([^}]+)?}smil$', smil.tag, 'namespace', default=None)
 
-    def _parse_smil_formats(self, smil, smil_url, video_id, namespace=None, f4m_params=None):
+    def _parse_smil_formats(self, smil, smil_url, video_id, namespace=None, f4m_params=None, transform_rtmp_url=None):
         base = smil_url
         for meta in smil.findall(self._xpath_ns('./head/meta', namespace)):
             b = meta.get('base') or meta.get('httpBase')
@@ -1091,6 +1097,12 @@ class InfoExtractor(object):
                     'width': width,
                     'height': height,
                 })
+                if transform_rtmp_url:
+                    streamer, src = transform_rtmp_url(streamer, src)
+                    formats[-1].update({
+                        'url': streamer,
+                        'play_path': src,
+                    })
                 continue
 
             src_url = src if src.startswith('http') else compat_urlparse.urljoin(base, src)
@@ -1129,7 +1141,7 @@ class InfoExtractor(object):
 
         return formats
 
-    def _parse_smil_subtitles(self, smil, namespace=None):
+    def _parse_smil_subtitles(self, smil, namespace=None, subtitles_lang='en'):
         subtitles = {}
         for num, textstream in enumerate(smil.findall(self._xpath_ns('.//textstream', namespace))):
             src = textstream.get('src')
@@ -1138,9 +1150,14 @@ class InfoExtractor(object):
             ext = textstream.get('ext') or determine_ext(src)
             if not ext:
                 type_ = textstream.get('type')
-                if type_ == 'text/srt':
-                    ext = 'srt'
-            lang = textstream.get('systemLanguage') or textstream.get('systemLanguageName')
+                SUBTITLES_TYPES = {
+                    'text/vtt': 'vtt',
+                    'text/srt': 'srt',
+                    'application/smptett+xml': 'tt',
+                }
+                if type_ in SUBTITLES_TYPES:
+                    ext = SUBTITLES_TYPES[type_]
+            lang = textstream.get('systemLanguage') or textstream.get('systemLanguageName') or textstream.get('lang') or subtitles_lang
             subtitles.setdefault(lang, []).append({
                 'url': src,
                 'ext': ext,
@@ -1268,6 +1285,23 @@ class InfoExtractor(object):
     def _get_subtitles(self, *args, **kwargs):
         raise NotImplementedError("This method must be implemented by subclasses")
 
+    @staticmethod
+    def _merge_subtitle_items(subtitle_list1, subtitle_list2):
+        """ Merge subtitle items for one language. Items with duplicated URLs
+        will be dropped. """
+        list1_urls = set([item['url'] for item in subtitle_list1])
+        ret = list(subtitle_list1)
+        ret.extend([item for item in subtitle_list2 if item['url'] not in list1_urls])
+        return ret
+
+    @classmethod
+    def _merge_subtitles(cls, subtitle_dict1, subtitle_dict2):
+        """ Merge two subtitle dictionaries, language by language. """
+        ret = dict(subtitle_dict1)
+        for lang in subtitle_dict2:
+            ret[lang] = cls._merge_subtitle_items(subtitle_dict1.get(lang, []), subtitle_dict2[lang])
+        return ret
+
     def extract_automatic_captions(self, *args, **kwargs):
         if (self._downloader.params.get('writeautomaticsub', False) or
                 self._downloader.params.get('listsubtitles')):

youtube_dl/extractor/crunchyroll.py

@@ -14,20 +14,40 @@ from ..compat import (
     compat_urllib_parse,
     compat_urllib_parse_unquote,
     compat_urllib_request,
+    compat_urlparse,
 )
 from ..utils import (
     ExtractorError,
     bytes_to_intlist,
     intlist_to_bytes,
+    int_or_none,
+    remove_end,
     unified_strdate,
     urlencode_postdata,
+    xpath_text,
 )
 from ..aes import (
     aes_cbc_decrypt,
 )
 
 
-class CrunchyrollIE(InfoExtractor):
+class CrunchyrollBaseIE(InfoExtractor):
+    def _download_webpage(self, url_or_request, video_id, note=None, errnote=None, fatal=True, tries=1, timeout=5, encoding=None):
+        request = (url_or_request if isinstance(url_or_request, compat_urllib_request.Request)
+                   else compat_urllib_request.Request(url_or_request))
+        # Accept-Language must be set explicitly to accept any language to avoid issues
+        # similar to https://github.com/rg3/youtube-dl/issues/6797.
+        # Along with IP address Crunchyroll uses Accept-Language to guess whether georestriction
+        # should be imposed or not (from what I can see it just takes the first language
+        # ignoring the priority and requires it to correspond the IP). By the way this causes
+        # Crunchyroll to not work in georestriction cases in some browsers that don't place
+        # the locale lang first in header. However allowing any language seems to workaround the issue.
+        request.add_header('Accept-Language', '*')
+        return super(CrunchyrollBaseIE, self)._download_webpage(
+            request, video_id, note, errnote, fatal, tries, timeout, encoding)
+
+
+class CrunchyrollIE(CrunchyrollBaseIE):
     _VALID_URL = r'https?://(?:(?P<prefix>www|m)\.)?(?P<url>crunchyroll\.(?:com|fr)/(?:media(?:-|/\?id=)|[^/]*/[^/?&]*?)(?P<video_id>[0-9]+))(?:[/?&]|$)'
     _NETRC_MACHINE = 'crunchyroll'
     _TESTS = [{
@@ -235,7 +255,9 @@ Format: Layer, Start, End, Style, Name, MarginL, MarginR, MarginV, Effect, Text
             webpage_url = 'http://www.' + mobj.group('url')
 
         webpage = self._download_webpage(webpage_url, video_id, 'Downloading webpage')
-        note_m = self._html_search_regex(r'<div class="showmedia-trailer-notice">(.+?)</div>', webpage, 'trailer-notice', default='')
+        note_m = self._html_search_regex(
+            r'<div class="showmedia-trailer-notice">(.+?)</div>',
+            webpage, 'trailer-notice', default='')
         if note_m:
             raise ExtractorError(note_m)
 
@@ -245,15 +267,22 @@ Format: Layer, Start, End, Style, Name, MarginL, MarginR, MarginV, Effect, Text
             if msg.get('type') == 'error':
                 raise ExtractorError('crunchyroll returned error: %s' % msg['message_body'], expected=True)
 
+        if 'To view this, please log in to verify you are 18 or older.' in webpage:
+            self.raise_login_required()
+
         video_title = self._html_search_regex(r'<h1[^>]*>(.+?)</h1>', webpage, 'video_title', flags=re.DOTALL)
         video_title = re.sub(r' {2,}', ' ', video_title)
         video_description = self._html_search_regex(r'"description":"([^"]+)', webpage, 'video_description', default='')
         if not video_description:
             video_description = None
-        video_upload_date = self._html_search_regex(r'<div>Availability for free users:(.+?)</div>', webpage, 'video_upload_date', fatal=False, flags=re.DOTALL)
+        video_upload_date = self._html_search_regex(
+            [r'<div>Availability for free users:(.+?)</div>', r'<div>[^<>]+<span>\s*(.+?\d{4})\s*</span></div>'],
+            webpage, 'video_upload_date', fatal=False, flags=re.DOTALL)
         if video_upload_date:
             video_upload_date = unified_strdate(video_upload_date)
-        video_uploader = self._html_search_regex(r'<div>\s*Publisher:(.+?)</div>', webpage, 'video_uploader', fatal=False, flags=re.DOTALL)
+        video_uploader = self._html_search_regex(
+            r'<a[^>]+href="/publisher/[^"]+"[^>]*>([^<]+)</a>', webpage,
+            'video_uploader', fatal=False)
 
         playerdata_url = compat_urllib_parse_unquote(self._html_search_regex(r'"config_url":"([^"]+)', webpage, 'playerdata_url'))
         playerdata_req = compat_urllib_request.Request(playerdata_url)
@@ -279,13 +308,33 @@ Format: Layer, Start, End, Style, Name, MarginL, MarginR, MarginV, Effect, Text
             stream_info = streamdata.find('./{default}preload/stream_info')
             video_url = stream_info.find('./host').text
             video_play_path = stream_info.find('./file').text
-            formats.append({
+            metadata = stream_info.find('./metadata')
+            format_info = {
+                'format': video_format,
+                'format_id': video_format,
+                'height': int_or_none(xpath_text(metadata, './height')),
+                'width': int_or_none(xpath_text(metadata, './width')),
+            }
+
+            if '.fplive.net/' in video_url:
+                video_url = re.sub(r'^rtmpe?://', 'http://', video_url.strip())
+                parsed_video_url = compat_urlparse.urlparse(video_url)
+                direct_video_url = compat_urlparse.urlunparse(parsed_video_url._replace(
+                    netloc='v.lvlt.crcdn.net',
+                    path='%s/%s' % (remove_end(parsed_video_url.path, '/'), video_play_path.split(':')[-1])))
+                if self._is_valid_url(direct_video_url, video_id, video_format):
+                    format_info.update({
+                        'url': direct_video_url,
+                    })
+                    formats.append(format_info)
+                    continue
+
+            format_info.update({
                 'url': video_url,
                 'play_path': video_play_path,
                 'ext': 'flv',
-                'format': video_format,
-                'format_id': video_format,
             })
+            formats.append(format_info)
 
         subtitles = self.extract_subtitles(video_id, webpage)
 
@@ -301,7 +350,7 @@ Format: Layer, Start, End, Style, Name, MarginL, MarginR, MarginV, Effect, Text
         }
 
 
-class CrunchyrollShowPlaylistIE(InfoExtractor):
+class CrunchyrollShowPlaylistIE(CrunchyrollBaseIE):
     IE_NAME = "crunchyroll:playlist"
     _VALID_URL = r'https?://(?:(?P<prefix>www|m)\.)?(?P<url>crunchyroll\.com/(?!(?:news|anime-news|library|forum|launchcalendar|lineup|store|comics|freetrial|login))(?P<id>[\w\-]+))/?$'
 

youtube_dl/extractor/dcn.py

@@ -44,8 +44,8 @@ class DCNIE(InfoExtractor):
         title = video.get('title_en') or video['title_ar']
 
         webpage = self._download_webpage(
-            'http://admin.mangomolo.com/analytics/index.php/customers/embed/video?'
-            + compat_urllib_parse.urlencode({
+            'http://admin.mangomolo.com/analytics/index.php/customers/embed/video?' +
+            compat_urllib_parse.urlencode({
                 'id': video['id'],
                 'user_id': video['user_id'],
                 'signature': video['signature'],

youtube_dl/extractor/dumpert.py

@@ -9,8 +9,8 @@ from ..utils import qualities
 
 
 class DumpertIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:www\.)?dumpert\.nl/mediabase/(?P<id>[0-9]+/[0-9a-zA-Z]+)'
-    _TEST = {
+    _VALID_URL = r'https?://(?:www\.)?dumpert\.nl/(?:mediabase|embed)/(?P<id>[0-9]+/[0-9a-zA-Z]+)'
+    _TESTS = [{
         'url': 'http://www.dumpert.nl/mediabase/6646981/951bc60f/',
         'md5': '1b9318d7d5054e7dcb9dc7654f21d643',
         'info_dict': {
@@ -20,11 +20,15 @@ class DumpertIE(InfoExtractor):
             'description': 'Niet schrikken hoor',
             'thumbnail': 're:^https?://.*\.jpg$',
         }
-    }
+    }, {
+        'url': 'http://www.dumpert.nl/embed/6675421/dc440fe7/',
+        'only_matching': True,
+    }]
 
     def _real_extract(self, url):
         video_id = self._match_id(url)
 
+        url = 'https://www.dumpert.nl/mediabase/' + video_id
         req = compat_urllib_request.Request(url)
         req.add_header('Cookie', 'nsfw=1; cpc=10')
         webpage = self._download_webpage(req, video_id)

youtube_dl/extractor/eagleplatform.py

@@ -79,7 +79,7 @@ class EaglePlatformIE(InfoExtractor):
             age_limit = 0 if age_restriction == 'allow_all' else 18
 
         m3u8_data = self._download_json(
-            media['sources']['secure_m3u8']['auto'],
+            self._proto_relative_url(media['sources']['secure_m3u8']['auto'], 'http:'),
             video_id, 'Downloading m3u8 JSON')
 
         formats = self._extract_m3u8_formats(

youtube_dl/extractor/eroprofile.py

@@ -71,8 +71,7 @@ class EroProfileIE(InfoExtractor):
 
         m = re.search(r'You must be logged in to view this video\.', webpage)
         if m:
-            raise ExtractorError(
-                'This video requires login. Please specify a username and password and try again.', expected=True)
+            self.raise_login_required('This video requires login')
 
         video_id = self._search_regex(
             [r"glbUpdViews\s*\('\d*','(\d+)'", r'p/report/video/(\d+)'],

youtube_dl/extractor/fc2.py

@@ -10,12 +10,13 @@ from ..compat import (
     compat_urlparse,
 )
 from ..utils import (
+    encode_dict,
     ExtractorError,
 )
 
 
 class FC2IE(InfoExtractor):
-    _VALID_URL = r'^http://video\.fc2\.com/(?:[^/]+/)?content/(?P<id>[^/]+)'
+    _VALID_URL = r'^http://video\.fc2\.com/(?:[^/]+/)*content/(?P<id>[^/]+)'
     IE_NAME = 'fc2'
     _NETRC_MACHINE = 'fc2'
     _TESTS = [{
@@ -37,6 +38,9 @@ class FC2IE(InfoExtractor):
             'password': '(snip)',
             'skip': 'requires actual password'
         }
+    }, {
+        'url': 'http://video.fc2.com/en/a/content/20130926eZpARwsF',
+        'only_matching': True,
     }]
 
     def _login(self):
@@ -52,10 +56,7 @@ class FC2IE(InfoExtractor):
             'Submit': ' Login ',
         }
 
-        # Convert to UTF-8 *before* urlencode because Python 2.x's urlencode
-        # chokes on unicode
-        login_form = dict((k.encode('utf-8'), v.encode('utf-8')) for k, v in login_form_strs.items())
-        login_data = compat_urllib_parse.urlencode(login_form).encode('utf-8')
+        login_data = compat_urllib_parse.urlencode(encode_dict(login_form_strs)).encode('utf-8')
         request = compat_urllib_request.Request(
             'https://secure.id.fc2.com/index.php?mode=login&switch_language=en', login_data)
 
@@ -80,7 +81,7 @@ class FC2IE(InfoExtractor):
 
         title = self._og_search_title(webpage)
         thumbnail = self._og_search_thumbnail(webpage)
-        refer = url.replace('/content/', '/a/content/')
+        refer = url.replace('/content/', '/a/content/') if '/a/content/' not in url else url
 
         mimi = hashlib.md5((video_id + '_gGddgPfeaf_gzyr').encode('utf-8')).hexdigest()
 

youtube_dl/extractor/folketinget.py

@@ -30,6 +30,10 @@ class FolketingetIE(InfoExtractor):
             'upload_date': '20141120',
             'duration': 3960,
         },
+        'params': {
+            # rtmp download
+            'skip_download': True,
+        },
     }
 
     def _real_extract(self, url):

youtube_dl/extractor/foxnews.py

@@ -1,5 +1,7 @@
 from __future__ import unicode_literals
 
+import re
+
 from .common import InfoExtractor
 from ..utils import (
     parse_iso8601,
@@ -8,7 +10,8 @@ from ..utils import (
 
 
 class FoxNewsIE(InfoExtractor):
-    _VALID_URL = r'https?://video\.foxnews\.com/v/(?:video-embed\.html\?video_id=)?(?P<id>\d+)'
+    IE_DESC = 'Fox News and Fox Business Video'
+    _VALID_URL = r'https?://(?P<host>video\.fox(?:news|business)\.com)/v/(?:video-embed\.html\?video_id=)?(?P<id>\d+)'
     _TESTS = [
         {
             'url': 'http://video.foxnews.com/v/3937480/frozen-in-time/#sp=show-clips',
@@ -42,13 +45,19 @@ class FoxNewsIE(InfoExtractor):
             'url': 'http://video.foxnews.com/v/video-embed.html?video_id=3937480&d=video.foxnews.com',
             'only_matching': True,
         },
+        {
+            'url': 'http://video.foxbusiness.com/v/4442309889001',
+            'only_matching': True,
+        },
     ]
 
     def _real_extract(self, url):
-        video_id = self._match_id(url)
+        mobj = re.match(self._VALID_URL, url)
+        video_id = mobj.group('id')
+        host = mobj.group('host')
 
         video = self._download_json(
-            'http://video.foxnews.com/v/feed/video/%s.js?template=fox' % video_id, video_id)
+            'http://%s/v/feed/video/%s.js?template=fox' % (host, video_id), video_id)
 
         item = video['channel']['item']
         title = item['title']

youtube_dl/extractor/francetv.py

@@ -78,9 +78,14 @@ class FranceTVBaseInfoExtractor(InfoExtractor):
                 })
         self._sort_formats(formats)
 
