[channel9] Add support for rss links (Closes #9673)

Added extractor of loc.gov, which closes #3188. I am not an experienced programmer, so I am sure I did a bunch of mistakes, but the extractor works (for me at least).

[LibraryOfCongress] don't use video_id for _search_regex()

[LibraryOfCongress] Improvements

.github/ISSUE_TEMPLATE.md

@@ -6,8 +6,8 @@
 
 ---
 
-### Make sure you are using the *latest* version: run `youtube-dl --version` and ensure your version is *2016.05.21.2*. If it's not read [this FAQ entry](https://github.com/rg3/youtube-dl/blob/master/README.md#how-do-i-update-youtube-dl) and update. Issues with outdated version will be rejected.
-- [ ] I've **verified** and **I assure** that I'm running youtube-dl **2016.05.21.2**
+### Make sure you are using the *latest* version: run `youtube-dl --version` and ensure your version is *2016.06.03*. If it's not read [this FAQ entry](https://github.com/rg3/youtube-dl/blob/master/README.md#how-do-i-update-youtube-dl) and update. Issues with outdated version will be rejected.
+- [ ] I've **verified** and **I assure** that I'm running youtube-dl **2016.06.03**
 
 ### Before submitting an *issue* make sure you have:
 - [ ] At least skimmed through [README](https://github.com/rg3/youtube-dl/blob/master/README.md) and **most notably** [FAQ](https://github.com/rg3/youtube-dl#faq) and [BUGS](https://github.com/rg3/youtube-dl#bugs) sections
@@ -35,7 +35,7 @@ $ youtube-dl -v <your command line>
 [debug] User config: []
 [debug] Command-line args: [u'-v', u'http://www.youtube.com/watch?v=BaW_jenozKcj']
 [debug] Encodings: locale cp1251, fs mbcs, out cp866, pref cp1251
-[debug] youtube-dl version 2016.05.21.2
+[debug] youtube-dl version 2016.06.03
 [debug] Python version 2.7.11 - Windows-2003Server-5.2.3790-SP2
 [debug] exe versions: ffmpeg N-75573-g1d0487f, ffprobe N-75573-g1d0487f, rtmpdump 2.4
 [debug] Proxy map: {}

.gitignore

@@ -28,12 +28,16 @@ updates_key.pem
 *.mp4
 *.m4a
 *.m4v
+*.mp3
 *.part
 *.swp
 test/testdata
 test/local_parameters.json
 .tox
 youtube-dl.zsh
+
+# IntelliJ related files
 .idea
-.idea/*
+*.iml
+
 tmp/

.travis.yml

@@ -14,7 +14,6 @@ script: nosetests test --verbose
 notifications:
   email:
     - filippo.valsorda@gmail.com
-    - phihag@phihag.de
     - yasoob.khld@gmail.com
 #  irc:
 #    channels:

AUTHORS

@@ -172,3 +172,4 @@ blahgeek
 Kevin Deldycke
 inondle
 Tomáš Čech
+Déstin Reed

Makefile

@@ -1,7 +1,7 @@
 all: youtube-dl README.md CONTRIBUTING.md README.txt youtube-dl.1 youtube-dl.bash-completion youtube-dl.zsh youtube-dl.fish supportedsites
 
 clean:
-	rm -rf youtube-dl.1.temp.md youtube-dl.1 youtube-dl.bash-completion README.txt MANIFEST build/ dist/ .coverage cover/ youtube-dl.tar.gz youtube-dl.zsh youtube-dl.fish youtube_dl/extractor/lazy_extractors.py *.dump *.part *.info.json *.mp4 *.flv *.mp3 *.avi *.mkv *.webm *.jpg *.png CONTRIBUTING.md.tmp ISSUE_TEMPLATE.md.tmp youtube-dl youtube-dl.exe
+	rm -rf youtube-dl.1.temp.md youtube-dl.1 youtube-dl.bash-completion README.txt MANIFEST build/ dist/ .coverage cover/ youtube-dl.tar.gz youtube-dl.zsh youtube-dl.fish youtube_dl/extractor/lazy_extractors.py *.dump *.part *.info.json *.mp4 *.m4a *.flv *.mp3 *.avi *.mkv *.webm *.jpg *.png CONTRIBUTING.md.tmp ISSUE_TEMPLATE.md.tmp youtube-dl youtube-dl.exe
 	find . -name "*.pyc" -delete
 	find . -name "*.class" -delete
 
@@ -69,7 +69,7 @@ README.txt: README.md
 	pandoc -f markdown -t plain README.md -o README.txt
 
 youtube-dl.1: README.md
-	$(PYTHON) devscripts/prepare_manpage.py >youtube-dl.1.temp.md
+	$(PYTHON) devscripts/prepare_manpage.py youtube-dl.1.temp.md
 	pandoc -s -f markdown -t man youtube-dl.1.temp.md -o youtube-dl.1
 	rm -f youtube-dl.1.temp.md
 

README.md

@@ -25,7 +25,7 @@ If you do not have curl, you can alternatively use a recent wget:
     sudo wget https://yt-dl.org/downloads/latest/youtube-dl -O /usr/local/bin/youtube-dl
     sudo chmod a+rx /usr/local/bin/youtube-dl
 
-Windows users can [download a .exe file](https://yt-dl.org/latest/youtube-dl.exe) and place it in their home directory or any other location on their [PATH](http://en.wikipedia.org/wiki/PATH_%28variable%29).
+Windows users can [download an .exe file](https://yt-dl.org/latest/youtube-dl.exe) and place it in any location on their [PATH](http://en.wikipedia.org/wiki/PATH_%28variable%29) except for `%SYSTEMROOT%\System32` (e.g. **do not** put in `C:\Windows\System32`).
 