+        title = info['titre']
+        subtitle = info.get('sous_titre')
+        if subtitle:
+            title += ' - %s' % subtitle
+
         return {
             'id': video_id,
-            'title': info['titre'],
+            'title': title,
             'description': clean_html(info['synopsis']),
             'thumbnail': compat_urlparse.urljoin('http://pluzz.francetv.fr', info['image']),
             'duration': int_or_none(info.get('real_duration')) or parse_duration(info['duree']),
@@ -214,15 +219,15 @@ class FranceTVIE(FranceTVBaseInfoExtractor):
         },
         # france5
         {
-            'url': 'http://www.france5.fr/emissions/c-a-dire/videos/92837968',
-            'md5': '78f0f4064f9074438e660785bbf2c5d9',
+            'url': 'http://www.france5.fr/emissions/c-a-dire/videos/quels_sont_les_enjeux_de_cette_rentree_politique__31-08-2015_908948?onglet=tous&page=1',
+            'md5': 'f6c577df3806e26471b3d21631241fd0',
             'info_dict': {
-                'id': '108961659',
+                'id': '123327454',
                 'ext': 'flv',
-                'title': 'C à dire ?!',
-                'description': 'md5:1a4aeab476eb657bf57c4ff122129f81',
-                'upload_date': '20140915',
-                'timestamp': 1410795000,
+                'title': 'C à dire ?! - Quels sont les enjeux de cette rentrée politique ?',
+                'description': 'md5:4a0d5cb5dce89d353522a84462bae5a4',
+                'upload_date': '20150831',
+                'timestamp': 1441035120,
             },
         },
         # franceo

youtube_dl/extractor/generic.py

@@ -48,6 +48,7 @@ from .vimeo import VimeoIE
 from .dailymotion import DailymotionCloudIE
 from .onionstudios import OnionStudiosIE
 from .snagfilms import SnagFilmsEmbedIE
+from .screenwavemedia import ScreenwaveMediaIE
 
 
 class GenericIE(InfoExtractor):
@@ -1001,6 +1002,16 @@ class GenericIE(InfoExtractor):
                 'description': 'New experience with Acrobat DC',
                 'duration': 248.667,
             },
+        },
+        # ScreenwaveMedia embed
+        {
+            'url': 'http://www.thecinemasnob.com/the-cinema-snob/a-nightmare-on-elm-street-2-freddys-revenge1',
+            'md5': '24ace5baba0d35d55c6810b51f34e9e0',
+            'info_dict': {
+                'id': 'cinemasnob-55d26273809dd',
+                'ext': 'mp4',
+                'title': 'cinemasnob',
+            },
         }
     ]
 
@@ -1718,6 +1729,11 @@ class GenericIE(InfoExtractor):
         if snagfilms_url:
             return self.url_result(snagfilms_url)
 
+        # Look for ScreenwaveMedia embeds
+        mobj = re.search(ScreenwaveMediaIE.EMBED_PATTERN, webpage)
+        if mobj is not None:
+            return self.url_result(unescapeHTML(mobj.group('url')), 'ScreenwaveMedia')
+
         # Look for AdobeTVVideo embeds
         mobj = re.search(
             r'<iframe[^>]+src=[\'"]((?:https?:)?//video\.tv\.adobe\.com/v/\d+[^"]+)[\'"]',
@@ -1781,7 +1797,7 @@ 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|audio)[^<]*(?:>.*?<source[^>]*)?\s+src=["\'](.*?)["\']', webpage)
         if not found:
             REDIRECT_REGEX = r'[0-9]{,2};\s*(?:URL|url)=\'?([^\'"]+)'
             found = re.search(

youtube_dl/extractor/globo.py

@@ -13,6 +13,7 @@ from ..compat import (
 from ..utils import (
     ExtractorError,
     float_or_none,
+    int_or_none,
 )
 
 
@@ -359,13 +360,8 @@ class GloboIE(InfoExtractor):
             self._API_URL_TEMPLATE % video_id, video_id)['videos'][0]
 
         title = video['title']
-        duration = float_or_none(video['duration'], 1000)
-        like_count = video['likes']
-        uploader = video['channel']
-        uploader_id = video['channel_id']
 
         formats = []
-
         for resource in video['resources']:
             resource_id = resource.get('_id')
             if not resource_id:
@@ -407,6 +403,11 @@ class GloboIE(InfoExtractor):
 
         self._sort_formats(formats)
 
+        duration = float_or_none(video.get('duration'), 1000)
+        like_count = int_or_none(video.get('likes'))
+        uploader = video.get('channel')
+        uploader_id = video.get('channel_id')
+
         return {
             'id': video_id,
             'title': title,

youtube_dl/extractor/gorillavid.py

@@ -10,15 +10,16 @@ from ..compat import (
 )
 from ..utils import (
     ExtractorError,
+    encode_dict,
     int_or_none,
 )
 
 
 class GorillaVidIE(InfoExtractor):
-    IE_DESC = 'GorillaVid.in, daclips.in, movpod.in, fastvideo.in and realvid.net'
+    IE_DESC = 'GorillaVid.in, daclips.in, movpod.in, fastvideo.in, realvid.net and filehoot.com'
     _VALID_URL = r'''(?x)
         https?://(?P<host>(?:www\.)?
-            (?:daclips\.in|gorillavid\.in|movpod\.in|fastvideo\.in|realvid\.net))/
+            (?:daclips\.in|gorillavid\.in|movpod\.in|fastvideo\.in|realvid\.net|filehoot\.com))/
         (?:embed-)?(?P<id>[0-9a-zA-Z]+)(?:-[0-9]+x[0-9]+\.html)?
     '''
 
@@ -67,13 +68,22 @@ class GorillaVidIE(InfoExtractor):
     }, {
         'url': 'http://movpod.in/0wguyyxi1yca',
         'only_matching': True,
+    }, {
+        'url': 'http://filehoot.com/3ivfabn7573c.html',
+        'info_dict': {
+            'id': '3ivfabn7573c',
+            'ext': 'mp4',
+            'title': 'youtube-dl test video \'äBaW_jenozKc.mp4.mp4',
+            'thumbnail': 're:http://.*\.jpg',
+        }
     }]
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
         video_id = mobj.group('id')
 
-        webpage = self._download_webpage('http://%s/%s' % (mobj.group('host'), video_id), video_id)
+        url = 'http://%s/%s' % (mobj.group('host'), video_id)
+        webpage = self._download_webpage(url, video_id)
 
         if re.search(self._FILE_NOT_FOUND_REGEX, webpage) is not None:
             raise ExtractorError('Video %s does not exist' % video_id, expected=True)
@@ -87,7 +97,7 @@ class GorillaVidIE(InfoExtractor):
             if countdown:
                 self._sleep(countdown, video_id)
 
-            post = compat_urllib_parse.urlencode(fields)
+            post = compat_urllib_parse.urlencode(encode_dict(fields))
 
             req = compat_urllib_request.Request(url, post)
             req.add_header('Content-type', 'application/x-www-form-urlencoded')
@@ -95,7 +105,7 @@ class GorillaVidIE(InfoExtractor):
             webpage = self._download_webpage(req, video_id, 'Downloading video page')
 
         title = self._search_regex(
-            [r'style="z-index: [0-9]+;">([^<]+)</span>', r'>Watch (.+) '],
+            [r'style="z-index: [0-9]+;">([^<]+)</span>', r'<td nowrap>([^<]+)</td>', r'>Watch (.+) '],
             webpage, 'title', default=None) or self._og_search_title(webpage)
         video_url = self._search_regex(
             r'file\s*:\s*["\'](http[^"\']+)["\'],', webpage, 'file url')

youtube_dl/extractor/imgur.py

@@ -13,7 +13,7 @@ from ..utils import (
 
 
 class ImgurIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:i\.)?imgur\.com/(?P<id>[a-zA-Z0-9]+)'
+    _VALID_URL = r'https?://(?:i\.)?imgur\.com/(?!gallery)(?P<id>[a-zA-Z0-9]+)'
 
     _TESTS = [{
         'url': 'https://i.imgur.com/A61SaA1.gifv',
@@ -97,3 +97,28 @@ class ImgurIE(InfoExtractor):
             'description': self._og_search_description(webpage),
             'title': self._og_search_title(webpage),
         }
+
+
+class ImgurAlbumIE(InfoExtractor):
+    _VALID_URL = r'https?://(?:i\.)?imgur\.com/gallery/(?P<id>[a-zA-Z0-9]+)'
+
+    _TEST = {
+        'url': 'http://imgur.com/gallery/Q95ko',
+        'info_dict': {
+            'id': 'Q95ko',
+        },
+        'playlist_count': 25,
+    }
+
+    def _real_extract(self, url):
+        album_id = self._match_id(url)
+
+        album_images = self._download_json(
+            'http://imgur.com/gallery/%s/album_images/hit.json?all=true' % album_id,
+            album_id)['data']['images']
+
+        entries = [
+            self.url_result('http://imgur.com/%s' % image['hash'])
+            for image in album_images if image.get('hash')]
+
+        return self.playlist_result(entries, album_id)

youtube_dl/extractor/kaltura.py

@@ -13,12 +13,24 @@ from ..utils import (
 
 class KalturaIE(InfoExtractor):
     _VALID_URL = r'''(?x)
-    (?:kaltura:|
-       https?://(:?(?:www|cdnapisec)\.)?kaltura\.com/index\.php/kwidget/(?:[^/]+/)*?wid/_
-    )(?P<partner_id>\d+)
-    (?::|
-       /(?:[^/]+/)*?entry_id/
-    )(?P<id>[0-9a-z_]+)'''
+                (?:
+                    kaltura:(?P<partner_id_s>\d+):(?P<id_s>[0-9a-z_]+)|
+                    https?://
+                        (:?(?:www|cdnapisec)\.)?kaltura\.com/
+                        (?:
+                            (?:
+                                # flash player
+                                index\.php/kwidget/
+                                (?:[^/]+/)*?wid/_(?P<partner_id>\d+)/
+                                (?:[^/]+/)*?entry_id/(?P<id>[0-9a-z_]+)|
+                                # html5 player
+                                html5/html5lib/
+                                (?:[^/]+/)*?entry_id/(?P<id_html5>[0-9a-z_]+)
+                                .*\?.*\bwid=_(?P<partner_id_html5>\d+)
+                            )
+                        )
+                )
+                '''
     _API_BASE = 'http://cdnapi.kaltura.com/api_v3/index.php?'
     _TESTS = [
         {
@@ -43,6 +55,10 @@ class KalturaIE(InfoExtractor):
             'url': 'https://cdnapisec.kaltura.com/index.php/kwidget/wid/_557781/uiconf_id/22845202/entry_id/1_plr1syf3',
             'only_matching': True,
         },
+        {
+            'url': 'https://cdnapisec.kaltura.com/html5/html5lib/v2.30.2/mwEmbedFrame.php/p/1337/uiconf_id/20540612/entry_id/1_sf5ovm7u?wid=_243342',
+            'only_matching': True,
+        }
     ]
 
     def _kaltura_api_call(self, video_id, actions, *args, **kwargs):
@@ -105,9 +121,9 @@ class KalturaIE(InfoExtractor):
             video_id, actions, note='Downloading video info JSON')
 
     def _real_extract(self, url):
-        video_id = self._match_id(url)
         mobj = re.match(self._VALID_URL, url)
-        partner_id, entry_id = mobj.group('partner_id'), mobj.group('id')
+        partner_id = mobj.group('partner_id_s') or mobj.group('partner_id') or mobj.group('partner_id_html5')
+        entry_id = mobj.group('id_s') or mobj.group('id') or mobj.group('id_html5')
 
         info, source_data = self._get_video_info(entry_id, partner_id)
 
@@ -126,7 +142,7 @@ class KalturaIE(InfoExtractor):
         self._sort_formats(formats)
 
         return {
-            'id': video_id,
+            'id': entry_id,
             'title': info['name'],
             'formats': formats,
             'description': info.get('description'),

youtube_dl/extractor/kontrtube.py

@@ -4,7 +4,10 @@ from __future__ import unicode_literals
 import re
 
 from .common import InfoExtractor
-from ..utils import int_or_none
+from ..utils import (
+    int_or_none,
+    parse_duration,
+)
 
 
 class KontrTubeIE(InfoExtractor):
@@ -34,33 +37,28 @@ class KontrTubeIE(InfoExtractor):
         webpage = self._download_webpage(
             url, display_id, 'Downloading page')
 
-        video_url = self._html_search_regex(
+        video_url = self._search_regex(
             r"video_url\s*:\s*'(.+?)/?',", webpage, 'video URL')
-        thumbnail = self._html_search_regex(
-            r"preview_url\s*:\s*'(.+?)/?',", webpage, 'video thumbnail', fatal=False)
+        thumbnail = self._search_regex(
+            r"preview_url\s*:\s*'(.+?)/?',", webpage, 'thumbnail', fatal=False)
         title = self._html_search_regex(
-            r'<title>(.+?)</title>', webpage, 'video title')
+            r'(?s)<h2>(.+?)</h2>', webpage, 'title')
         description = self._html_search_meta(
-            'description', webpage, 'video description')
+            'description', webpage, 'description')
 
-        mobj = re.search(
-            r'<div class="col_2">Длительность: <span>(?P<minutes>\d+)м:(?P<seconds>\d+)с</span></div>',
-            webpage)
-        duration = int(mobj.group('minutes')) * 60 + int(mobj.group('seconds')) if mobj else None
+        duration = self._search_regex(
+            r'Длительность: <em>([^<]+)</em>', webpage, 'duration', fatal=False)
+        if duration:
+            duration = parse_duration(duration.replace('мин', 'min').replace('сек', 'sec'))
 
-        view_count = self._html_search_regex(
-            r'<div class="col_2">Просмотров: <span>(\d+)</span></div>',
+        view_count = self._search_regex(
+            r'Просмотров: <em>([^<]+)</em>',
             webpage, 'view count', fatal=False)
+        if view_count:
+            view_count = int_or_none(view_count.replace(' ', ''))
 
-        comment_count = None
-        comment_str = self._html_search_regex(
-            r'Комментарии: <span>([^<]+)</span>', webpage, 'comment count', fatal=False)
-        if comment_str.startswith('комментариев нет'):
-            comment_count = 0
-        else:
-            mobj = re.search(r'\d+ из (?P<total>\d+) комментариев', comment_str)
-            if mobj:
-                comment_count = mobj.group('total')
+        comment_count = int_or_none(self._search_regex(
+            r'Комментарии \((\d+)\)<', webpage, ' comment count', fatal=False))
 
         return {
             'id': video_id,

youtube_dl/extractor/krasview.py

@@ -25,6 +25,9 @@ class KrasViewIE(InfoExtractor):
             'duration': 27,
             'thumbnail': 're:^https?://.*\.jpg',
         },
+        'params': {
+            'skip_download': 'Not accessible from Travis CI server',
+        },
     }
 
     def _real_extract(self, url):

youtube_dl/extractor/kuwo.py

@@ -202,6 +202,7 @@ class KuwoSingerIE(InfoExtractor):
             'title': 'Ali',
         },
         'playlist_mincount': 95,
+        'skip': 'Regularly stalls travis build',  # See https://travis-ci.org/rg3/youtube-dl/jobs/78878540
     }]
 
     def _real_extract(self, url):

youtube_dl/extractor/libsyn.py

@@ -8,9 +8,9 @@ from ..utils import unified_strdate
 
 
 class LibsynIE(InfoExtractor):
-    _VALID_URL = r'https?://html5-player\.libsyn\.com/embed/episode/id/(?P<id>[0-9]+)'
+    _VALID_URL = r'(?P<mainurl>https?://html5-player\.libsyn\.com/embed/episode/id/(?P<id>[0-9]+))'
 
-    _TEST = {
+    _TESTS = [{
         'url': 'http://html5-player.libsyn.com/embed/episode/id/3377616/',
         'md5': '443360ee1b58007bc3dcf09b41d093bb',
         'info_dict': {
@@ -19,12 +19,24 @@ class LibsynIE(InfoExtractor):
             'title': "The Daily Show Podcast without Jon Stewart - Episode 12: Bassem Youssef: Egypt's Jon Stewart",
             'description': 'md5:601cb790edd05908957dae8aaa866465',
             'upload_date': '20150220',
+            'thumbnail': 're:^https?://.*',
         },
-    }
+    }, {
+        'url': 'https://html5-player.libsyn.com/embed/episode/id/3727166/height/75/width/200/theme/standard/direction/no/autoplay/no/autonext/no/thumbnail/no/preload/no/no_addthis/no/',
+        'md5': '6c5cb21acd622d754d3b1a92b582ce42',
+        'info_dict': {
+            'id': '3727166',
+            'ext': 'mp3',
+            'title': 'Clients From Hell Podcast - How a Sex Toy Company Kickstarted my Freelance Career',
+            'upload_date': '20150818',
+            'thumbnail': 're:^https?://.*',
+        }
+    }]
 
     def _real_extract(self, url):
-        video_id = self._match_id(url)
-
+        m = re.match(self._VALID_URL, url)
+        video_id = m.group('id')
+        url = m.group('mainurl')
         webpage = self._download_webpage(url, video_id)
 
         formats = [{
@@ -32,20 +44,18 @@ class LibsynIE(InfoExtractor):
         } for media_url in set(re.findall('var\s+mediaURL(?:Libsyn)?\s*=\s*"([^"]+)"', webpage))]
 
         podcast_title = self._search_regex(
-            r'<h2>([^<]+)</h2>', webpage, 'title')
+            r'<h2>([^<]+)</h2>', webpage, 'podcast title', default=None)
         episode_title = self._search_regex(
-            r'<h3>([^<]+)</h3>', webpage, 'title', default=None)
+            r'(?:<div class="episode-title">|<h3>)([^<]+)</', webpage, 'episode title')
 
         title = '%s - %s' % (podcast_title, episode_title) if podcast_title else episode_title
 
         description = self._html_search_regex(
             r'<div id="info_text_body">(.+?)</div>', webpage,
-            'description', fatal=False)
-
+            'description', default=None)
         thumbnail = self._search_regex(
             r'<img[^>]+class="info-show-icon"[^>]+src="([^"]+)"',
             webpage, 'thumbnail', fatal=False)
-
         release_date = unified_strdate(self._search_regex(
             r'<div class="release_date">Released: ([^<]+)<', webpage, 'release date', fatal=False))
 

youtube_dl/extractor/lynda.py

@@ -118,9 +118,7 @@ class LyndaIE(LyndaBaseIE):
                 'lynda returned error: %s' % video_json['Message'], expected=True)
 
         if video_json['HasAccess'] is False:
-            raise ExtractorError(
-                'Video %s is only available for members. '
-                % video_id + self._ACCOUNT_CREDENTIALS_HINT, expected=True)
+            self.raise_login_required('Video %s is only available for members' % video_id)
 
         video_id = compat_str(video_json['ID'])
         duration = video_json['DurationInSeconds']

youtube_dl/extractor/mailru.py

@@ -25,6 +25,7 @@ class MailRuIE(InfoExtractor):
                 'uploader_id': 'sonypicturesrus@mail.ru',
                 'duration': 184,
             },
+            'skip': 'Not accessible from Travis CI server',
         },
         {
             'url': 'http://my.mail.ru/corp/hitech/video/news_hi-tech_mail_ru/1263.html',
@@ -39,6 +40,7 @@ class MailRuIE(InfoExtractor):
                 'uploader_id': 'hitech@corp.mail.ru',
                 'duration': 245,
             },
+            'skip': 'Not accessible from Travis CI server',
         },
     ]
 

youtube_dl/extractor/mit.py

@@ -18,12 +18,12 @@ class TechTVMITIE(InfoExtractor):
 