 OS X users can install **youtube-dl** with [Homebrew](http://brew.sh/).
 
@@ -73,8 +73,8 @@ which means you can modify it, redistribute it or use it however you like.
                                      repairs broken URLs, but emits an error if
                                      this is not possible instead of searching.
     --ignore-config                  Do not read configuration files. When given
-                                     in the global configuration file /etc
-                                     /youtube-dl.conf: Do not read the user
+                                     in the global configuration file
+                                     /etc/youtube-dl.conf: Do not read the user
                                      configuration in ~/.config/youtube-
                                      dl/config (%APPDATA%/youtube-dl/config.txt
                                      on Windows)
@@ -162,7 +162,7 @@ which means you can modify it, redistribute it or use it however you like.
                                      (experimental)
 
 ## Download Options:
-    -r, --rate-limit LIMIT           Maximum download rate in bytes per second
+    -r, --limit-rate RATE            Maximum download rate in bytes per second
                                      (e.g. 50K or 4.2M)
     -R, --retries RETRIES            Number of retries (default is 10), or
                                      "infinite".
@@ -256,11 +256,12 @@ which means you can modify it, redistribute it or use it however you like.
                                      jar in
     --cache-dir DIR                  Location in the filesystem where youtube-dl
                                      can store some downloaded information
-                                     permanently. By default $XDG_CACHE_HOME
-                                     /youtube-dl or ~/.cache/youtube-dl . At the
-                                     moment, only YouTube player files (for
-                                     videos with obfuscated signatures) are
-                                     cached, but that may change.
+                                     permanently. By default
+                                     $XDG_CACHE_HOME/youtube-dl or
+                                     ~/.cache/youtube-dl . At the moment, only
+                                     YouTube player files (for videos with
+                                     obfuscated signatures) are cached, but that
+                                     may change.
     --no-cache-dir                   Disable filesystem caching
     --rm-cache-dir                   Delete all filesystem cache files
 
@@ -433,7 +434,7 @@ You can use `--ignore-config` if you want to disable the configuration file for
 
 ### Authentication with `.netrc` file
 
-You may also want to configure automatic credentials storage for extractors that support authentication (by providing login and password with `--username` and `--password`) in order not to pass credentials as command line arguments on every youtube-dl execution and prevent tracking plain text passwords in the shell command history. You can achieve this using a [`.netrc` file](http://stackoverflow.com/tags/.netrc/info) on per extractor basis. For that you will need to create a`.netrc` file in your `$HOME` and restrict permissions to read/write by you only:
+You may also want to configure automatic credentials storage for extractors that support authentication (by providing login and password with `--username` and `--password`) in order not to pass credentials as command line arguments on every youtube-dl execution and prevent tracking plain text passwords in the shell command history. You can achieve this using a [`.netrc` file](http://stackoverflow.com/tags/.netrc/info) on per extractor basis. For that you will need to create a `.netrc` file in your `$HOME` and restrict permissions to read/write by you only:
 ```
 touch $HOME/.netrc
 chmod a-rwx,u+rw $HOME/.netrc
@@ -693,6 +694,10 @@ hash -r
 
 Again, from then on you'll be able to update with `sudo youtube-dl -U`.
 
+### youtube-dl is extremely slow to start on Windows
+
+Add a file exclusion for `youtube-dl.exe` in Windows Defender settings.
+
 ### I'm getting an error `Unable to extract OpenGraph title` on YouTube playlists
 
 YouTube changed their playlist format in March 2014 and later on, so you'll need at least youtube-dl 2014.07.25 to download all YouTube videos.
@@ -780,9 +785,9 @@ means you're using an outdated version of Python. Please update to Python 2.6 or
 
 Since June 2012 ([#342](https://github.com/rg3/youtube-dl/issues/342)) youtube-dl is packed as an executable zipfile, simply unzip it (might need renaming to `youtube-dl.zip` first on some systems) or clone the git repository, as laid out above. If you modify the code, you can run it by executing the `__main__.py` file. To recompile the executable, run `make youtube-dl`.
 
-### The exe throws a *Runtime error from Visual C++*
+### The exe throws an error due to missing `MSVCR100.dll`
 
-To run the exe you need to install first the [Microsoft Visual C++ 2008 Redistributable Package](http://www.microsoft.com/en-us/download/details.aspx?id=29).
+To run the exe you need to install first the [Microsoft Visual C++ 2010 Redistributable Package (x86)](https://www.microsoft.com/en-US/download/details.aspx?id=5555).
 
 ### On Windows, how should I set up ffmpeg and youtube-dl? Where should I put the exe files?
 

devscripts/buildserver.py

@@ -1,17 +1,42 @@
 #!/usr/bin/python3
 
-from http.server import HTTPServer, BaseHTTPRequestHandler
-from socketserver import ThreadingMixIn
 import argparse
 import ctypes
 import functools
+import shutil
+import subprocess
 import sys
+import tempfile
 import threading
 import traceback
 import os.path
 
+sys.path.insert(0, os.path.dirname(os.path.dirname((os.path.abspath(__file__)))))
+from youtube_dl.compat import (
+    compat_http_server,
+    compat_str,
+    compat_urlparse,
+)
 
-class BuildHTTPServer(ThreadingMixIn, HTTPServer):
+# These are not used outside of buildserver.py thus not in compat.py
+
+try:
+    import winreg as compat_winreg
+except ImportError:  # Python 2
+    import _winreg as compat_winreg
+
+try:
+    import socketserver as compat_socketserver
+except ImportError:  # Python 2
+    import SocketServer as compat_socketserver
+
+try:
+    compat_input = raw_input
+except NameError:  # Python 3
+    compat_input = input
+
+
+class BuildHTTPServer(compat_socketserver.ThreadingMixIn, compat_http_server.HTTPServer):
     allow_reuse_address = True
 
 
@@ -191,7 +216,7 @@ def main(args=None):
                         action='store_const', dest='action', const='service',
                         help='Run as a Windows service')
     parser.add_argument('-b', '--bind', metavar='<host:port>',
-                        action='store', default='localhost:8142',
+                        action='store', default='0.0.0.0:8142',
                         help='Bind to host:port (default %default)')
     options = parser.parse_args(args=args)
 
@@ -216,7 +241,7 @@ def main(args=None):
     srv = BuildHTTPServer((host, port), BuildHTTPRequestHandler)
     thr = threading.Thread(target=srv.serve_forever)
     thr.start()
-    input('Press ENTER to shut down')
+    compat_input('Press ENTER to shut down')
     srv.shutdown()
     thr.join()
 
@@ -231,8 +256,6 @@ def rmtree(path):
             os.remove(fname)
     os.rmdir(path)
 
-#==============================================================================
-
 
 class BuildError(Exception):
     def __init__(self, output, code=500):
@@ -249,15 +272,25 @@ class HTTPError(BuildError):
 
 class PythonBuilder(object):
     def __init__(self, **kwargs):
-        pythonVersion = kwargs.pop('python', '2.7')
-        try:
-            key = _winreg.OpenKey(_winreg.HKEY_LOCAL_MACHINE, r'SOFTWARE\Python\PythonCore\%s\InstallPath' % pythonVersion)
+        python_version = kwargs.pop('python', '3.4')