     _TEST = {
         'url': 'http://techtv.mit.edu/videos/25418-mit-dna-learning-center-set',
-        'md5': '1f8cb3e170d41fd74add04d3c9330e5f',
+        'md5': '00a3a27ee20d44bcaa0933ccec4a2cf7',
         'info_dict': {
             'id': '25418',
             'ext': 'mp4',
-            'title': 'MIT DNA Learning Center Set',
-            'description': 'md5:82313335e8a8a3f243351ba55bc1b474',
+            'title': 'MIT DNA and Protein Sets',
+            'description': 'md5:46f5c69ce434f0a97e7c628cc142802d',
         },
     }
 
@@ -33,8 +33,8 @@ class TechTVMITIE(InfoExtractor):
             'http://techtv.mit.edu/videos/%s' % video_id, video_id)
         clean_page = re.compile(r'<!--.*?-->', re.S).sub('', raw_page)
 
-        base_url = self._search_regex(
-            r'ipadUrl: \'(.+?cloudfront.net/)', raw_page, 'base url')
+        base_url = self._proto_relative_url(self._search_regex(
+            r'ipadUrl: \'(.+?cloudfront.net/)', raw_page, 'base url'), 'http:')
         formats_json = self._search_regex(
             r'bitrates: (\[.+?\])', raw_page, 'video formats')
         formats_mit = json.loads(formats_json)

youtube_dl/extractor/mtv.py

@@ -67,7 +67,7 @@ class MTVServicesInfoExtractor(InfoExtractor):
         return [{'url': url, 'ext': 'mp4'}]
 
     def _extract_video_formats(self, mdoc, mtvn_id):
-        if re.match(r'.*/(error_country_block\.swf|geoblock\.mp4)$', mdoc.find('.//src').text) is not None:
+        if re.match(r'.*/(error_country_block\.swf|geoblock\.mp4|copyright_error\.flv(?:\?geo\b.+?)?)$', mdoc.find('.//src').text) is not None:
             if mtvn_id is not None and self._MOBILE_TEMPLATE is not None:
                 self.to_screen('The normal version is not available from your '
                                'country, trying with the mobile version')
@@ -114,7 +114,8 @@ class MTVServicesInfoExtractor(InfoExtractor):
         # Remove the templates, like &device={device}
         mediagen_url = re.sub(r'&[^=]*?={.*?}(?=(&|$))', '', mediagen_url)
         if 'acceptMethods' not in mediagen_url:
-            mediagen_url += '&acceptMethods=fms'
+            mediagen_url += '&' if '?' in mediagen_url else '?'
+            mediagen_url += 'acceptMethods=fms'
 
         mediagen_doc = self._download_xml(mediagen_url, video_id,
                                           'Downloading video urls')
@@ -141,7 +142,7 @@ class MTVServicesInfoExtractor(InfoExtractor):
         if title_el is None:
             title_el = itemdoc.find('.//{http://search.yahoo.com/mrss/}title')
         if title_el is None:
-            title_el = itemdoc.find('.//title')
+            title_el = itemdoc.find('.//title') or itemdoc.find('./title')
             if title_el.text is None:
                 title_el = None
 
@@ -174,8 +175,11 @@ class MTVServicesInfoExtractor(InfoExtractor):
         if self._LANG:
             info_url += 'lang=%s&' % self._LANG
         info_url += data
+        return self._get_videos_info_from_url(info_url, video_id)
+
+    def _get_videos_info_from_url(self, url, video_id):
         idoc = self._download_xml(
-            info_url, video_id,
+            url, video_id,
             'Downloading info', transform_source=fix_xml_ampersands)
         return self.playlist_result(
             [self._get_video_info(item) for item in idoc.findall('.//item')])
@@ -288,3 +292,65 @@ class MTVIggyIE(MTVServicesInfoExtractor):
         }
     }
     _FEED_URL = 'http://all.mtvworldverticals.com/feed-xml/'
+
+
+class MTVDEIE(MTVServicesInfoExtractor):
+    IE_NAME = 'mtv.de'
+    _VALID_URL = r'https?://(?:www\.)?mtv\.de/(?:artists|shows|news)/(?:[^/]+/)*(?P<id>\d+)-[^/#?]+/*(?:[#?].*)?$'
+    _TESTS = [{
+        'url': 'http://www.mtv.de/artists/10571-cro/videos/61131-traum',
+        'info_dict': {
+            'id': 'music_video-a50bc5f0b3aa4b3190aa',
+            'ext': 'mp4',
+            'title': 'MusicVideo_cro-traum',
+            'description': 'Cro - Traum',
+        },
+        'params': {
+            # rtmp download
+            'skip_download': True,
+        },
+    }, {
+        # mediagen URL without query (e.g. http://videos.mtvnn.com/mediagen/e865da714c166d18d6f80893195fcb97)
+        'url': 'http://www.mtv.de/shows/933-teen-mom-2/staffeln/5353/folgen/63565-enthullungen',
+        'info_dict': {
+            'id': 'local_playlist-f5ae778b9832cc837189',
+            'ext': 'mp4',
+            'title': 'Episode_teen-mom-2_shows_season-5_episode-1_full-episode_part1',
+        },
+        'params': {
+            # rtmp download
+            'skip_download': True,
+        },
+    }, {
+        # single video in pagePlaylist with different id
+        'url': 'http://www.mtv.de/news/77491-mtv-movies-spotlight-pixels-teil-3',
+        'info_dict': {
+            'id': 'local_playlist-4e760566473c4c8c5344',
+            'ext': 'mp4',
+            'title': 'Article_mtv-movies-spotlight-pixels-teil-3_short-clips_part1',
+            'description': 'MTV Movies Supercut',
+        },
+        'params': {
+            # rtmp download
+            'skip_download': True,
+        },
+    }]
+
+    def _real_extract(self, url):
+        video_id = self._match_id(url)
+
+        webpage = self._download_webpage(url, video_id)
+
+        playlist = self._parse_json(
+            self._search_regex(
+                r'window\.pagePlaylist\s*=\s*(\[.+?\]);\n', webpage, 'page playlist'),
+            video_id)
+
+        # news pages contain single video in playlist with different id
+        if len(playlist) == 1:
+            return self._get_videos_info_from_url(playlist[0]['mrss'], video_id)
+
+        for item in playlist:
+            item_id = item.get('id')
+            if item_id and compat_str(item_id) == video_id:
+                return self._get_videos_info_from_url(item['mrss'], video_id)

youtube_dl/extractor/musicvault.py

@@ -1,63 +0,0 @@
-from __future__ import unicode_literals
-
-import re
-
-from .common import InfoExtractor
-
-
-class MusicVaultIE(InfoExtractor):
-    _VALID_URL = r'https?://www\.musicvault\.com/(?P<uploader_id>[^/?#]*)/video/(?P<display_id>[^/?#]*)_(?P<id>[0-9]+)\.html'
-    _TEST = {
-        'url': 'http://www.musicvault.com/the-allman-brothers-band/video/straight-from-the-heart_1010863.html',
-        'md5': '3adcbdb3dcc02d647539e53f284ba171',
-        'info_dict': {
-            'id': '1010863',
-            'ext': 'mp4',
-            'uploader_id': 'the-allman-brothers-band',
-            'title': 'Straight from the Heart',
-            'duration': 244,
-            'uploader': 'The Allman Brothers Band',
-            'thumbnail': 're:^https?://.*/thumbnail/.*',
-            'upload_date': '20131219',
-            'location': 'Capitol Theatre (Passaic, NJ)',
-            'description': 'Listen to The Allman Brothers Band perform Straight from the Heart at Capitol Theatre (Passaic, NJ) on Dec 16, 1981',
-            'timestamp': int,
-        }
-    }
-
-    def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
-        display_id = mobj.group('display_id')
-        webpage = self._download_webpage(url, display_id)
-
-        thumbnail = self._search_regex(
-            r'<meta itemprop="thumbnail" content="([^"]+)"',
-            webpage, 'thumbnail', fatal=False)
-
-        data_div = self._search_regex(
-            r'(?s)<div class="data">(.*?)</div>', webpage, 'data fields')
-        uploader = self._html_search_regex(
-            r'<h1.*?>(.*?)</h1>', data_div, 'uploader', fatal=False)
-        title = self._html_search_regex(
-            r'<h2.*?>(.*?)</h2>', data_div, 'title')
-        location = self._html_search_regex(
-            r'<h4.*?>(.*?)</h4>', data_div, 'location', fatal=False)
-
-        kaltura_id = self._search_regex(
-            r'<div id="video-detail-player" data-kaltura-id="([^"]+)"',
-            webpage, 'kaltura ID')
-        wid = self._search_regex(r'/wid/_([0-9]+)/', webpage, 'wid')
-
-        return {
-            'id': mobj.group('id'),
-            '_type': 'url_transparent',
-            'url': 'kaltura:%s:%s' % (wid, kaltura_id),
-            'ie_key': 'Kaltura',
-            'display_id': display_id,
-            'uploader_id': mobj.group('uploader_id'),
-            'thumbnail': thumbnail,
-            'description': self._html_search_meta('description', webpage),
-            'location': location,
-            'title': title,
-            'uploader': uploader,
-        }

youtube_dl/extractor/mwave.py

@@ -0,0 +1,58 @@
+from __future__ import unicode_literals
+
+from .common import InfoExtractor
+from ..compat import compat_str
+from ..utils import (
+    int_or_none,
+    parse_duration,
+)
+
+
+class MwaveIE(InfoExtractor):
+    _VALID_URL = r'https?://mwave\.interest\.me/mnettv/videodetail\.m\?searchVideoDetailVO\.clip_id=(?P<id>[0-9]+)'
+    _TEST = {
+        'url': 'http://mwave.interest.me/mnettv/videodetail.m?searchVideoDetailVO.clip_id=168859',
+        'md5': 'c930e27b7720aaa3c9d0018dfc8ff6cc',
+        'info_dict': {
+            'id': '168859',
+            'ext': 'flv',
+            'title': '[M COUNTDOWN] SISTAR - SHAKE IT',
+            'thumbnail': 're:^https?://.*\.jpg$',
+            'uploader': 'M COUNTDOWN',
+            'duration': 206,
+            'view_count': int,
+        }
+    }
+
+    def _real_extract(self, url):
+        video_id = self._match_id(url)
+
+        vod_info = self._download_json(
+            'http://mwave.interest.me/onair/vod_info.m?vodtype=CL&sectorid=&endinfo=Y&id=%s' % video_id,
+            video_id, 'Download vod JSON')
+
+        formats = []
+        for num, cdn_info in enumerate(vod_info['cdn']):
+            stream_url = cdn_info.get('url')
+            if not stream_url:
+                continue
+            stream_name = cdn_info.get('name') or compat_str(num)
+            f4m_stream = self._download_json(
+                stream_url, video_id,
+                'Download %s stream JSON' % stream_name)
+            f4m_url = f4m_stream.get('fileurl')
+            if not f4m_url:
+                continue
+            formats.extend(
+                self._extract_f4m_formats(f4m_url + '&hdcore=3.0.3', video_id, f4m_id=stream_name))
+        self._sort_formats(formats)
+
+        return {
+            'id': video_id,
+            'title': vod_info['title'],
+            'thumbnail': vod_info.get('cover'),
+            'uploader': vod_info.get('program_title'),
+            'duration': parse_duration(vod_info.get('time')),
+            'view_count': int_or_none(vod_info.get('hit')),
+            'formats': formats,
+        }

youtube_dl/extractor/nbc.py

@@ -236,3 +236,28 @@ class NBCNewsIE(InfoExtractor):
                 'url': info['videoAssets'][-1]['publicUrl'],
                 'ie_key': 'ThePlatform',
             }
+
+
+class MSNBCIE(InfoExtractor):
+    # https URLs redirect to corresponding http ones
+    _VALID_URL = r'http://www\.msnbc\.com/[^/]+/watch/(?P<id>[^/]+)'
+    _TEST = {
+        'url': 'http://www.msnbc.com/all-in-with-chris-hayes/watch/the-chaotic-gop-immigration-vote-314487875924',
+        'md5': '6d236bf4f3dddc226633ce6e2c3f814d',
+        'info_dict': {
+            'id': 'n_hayes_Aimm_140801_272214',
+            'ext': 'mp4',
+            'title': 'The chaotic GOP immigration vote',
+            'description': 'The Republican House votes on a border bill that has no chance of getting through the Senate or signed by the President and is drawing criticism from all sides.',
+            'thumbnail': 're:^https?://.*\.jpg$',
+            'timestamp': 1406937606,
+            'upload_date': '20140802',
+            'categories': ['MSNBC/Topics/Franchise/Best of last night', 'MSNBC/Topics/General/Congress'],
+        },
+    }
+
+    def _real_extract(self, url):
+        video_id = self._match_id(url)
+        webpage = self._download_webpage(url, video_id)
+        embed_url = self._html_search_meta('embedURL', webpage)
+        return self.url_result(embed_url)

youtube_dl/extractor/niconico.py

@@ -12,6 +12,7 @@ from ..compat import (
     compat_urlparse,
 )
 from ..utils import (
+    encode_dict,
     ExtractorError,
     int_or_none,
     parse_duration,
@@ -100,10 +101,7 @@ class NiconicoIE(InfoExtractor):
             'mail': username,
             'password': password,
         }
-        # Convert to UTF-8 *before* urlencode because Python 2.x's urlencode
-        # chokes on unicode
-        login_form = dict((k.encode('utf-8'), v.encode('utf-8')) for k, v in login_form_strs.items())
-        login_data = compat_urllib_parse.urlencode(login_form).encode('utf-8')
+        login_data = compat_urllib_parse.urlencode(encode_dict(login_form_strs)).encode('utf-8')
         request = compat_urllib_request.Request(
             'https://secure.nicovideo.jp/secure/login', login_data)
         login_results = self._download_webpage(

youtube_dl/extractor/nowtv.py

@@ -130,10 +130,16 @@ class NowTVIE(InfoExtractor):
     }, {
         'url': 'http://www.nowtv.at/rtl/bauer-sucht-frau/die-neuen-bauern-und-eine-hochzeit/preview?return=/rtl/bauer-sucht-frau/die-neuen-bauern-und-eine-hochzeit',
         'only_matching': True,
+    }, {
+        'url': 'http://www.nowtv.de/rtl2/echtzeit/list/aktuell/schnelles-geld-am-ende-der-welt/player',
+        'only_matching': True,
     }]
 
     def _real_extract(self, url):
         display_id = self._match_id(url)
+        display_id_split = display_id.split('/')
+        if len(display_id) > 2:
+            display_id = '/'.join((display_id_split[0], display_id_split[-1]))
 
         info = self._download_json(
             'https://api.nowtv.de/v3/movies/%s?fields=id,title,free,geoblocked,articleLong,articleShort,broadcastStartDate,seoUrl,duration,format,files' % display_id,

youtube_dl/extractor/nowvideo.py

@@ -7,7 +7,7 @@ class NowVideoIE(NovaMovIE):
     IE_NAME = 'nowvideo'
     IE_DESC = 'NowVideo'
 
-    _VALID_URL = NovaMovIE._VALID_URL_TEMPLATE % {'host': 'nowvideo\.(?:ch|sx|eu|at|ag|co|li)'}
+    _VALID_URL = NovaMovIE._VALID_URL_TEMPLATE % {'host': 'nowvideo\.(?:ch|ec|sx|eu|at|ag|co|li)'}
 
     _HOST = 'www.nowvideo.ch'
 

youtube_dl/extractor/odnoklassniki.py

@@ -12,7 +12,7 @@ from ..utils import (
 
 
 class OdnoklassnikiIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:odnoklassniki|ok)\.ru/(?:video|web-api/video/moviePlayer)/(?P<id>[\d-]+)'
+    _VALID_URL = r'https?://(?:www\.)?(?:odnoklassniki|ok)\.ru/(?:video|web-api/video/moviePlayer)/(?P<id>[\d-]+)'
     _TESTS = [{
         # metadata in JSON
         'url': 'http://ok.ru/video/20079905452',
@@ -43,9 +43,27 @@ class OdnoklassnikiIE(InfoExtractor):
             'like_count': int,
             'age_limit': 0,
         },
+    }, {
+        # YouTube embed (metadataUrl, provider == USER_YOUTUBE)
+        'url': 'http://ok.ru/video/64211978996595-1',
+        'md5': '5d7475d428845cd2e13bae6f1a992278',
+        'info_dict': {
+            'id': '64211978996595-1',
+            'ext': 'mp4',
+            'title': 'Космическая среда от 26 августа 2015',
+            'description': 'md5:848eb8b85e5e3471a3a803dae1343ed0',
+            'duration': 440,
+            'upload_date': '20150826',
+            'uploader_id': '750099571',
+            'uploader': 'Алина П',
+            'age_limit': 0,
+        },
     }, {
         'url': 'http://ok.ru/web-api/video/moviePlayer/20079905452',
         'only_matching': True,
+    }, {
+        'url': 'http://www.ok.ru/video/20648036891',
+        'only_matching': True,
     }]
 
     def _real_extract(self, url):
@@ -56,7 +74,8 @@ class OdnoklassnikiIE(InfoExtractor):
 
         player = self._parse_json(
             unescapeHTML(self._search_regex(
-                r'data-attributes="([^"]+)"', webpage, 'player')),
+                r'data-options=(?P<quote>["\'])(?P<player>{.+?%s.+?})(?P=quote)' % video_id,
+                webpage, 'player', group='player')),
             video_id)
 
         flashvars = player['flashvars']
@@ -89,16 +108,7 @@ class OdnoklassnikiIE(InfoExtractor):
 
         like_count = int_or_none(metadata.get('likeCount'))
 
-        quality = qualities(('mobile', 'lowest', 'low', 'sd', 'hd'))
-
-        formats = [{
-            'url': f['url'],
-            'ext': 'mp4',
-            'format_id': f['name'],
-            'quality': quality(f['name']),
-        } for f in metadata['videos']]
-
-        return {
+        info = {
             'id': video_id,
             'title': title,
             'thumbnail': thumbnail,
@@ -108,5 +118,24 @@ class OdnoklassnikiIE(InfoExtractor):
             'uploader_id': uploader_id,
             'like_count': like_count,
             'age_limit': age_limit,
-            'formats': formats,
         }
+
+        if metadata.get('provider') == 'USER_YOUTUBE':
+            info.update({
+                '_type': 'url_transparent',
+                'url': movie['contentId'],
+            })
+            return info
+
+        quality = qualities(('mobile', 'lowest', 'low', 'sd', 'hd'))
+
+        formats = [{
+            'url': f['url'],
+            'ext': 'mp4',
+            'format_id': f['name'],
+            'quality': quality(f['name']),
+        } for f in metadata['videos']]
+        self._sort_formats(formats)
+
+        info['formats'] = formats
+        return info

youtube_dl/extractor/playtvak.py

@@ -106,7 +106,7 @@ class PlaytvakIE(InfoExtractor):
         })
 
         info_url = compat_urlparse.urlunparse(
-            parsed_url._replace(query = compat_urllib_parse.urlencode(qs, True)))
+            parsed_url._replace(query=compat_urllib_parse.urlencode(qs, True)))
 
         json_info = self._download_json(
             info_url, video_id,

youtube_dl/extractor/pluralsight.py

@@ -0,0 +1,207 @@
+from __future__ import unicode_literals
+
+import re
+import json
+
+from .common import InfoExtractor
+from ..compat import (
+    compat_str,
+    compat_urllib_parse,
+    compat_urllib_request,
+    compat_urlparse,
+)
+from ..utils import (
+    ExtractorError,
+    int_or_none,
+    parse_duration,
+)
+
+
+class PluralsightIE(InfoExtractor):
+    IE_NAME = 'pluralsight'
+    _VALID_URL = r'https?://(?:www\.)?pluralsight\.com/training/player\?author=(?P<author>[^&]+)&name=(?P<name>[^&]+)(?:&mode=live)?&clip=(?P<clip>\d+)&course=(?P<course>[^&]+)'
+    _LOGIN_URL = 'https://www.pluralsight.com/id/'
+    _NETRC_MACHINE = 'pluralsight'
+
+    _TEST = {
+        'url': 'http://www.pluralsight.com/training/player?author=mike-mckeown&name=hosting-sql-server-windows-azure-iaas-m7-mgmt&mode=live&clip=3&course=hosting-sql-server-windows-azure-iaas',
+        'md5': '4d458cf5cf4c593788672419a8dd4cf8',
+        'info_dict': {
+            'id': 'hosting-sql-server-windows-azure-iaas-m7-mgmt-04',
+            'ext': 'mp4',
+            'title': 'Management of SQL Server - Demo Monitoring',
+            'duration': 338,
+        },
+        'skip': 'Requires pluralsight account credentials',
+    }
+
+    def _real_initialize(self):
+        self._login()
+
+    def _login(self):
+        (username, password) = self._get_login_info()
+        if username is None:
+            self.raise_login_required('Pluralsight account is required')
+
+        login_page = self._download_webpage(
+            self._LOGIN_URL, None, 'Downloading login page')
+
+        login_form = self._hidden_inputs(login_page)
+
+        login_form.update({
+            'Username': username.encode('utf-8'),
+            'Password': password.encode('utf-8'),
+        })
+
+        post_url = self._search_regex(
+            r'<form[^>]+action=(["\'])(?P<url>.+?)\1', login_page,
+            'post url', default=self._LOGIN_URL, group='url')
+
+        if not post_url.startswith('http'):
+            post_url = compat_urlparse.urljoin(self._LOGIN_URL, post_url)
+
+        request = compat_urllib_request.Request(
+            post_url, compat_urllib_parse.urlencode(login_form).encode('utf-8'))
+        request.add_header('Content-Type', 'application/x-www-form-urlencoded')
+
+        response = self._download_webpage(
+            request, None, 'Logging in as %s' % username)
+
+        error = self._search_regex(
+            r'<span[^>]+class="field-validation-error"[^>]*>([^<]+)</span>',
+            response, 'error message', default=None)
+        if error:
+            raise ExtractorError('Unable to login: %s' % error, expected=True)
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        author = mobj.group('author')
+        name = mobj.group('name')
+        clip_id = mobj.group('clip')
+        course = mobj.group('course')
+
+        display_id = '%s-%s' % (name, clip_id)
+
+        webpage = self._download_webpage(url, display_id)
+
+        collection = self._parse_json(
+            self._search_regex(
+                r'moduleCollection\s*:\s*new\s+ModuleCollection\((\[.+?\])\s*,\s*\$rootScope\)',
+                webpage, 'modules'),
+            display_id)
+
+        module, clip = None, None
+
+        for module_ in collection:
+            if module_.get('moduleName') == name:
+                module = module_
+                for clip_ in module_.get('clips', []):
+                    clip_index = clip_.get('clipIndex')
+                    if clip_index is None:
+                        continue
+                    if compat_str(clip_index) == clip_id:
+                        clip = clip_
+                        break
+
+        if not clip:
+            raise ExtractorError('Unable to resolve clip')
+
+        QUALITIES = {
+            'low': {'width': 640, 'height': 480},
+            'medium': {'width': 848, 'height': 640},
+            'high': {'width': 1024, 'height': 768},
+        }
+
+        ALLOWED_QUALITIES = (
+            ('webm', ('high',)),
+            ('mp4', ('low', 'medium', 'high',)),
+        )
+
+        formats = []
+        for ext, qualities in ALLOWED_QUALITIES:
+            for quality in qualities:
+                f = QUALITIES[quality].copy()
+                clip_post = {
+                    'a': author,
+                    'cap': 'false',
+                    'cn': clip_id,
+                    'course': course,
+                    'lc': 'en',
+                    'm': name,
+                    'mt': ext,
+                    'q': '%dx%d' % (f['width'], f['height']),
+                }
+                request = compat_urllib_request.Request(
+                    'http://www.pluralsight.com/training/Player/ViewClip',
+                    json.dumps(clip_post).encode('utf-8'))
+                request.add_header('Content-Type', 'application/json;charset=utf-8')
+                format_id = '%s-%s' % (ext, quality)
+                clip_url = self._download_webpage(
+                    request, display_id, 'Downloading %s URL' % format_id, fatal=False)
+                if not clip_url:
+                    continue
+                f.update({
+                    'url': clip_url,
+                    'ext': ext,
+                    'format_id': format_id,
+                })
+                formats.append(f)
+        self._sort_formats(formats)
+
+        # TODO: captions
+        # http://www.pluralsight.com/training/Player/ViewClip + cap = true
+        # or
+        # http://www.pluralsight.com/training/Player/Captions
+        # { a = author, cn = clip_id, lc = end, m = name }
+
+        return {
+            'id': clip['clipName'],
+            'title': '%s - %s' % (module['title'], clip['title']),
+            'duration': int_or_none(clip.get('duration')) or parse_duration(clip.get('formattedDuration')),
+            'creator': author,
+            'formats': formats
+        }
+
+
+class PluralsightCourseIE(InfoExtractor):
+    IE_NAME = 'pluralsight:course'
+    _VALID_URL = r'https?://(?:www\.)?pluralsight\.com/courses/(?P<id>[^/]+)'
+    _TEST = {
+        # Free course from Pluralsight Starter Subscription for Microsoft TechNet
+        # https://offers.pluralsight.com/technet?loc=zTS3z&prod=zOTprodz&tech=zOttechz&prog=zOTprogz&type=zSOz&media=zOTmediaz&country=zUSz
+        'url': 'http://www.pluralsight.com/courses/hosting-sql-server-windows-azure-iaas',
+        'info_dict': {
+            'id': 'hosting-sql-server-windows-azure-iaas',
+            'title': 'Hosting SQL Server in Microsoft Azure IaaS Fundamentals',
+            'description': 'md5:61b37e60f21c4b2f91dc621a977d0986',
+        },
+        'playlist_count': 31,
+    }
+
+    def _real_extract(self, url):
+        course_id = self._match_id(url)
+
+        # TODO: PSM cookie
+
+        course = self._download_json(
+            'http://www.pluralsight.com/data/course/%s' % course_id,
+            course_id, 'Downloading course JSON')
+
+        title = course['title']
+        description = course.get('description') or course.get('shortDescription')
+
+        course_data = self._download_json(
+            'http://www.pluralsight.com/data/course/content/%s' % course_id,
+            course_id, 'Downloading course data JSON')
+
+        entries = []
+        for module in course_data:
+            for clip in module.get('clips', []):
+                player_parameters = clip.get('playerParameters')
+                if not player_parameters:
+                    continue
+                entries.append(self.url_result(
+                    'http://www.pluralsight.com/training/player?%s' % player_parameters,
+                    'Pluralsight'))
+
+        return self.playlist_result(entries, course_id, title, description)

youtube_dl/extractor/rtl2.py

@@ -1,6 +1,7 @@
 # encoding: utf-8
 from __future__ import unicode_literals
 
+import re
 from .common import InfoExtractor
 
 
@@ -8,22 +9,28 @@ class RTL2IE(InfoExtractor):
     _VALID_URL = r'http?://(?:www\.)?rtl2\.de/[^?#]*?/(?P<id>[^?#/]*?)(?:$|/(?:$|[?#]))'
     _TESTS = [{
         'url': 'http://www.rtl2.de/sendung/grip-das-motormagazin/folge/folge-203-0',
-        'md5': 'bfcc179030535b08dc2b36b469b5adc7',
         'info_dict': {
             'id': 'folge-203-0',
             'ext': 'f4v',
             'title': 'GRIP sucht den Sommerkönig',
             'description': 'Matthias, Det und Helge treten gegeneinander an.'
         },
+        'params': {
+            # rtmp download
+            'skip_download': True,
+        },
     }, {
         'url': 'http://www.rtl2.de/sendung/koeln-50667/video/5512-anna/21040-anna-erwischt-alex/',
-        'md5': 'ffcd517d2805b57ce11a58a2980c2b02',
         'info_dict': {
             'id': '21040-anna-erwischt-alex',
             'ext': 'mp4',
             'title': 'Anna erwischt Alex!',
             'description': 'Anna ist Alex\' Tochter bei Köln 50667.'
         },
+        'params': {
+            # rtmp download
+            'skip_download': True,
+        },
     }]
 
     def _real_extract(self, url):
@@ -34,12 +41,18 @@ class RTL2IE(InfoExtractor):
         video_id = self._match_id(url)
         webpage = self._download_webpage(url, video_id)
 
-        vico_id = self._html_search_regex(
-            r'vico_id\s*:\s*([0-9]+)', webpage, 'vico_id')
-        vivi_id = self._html_search_regex(
-            r'vivi_id\s*:\s*([0-9]+)', webpage, 'vivi_id')
+        mobj = re.search(
+            r'<div[^>]+data-collection="(?P<vico_id>\d+)"[^>]+data-video="(?P<vivi_id>\d+)"',
+            webpage)
+        if mobj:
+            vico_id = mobj.group('vico_id')
+            vivi_id = mobj.group('vivi_id')
+        else:
+            vico_id = self._html_search_regex(
+                r'vico_id\s*:\s*([0-9]+)', webpage, 'vico_id')
+            vivi_id = self._html_search_regex(
+                r'vivi_id\s*:\s*([0-9]+)', webpage, 'vivi_id')
         info_url = 'http://www.rtl2.de/video/php/get_video.php?vico_id=' + vico_id + '&vivi_id=' + vivi_id
-        webpage = self._download_webpage(info_url, '')
 
         info = self._download_json(info_url, video_id)
         video_info = info['video']

youtube_dl/extractor/rtp.py

@@ -18,6 +18,10 @@ class RTPIE(InfoExtractor):
             'description': 'As paixões musicais de António Cartaxo e António Macedo',
             'thumbnail': 're:^https?://.*\.jpg',
         },
+        'params': {
+            # rtmp download
+            'skip_download': True,
+        },
     }, {
         'url': 'http://www.rtp.pt/play/p831/a-quimica-das-coisas',
         'only_matching': True,

youtube_dl/extractor/rtve.py

@@ -6,7 +6,7 @@ import re
 import time
 
 from .common import InfoExtractor
-from ..compat import compat_urlparse
+from ..compat import compat_urllib_request, compat_urlparse
 from ..utils import (
     ExtractorError,
     float_or_none,
@@ -102,7 +102,9 @@ class RTVEALaCartaIE(InfoExtractor):
         if info['state'] == 'DESPU':
             raise ExtractorError('The video is no longer available', expected=True)
         png_url = 'http://www.rtve.es/ztnr/movil/thumbnail/%s/videos/%s.png' % (self._manager, video_id)
-        png = self._download_webpage(png_url, video_id, 'Downloading url information')
+        png_request = compat_urllib_request.Request(png_url)
+        png_request.add_header('Referer', url)
+        png = self._download_webpage(png_request, video_id, 'Downloading url information')
         video_url = _decrypt_url(png)
         if not video_url.endswith('.f4m'):
             auth_url = video_url.replace(

youtube_dl/extractor/ruutu.py

@@ -6,19 +6,19 @@ from ..compat import compat_urllib_parse_urlparse
 from ..utils import (
     determine_ext,
     int_or_none,
+    xpath_attr,
     xpath_text,
 )
 
 
 class RuutuIE(InfoExtractor):
-    _VALID_URL = r'http://(?:www\.)?ruutu\.fi/ohjelmat/(?:[^/?#]+/)*(?P<id>[^/?#]+)'
+    _VALID_URL = r'https?://(?:www\.)?ruutu\.fi/video/(?P<id>\d+)'
     _TESTS = [
         {
-            'url': 'http://www.ruutu.fi/ohjelmat/oletko-aina-halunnut-tietaa-mita-tapahtuu-vain-hetki-ennen-lahetysta-nyt-se-selvisi',
+            'url': 'http://www.ruutu.fi/video/2058907',
             'md5': 'ab2093f39be1ca8581963451b3c0234f',
             'info_dict': {
                 'id': '2058907',
-                'display_id': 'oletko-aina-halunnut-tietaa-mita-tapahtuu-vain-hetki-ennen-lahetysta-nyt-se-selvisi',
                 'ext': 'mp4',
                 'title': 'Oletko aina halunnut tietää mitä tapahtuu vain hetki ennen lähetystä? - Nyt se selvisi!',
                 'description': 'md5:cfc6ccf0e57a814360df464a91ff67d6',
@@ -28,14 +28,13 @@ class RuutuIE(InfoExtractor):
             },
         },
         {
-            'url': 'http://www.ruutu.fi/ohjelmat/superpesis/superpesis-katso-koko-kausi-ruudussa',
+            'url': 'http://www.ruutu.fi/video/2057306',
             'md5': '065a10ae4d5b8cfd9d0c3d332465e3d9',
             'info_dict': {
                 'id': '2057306',
-                'display_id': 'superpesis-katso-koko-kausi-ruudussa',
                 'ext': 'mp4',
                 'title': 'Superpesis: katso koko kausi Ruudussa',
-                'description': 'md5:44c44a99fdbe5b380ab74ebd75f0af77',
+                'description': 'md5:da2736052fef3b2bd5e0005e63c25eac',
                 'thumbnail': 're:^https?://.*\.jpg$',
                 'duration': 40,
                 'age_limit': 0,
@@ -44,29 +43,10 @@ class RuutuIE(InfoExtractor):
     ]
 
     def _real_extract(self, url):
-        display_id = self._match_id(url)
+        video_id = self._match_id(url)
 
-        webpage = self._download_webpage(url, display_id)
-
-        video_id = self._search_regex(
-            r'data-media-id="(\d+)"', webpage, 'media id')
-
-        video_xml_url = None
-
-        media_data = self._search_regex(
-            r'jQuery\.extend\([^,]+,\s*(.+?)\);', webpage,
-            'media data', default=None)
-        if media_data:
-            media_json = self._parse_json(media_data, display_id, fatal=False)
-            if media_json:
-                xml_url = media_json.get('ruutuplayer', {}).get('xmlUrl')
-                if xml_url:
-                    video_xml_url = xml_url.replace('{ID}', video_id)
-
-        if not video_xml_url:
-            video_xml_url = 'http://gatling.ruutu.fi/media-xml-cache?id=%s' % video_id
-
-        video_xml = self._download_xml(video_xml_url, video_id)
+        video_xml = self._download_xml(
+            'http://gatling.ruutu.fi/media-xml-cache?id=%s' % video_id, video_id)
 
         formats = []
         processed_urls = []
@@ -109,10 +89,9 @@ class RuutuIE(InfoExtractor):
 
         return {
             'id': video_id,
-            'display_id': display_id,
-            'title': self._og_search_title(webpage),
-            'description': self._og_search_description(webpage),
-            'thumbnail': self._og_search_thumbnail(webpage),
+            'title': xpath_attr(video_xml, './/Behavior/Program', 'program_name', 'title', fatal=True),
+            'description': xpath_attr(video_xml, './/Behavior/Program', 'description', 'description'),
+            'thumbnail': xpath_attr(video_xml, './/Behavior/Startpicture', 'href', 'thumbnail'),
             'duration': int_or_none(xpath_text(video_xml, './/Runtime', 'duration')),
             'age_limit': int_or_none(xpath_text(video_xml, './/AgeLimit', 'age limit')),
             'formats': formats,

youtube_dl/extractor/safari.py

@@ -20,7 +20,6 @@ from ..utils import (
 class SafariBaseIE(InfoExtractor):
     _LOGIN_URL = 'https://www.safaribooksonline.com/accounts/login/'
     _SUCCESSFUL_LOGIN_REGEX = r'<a href="/accounts/logout/"[^>]*>Sign Out</a>'
-    _ACCOUNT_CREDENTIALS_HINT = 'Use --username and --password options to supply credentials for safaribooksonline.com'
     _NETRC_MACHINE = 'safari'
 