+        python_path = None
+        for node in ('Wow6432Node\\', ''):
             try:
-                self.pythonPath, _ = _winreg.QueryValueEx(key, '')
-            finally:
-                _winreg.CloseKey(key)
-        except Exception:
-            raise BuildError('No such Python version: %s' % pythonVersion)
+                key = compat_winreg.OpenKey(
+                    compat_winreg.HKEY_LOCAL_MACHINE,
+                    r'SOFTWARE\%sPython\PythonCore\%s\InstallPath' % (node, python_version))
+                try:
+                    python_path, _ = compat_winreg.QueryValueEx(key, '')
+                finally:
+                    compat_winreg.CloseKey(key)
+                break
+            except Exception:
+                pass
+
+        if not python_path:
+            raise BuildError('No such Python version: %s' % python_version)
+
+        self.pythonPath = python_path
 
         super(PythonBuilder, self).__init__(**kwargs)
 
@@ -305,8 +338,10 @@ class YoutubeDLBuilder(object):
 
     def build(self):
         try:
-            subprocess.check_output([os.path.join(self.pythonPath, 'python.exe'), 'setup.py', 'py2exe'],
-                                    cwd=self.buildPath)
+            proc = subprocess.Popen([os.path.join(self.pythonPath, 'python.exe'), 'setup.py', 'py2exe'], stdin=subprocess.PIPE, cwd=self.buildPath)
+            proc.wait()
+            #subprocess.check_output([os.path.join(self.pythonPath, 'python.exe'), 'setup.py', 'py2exe'],
+            #                        cwd=self.buildPath)
         except subprocess.CalledProcessError as e:
             raise BuildError(e.output)
 
@@ -369,12 +404,12 @@ class Builder(PythonBuilder, GITBuilder, YoutubeDLBuilder, DownloadBuilder, Clea
     pass
 
 
-class BuildHTTPRequestHandler(BaseHTTPRequestHandler):
+class BuildHTTPRequestHandler(compat_http_server.BaseHTTPRequestHandler):
     actionDict = {'build': Builder, 'download': Builder}  # They're the same, no more caching.
 
     def do_GET(self):
-        path = urlparse.urlparse(self.path)
-        paramDict = dict([(key, value[0]) for key, value in urlparse.parse_qs(path.query).items()])
+        path = compat_urlparse.urlparse(self.path)
+        paramDict = dict([(key, value[0]) for key, value in compat_urlparse.parse_qs(path.query).items()])
         action, _, path = path.path.strip('/').partition('/')
         if path:
             path = path.split('/')
@@ -388,7 +423,7 @@ class BuildHTTPRequestHandler(BaseHTTPRequestHandler):
                         builder.close()
                 except BuildError as e:
                     self.send_response(e.code)
-                    msg = unicode(e).encode('UTF-8')
+                    msg = compat_str(e).encode('UTF-8')
                     self.send_header('Content-Type', 'text/plain; charset=UTF-8')
                     self.send_header('Content-Length', len(msg))
                     self.end_headers()
@@ -400,7 +435,5 @@ class BuildHTTPRequestHandler(BaseHTTPRequestHandler):
         else:
             self.send_response(500, 'Malformed URL')
 
-#==============================================================================
-
 if __name__ == '__main__':
     main()

devscripts/prepare_manpage.py

@@ -1,13 +1,46 @@
 from __future__ import unicode_literals
 
 import io
+import optparse
 import os.path
-import sys
 import re
 
 ROOT_DIR = os.path.dirname(os.path.dirname(os.path.abspath(__file__)))
 README_FILE = os.path.join(ROOT_DIR, 'README.md')
 
+PREFIX = '''%YOUTUBE-DL(1)
+
+# NAME
+
+youtube\-dl \- download videos from youtube.com or other video platforms
+
+# SYNOPSIS
+
+**youtube-dl** \[OPTIONS\] URL [URL...]
+
+'''
+
+
+def main():
+    parser = optparse.OptionParser(usage='%prog OUTFILE.md')
+    options, args = parser.parse_args()
+    if len(args) != 1:
+        parser.error('Expected an output filename')
+
+    outfile, = args
+
+    with io.open(README_FILE, encoding='utf-8') as f:
+        readme = f.read()
+
+    readme = re.sub(r'(?s)^.*?(?=# DESCRIPTION)', '', readme)
+    readme = re.sub(r'\s+youtube-dl \[OPTIONS\] URL \[URL\.\.\.\]', '', readme)
+    readme = PREFIX + readme
+
+    readme = filter_options(readme)
+
+    with io.open(outfile, 'w', encoding='utf-8') as outf:
+        outf.write(readme)
+
 
 def filter_options(readme):
     ret = ''
@@ -37,27 +70,5 @@ def filter_options(readme):
 
     return ret
 
-with io.open(README_FILE, encoding='utf-8') as f:
-    readme = f.read()
-
-PREFIX = '''%YOUTUBE-DL(1)
-
-# NAME
-
-youtube\-dl \- download videos from youtube.com or other video platforms
-
-# SYNOPSIS
-
-**youtube-dl** \[OPTIONS\] URL [URL...]
-
-'''
-readme = re.sub(r'(?s)^.*?(?=# DESCRIPTION)', '', readme)
-readme = re.sub(r'\s+youtube-dl \[OPTIONS\] URL \[URL\.\.\.\]', '', readme)
-readme = PREFIX + readme
-
-readme = filter_options(readme)
-
-if sys.version_info < (3, 0):
-    print(readme.encode('utf-8'))
-else:
-    print(readme)
+if __name__ == '__main__':
+    main()

devscripts/release.sh

@@ -6,7 +6,7 @@
 # * the git config user.signingkey is properly set
 
 # You will need
-# pip install coverage nose rsa
+# pip install coverage nose rsa wheel
 
 # TODO
 # release notes
@@ -15,10 +15,28 @@