     _API_BASE = 'https://www.safaribooksonline.com/api/v1/book'
@@ -37,9 +36,7 @@ class SafariBaseIE(InfoExtractor):
     def _login(self):
         (username, password) = self._get_login_info()
         if username is None:
-            raise ExtractorError(
-                self._ACCOUNT_CREDENTIALS_HINT,
-                expected=True)
+            self.raise_login_required('safaribooksonline.com account is required')
 
         headers = std_headers
         if 'Referer' not in headers:

youtube_dl/extractor/screenwavemedia.py

@@ -12,8 +12,8 @@ from ..utils import (
 
 
 class ScreenwaveMediaIE(InfoExtractor):
-    _VALID_URL = r'http://player\d?\.screenwavemedia\.com/(?:play/)?[a-zA-Z]+\.php\?[^"]*\bid=(?P<id>.+)'
-
+    _VALID_URL = r'https?://player\d?\.screenwavemedia\.com/(?:play/)?[a-zA-Z]+\.php\?.*\bid=(?P<id>[A-Za-z0-9-]+)'
+    EMBED_PATTERN = r'src=(["\'])(?P<url>(?:https?:)?//player\d?\.screenwavemedia\.com/(?:play/)?[a-zA-Z]+\.php\?.*\bid=.+?)\1'
     _TESTS = [{
         'url': 'http://player.screenwavemedia.com/play/play.php?playerdiv=videoarea&companiondiv=squareAd&id=Cinemassacre-19911',
         'only_matching': True,
@@ -33,7 +33,7 @@ class ScreenwaveMediaIE(InfoExtractor):
             'http://player.screenwavemedia.com/player.js',
             video_id, 'Downloading playerconfig webpage')
 
-        videoserver = self._search_regex(r"\[ipaddress\]\s*=>\s*([\d\.]+)", playerdata, 'videoserver')
+        videoserver = self._search_regex(r'SWMServer\s*=\s*"([\d\.]+)"', playerdata, 'videoserver')
 
         sources = self._parse_json(
             js_to_json(
@@ -56,6 +56,7 @@ class ScreenwaveMediaIE(InfoExtractor):
 
         # Fallback to hardcoded sources if JS changes again
         if not sources:
+            self.report_warning('Falling back to a hardcoded list of streams')
             sources = [{
                 'file': 'http://%s/vod/%s_%s.mp4' % (videoserver, video_id, format_id),
                 'type': 'mp4',

youtube_dl/extractor/shared.py

@@ -14,17 +14,28 @@ from ..utils import (
 
 
 class SharedIE(InfoExtractor):
-    _VALID_URL = r'http://shared\.sx/(?P<id>[\da-z]{10})'
+    IE_DESC = 'shared.sx and vivo.sx'
+    _VALID_URL = r'http://(?:shared|vivo)\.sx/(?P<id>[\da-z]{10})'
 
-    _TEST = {
+    _TESTS = [{
         'url': 'http://shared.sx/0060718775',
         'md5': '106fefed92a8a2adb8c98e6a0652f49b',
         'info_dict': {
             'id': '0060718775',
             'ext': 'mp4',
             'title': 'Bmp4',
+            'filesize': 1720110,
         },
-    }
+    }, {
+        'url': 'http://vivo.sx/d7ddda0e78',
+        'md5': '15b3af41be0b4fe01f4df075c2678b2c',
+        'info_dict': {
+            'id': 'd7ddda0e78',
+            'ext': 'mp4',
+            'title': 'Chicken',
+            'filesize': 528031,
+        },
+    }]
 
     def _real_extract(self, url):
         video_id = self._match_id(url)

youtube_dl/extractor/smotri.py

@@ -330,10 +330,7 @@ class SmotriBroadcastIE(InfoExtractor):
 
             (username, password) = self._get_login_info()
             if username is None:
-                raise ExtractorError(
-                    'Erotic broadcasts allowed only for registered users, '
-                    'use --username and --password options to provide account credentials.',
-                    expected=True)
+                self.raise_login_required('Erotic broadcasts allowed only for registered users')
 
             login_form = {
                 'login-hint53': '1',

youtube_dl/extractor/soundcloud.py

@@ -309,7 +309,7 @@ class SoundcloudUserIE(SoundcloudIE):
             'id': '114582580',
             'title': 'The Akashic Chronicler (All)',
         },
-        'playlist_mincount': 112,
+        'playlist_mincount': 111,
     }, {
         'url': 'https://soundcloud.com/the-akashic-chronicler/tracks',
         'info_dict': {
@@ -330,14 +330,14 @@ class SoundcloudUserIE(SoundcloudIE):
             'id': '114582580',
             'title': 'The Akashic Chronicler (Reposts)',
         },
-        'playlist_mincount': 9,
+        'playlist_mincount': 7,
     }, {
         'url': 'https://soundcloud.com/the-akashic-chronicler/likes',
         'info_dict': {
             'id': '114582580',
             'title': 'The Akashic Chronicler (Likes)',
         },
-        'playlist_mincount': 333,
+        'playlist_mincount': 321,
     }, {
         'url': 'https://soundcloud.com/grynpyret/spotlight',
         'info_dict': {

youtube_dl/extractor/spankwire.py

@@ -16,8 +16,9 @@ from ..aes import aes_decrypt_text
 
 
 class SpankwireIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:www\.)?(?P<url>spankwire\.com/[^/]*/video(?P<videoid>[0-9]+)/?)'
-    _TEST = {
+    _VALID_URL = r'https?://(?:www\.)?(?P<url>spankwire\.com/[^/]*/video(?P<id>[0-9]+)/?)'
+    _TESTS = [{
+        # download URL pattern: */<height>P_<tbr>K_<video_id>.mp4
         'url': 'http://www.spankwire.com/Buckcherry-s-X-Rated-Music-Video-Crazy-Bitch/video103545/',
         'md5': '8bbfde12b101204b39e4b9fe7eb67095',
         'info_dict': {
@@ -30,14 +31,27 @@ class SpankwireIE(InfoExtractor):
             'upload_date': '20070507',
             'age_limit': 18,
         }
-    }
+    }, {
+        # download URL pattern: */mp4_<format_id>_<video_id>.mp4
+        'url': 'http://www.spankwire.com/Titcums-Compiloation-I/video1921551/',
+        'md5': '09b3c20833308b736ae8902db2f8d7e6',
+        'info_dict': {
+            'id': '1921551',
+            'ext': 'mp4',
+            'title': 'Titcums Compiloation I',
+            'description': 'cum on tits',
+            'uploader': 'dannyh78999',
+            'uploader_id': '3056053',
+            'upload_date': '20150822',
+            'age_limit': 18,
+        },
+    }]
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
-        video_id = mobj.group('videoid')
-        url = 'http://www.' + mobj.group('url')
+        video_id = mobj.group('id')
 
-        req = compat_urllib_request.Request(url)
+        req = compat_urllib_request.Request('http://www.' + mobj.group('url'))
         req.add_header('Cookie', 'age_verified=1')
         webpage = self._download_webpage(req, video_id)
 
@@ -54,7 +68,7 @@ class SpankwireIE(InfoExtractor):
             r'by:\s*<a [^>]*>(.+?)</a>',
             webpage, 'uploader', fatal=False)
         uploader_id = self._html_search_regex(
-            r'by:\s*<a href="/Profile\.aspx\?.*?UserId=(\d+).*?"',
+            r'by:\s*<a href="/(?:user/viewProfile|Profile\.aspx)\?.*?UserId=(\d+).*?"',
             webpage, 'uploader id', fatal=False)
         upload_date = unified_strdate(self._html_search_regex(
             r'</a> on (.+?) at \d+:\d+',
@@ -67,9 +81,10 @@ class SpankwireIE(InfoExtractor):
             r'<span\s+id="spCommentCount"[^>]*>([\d,\.]+)</span>',
             webpage, 'comment count', fatal=False))
 
-        video_urls = list(map(
-            compat_urllib_parse_unquote,
-            re.findall(r'playerData\.cdnPath[0-9]{3,}\s*=\s*(?:encodeURIComponent\()?["\']([^"\']+)["\']', webpage)))
+        videos = re.findall(
+            r'playerData\.cdnPath([0-9]{3,})\s*=\s*(?:encodeURIComponent\()?["\']([^"\']+)["\']', webpage)
+        heights = [int(video[0]) for video in videos]
+        video_urls = list(map(compat_urllib_parse_unquote, [video[1] for video in videos]))
         if webpage.find('flashvars\.encrypted = "true"') != -1:
             password = self._search_regex(
                 r'flashvars\.video_title = "([^"]+)',
@@ -79,21 +94,22 @@ class SpankwireIE(InfoExtractor):
                 video_urls))
 
         formats = []
-        for video_url in video_urls:
+        for height, video_url in zip(heights, video_urls):
             path = compat_urllib_parse_urlparse(video_url).path
-            format = path.split('/')[4].split('_')[:2]
-            resolution, bitrate_str = format
-            format = "-".join(format)
-            height = int(resolution.rstrip('Pp'))
-            tbr = int(bitrate_str.rstrip('Kk'))
-            formats.append({
+            _, quality = path.split('/')[4].split('_')[:2]
+            f = {
                 'url': video_url,
-                'resolution': resolution,
-                'format': format,
-                'tbr': tbr,
                 'height': height,
-                'format_id': format,
-            })
+            }
+            tbr = self._search_regex(r'^(\d+)[Kk]$', quality, 'tbr', default=None)
+            if tbr:
+                f.update({
+                    'tbr': int(tbr),
+                    'format_id': '%dp' % height,
+                })
+            else:
+                f['format_id'] = quality
+            formats.append(f)
         self._sort_formats(formats)
 
         age_limit = self._rta_search(webpage)

youtube_dl/extractor/telecinco.py

@@ -6,7 +6,7 @@ from .mitele import MiTeleIE
 
 class TelecincoIE(MiTeleIE):
     IE_NAME = 'telecinco.es'
-    _VALID_URL = r'https?://www\.telecinco\.es/[^/]+/[^/]+/(?:[^/]+/)?(?P<id>.*?)\.html'
+    _VALID_URL = r'https?://www\.telecinco\.es/(?:[^/]+/)+(?P<id>.+?)\.html'
 
     _TESTS = [{
         'url': 'http://www.telecinco.es/robinfood/temporada-01/t01xp14/Bacalao-cocochas-pil-pil_0_1876350223.html',
@@ -23,4 +23,7 @@ class TelecincoIE(MiTeleIE):
     }, {
         'url': 'http://www.telecinco.es/informativos/nacional/Pablo_Iglesias-Informativos_Telecinco-entrevista-Pedro_Piqueras_2_1945155182.html',
         'only_matching': True,
+    }, {
+        'url': 'http://www.telecinco.es/espanasinirmaslejos/Espana-gran-destino-turistico_2_1240605043.html',
+        'only_matching': True,
     }]

youtube_dl/extractor/theplatform.py

@@ -1,7 +1,7 @@
+# -*- coding: utf-8 -*-
 from __future__ import unicode_literals
 
 import re
-import json
 import time
 import hmac
 import binascii
@@ -10,7 +10,8 @@ import hashlib
 
 from .common import InfoExtractor
 from ..compat import (
-    compat_str,
+    compat_parse_qs,
+    compat_urllib_parse_urlparse,
 )
 from ..utils import (
     determine_ext,
@@ -18,12 +19,69 @@ from ..utils import (
     xpath_with_ns,
     unsmuggle_url,
     int_or_none,
+    url_basename,
+    float_or_none,
 )
 
-_x = lambda p: xpath_with_ns(p, {'smil': 'http://www.w3.org/2005/SMIL21/Language'})
+default_ns = 'http://www.w3.org/2005/SMIL21/Language'
+_x = lambda p: xpath_with_ns(p, {'smil': default_ns})
 
 
-class ThePlatformIE(InfoExtractor):
+class ThePlatformBaseIE(InfoExtractor):
+    def _extract_theplatform_smil(self, smil_url, video_id, note='Downloading SMIL data'):
+        meta = self._download_xml(smil_url, video_id, note=note)
+        try:
+            error_msg = next(
+                n.attrib['abstract']
+                for n in meta.findall(_x('.//smil:ref'))
+                if n.attrib.get('title') == 'Geographic Restriction' or n.attrib.get('title') == 'Expired')
+        except StopIteration:
+            pass
+        else:
+            raise ExtractorError(error_msg, expected=True)
+
+        formats = self._parse_smil_formats(
+            meta, smil_url, video_id, namespace=default_ns,
+            # the parameters are from syfy.com, other sites may use others,
+            # they also work for nbc.com
+            f4m_params={'g': 'UXWGVKRWHFSP', 'hdcore': '3.0.3'},
+            transform_rtmp_url=lambda streamer, src: (streamer, 'mp4:' + src))
+
+        for _format in formats:
+            ext = determine_ext(_format['url'])
+            if ext == 'once':
+                _format['ext'] = 'mp4'
+
+        self._sort_formats(formats)
+
+        subtitles = self._parse_smil_subtitles(meta, default_ns)
+
+        return formats, subtitles
+
+    def get_metadata(self, path, video_id):
+        info_url = 'http://link.theplatform.com/s/%s?format=preview' % path
+        info = self._download_json(info_url, video_id)
+
+        subtitles = {}
+        captions = info.get('captions')
+        if isinstance(captions, list):
+            for caption in captions:
+                lang, src, mime = caption.get('lang', 'en'), caption.get('src'), caption.get('type')
+                subtitles[lang] = [{
+                    'ext': 'srt' if mime == 'text/srt' else 'ttml',
+                    'url': src,
+                }]
+
+        return {
+            'title': info['title'],
+            'subtitles': subtitles,
+            'description': info['description'],
+            'thumbnail': info['defaultThumbnailUrl'],
+            'duration': int_or_none(info.get('duration'), 1000),
+        }
+
+
+class ThePlatformIE(ThePlatformBaseIE):
     _VALID_URL = r'''(?x)
         (?:https?://(?:link|player)\.theplatform\.com/[sp]/(?P<provider_id>[^/]+)/
            (?:(?P<media>(?:[^/]+/)+select/media/)|(?P<config>(?:[^/\?]+/(?:swf|config)|onsite)/select/))?
@@ -67,6 +125,20 @@ class ThePlatformIE(InfoExtractor):
     }, {
         'url': 'http://player.theplatform.com/p/NnzsPC/widget/select/media/4Y0TlYUr_ZT7',
         'only_matching': True,
+    }, {
+        'url': 'http://player.theplatform.com/p/2E2eJC/nbcNewsOffsite?guid=tdy_or_siri_150701',
+        'md5': '734f3790fb5fc4903da391beeebc4836',
+        'info_dict': {
+            'id': 'tdy_or_siri_150701',
+            'ext': 'mp4',
+            'title': 'iPhone Siri’s sassy response to a math question has people talking',
+            'description': 'md5:a565d1deadd5086f3331d57298ec6333',
+            'duration': 83.0,
+            'thumbnail': 're:^https?://.*\.jpg$',
+            'timestamp': 1435752600,
+            'upload_date': '20150701',
+            'categories': ['Today/Shows/Orange Room', 'Today/Sections/Money', 'Today/Topics/Tech', "Today/Topics/Editor's picks"],
+        },
     }]
 
     @staticmethod
@@ -101,6 +173,24 @@ class ThePlatformIE(InfoExtractor):
             path += '/media'
         path += '/' + video_id
 
+        qs_dict = compat_parse_qs(compat_urllib_parse_urlparse(url).query)
+        if 'guid' in qs_dict:
+            webpage = self._download_webpage(url, video_id)
+            scripts = re.findall(r'<script[^>]+src="([^"]+)"', webpage)
+            feed_id = None
+            # feed id usually locates in the last script.
+            # Seems there's no pattern for the interested script filename, so
+            # I try one by one
+            for script in reversed(scripts):
+                feed_script = self._download_webpage(script, video_id, 'Downloading feed script')
+                feed_id = self._search_regex(r'defaultFeedId\s*:\s*"([^"]+)"', feed_script, 'default feed id', default=None)
+                if feed_id is not None:
+                    break
+            if feed_id is None:
+                raise ExtractorError('Unable to find feed id')
+            return self.url_result('http://feed.theplatform.com/f/%s/%s?byGuid=%s' % (
+                provider_id, feed_id, qs_dict['guid'][0]))
+
         if smuggled_data.get('force_smil_url', False):
             smil_url = url
         elif mobj.group('config'):
@@ -120,95 +210,85 @@ class ThePlatformIE(InfoExtractor):
         if sig:
             smil_url = self._sign_url(smil_url, sig['key'], sig['secret'])
 
-        meta = self._download_xml(smil_url, video_id)
-        try:
-            error_msg = next(
-                n.attrib['abstract']
-                for n in meta.findall(_x('.//smil:ref'))
-                if n.attrib.get('title') == 'Geographic Restriction' or n.attrib.get('title') == 'Expired')
-        except StopIteration:
-            pass
-        else:
-            raise ExtractorError(error_msg, expected=True)
+        formats, subtitles = self._extract_theplatform_smil(smil_url, video_id)
 