 set -e
 
 skip_tests=true
-if [ "$1" = '--run-tests' ]; then
-    skip_tests=false
-    shift
-fi
+buildserver='localhost:8142'
+
+while true
+do
+case "$1" in
+    --run-tests)
+        skip_tests=false
+        shift
+    ;;
+    --buildserver)
+        buildserver="$2"
+        shift 2
+    ;;
+    --*)
+        echo "ERROR: unknown option $1"
+        exit 1
+    ;;
+    *)
+        break
+    ;;
+esac
+done
 
 if [ -z "$1" ]; then echo "ERROR: specify version number like this: $0 1994.09.06"; exit 1; fi
 version="$1"
@@ -35,6 +53,7 @@ if [ ! -z "$useless_files" ]; then echo "ERROR: Non-.py files in youtube_dl: $us
 if [ ! -f "updates_key.pem" ]; then echo 'ERROR: updates_key.pem missing'; exit 1; fi
 if ! type pandoc >/dev/null 2>/dev/null; then echo 'ERROR: pandoc is missing'; exit 1; fi
 if ! python3 -c 'import rsa' 2>/dev/null; then echo 'ERROR: python3-rsa is missing'; exit 1; fi
+if ! python3 -c 'import wheel' 2>/dev/null; then echo 'ERROR: wheel is missing'; exit 1; fi
 
 /bin/echo -e "\n### First of all, testing..."
 make clean
@@ -66,7 +85,7 @@ git push origin "$version"
 REV=$(git rev-parse HEAD)
 make youtube-dl youtube-dl.tar.gz
 read -p "VM running? (y/n) " -n 1
-wget "http://localhost:8142/build/rg3/youtube-dl/youtube-dl.exe?rev=$REV" -O youtube-dl.exe
+wget "http://$buildserver/build/rg3/youtube-dl/youtube-dl.exe?rev=$REV" -O youtube-dl.exe
 mkdir -p "build/$version"
 mv youtube-dl youtube-dl.exe "build/$version"
 mv youtube-dl.tar.gz "build/$version/youtube-dl-$version.tar.gz"

docs/supportedsites.md

@@ -55,6 +55,7 @@
  - **arte.tv:future**
  - **arte.tv:info**
  - **arte.tv:magazine**
+ - **arte.tv:playlist**
  - **AtresPlayer**
  - **ATTTechChannel**
  - **AudiMedia**
@@ -136,6 +137,7 @@
  - **ComedyCentral**
  - **ComedyCentralShows**: The Daily Show / The Colbert Report
  - **CondeNast**: Condé Nast media group: Allure, Architectural Digest, Ars Technica, Bon Appétit, Brides, Condé Nast, Condé Nast Traveler, Details, Epicurious, GQ, Glamour, Golf Digest, SELF, Teen Vogue, The New Yorker, Vanity Fair, Vogue, W Magazine, WIRED
+ - **Coub**
  - **Cracked**
  - **Crackle**
  - **Criterion**
@@ -205,6 +207,7 @@
  - **exfm**: ex.fm
  - **ExpoTV**
  - **ExtremeTube**
+ - **EyedoTV**
  - **facebook**
  - **faz.net**
  - **fc2**
@@ -326,8 +329,8 @@
  - **LePlaylist**
  - **LetvCloud**: 乐视云
  - **Libsyn**
+ - **life**: Life.ru
  - **life:embed**
- - **lifenews**: LIFE | NEWS
  - **limelight**
  - **limelight:channel**
  - **limelight:channel_list**
@@ -336,6 +339,7 @@
  - **livestream**
  - **livestream:original**
  - **LnkGo**
+ - **loc**: Library of Congress
  - **LocalNews8**
  - **LoveHomePorn**
  - **lrt.lt**
@@ -512,6 +516,8 @@
  - **R7**
  - **radio.de**
  - **radiobremen**
+ - **radiocanada**
+ - **RadioCanadaAudioVideo**
  - **radiofrance**
  - **RadioJavan**
  - **Rai**
@@ -521,8 +527,10 @@
  - **RedTube**
  - **RegioTV**
  - **Restudy**
+ - **Reuters**
  - **ReverbNation**
- - **Revision3**
+ - **revision**
+ - **revision3:embed**
  - **RICE**
  - **RingTV**
  - **RottenTomatoes**
@@ -561,6 +569,7 @@
  - **ScreencastOMatic**
  - **ScreenJunkies**
  - **ScreenwaveMedia**
+ - **Seeker**
  - **SenateISVP**
  - **SendtoNews**
  - **ServingSys**
@@ -682,8 +691,8 @@
  - **TVCArticle**
  - **tvigle**: Интернет-телевидение Tvigle.ru
  - **tvland.com**
- - **tvp.pl**
- - **tvp.pl:Series**
+ - **tvp**: Telewizja Polska
+ - **tvp:series**
  - **TVPlay**: TV3Play and related services
  - **Tweakers**
  - **twitch:chapter**
@@ -766,7 +775,8 @@
  - **VuClip**
  - **vulture.com**
  - **Walla**
- - **WashingtonPost**
+ - **washingtonpost**
+ - **washingtonpost:article**
  - **wat.tv**
  - **WatchIndianPorn**: Watch Indian Porn
  - **WDR**

test/test_compat.py

@@ -103,6 +103,12 @@ class TestCompat(unittest.TestCase):
         self.assertTrue(isinstance(doc.find('chinese').text, compat_str))
         self.assertTrue(isinstance(doc.find('foo/bar').text, compat_str))
 
+    def test_compat_etree_fromstring_doctype(self):
+        xml = '''<?xml version="1.0"?>
+<!DOCTYPE smil PUBLIC "-//W3C//DTD SMIL 2.0//EN" "http://www.w3.org/2001/SMIL20/SMIL20.dtd">
+<smil xmlns="http://www.w3.org/2001/SMIL20/Language"></smil>'''
+        compat_etree_fromstring(xml)
+
     def test_struct_unpack(self):
         self.assertEqual(compat_struct_unpack('!B', b'\x00'), (0,))
 

test/test_http.py

@@ -16,6 +16,15 @@ import threading
 TEST_DIR = os.path.dirname(os.path.abspath(__file__))
 
 
+def http_server_port(httpd):
+    if os.name == 'java' and isinstance(httpd.socket, ssl.SSLSocket):
+        # In Jython SSLSocket is not a subclass of socket.socket
+        sock = httpd.socket.sock
+    else:
+        sock = httpd.socket
+    return sock.getsockname()[1]
+
+
 class HTTPTestRequestHandler(compat_http_server.BaseHTTPRequestHandler):
     def log_message(self, format, *args):
         pass
@@ -31,6 +40,22 @@ class HTTPTestRequestHandler(compat_http_server.BaseHTTPRequestHandler):
             self.send_header('Content-Type', 'video/mp4')
             self.end_headers()
             self.wfile.write(b'\x00\x00\x00\x00\x20\x66\x74[video]')
+        elif self.path == '/302':
+            if sys.version_info[0] == 3:
+                # XXX: Python 3 http server does not allow non-ASCII header values
+                self.send_response(404)
+                self.end_headers()
+                return
+
+            new_url = 'http://localhost:%d/中文.html' % http_server_port(self.server)
+            self.send_response(302)
+            self.send_header(b'Location', new_url.encode('utf-8'))
+            self.end_headers()
+        elif self.path == '/%E4%B8%AD%E6%96%87.html':
+            self.send_response(200)
+            self.send_header('Content-Type', 'text/html; charset=utf-8')
+            self.end_headers()
+            self.wfile.write(b'<html><video src="/vid.mp4" /></html>')
         else:
             assert False
 
@@ -47,18 +72,32 @@ class FakeLogger(object):
 
 
 class TestHTTP(unittest.TestCase):
+    def setUp(self):
+        self.httpd = compat_http_server.HTTPServer(
+            ('localhost', 0), HTTPTestRequestHandler)
+        self.port = http_server_port(self.httpd)
+        self.server_thread = threading.Thread(target=self.httpd.serve_forever)
+        self.server_thread.daemon = True
+        self.server_thread.start()
+
+    def test_unicode_path_redirection(self):
+        # XXX: Python 3 http server does not allow non-ASCII header values
+        if sys.version_info[0] == 3:
+            return
+
+        ydl = YoutubeDL({'logger': FakeLogger()})
+        r = ydl.extract_info('http://localhost:%d/302' % self.port)
+        self.assertEqual(r['url'], 'http://localhost:%d/vid.mp4' % self.port)
+
+
+class TestHTTPS(unittest.TestCase):
     def setUp(self):
         certfn = os.path.join(TEST_DIR, 'testcert.pem')
         self.httpd = compat_http_server.HTTPServer(
             ('localhost', 0), HTTPTestRequestHandler)
         self.httpd.socket = ssl.wrap_socket(
             self.httpd.socket, certfile=certfn, server_side=True)
-        if os.name == 'java':
-            # In Jython SSLSocket is not a subclass of socket.socket
-            sock = self.httpd.socket.sock
-        else:
-            sock = self.httpd.socket
-        self.port = sock.getsockname()[1]
+        self.port = http_server_port(self.httpd)
         self.server_thread = threading.Thread(target=self.httpd.serve_forever)
         self.server_thread.daemon = True
         self.server_thread.start()
@@ -94,14 +133,14 @@ class TestProxy(unittest.TestCase):
     def setUp(self):
         self.proxy = compat_http_server.HTTPServer(
             ('localhost', 0), _build_proxy_handler('normal'))
-        self.port = self.proxy.socket.getsockname()[1]
+        self.port = http_server_port(self.proxy)
         self.proxy_thread = threading.Thread(target=self.proxy.serve_forever)
         self.proxy_thread.daemon = True
         self.proxy_thread.start()
 
         self.cn_proxy = compat_http_server.HTTPServer(
             ('localhost', 0), _build_proxy_handler('cn'))
-        self.cn_port = self.cn_proxy.socket.getsockname()[1]
+        self.cn_port = http_server_port(self.cn_proxy)
         self.cn_proxy_thread = threading.Thread(target=self.cn_proxy.serve_forever)
         self.cn_proxy_thread.daemon = True
         self.cn_proxy_thread.start()

test/test_utils.py

@@ -157,8 +157,8 @@ class TestUtil(unittest.TestCase):
         self.assertTrue(sanitize_filename(':', restricted=True) != '')
 
         self.assertEqual(sanitize_filename(
-            'ÂÃÄÀÁÅÆÇÈÉÊËÌÍÎÏÐÑÒÓÔÕÖØŒÙÚÛÜÝÞßàáâãäåæçèéêëìíîïðñòóôõöøœùúûüýþÿ', restricted=True),
-            'AAAAAAAECEEEEIIIIDNOOOOOOOEUUUUYPssaaaaaaaeceeeeiiiionoooooooeuuuuypy')
+            'ÂÃÄÀÁÅÆÇÈÉÊËÌÍÎÏÐÑÒÓÔÕÖŐØŒÙÚÛÜŰÝÞßàáâãäåæçèéêëìíîïðñòóôõöőøœùúûüűýþÿ', restricted=True),
+            'AAAAAAAECEEEEIIIIDNOOOOOOOOEUUUUUYPssaaaaaaaeceeeeiiiionooooooooeuuuuuypy')
 
     def test_sanitize_ids(self):
         self.assertEqual(sanitize_filename('_n_cd26wFpw', is_id=True), '_n_cd26wFpw')

youtube_dl/compat.py

@@ -245,13 +245,20 @@ try:
 except ImportError:  # Python 2.6
     from xml.parsers.expat import ExpatError as compat_xml_parse_error
 
+
+etree = xml.etree.ElementTree
+
+
+class _TreeBuilder(etree.TreeBuilder):
+    def doctype(self, name, pubid, system):
+        pass
+
 if sys.version_info[0] >= 3:
-    compat_etree_fromstring = xml.etree.ElementTree.fromstring
+    def compat_etree_fromstring(text):
+        return etree.XML(text, parser=etree.XMLParser(target=_TreeBuilder()))
 else:
     # python 2.x tries to encode unicode strings with ascii (see the
     # XMLParser._fixtext method)
-    etree = xml.etree.ElementTree
-
     try:
         _etree_iter = etree.Element.iter
     except AttributeError:  # Python <=2.6
@@ -265,7 +272,7 @@ else:
     # 2.7 source
     def _XML(text, parser=None):
         if not parser:
-            parser = etree.XMLParser(target=etree.TreeBuilder())
+            parser = etree.XMLParser(target=_TreeBuilder())
         parser.feed(text)
         return parser.close()
 
@@ -277,7 +284,7 @@ else:
         return el
 
     def compat_etree_fromstring(text):
-        doc = _XML(text, parser=etree.XMLParser(target=etree.TreeBuilder(element_factory=_element_factory)))
+        doc = _XML(text, parser=etree.XMLParser(target=_TreeBuilder(element_factory=_element_factory)))