-        info_url = 'http://link.theplatform.com/s/%s?format=preview' % path
-        info_json = self._download_webpage(info_url, video_id)
-        info = json.loads(info_json)
-
-        subtitles = {}
-        captions = info.get('captions')
-        if isinstance(captions, list):
-            for caption in captions:
-                lang, src, mime = caption.get('lang', 'en'), caption.get('src'), caption.get('type')
-                subtitles[lang] = [{
-                    'ext': 'srt' if mime == 'text/srt' else 'ttml',
-                    'url': src,
-                }]
-
-        head = meta.find(_x('smil:head'))
-        body = meta.find(_x('smil:body'))
-
-        f4m_node = body.find(_x('smil:seq//smil:video'))
-        if f4m_node is None:
-            f4m_node = body.find(_x('smil:seq/smil:video'))
-        if f4m_node is not None and '.f4m' in f4m_node.attrib['src']:
-            f4m_url = f4m_node.attrib['src']
-            if 'manifest.f4m?' not in f4m_url:
-                f4m_url += '?'
-            # the parameters are from syfy.com, other sites may use others,
-            # they also work for nbc.com
-            f4m_url += '&g=UXWGVKRWHFSP&hdcore=3.0.3'
-            formats = self._extract_f4m_formats(f4m_url, video_id)
-        else:
-            formats = []
-            switch = body.find(_x('smil:switch'))
-            if switch is None:
-                switch = body.find(_x('smil:par//smil:switch'))
-            if switch is None:
-                switch = body.find(_x('smil:par/smil:switch'))
-            if switch is None:
-                switch = body.find(_x('smil:par'))
-            if switch is not None:
-                base_url = head.find(_x('smil:meta')).attrib['base']
-                for f in switch.findall(_x('smil:video')):
-                    attr = f.attrib
-                    width = int_or_none(attr.get('width'))
-                    height = int_or_none(attr.get('height'))
-                    vbr = int_or_none(attr.get('system-bitrate'), 1000)
-                    format_id = '%dx%d_%dk' % (width, height, vbr)
-                    formats.append({
-                        'format_id': format_id,
-                        'url': base_url,
-                        'play_path': 'mp4:' + attr['src'],
-                        'ext': 'flv',
-                        'width': width,
-                        'height': height,
-                        'vbr': vbr,
-                    })
-            else:
-                switch = body.find(_x('smil:seq//smil:switch'))
-                if switch is None:
-                    switch = body.find(_x('smil:seq/smil:switch'))
-                for f in switch.findall(_x('smil:video')):
-                    attr = f.attrib
-                    vbr = int_or_none(attr.get('system-bitrate'), 1000)
-                    ext = determine_ext(attr['src'])
-                    if ext == 'once':
-                        ext = 'mp4'
-                    formats.append({
-                        'format_id': compat_str(vbr),
-                        'url': attr['src'],
-                        'vbr': vbr,
-                        'ext': ext,
-                    })
-            self._sort_formats(formats)
-
-        return {
+        ret = self.get_metadata(path, video_id)
+        combined_subtitles = self._merge_subtitles(ret.get('subtitles', {}), subtitles)
+        ret.update({
             'id': video_id,
-            'title': info['title'],
-            'subtitles': subtitles,
             'formats': formats,
-            'description': info['description'],
-            'thumbnail': info['defaultThumbnailUrl'],
-            'duration': int_or_none(info.get('duration'), 1000),
-        }
+            'subtitles': combined_subtitles,
+        })
+
+        return ret
+
+
+class ThePlatformFeedIE(ThePlatformBaseIE):
+    _URL_TEMPLATE = '%s//feed.theplatform.com/f/%s/%s?form=json&byGuid=%s'
+    _VALID_URL = r'https?://feed\.theplatform\.com/f/(?P<provider_id>[^/]+)/(?P<feed_id>[^?/]+)\?(?:[^&]+&)*byGuid=(?P<id>[a-zA-Z0-9_]+)'
+    _TEST = {
+        # From http://player.theplatform.com/p/7wvmTC/MSNBCEmbeddedOffSite?guid=n_hardball_5biden_140207
+        'url': 'http://feed.theplatform.com/f/7wvmTC/msnbc_video-p-test?form=json&pretty=true&range=-40&byGuid=n_hardball_5biden_140207',
+        'md5': '22d2b84f058d3586efcd99e57d59d314',
+        'info_dict': {
+            'id': 'n_hardball_5biden_140207',
+            'ext': 'mp4',
+            'title': 'The Biden factor: will Joe run in 2016?',
+            'description': 'Could Vice President Joe Biden be preparing a 2016 campaign? Mark Halperin and Sam Stein weigh in.',
+            'thumbnail': 're:^https?://.*\.jpg$',
+            'upload_date': '20140208',
+            'timestamp': 1391824260,
+            'duration': 467.0,
+            'categories': ['MSNBC/Issues/Democrats', 'MSNBC/Issues/Elections/Election 2016'],
+        },
+    }
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+
+        video_id = mobj.group('id')
+        provider_id = mobj.group('provider_id')
+        feed_id = mobj.group('feed_id')
+
+        real_url = self._URL_TEMPLATE % (self.http_scheme(), provider_id, feed_id, video_id)
+        feed = self._download_json(real_url, video_id)
+        entry = feed['entries'][0]
+
+        formats = []
+        subtitles = {}
+        first_video_id = None
+        duration = None
+        for item in entry['media$content']:
+            smil_url = item['plfile$url'] + '&format=SMIL&Tracking=true&Embedded=true&formats=MPEG4,F4M'
+            cur_video_id = url_basename(smil_url)
+            if first_video_id is None:
+                first_video_id = cur_video_id
+                duration = float_or_none(item.get('plfile$duration'))
+            cur_formats, cur_subtitles = self._extract_theplatform_smil(smil_url, video_id, 'Downloading SMIL data for %s' % cur_video_id)
+            formats.extend(cur_formats)
+            subtitles = self._merge_subtitles(subtitles, cur_subtitles)
+
+        self._sort_formats(formats)
+
+        thumbnails = [{
+            'url': thumbnail['plfile$url'],
+            'width': int_or_none(thumbnail.get('plfile$width')),
+            'height': int_or_none(thumbnail.get('plfile$height')),
+        } for thumbnail in entry.get('media$thumbnails', [])]
+
+        timestamp = int_or_none(entry.get('media$availableDate'), scale=1000)
+        categories = [item['media$name'] for item in entry.get('media$categories', [])]
+
+        ret = self.get_metadata('%s/%s' % (provider_id, first_video_id), video_id)
+        subtitles = self._merge_subtitles(subtitles, ret['subtitles'])
+        ret.update({
+            'id': video_id,
+            'formats': formats,
+            'subtitles': subtitles,
+            'thumbnails': thumbnails,
+            'duration': duration,
+            'timestamp': timestamp,
+            'categories': categories,
+        })
+
+        return ret

youtube_dl/extractor/tubitv.py

@@ -60,9 +60,7 @@ class TubiTvIE(InfoExtractor):
 
         webpage = self._download_webpage(url, video_id)
         if re.search(r"<(?:DIV|div) class='login-required-screen'>", webpage):
-            raise ExtractorError(
-                'This video requires login, use --username and --password '
-                'options to provide account credentials.', expected=True)
+            self.raise_login_required('This video requires login')
 
         title = self._og_search_title(webpage)
         description = self._og_search_description(webpage)

youtube_dl/extractor/twitch.py

@@ -12,9 +12,11 @@ from ..compat import (
     compat_urllib_parse,
     compat_urllib_parse_urlparse,
     compat_urllib_request,
+    compat_urlparse,
 )
 from ..utils import (
     ExtractorError,
+    int_or_none,
     parse_duration,
     parse_iso8601,
 )
@@ -26,7 +28,7 @@ class TwitchBaseIE(InfoExtractor):
     _API_BASE = 'https://api.twitch.tv'
     _USHER_BASE = 'http://usher.twitch.tv'
     _LOGIN_URL = 'https://secure.twitch.tv/login'
-    _LOGIN_POST_URL = 'https://passport.twitch.tv/authorize'
+    _LOGIN_POST_URL = 'https://passport.twitch.tv/authentications/new'
     _NETRC_MACHINE = 'twitch'
 
     def _handle_error(self, response):
@@ -69,8 +71,15 @@ class TwitchBaseIE(InfoExtractor):
             'password': password.encode('utf-8'),
         })
 
+        post_url = self._search_regex(
+            r'<form[^>]+action=(["\'])(?P<url>.+?)\1', login_page,
+            'post url', default=self._LOGIN_POST_URL, group='url')
+
+        if not post_url.startswith('http'):
+            post_url = compat_urlparse.urljoin(self._LOGIN_URL, post_url)
+
         request = compat_urllib_request.Request(
-            self._LOGIN_POST_URL, compat_urllib_parse.urlencode(login_form).encode('utf-8'))
+            post_url, compat_urllib_parse.urlencode(login_form).encode('utf-8'))
         request.add_header('Referer', self._LOGIN_URL)
         response = self._download_webpage(
             request, None, 'Logging in as %s' % username)
@@ -132,14 +141,14 @@ class TwitchItemBaseIE(TwitchBaseIE):
     def _extract_info(self, info):
         return {
             'id': info['_id'],
-            'title': info['title'],
-            'description': info['description'],
-            'duration': info['length'],
-            'thumbnail': info['preview'],
-            'uploader': info['channel']['display_name'],
-            'uploader_id': info['channel']['name'],
-            'timestamp': parse_iso8601(info['recorded_at']),
-            'view_count': info['views'],
+            'title': info.get('title') or 'Untitled Broadcast',
+            'description': info.get('description'),
+            'duration': int_or_none(info.get('length')),
+            'thumbnail': info.get('preview'),
+            'uploader': info.get('channel', {}).get('display_name'),
+            'uploader_id': info.get('channel', {}).get('name'),
+            'timestamp': parse_iso8601(info.get('recorded_at')),
+            'view_count': int_or_none(info.get('views')),
         }
 
     def _real_extract(self, url):
@@ -187,7 +196,7 @@ class TwitchVodIE(TwitchItemBaseIE):
     _ITEM_TYPE = 'vod'
     _ITEM_SHORTCUT = 'v'
 
-    _TEST = {
+    _TESTS = [{
         'url': 'http://www.twitch.tv/riotgames/v/6528877?t=5m10s',
         'info_dict': {
             'id': 'v6528877',
@@ -206,7 +215,26 @@ class TwitchVodIE(TwitchItemBaseIE):
             # m3u8 download
             'skip_download': True,
         },
-    }
+    }, {
+        # Untitled broadcast (title is None)
+        'url': 'http://www.twitch.tv/belkao_o/v/11230755',
+        'info_dict': {
+            'id': 'v11230755',
+            'ext': 'mp4',
+            'title': 'Untitled Broadcast',
+            'thumbnail': 're:^https?://.*\.jpg$',
+            'duration': 1638,
+            'timestamp': 1439746708,
+            'upload_date': '20150816',
+            'uploader': 'BelkAO_o',
+            'uploader_id': 'belkao_o',
+            'view_count': int,
+        },
+        'params': {
+            # m3u8 download
+            'skip_download': True,
+        },
+    }]
 
     def _real_extract(self, url):
         item_id = self._match_id(url)

youtube_dl/extractor/udemy.py

@@ -70,9 +70,7 @@ class UdemyIE(InfoExtractor):
     def _login(self):
         (username, password) = self._get_login_info()
         if username is None:
-            raise ExtractorError(
-                'Udemy account is required, use --username and --password options to provide account credentials.',
-                expected=True)
+            self.raise_login_required('Udemy account is required')
 
         login_popup = self._download_webpage(
             self._LOGIN_URL, None, 'Downloading login popup')

youtube_dl/extractor/videobam.py

@@ -1,81 +0,0 @@
-from __future__ import unicode_literals
-
-import re
-import json
-
-from .common import InfoExtractor
-from ..utils import int_or_none
-
-
-class VideoBamIE(InfoExtractor):
-    _VALID_URL = r'http://(?:www\.)?videobam\.com/(?:videos/download/)?(?P<id>[a-zA-Z]+)'
-
-    _TESTS = [
-        {
-            'url': 'http://videobam.com/OiJQM',
-            'md5': 'db471f27763a531f10416a0c58b5a1e0',
-            'info_dict': {
-                'id': 'OiJQM',
-                'ext': 'mp4',
-                'title': 'Is Alcohol Worse Than Ecstasy?',
-                'description': 'md5:d25b96151515c91debc42bfbb3eb2683',
-                'uploader': 'frihetsvinge',
-            },
-        },
-        {
-            'url': 'http://videobam.com/pqLvq',
-            'md5': 'd9a565b5379a99126ef94e1d7f9a383e',
-            'note': 'HD video',
-            'info_dict': {
-                'id': 'pqLvq',
-                'ext': 'mp4',
-                'title': '_',
-            }
-        },
-    ]
-
-    def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
-        video_id = mobj.group('id')
-
-        page = self._download_webpage('http://videobam.com/%s' % video_id, video_id, 'Downloading page')
-
-        formats = []
-
-        for preference, format_id in enumerate(['low', 'high']):
-            mobj = re.search(r"%s: '(?P<url>[^']+)'" % format_id, page)
-            if not mobj:
-                continue
-            formats.append({
-                'url': mobj.group('url'),
-                'ext': 'mp4',
-                'format_id': format_id,
-                'preference': preference,
-            })
-
-        if not formats:
-            player_config = json.loads(self._html_search_regex(r'var player_config = ({.+?});', page, 'player config'))
-            formats = [{
-                'url': item['url'],
-                'ext': 'mp4',
-            } for item in player_config['playlist'] if 'autoPlay' in item]
-
-        self._sort_formats(formats)
-
-        title = self._og_search_title(page, default='_', fatal=False)
-        description = self._og_search_description(page, default=None)
-        thumbnail = self._og_search_thumbnail(page)
-        uploader = self._html_search_regex(r'Upload by ([^<]+)</a>', page, 'uploader', fatal=False, default=None)
-        view_count = int_or_none(
-            self._html_search_regex(r'<strong>Views:</strong> (\d+) ', page, 'view count', fatal=False))
-
-        return {
-            'id': video_id,
-            'title': title,
-            'description': description,
-            'thumbnail': thumbnail,
-            'uploader': uploader,
-            'view_count': view_count,
-            'formats': formats,
-            'age_limit': 18,
-        }

youtube_dl/extractor/vidme.py

@@ -1,10 +1,12 @@
 from __future__ import unicode_literals
 
 from .common import InfoExtractor
+from ..compat import compat_HTTPError
 from ..utils import (
+    ExtractorError,
     int_or_none,
     float_or_none,
-    str_to_int,
+    parse_iso8601,
 )
 
 
@@ -12,18 +14,41 @@ class VidmeIE(InfoExtractor):
     _VALID_URL = r'https?://vid\.me/(?:e/)?(?P<id>[\da-zA-Z]+)'
     _TESTS = [{
         'url': 'https://vid.me/QNB',
-        'md5': 'f42d05e7149aeaec5c037b17e5d3dc82',
+        'md5': 'c62f1156138dc3323902188c5b5a8bd6',
         'info_dict': {
             'id': 'QNB',
             'ext': 'mp4',
             'title': 'Fishing for piranha - the easy way',
             'description': 'source: https://www.facebook.com/photo.php?v=312276045600871',
-            'duration': 119.92,
+            'thumbnail': 're:^https?://.*\.jpg',
             'timestamp': 1406313244,
             'upload_date': '20140725',
-            'thumbnail': 're:^https?://.*\.jpg',
+            'age_limit': 0,
+            'duration': 119.92,
             'view_count': int,
             'like_count': int,
+            'comment_count': int,
+        },
+    }, {
+        'url': 'https://vid.me/Gc6M',
+        'md5': 'f42d05e7149aeaec5c037b17e5d3dc82',
+        'info_dict': {
+            'id': 'Gc6M',
+            'ext': 'mp4',
+            'title': 'O Mere Dil ke chain - Arnav and Khushi VM',
+            'thumbnail': 're:^https?://.*\.jpg',
+            'timestamp': 1441211642,
+            'upload_date': '20150902',
+            'uploader': 'SunshineM',
+            'uploader_id': '3552827',
+            'age_limit': 0,
+            'duration': 223.72,
+            'view_count': int,
+            'like_count': int,
+            'comment_count': int,
+        },
+        'params': {
+            'skip_download': True,
         },
     }, {
         # tests uploader field
@@ -33,63 +58,94 @@ class VidmeIE(InfoExtractor):
             'ext': 'mp4',
             'title': 'The Carver',
             'description': 'md5:e9c24870018ae8113be936645b93ba3c',
-            'duration': 97.859999999999999,
+            'thumbnail': 're:^https?://.*\.jpg',
             'timestamp': 1433203629,
             'upload_date': '20150602',
             'uploader': 'Thomas',
-            'thumbnail': 're:^https?://.*\.jpg',
+            'uploader_id': '109747',
+            'age_limit': 0,
+            'duration': 97.859999999999999,
             'view_count': int,
             'like_count': int,
+            'comment_count': int,
         },
         'params': {
             'skip_download': True,
         },
     }, {
-        # From http://naked-yogi.tumblr.com/post/118312946248/naked-smoking-stretching
+        # nsfw test from http://naked-yogi.tumblr.com/post/118312946248/naked-smoking-stretching
         'url': 'https://vid.me/e/Wmur',
-        'only_matching': True,
+        'info_dict': {
+            'id': 'Wmur',
+            'ext': 'mp4',
+            'title': 'naked smoking & stretching',
+            'thumbnail': 're:^https?://.*\.jpg',
+            'timestamp': 1430931613,
+            'upload_date': '20150506',
+            'uploader': 'naked-yogi',
+            'uploader_id': '1638622',
+            'age_limit': 18,
+            'duration': 653.26999999999998,
+            'view_count': int,
+            'like_count': int,
+            'comment_count': int,
+        },
+        'params': {
+            'skip_download': True,
+        },
     }]
 
     def _real_extract(self, url):
-        url = url.replace('vid.me/e/', 'vid.me/')
         video_id = self._match_id(url)
-        webpage = self._download_webpage(url, video_id)
 
-        video_url = self._html_search_regex(
-            r'<source src="([^"]+)"', webpage, 'video URL')
+        try:
+            response = self._download_json(
+                'https://api.vid.me/videoByUrl/%s' % video_id, video_id)
+        except ExtractorError as e:
+            if isinstance(e.cause, compat_HTTPError) and e.cause.code == 400:
+                response = self._parse_json(e.cause.read(), video_id)
+            else:
+                raise
 
-        title = self._og_search_title(webpage)
-        description = self._og_search_description(webpage, default='')
-        thumbnail = self._og_search_thumbnail(webpage)
-        timestamp = int_or_none(self._og_search_property(
-            'updated_time', webpage, fatal=False))
-        width = int_or_none(self._og_search_property(
-            'video:width', webpage, fatal=False))
-        height = int_or_none(self._og_search_property(
-            'video:height', webpage, fatal=False))
-        duration = float_or_none(self._html_search_regex(
-            r'data-duration="([^"]+)"', webpage, 'duration', fatal=False))
-        view_count = str_to_int(self._html_search_regex(
-            r'<(?:li|span) class="video_views">\s*([\d,\.]+)\s*plays?',
-            webpage, 'view count', fatal=False))
-        like_count = str_to_int(self._html_search_regex(
-            r'class="score js-video-vote-score"[^>]+data-score="([\d,\.\s]+)">',
-            webpage, 'like count', fatal=False))
-        uploader = self._html_search_regex(
-            'class="video_author_username"[^>]*>([^<]+)',
-            webpage, 'uploader', default=None)
+        error = response.get('error')
+        if error:
+            raise ExtractorError(
+                '%s returned error: %s' % (self.IE_NAME, error), expected=True)
+
+        video = response['video']
+
+        formats = [{
+            'format_id': f.get('type'),
+            'url': f['uri'],
+            'width': int_or_none(f.get('width')),
+            'height': int_or_none(f.get('height')),
+        } for f in video.get('formats', []) if f.get('uri')]
+        self._sort_formats(formats)
+
+        title = video['title']
+        description = video.get('description')
+        thumbnail = video.get('thumbnail_url')
+        timestamp = parse_iso8601(video.get('date_created'), ' ')
+        uploader = video.get('user', {}).get('username')
+        uploader_id = video.get('user', {}).get('user_id')
+        age_limit = 18 if video.get('nsfw') is True else 0
+        duration = float_or_none(video.get('duration'))
+        view_count = int_or_none(video.get('view_count'))
+        like_count = int_or_none(video.get('likes_count'))
+        comment_count = int_or_none(video.get('comment_count'))
 
         return {
             'id': video_id,
-            'url': video_url,
             'title': title,
             'description': description,
             'thumbnail': thumbnail,
+            'uploader': uploader,
+            'uploader_id': uploader_id,
+            'age_limit': age_limit,
             'timestamp': timestamp,
-            'width': width,
-            'height': height,
             'duration': duration,
             'view_count': view_count,
             'like_count': like_count,
-            'uploader': uploader,
+            'comment_count': comment_count,
+            'formats': formats,
         }

youtube_dl/extractor/vier.py

@@ -2,6 +2,7 @@
 from __future__ import unicode_literals
 
 import re
+import itertools
 
 from .common import InfoExtractor
 
@@ -91,31 +92,27 @@ class VierVideosIE(InfoExtractor):
         mobj = re.match(self._VALID_URL, url)
         program = mobj.group('program')
 
-        webpage = self._download_webpage(url, program)
-
         page_id = mobj.group('page')
         if page_id:
             page_id = int(page_id)
             start_page = page_id
-            last_page = start_page + 1
             playlist_id = '%s-page%d' % (program, page_id)
         else:
             start_page = 0
-            last_page = int(self._search_regex(
-                r'videos\?page=(\d+)">laatste</a>',
-                webpage, 'last page', default=0)) + 1
             playlist_id = program
 
         entries = []
-        for current_page_id in range(start_page, last_page):
+        for current_page_id in itertools.count(start_page):
             current_page = self._download_webpage(
                 'http://www.vier.be/%s/videos?page=%d' % (program, current_page_id),
                 program,
-                'Downloading page %d' % (current_page_id + 1)) if current_page_id != page_id else webpage
+                'Downloading page %d' % (current_page_id + 1))
             page_entries = [
                 self.url_result('http://www.vier.be' + video_url, 'Vier')
                 for video_url in re.findall(
                     r'<h3><a href="(/[^/]+/videos/[^/]+(?:/\d+)?)">', current_page)]
             entries.extend(page_entries)
+            if page_id or '>Meer<' not in current_page:
+                break
 
         return self.playlist_result(entries, playlist_id)

youtube_dl/extractor/vlive.py

@@ -0,0 +1,86 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+import hmac
+from hashlib import sha1
+from base64 import b64encode
+from time import time
+
+from .common import InfoExtractor
+from ..utils import (
+    ExtractorError,
+    determine_ext
+)
+from ..compat import compat_urllib_parse
+
+
+class VLiveIE(InfoExtractor):
+    IE_NAME = 'vlive'
+    # www.vlive.tv/video/ links redirect to m.vlive.tv/video/ for mobile devices
+    _VALID_URL = r'https?://(?:(www|m)\.)?vlive\.tv/video/(?P<id>[0-9]+)'
+    _TEST = {
+        'url': 'http://m.vlive.tv/video/1326',
+        'md5': 'cc7314812855ce56de70a06a27314983',
+        'info_dict': {
+            'id': '1326',
+            'ext': 'mp4',
+            'title': '[V] Girl\'s Day\'s Broadcast',
+            'creator': 'Girl\'s Day',
+        },
+    }
+    _SECRET = 'rFkwZet6pqk1vQt6SxxUkAHX7YL3lmqzUMrU4IDusTo4jEBdtOhNfT4BYYAdArwH'
+
+    def _real_extract(self, url):
+        video_id = self._match_id(url)
+
+        webpage = self._download_webpage(
+            'http://m.vlive.tv/video/%s' % video_id,
+            video_id, note='Download video page')
+
+        title = self._og_search_title(webpage)
+        thumbnail = self._og_search_thumbnail(webpage)
+        creator = self._html_search_regex(
+            r'<span[^>]+class="name">([^<>]+)</span>', webpage, 'creator')
+
+        url = 'http://global.apis.naver.com/globalV/globalV/vod/%s/playinfo?' % video_id
+        msgpad = '%.0f' % (time() * 1000)
+        md = b64encode(
+            hmac.new(self._SECRET.encode('ascii'),
+                     (url[:255] + msgpad).encode('ascii'), sha1).digest()
+        )
+        url += '&' + compat_urllib_parse.urlencode({'msgpad': msgpad, 'md': md})
+        playinfo = self._download_json(url, video_id, 'Downloading video json')
+
+        if playinfo.get('message', '') != 'success':
+            raise ExtractorError(playinfo.get('message', 'JSON request unsuccessful'))
+
+        if not playinfo.get('result'):
+            raise ExtractorError('No videos found.')
+
+        formats = []
+        for vid in playinfo['result'].get('videos', {}).get('list', []):
+            formats.append({
+                'url': vid['source'],
+                'ext': 'mp4',
+                'abr': vid.get('bitrate', {}).get('audio'),
+                'vbr': vid.get('bitrate', {}).get('video'),
+                'format_id': vid['encodingOption']['name'],
+                'height': vid.get('height'),
+                'width': vid.get('width'),
+            })
+        self._sort_formats(formats)
+
+        subtitles = {}
+        for caption in playinfo['result'].get('captions', {}).get('list', []):
+            subtitles[caption['language']] = [
+                {'ext': determine_ext(caption['source'], default_ext='vtt'),
+                 'url': caption['source']}]
+
+        return {
+            'id': video_id,
+            'title': title,
+            'creator': creator,
+            'thumbnail': thumbnail,
+            'formats': formats,
+            'subtitles': subtitles,
+        }

youtube_dl/extractor/washingtonpost.py

@@ -19,25 +19,25 @@ class WashingtonPostIE(InfoExtractor):
             'title': 'Sinkhole of bureaucracy',
         },
         'playlist': [{
-            'md5': '79132cc09ec5309fa590ae46e4cc31bc',
+            'md5': 'b9be794ceb56c7267d410a13f99d801a',
             'info_dict': {
                 'id': 'fc433c38-b146-11e3-b8b3-44b1d1cd4c1f',
                 'ext': 'mp4',
                 'title': 'Breaking Points: The Paper Mine',
-                'duration': 1287,
+                'duration': 1290,
                 'description': 'Overly complicated paper pushing is nothing new to government bureaucracy. But the way federal retirement applications are filed may be the most outdated. David Fahrenthold explains.',
                 'uploader': 'The Washington Post',
                 'timestamp': 1395527908,
                 'upload_date': '20140322',
             },
         }, {
-            'md5': 'e1d5734c06865cc504ad99dc2de0d443',
+            'md5': '1fff6a689d8770966df78c8cb6c8c17c',
             'info_dict': {
                 'id': '41255e28-b14a-11e3-b8b3-44b1d1cd4c1f',
                 'ext': 'mp4',
                 'title': 'The town bureaucracy sustains',
                 'description': 'Underneath the friendly town of Boyers is a sea of government paperwork. In a disused limestone mine, hundreds of locals now track, file and process retirement applications for the federal government. We set out to find out what it\'s like to do paperwork 230 feet underground.',
-                'duration': 2217,
+                'duration': 2220,
                 'timestamp': 1395528005,
                 'upload_date': '20140322',
                 'uploader': 'The Washington Post',

youtube_dl/extractor/wimp.py

@@ -1,40 +1,33 @@
 from __future__ import unicode_literals
 
-import re
-
 from .common import InfoExtractor
 from .youtube import YoutubeIE
 
 
 class WimpIE(InfoExtractor):
-    _VALID_URL = r'http://(?:www\.)?wimp\.com/([^/]+)/'
+    _VALID_URL = r'http://(?:www\.)?wimp\.com/(?P<id>[^/]+)/'
     _TESTS = [{
         'url': 'http://www.wimp.com/maruexhausted/',
-        'md5': 'f1acced123ecb28d9bb79f2479f2b6a1',
+        'md5': 'ee21217ffd66d058e8b16be340b74883',
         'info_dict': {
             'id': 'maruexhausted',
-            'ext': 'flv',
+            'ext': 'mp4',
             'title': 'Maru is exhausted.',
             'description': 'md5:57e099e857c0a4ea312542b684a869b8',
         }
     }, {
-        # youtube video
         'url': 'http://www.wimp.com/clowncar/',
+        'md5': '4e2986c793694b55b37cf92521d12bb4',
         'info_dict': {
-            'id': 'cG4CEr2aiSg',
+            'id': 'clowncar',
             'ext': 'mp4',
-            'title': 'Basset hound clown car...incredible!',
-            'description': 'md5:8d228485e0719898c017203f900b3a35',
-            'uploader': 'Gretchen Hoey',
-            'uploader_id': 'gretchenandjeff1',
-            'upload_date': '20140303',
+            'title': 'It\'s like a clown car.',
+            'description': 'md5:0e56db1370a6e49c5c1d19124c0d2fb2',
         },
-        'add_ie': ['Youtube'],
     }]
 
     def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
-        video_id = mobj.group(1)
+        video_id = self._match_id(url)
         webpage = self._download_webpage(url, video_id)
         video_url = self._search_regex(
             [r"[\"']file[\"']\s*[:,]\s*[\"'](.+?)[\"']", r"videoId\s*:\s*[\"']([^\"']+)[\"']"],

youtube_dl/extractor/xuite.py

@@ -19,7 +19,7 @@ class XuiteIE(InfoExtractor):
     _TESTS = [{
         # Audio
         'url': 'http://vlog.xuite.net/play/RGkzc1ZULTM4NjA5MTQuZmx2',
-        'md5': '63a42c705772aa53fd4c1a0027f86adf',
+        'md5': 'e79284c87b371424885448d11f6398c8',
         'info_dict': {
             'id': '3860914',
             'ext': 'mp3',

youtube_dl/extractor/yahoo.py

@@ -101,7 +101,7 @@ class YahooIE(InfoExtractor):
             }
         }, {
             'url': 'http://news.yahoo.com/video/china-moses-crazy-blues-104538833.html',
-            'md5': '67010fdf3a08d290e060a4dd96baa07b',
+            'md5': '88e209b417f173d86186bef6e4d1f160',
             'info_dict': {
                 'id': 'f885cf7f-43d4-3450-9fac-46ac30ece521',
                 'ext': 'mp4',

youtube_dl/extractor/yandexmusic.py

@@ -1,18 +1,38 @@
-# coding=utf-8
+# coding: utf-8
 from __future__ import unicode_literals
 
 import re
 import hashlib
 
 from .common import InfoExtractor
-from ..compat import compat_str
+from ..compat import (
+    compat_str,
+    compat_urllib_parse,
+    compat_urllib_request,
+)
 from ..utils import (
     int_or_none,
     float_or_none,
 )
 
 
-class YandexMusicBaseIE(InfoExtractor):
+class YandexMusicTrackIE(InfoExtractor):
+    IE_NAME = 'yandexmusic:track'
+    IE_DESC = 'Яндекс.Музыка - Трек'
+    _VALID_URL = r'https?://music\.yandex\.(?:ru|kz|ua|by)/album/(?P<album_id>\d+)/track/(?P<id>\d+)'
+
+    _TEST = {
+        'url': 'http://music.yandex.ru/album/540508/track/4878838',
+        'md5': 'f496818aa2f60b6c0062980d2e00dc20',
+        'info_dict': {
+            'id': '4878838',
+            'ext': 'mp3',
+            'title': 'Carlo Ambrosio - Gypsy Eyes 1',
+            'filesize': 4628061,
+            'duration': 193.04,
+        }
+    }
+
     def _get_track_url(self, storage_dir, track_id):
         data = self._download_json(
             'http://music.yandex.ru/api/v1.5/handlers/api-jsonp.jsx?action=getTrackSrc&p=download-info/%s'
@@ -35,24 +55,6 @@ class YandexMusicBaseIE(InfoExtractor):
             'duration': float_or_none(track.get('durationMs'), 1000),
         }
 
-
-class YandexMusicTrackIE(YandexMusicBaseIE):
-    IE_NAME = 'yandexmusic:track'
-    IE_DESC = 'Яндекс.Музыка - Трек'
-    _VALID_URL = r'https?://music\.yandex\.(?:ru|kz|ua|by)/album/(?P<album_id>\d+)/track/(?P<id>\d+)'
-
-    _TEST = {
-        'url': 'http://music.yandex.ru/album/540508/track/4878838',
-        'md5': 'f496818aa2f60b6c0062980d2e00dc20',
-        'info_dict': {
-            'id': '4878838',
-            'ext': 'mp3',
-            'title': 'Carlo Ambrosio - Gypsy Eyes 1',
-            'filesize': 4628061,
-            'duration': 193.04,
-        }
-    }
-
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
         album_id, track_id = mobj.group('album_id'), mobj.group('id')
@@ -64,7 +66,15 @@ class YandexMusicTrackIE(YandexMusicBaseIE):
         return self._get_track_info(track)
 
 
-class YandexMusicAlbumIE(YandexMusicBaseIE):
+class YandexMusicPlaylistBaseIE(InfoExtractor):
+    def _build_playlist(self, tracks):
+        return [
+            self.url_result(
+                'http://music.yandex.ru/album/%s/track/%s' % (track['albums'][0]['id'], track['id']))
+            for track in tracks if track.get('albums') and isinstance(track.get('albums'), list)]
+
+
+class YandexMusicAlbumIE(YandexMusicPlaylistBaseIE):
     IE_NAME = 'yandexmusic:album'
     IE_DESC = 'Яндекс.Музыка - Альбом'
     _VALID_URL = r'https?://music\.yandex\.(?:ru|kz|ua|by)/album/(?P<id>\d+)/?(\?|$)'
@@ -85,7 +95,7 @@ class YandexMusicAlbumIE(YandexMusicBaseIE):
             'http://music.yandex.ru/handlers/album.jsx?album=%s' % album_id,
             album_id, 'Downloading album JSON')
 
-        entries = [self._get_track_info(track) for track in album['volumes'][0]]
+        entries = self._build_playlist(album['volumes'][0])
 
         title = '%s - %s' % (album['artists'][0]['name'], album['title'])
         year = album.get('year')
@@ -95,12 +105,12 @@ class YandexMusicAlbumIE(YandexMusicBaseIE):
         return self.playlist_result(entries, compat_str(album['id']), title)
 
 
-class YandexMusicPlaylistIE(YandexMusicBaseIE):
+class YandexMusicPlaylistIE(YandexMusicPlaylistBaseIE):
     IE_NAME = 'yandexmusic:playlist'
     IE_DESC = 'Яндекс.Музыка - Плейлист'
     _VALID_URL = r'https?://music\.yandex\.(?:ru|kz|ua|by)/users/[^/]+/playlists/(?P<id>\d+)'
 
-    _TEST = {
+    _TESTS = [{
         'url': 'http://music.yandex.ru/users/music.partners/playlists/1245',
         'info_dict': {
             'id': '1245',
@@ -108,20 +118,54 @@ class YandexMusicPlaylistIE(YandexMusicBaseIE):
             'description': 'md5:3b9f27b0efbe53f2ee1e844d07155cc9',
         },
         'playlist_count': 6,
-    }
+    }, {
+        # playlist exceeding the limit of 150 tracks shipped with webpage (see
+        # https://github.com/rg3/youtube-dl/issues/6666)
+        'url': 'https://music.yandex.ru/users/ya.playlist/playlists/1036',
+        'info_dict': {
+            'id': '1036',
+            'title': 'Музыка 90-х',
+        },
+        'playlist_count': 310,
+    }]
 
     def _real_extract(self, url):
         playlist_id = self._match_id(url)
 
         webpage = self._download_webpage(url, playlist_id)
 
-        playlist = self._parse_json(
+        mu = self._parse_json(
             self._search_regex(
                 r'var\s+Mu\s*=\s*({.+?});\s*</script>', webpage, 'player'),
-            playlist_id)['pageData']['playlist']
+            playlist_id)
 