         for el in _etree_iter(doc):
             if el.text is not None and isinstance(el.text, bytes):
                 el.text = el.text.decode('utf-8')

youtube_dl/downloader/f4m.py

@@ -319,7 +319,7 @@ class F4mFD(FragmentFD):
         doc = compat_etree_fromstring(manifest)
         formats = [(int(f.attrib.get('bitrate', -1)), f)
                    for f in self._get_unencrypted_media(doc)]
-        if requested_bitrate is None:
+        if requested_bitrate is None or len(formats) == 1:
             # get the best format
             formats = sorted(formats, key=lambda f: f[0])
             rate, media = formats[-1]

youtube_dl/extractor/arte.py

@@ -61,10 +61,7 @@ class ArteTvIE(InfoExtractor):
         }
 
 
-class ArteTVPlus7IE(InfoExtractor):
-    IE_NAME = 'arte.tv:+7'
-    _VALID_URL = r'https?://(?:www\.)?arte\.tv/guide/(?P<lang>fr|de|en|es)/(?:(?:sendungen|emissions|embed)/)?(?P<id>[^/]+)/(?P<name>[^/?#&]+)'
-
+class ArteTVBaseIE(InfoExtractor):
     @classmethod
     def _extract_url_info(cls, url):
         mobj = re.match(cls._VALID_URL, url)
@@ -78,60 +75,6 @@ class ArteTVPlus7IE(InfoExtractor):
             video_id = mobj.group('id')
         return video_id, lang
 
-    def _real_extract(self, url):
-        video_id, lang = self._extract_url_info(url)
-        webpage = self._download_webpage(url, video_id)
-        return self._extract_from_webpage(webpage, video_id, lang)
-
-    def _extract_from_webpage(self, webpage, video_id, lang):
-        patterns_templates = (r'arte_vp_url=["\'](.*?%s.*?)["\']', r'data-url=["\']([^"]+%s[^"]+)["\']')
-        ids = (video_id, '')
-        # some pages contain multiple videos (like
-        # http://www.arte.tv/guide/de/sendungen/XEN/xenius/?vid=055918-015_PLUS7-D),
-        # so we first try to look for json URLs that contain the video id from
-        # the 'vid' parameter.
-        patterns = [t % re.escape(_id) for _id in ids for t in patterns_templates]
-        json_url = self._html_search_regex(
-            patterns, webpage, 'json vp url', default=None)
-        if not json_url:
-            def find_iframe_url(webpage, default=NO_DEFAULT):
-                return self._html_search_regex(
-                    r'<iframe[^>]+src=(["\'])(?P<url>.+\bjson_url=.+?)\1',
-                    webpage, 'iframe url', group='url', default=default)
-
-            iframe_url = find_iframe_url(webpage, None)
-            if not iframe_url:
-                embed_url = self._html_search_regex(
-                    r'arte_vp_url_oembed=\'([^\']+?)\'', webpage, 'embed url', default=None)
-                if embed_url:
-                    player = self._download_json(
-                        embed_url, video_id, 'Downloading player page')
-                    iframe_url = find_iframe_url(player['html'])
-            # en and es URLs produce react-based pages with different layout (e.g.
-            # http://www.arte.tv/guide/en/053330-002-A/carnival-italy?zone=world)
-            if not iframe_url:
-                program = self._search_regex(
-                    r'program\s*:\s*({.+?["\']embed_html["\'].+?}),?\s*\n',
-                    webpage, 'program', default=None)
-                if program:
-                    embed_html = self._parse_json(program, video_id)
-                    if embed_html:
-                        iframe_url = find_iframe_url(embed_html['embed_html'])
-            if iframe_url:
-                json_url = compat_parse_qs(
-                    compat_urllib_parse_urlparse(iframe_url).query)['json_url'][0]
-        if json_url:
-            title = self._search_regex(
-                r'<h3[^>]+title=(["\'])(?P<title>.+?)\1',
-                webpage, 'title', default=None, group='title')
-            return self._extract_from_json_url(json_url, video_id, lang, title=title)
-        # Different kind of embed URL (e.g.
-        # http://www.arte.tv/magazine/trepalium/fr/episode-0406-replay-trepalium)
-        embed_url = self._search_regex(
-            r'<iframe[^>]+src=(["\'])(?P<url>.+?)\1',
-            webpage, 'embed url', group='url')
-        return self.url_result(embed_url)
-
     def _extract_from_json_url(self, json_url, video_id, lang, title=None):
         info = self._download_json(json_url, video_id)
         player_info = info['videoJsonPlayer']
@@ -235,6 +178,74 @@ class ArteTVPlus7IE(InfoExtractor):
         return info_dict
 
 
+class ArteTVPlus7IE(ArteTVBaseIE):
+    IE_NAME = 'arte.tv:+7'
+    _VALID_URL = r'https?://(?:www\.)?arte\.tv/guide/(?P<lang>fr|de|en|es)/(?:(?:sendungen|emissions|embed)/)?(?P<id>[^/]+)/(?P<name>[^/?#&]+)'
+
+    _TESTS = [{
+        'url': 'http://www.arte.tv/guide/de/sendungen/XEN/xenius/?vid=055918-015_PLUS7-D',
+        'only_matching': True,
+    }]
+
+    @classmethod
+    def suitable(cls, url):
+        return False if ArteTVPlaylistIE.suitable(url) else super(ArteTVPlus7IE, cls).suitable(url)
+
+    def _real_extract(self, url):
+        video_id, lang = self._extract_url_info(url)
+        webpage = self._download_webpage(url, video_id)
+        return self._extract_from_webpage(webpage, video_id, lang)
+
+    def _extract_from_webpage(self, webpage, video_id, lang):