-        entries = [self._get_track_info(track) for track in playlist['tracks']]
+        playlist = mu['pageData']['playlist']
+        tracks, track_ids = playlist['tracks'], playlist['trackIds']
+
+        # tracks dictionary shipped with webpage is limited to 150 tracks,
+        # missing tracks should be retrieved manually.
+        if len(tracks) < len(track_ids):
+            present_track_ids = set([compat_str(track['id']) for track in tracks if track.get('id')])
+            missing_track_ids = set(map(compat_str, track_ids)) - set(present_track_ids)
+            request = compat_urllib_request.Request(
+                'https://music.yandex.ru/handlers/track-entries.jsx',
+                compat_urllib_parse.urlencode({
+                    'entries': ','.join(missing_track_ids),
+                    'lang': mu.get('settings', {}).get('lang', 'en'),
+                    'external-domain': 'music.yandex.ru',
+                    'overembed': 'false',
+                    'sign': mu.get('authData', {}).get('user', {}).get('sign'),
+                    'strict': 'true',
+                }).encode('utf-8'))
+            request.add_header('Referer', url)
+            request.add_header('X-Requested-With', 'XMLHttpRequest')
+
+            missing_tracks = self._download_json(
+                request, playlist_id, 'Downloading missing tracks JSON', fatal=False)
+            if missing_tracks:
+                tracks.extend(missing_tracks)
 
         return self.playlist_result(
-            entries, compat_str(playlist_id),
+            self._build_playlist(tracks),
+            compat_str(playlist_id),
             playlist['title'], playlist.get('description'))

youtube_dl/extractor/youku.py

@@ -49,6 +49,17 @@ class YoukuIE(InfoExtractor):
         },
         'playlist_count': 13,
         'skip': 'Available in China only',
+    }, {
+        'url': 'http://v.youku.com/v_show/id_XNjA1NzA2Njgw.html',
+        'note': 'Video protected with password',
+        'info_dict': {
+            'id': 'XNjA1NzA2Njgw',
+            'title': '邢義田复旦讲座之想象中的胡人—从“左衽孔子”说起',
+        },
+        'playlist_count': 19,
+        'params': {
+            'videopassword': '100600',
+        },
     }]
 
     def construct_video_urls(self, data1, data2):
@@ -185,9 +196,15 @@ class YoukuIE(InfoExtractor):
             raw_data = self._download_json(req, video_id, note=note)
             return raw_data['data'][0]
 
+        video_password = self._downloader.params.get('videopassword', None)
+
         # request basic data
+        basic_data_url = 'http://v.youku.com/player/getPlayList/VideoIDS/%s' % video_id
+        if video_password:
+            basic_data_url += '?password=%s' % video_password
+
         data1 = retrieve_data(
-            'http://v.youku.com/player/getPlayList/VideoIDS/%s' % video_id,
+            basic_data_url,
             'Downloading JSON metadata 1')
         data2 = retrieve_data(
             'http://v.youku.com/player/getPlayList/VideoIDS/%s/Pf/4/ctype/12/ev/1' % video_id,

youtube_dl/extractor/youtube.py

@@ -26,6 +26,7 @@ from ..compat import (
 )
 from ..utils import (
     clean_html,
+    encode_dict,
     ExtractorError,
     float_or_none,
     get_element_by_attribute,
@@ -111,10 +112,7 @@ class YoutubeBaseInfoExtractor(InfoExtractor):
             'hl': 'en_US',
         }
 
-        # Convert to UTF-8 *before* urlencode because Python 2.x's urlencode
-        # chokes on unicode
-        login_form = dict((k.encode('utf-8'), v.encode('utf-8')) for k, v in login_form_strs.items())
-        login_data = compat_urllib_parse.urlencode(login_form).encode('ascii')
+        login_data = compat_urllib_parse.urlencode(encode_dict(login_form_strs)).encode('ascii')
 
         req = compat_urllib_request.Request(self._LOGIN_URL, login_data)
         login_results = self._download_webpage(
@@ -147,8 +145,7 @@ class YoutubeBaseInfoExtractor(InfoExtractor):
                 'TrustDevice': 'on',
             })
 
-            tfa_form = dict((k.encode('utf-8'), v.encode('utf-8')) for k, v in tfa_form_strs.items())
-            tfa_data = compat_urllib_parse.urlencode(tfa_form).encode('ascii')
+            tfa_data = compat_urllib_parse.urlencode(encode_dict(tfa_form_strs)).encode('ascii')
 
             tfa_req = compat_urllib_request.Request(self._TWOFACTOR_URL, tfa_data)
             tfa_results = self._download_webpage(
@@ -660,7 +657,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
 
     def _extract_signature_function(self, video_id, player_url, example_sig):
         id_m = re.match(
-            r'.*?-(?P<id>[a-zA-Z0-9_-]+)(?:/watch_as3|/html5player)?\.(?P<ext>[a-z]+)$',
+            r'.*?-(?P<id>[a-zA-Z0-9_-]+)(?:/watch_as3|/html5player(?:-new)?)?\.(?P<ext>[a-z]+)$',
             player_url)
         if not id_m:
             raise ExtractorError('Cannot identify player %r' % player_url)
@@ -1243,7 +1240,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
             encoded_url_map = video_info.get('url_encoded_fmt_stream_map', [''])[0] + ',' + video_info.get('adaptive_fmts', [''])[0]
             if 'rtmpe%3Dyes' in encoded_url_map:
                 raise ExtractorError('rtmpe downloads are not supported, see https://github.com/rg3/youtube-dl/issues/343 for more information.', expected=True)
-            url_map = {}
+            formats = []
             for url_data_str in encoded_url_map.split(','):
                 url_data = compat_parse_qs(url_data_str)
                 if 'itag' not in url_data or 'url' not in url_data:
@@ -1289,7 +1286,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
                                 player_desc = 'flash player %s' % player_version
                             else:
                                 player_version = self._search_regex(
-                                    r'html5player-([^/]+?)(?:/html5player)?\.js',
+                                    r'html5player-([^/]+?)(?:/html5player(?:-new)?)?\.js',
                                     player_url,
                                     'html5 player', fatal=False)
                                 player_desc = 'html5 player %s' % player_version
@@ -1303,8 +1300,50 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
                     url += '&signature=' + signature
                 if 'ratebypass' not in url:
                     url += '&ratebypass=yes'
-                url_map[format_id] = url
-            formats = _map_to_format_list(url_map)
+
+                # Some itags are not included in DASH manifest thus corresponding formats will
+                # lack metadata (see https://github.com/rg3/youtube-dl/pull/5993).
+                # Trying to extract metadata from url_encoded_fmt_stream_map entry.
+                mobj = re.search(r'^(?P<width>\d+)[xX](?P<height>\d+)$', url_data.get('size', [''])[0])
+                width, height = (int(mobj.group('width')), int(mobj.group('height'))) if mobj else (None, None)
+                dct = {
+                    'format_id': format_id,
+                    'url': url,
+                    'player_url': player_url,
+                    'filesize': int_or_none(url_data.get('clen', [None])[0]),
+                    'tbr': float_or_none(url_data.get('bitrate', [None])[0], 1000),
+                    'width': width,
+                    'height': height,
+                    'fps': int_or_none(url_data.get('fps', [None])[0]),
+                    'format_note': url_data.get('quality_label', [None])[0] or url_data.get('quality', [None])[0],
+                }
+                type_ = url_data.get('type', [None])[0]
+                if type_:
+                    type_split = type_.split(';')
+                    kind_ext = type_split[0].split('/')
+                    if len(kind_ext) == 2:
+                        kind, ext = kind_ext
+                        dct['ext'] = ext
+                        if kind in ('audio', 'video'):
+                            codecs = None
+                            for mobj in re.finditer(
+                                    r'(?P<key>[a-zA-Z_-]+)=(?P<quote>["\']?)(?P<val>.+?)(?P=quote)(?:;|$)', type_):
+                                if mobj.group('key') == 'codecs':
+                                    codecs = mobj.group('val')
+                                    break
+                            if codecs:
+                                codecs = codecs.split(',')
+                                if len(codecs) == 2:
+                                    acodec, vcodec = codecs[0], codecs[1]
+                                else:
+                                    acodec, vcodec = (codecs[0], 'none') if kind == 'audio' else ('none', codecs[0])
+                                dct.update({
+                                    'acodec': acodec,
+                                    'vcodec': vcodec,
+                                })
+                if format_id in self._formats:
+                    dct.update(self._formats[format_id])
+                formats.append(dct)
         elif video_info.get('hlsvp'):
             manifest_url = video_info['hlsvp'][0]
             url_map = self._extract_from_m3u8(manifest_url, video_id)
@@ -1796,8 +1835,8 @@ class YoutubeShowIE(InfoExtractor):
     _VALID_URL = r'https?://www\.youtube\.com/show/(?P<id>[^?#]*)'
     IE_NAME = 'youtube:show'
     _TESTS = [{
-        'url': 'http://www.youtube.com/show/airdisasters',
-        'playlist_mincount': 3,
+        'url': 'https://www.youtube.com/show/airdisasters',
+        'playlist_mincount': 5,
         'info_dict': {
             'id': 'airdisasters',
             'title': 'Air Disasters',
@@ -1808,7 +1847,7 @@ class YoutubeShowIE(InfoExtractor):
         mobj = re.match(self._VALID_URL, url)
         playlist_id = mobj.group('id')
         webpage = self._download_webpage(
-            url, playlist_id, 'Downloading show webpage')
+            'https://www.youtube.com/show/%s/playlists' % playlist_id, playlist_id, 'Downloading show webpage')
         # There's one playlist for each season of the show
         m_seasons = list(re.finditer(r'href="(/playlist\?list=.*?)"', webpage))
         self.to_screen('%s: Found %s seasons' % (playlist_id, len(m_seasons)))

youtube_dl/options.py

@@ -2,7 +2,6 @@ from __future__ import unicode_literals
 
 import os.path
 import optparse
-import shlex
 import sys
 
 from .downloader.external import list_external_downloaders
@@ -11,6 +10,7 @@ from .compat import (
     compat_get_terminal_size,
     compat_getenv,
     compat_kwargs,
+    compat_shlex_split,
 )
 from .utils import (
     preferredencoding,
@@ -28,7 +28,7 @@ def parseOpts(overrideArguments=None):
         try:
             res = []
             for l in optionf:
-                res += shlex.split(l, comments=True)
+                res += compat_shlex_split(l, comments=True)
         finally:
             optionf.close()
         return res
@@ -320,7 +320,7 @@ def parseOpts(overrideArguments=None):
     authentication.add_option(
         '--video-password',
         dest='videopassword', metavar='PASSWORD',
-        help='Video password (vimeo, smotri)')
+        help='Video password (vimeo, smotri, youku)')
 
     video_format = optparse.OptionGroup(parser, 'Video Format Options')
     video_format.add_option(

youtube_dl/postprocessor/common.py

@@ -4,6 +4,7 @@ import os
 
 from ..utils import (
     PostProcessingError,
+    cli_configuration_args,
     encodeFilename,
 )
 
@@ -61,11 +62,7 @@ class PostProcessor(object):
             self._downloader.report_warning(errnote)
 
     def _configuration_args(self, default=[]):
-        pp_args = self._downloader.params.get('postprocessor_args')
-        if pp_args is None:
-            return default
-        assert isinstance(pp_args, list)
-        return pp_args
+        return cli_configuration_args(self._downloader.params, 'postprocessor_args', default)
 
 
 class AudioConversionError(PostProcessingError):

youtube_dl/utils.py

@@ -141,7 +141,7 @@ def write_json_file(obj, fn):
 if sys.version_info >= (2, 7):
     def find_xpath_attr(node, xpath, key, val=None):
         """ Find the xpath xpath[@key=val] """
-        assert re.match(r'^[a-zA-Z-]+$', key)
+        assert re.match(r'^[a-zA-Z_-]+$', key)
         if val:
             assert re.match(r'^[a-zA-Z0-9@\s:._-]*$', val)
         expr = xpath + ('[@%s]' % key if val is None else "[@%s='%s']" % (key, val))
@@ -176,12 +176,12 @@ def xpath_with_ns(path, ns_map):
     return '/'.join(replaced)
 
 
-def xpath_text(node, xpath, name=None, fatal=False, default=NO_DEFAULT):
+def xpath_element(node, xpath, name=None, fatal=False, default=NO_DEFAULT):
     if sys.version_info < (2, 7):  # Crazy 2.6
         xpath = xpath.encode('ascii')
 
     n = node.find(xpath)
-    if n is None or n.text is None:
+    if n is None:
         if default is not NO_DEFAULT:
             return default
         elif fatal:
@@ -189,9 +189,37 @@ def xpath_text(node, xpath, name=None, fatal=False, default=NO_DEFAULT):
             raise ExtractorError('Could not find XML element %s' % name)
         else:
             return None
+    return n
+
+
+def xpath_text(node, xpath, name=None, fatal=False, default=NO_DEFAULT):
+    n = xpath_element(node, xpath, name, fatal=fatal, default=default)
+    if n is None or n == default:
+        return n
+    if n.text is None:
+        if default is not NO_DEFAULT:
+            return default
+        elif fatal:
+            name = xpath if name is None else name
+            raise ExtractorError('Could not find XML element\'s text %s' % name)
+        else:
+            return None
     return n.text
 
 
+def xpath_attr(node, xpath, key, name=None, fatal=False, default=NO_DEFAULT):
+    n = find_xpath_attr(node, xpath, key)
+    if n is None:
+        if default is not NO_DEFAULT:
+            return default
+        elif fatal:
+            name = '%s[@%s]' % (xpath, key) if name is None else name
+            raise ExtractorError('Could not find XML attribute %s' % name)
+        else:
+            return None
+    return n.attrib[key]
+
+
 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)
@@ -587,6 +615,11 @@ class ContentTooShortError(Exception):
 
 
 def _create_http_connection(ydl_handler, http_class, is_https, *args, **kwargs):
+    # Working around python 2 bug (see http://bugs.python.org/issue17849) by limiting
+    # expected HTTP responses to meet HTTP/1.0 or later (see also
+    # https://github.com/rg3/youtube-dl/issues/6727)
+    if sys.version_info < (3, 0):
+        kwargs['strict'] = True
     hc = http_class(*args, **kwargs)
     source_address = ydl_handler._params.get('source_address')
     if source_address is not None:
@@ -715,7 +748,8 @@ class YoutubeDLHandler(compat_urllib_request.HTTPHandler):
             gz = io.BytesIO(self.deflate(resp.read()))
             resp = self.addinfourl_wrapper(gz, old_resp.headers, old_resp.url, old_resp.code)
             resp.msg = old_resp.msg
-        # Percent-encode redirect URL of Location HTTP header to satisfy RFC 3986
+        # Percent-encode redirect URL of Location HTTP header to satisfy RFC 3986 (see
+        # https://github.com/rg3/youtube-dl/issues/6457).
         if 300 <= resp.code < 400:
             location = resp.headers.get('Location')
             if location:
@@ -749,6 +783,30 @@ class YoutubeDLHTTPSHandler(compat_urllib_request.HTTPSHandler):
             req, **kwargs)
 
 
+class YoutubeDLCookieProcessor(compat_urllib_request.HTTPCookieProcessor):
+    def __init__(self, cookiejar=None):
+        compat_urllib_request.HTTPCookieProcessor.__init__(self, cookiejar)
+
+    def http_response(self, request, response):
+        # Python 2 will choke on next HTTP request in row if there are non-ASCII
+        # characters in Set-Cookie HTTP header of last response (see
+        # https://github.com/rg3/youtube-dl/issues/6769).
+        # In order to at least prevent crashing we will percent encode Set-Cookie
+        # header before HTTPCookieProcessor starts processing it.
+        # if sys.version_info < (3, 0) and response.headers:
+        #     for set_cookie_header in ('Set-Cookie', 'Set-Cookie2'):
+        #         set_cookie = response.headers.get(set_cookie_header)
+        #         if set_cookie:
+        #             set_cookie_escaped = compat_urllib_parse.quote(set_cookie, b"%/;:@&=+$,!~*'()?#[] ")
+        #             if set_cookie != set_cookie_escaped:
+        #                 del response.headers[set_cookie_header]
+        #                 response.headers[set_cookie_header] = set_cookie_escaped
+        return compat_urllib_request.HTTPCookieProcessor.http_response(self, request, response)
+
+    https_request = compat_urllib_request.HTTPCookieProcessor.http_request
+    https_response = http_response
+
+
 def parse_iso8601(date_str, delimiter='T', timezone=None):
     """ Return a UNIX timestamp from the given date """
 
@@ -1578,6 +1636,10 @@ def urlencode_postdata(*args, **kargs):
     return compat_urllib_parse.urlencode(*args, **kargs).encode('ascii')
 
 
+def encode_dict(d, encoding='utf-8'):
+    return dict((k.encode(encoding), v.encode(encoding)) for k, v in d.items())
+
+
 try:
     etree_iter = xml.etree.ElementTree.Element.iter
 except AttributeError:  # Python <=2.6
@@ -1918,6 +1980,32 @@ def dfxp2srt(dfxp_data):
     return ''.join(out)
 
 
+def cli_option(params, command_option, param):
+    param = params.get(param)
+    return [command_option, param] if param is not None else []
+
+
+def cli_bool_option(params, command_option, param, true_value='true', false_value='false', separator=None):
+    param = params.get(param)
+    assert isinstance(param, bool)
+    if separator:
+        return [command_option + separator + (true_value if param else false_value)]
+    return [command_option, true_value if param else false_value]
+
+
+def cli_valueless_option(params, command_option, param, expected_value=True):
+    param = params.get(param)
+    return [command_option] if param == expected_value else []
+
+
+def cli_configuration_args(params, param, default=[]):
+    ex_args = params.get(param)
+    if ex_args is None:
+        return default
+    assert isinstance(ex_args, list)
+    return ex_args
+
+
 class ISO639Utils(object):
     # See http://www.loc.gov/standards/iso639-2/ISO-639-2_utf-8.txt
     _lang_map = {

youtube_dl/version.py

@@ -1,3 +1,3 @@
 from __future__ import unicode_literals
 
-__version__ = '2015.08.16.1'
+__version__ = '2015.09.09'