+        patterns_templates = (r'arte_vp_url=["\'](.*?%s.*?)["\']', r'data-url=["\']([^"]+%s[^"]+)["\']')
+        ids = (video_id, '')
+        # some pages contain multiple videos (like
+        # http://www.arte.tv/guide/de/sendungen/XEN/xenius/?vid=055918-015_PLUS7-D),
+        # so we first try to look for json URLs that contain the video id from
+        # the 'vid' parameter.
+        patterns = [t % re.escape(_id) for _id in ids for t in patterns_templates]
+        json_url = self._html_search_regex(
+            patterns, webpage, 'json vp url', default=None)
+        if not json_url:
+            def find_iframe_url(webpage, default=NO_DEFAULT):
+                return self._html_search_regex(
+                    r'<iframe[^>]+src=(["\'])(?P<url>.+\bjson_url=.+?)\1',
+                    webpage, 'iframe url', group='url', default=default)
+
+            iframe_url = find_iframe_url(webpage, None)
+            if not iframe_url:
+                embed_url = self._html_search_regex(
+                    r'arte_vp_url_oembed=\'([^\']+?)\'', webpage, 'embed url', default=None)
+                if embed_url:
+                    player = self._download_json(
+                        embed_url, video_id, 'Downloading player page')
+                    iframe_url = find_iframe_url(player['html'])
+            # en and es URLs produce react-based pages with different layout (e.g.
+            # http://www.arte.tv/guide/en/053330-002-A/carnival-italy?zone=world)
+            if not iframe_url:
+                program = self._search_regex(
+                    r'program\s*:\s*({.+?["\']embed_html["\'].+?}),?\s*\n',
+                    webpage, 'program', default=None)
+                if program:
+                    embed_html = self._parse_json(program, video_id)
+                    if embed_html:
+                        iframe_url = find_iframe_url(embed_html['embed_html'])
+            if iframe_url:
+                json_url = compat_parse_qs(
+                    compat_urllib_parse_urlparse(iframe_url).query)['json_url'][0]
+        if json_url:
+            title = self._search_regex(
+                r'<h3[^>]+title=(["\'])(?P<title>.+?)\1',
+                webpage, 'title', default=None, group='title')
+            return self._extract_from_json_url(json_url, video_id, lang, title=title)
+        # Different kind of embed URL (e.g.
+        # http://www.arte.tv/magazine/trepalium/fr/episode-0406-replay-trepalium)
+        embed_url = self._search_regex(
+            r'<iframe[^>]+src=(["\'])(?P<url>.+?)\1',
+            webpage, 'embed url', group='url')
+        return self.url_result(embed_url)
+
+
 # It also uses the arte_vp_url url from the webpage to extract the information
 class ArteTVCreativeIE(ArteTVPlus7IE):
     IE_NAME = 'arte.tv:creative'
@@ -267,7 +278,7 @@ class ArteTVInfoIE(ArteTVPlus7IE):
     IE_NAME = 'arte.tv:info'
     _VALID_URL = r'https?://info\.arte\.tv/(?P<lang>fr|de|en|es)/(?:[^/]+/)*(?P<id>[^/?#&]+)'
 
-    _TEST = {
+    _TESTS = [{
         'url': 'http://info.arte.tv/fr/service-civique-un-cache-misere',
         'info_dict': {
             'id': '067528-000-A',
@@ -275,7 +286,7 @@ class ArteTVInfoIE(ArteTVPlus7IE):
             'title': 'Service civique, un cache misère ?',
             'upload_date': '20160403',
         },
-    }
+    }]
 
 
 class ArteTVFutureIE(ArteTVPlus7IE):
@@ -300,6 +311,8 @@ class ArteTVDDCIE(ArteTVPlus7IE):
     IE_NAME = 'arte.tv:ddc'
     _VALID_URL = r'https?://ddc\.arte\.tv/(?P<lang>emission|folge)/(?P<id>[^/?#&]+)'
 
+    _TESTS = []
+
     def _real_extract(self, url):
         video_id, lang = self._extract_url_info(url)
         if lang == 'folge':
@@ -318,7 +331,7 @@ class ArteTVConcertIE(ArteTVPlus7IE):
     IE_NAME = 'arte.tv:concert'
     _VALID_URL = r'https?://concert\.arte\.tv/(?P<lang>fr|de|en|es)/(?P<id>[^/?#&]+)'
 
-    _TEST = {
+    _TESTS = [{
         'url': 'http://concert.arte.tv/de/notwist-im-pariser-konzertclub-divan-du-monde',
         'md5': '9ea035b7bd69696b67aa2ccaaa218161',
         'info_dict': {
@@ -328,14 +341,14 @@ class ArteTVConcertIE(ArteTVPlus7IE):
             'upload_date': '20140128',
             'description': 'md5:486eb08f991552ade77439fe6d82c305',
         },
-    }
+    }]
 
 
 class ArteTVCinemaIE(ArteTVPlus7IE):
     IE_NAME = 'arte.tv:cinema'
     _VALID_URL = r'https?://cinema\.arte\.tv/(?P<lang>fr|de|en|es)/(?P<id>.+)'
 
-    _TEST = {
+    _TESTS = [{
         'url': 'http://cinema.arte.tv/de/node/38291',
         'md5': '6b275511a5107c60bacbeeda368c3aa1',
         'info_dict': {
@@ -345,7 +358,7 @@ class ArteTVCinemaIE(ArteTVPlus7IE):
             'upload_date': '20160122',
             'description': 'md5:7f749bbb77d800ef2be11d54529b96bc',
         },
-    }
+    }]
 
 
 class ArteTVMagazineIE(ArteTVPlus7IE):
@@ -390,9 +403,41 @@ class ArteTVEmbedIE(ArteTVPlus7IE):
         )
     '''
 
+    _TESTS = []
+
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
         video_id = mobj.group('id')
         lang = mobj.group('lang')
         json_url = mobj.group('json_url')
         return self._extract_from_json_url(json_url, video_id, lang)
+
+
+class ArteTVPlaylistIE(ArteTVBaseIE):
+    IE_NAME = 'arte.tv:playlist'
+    _VALID_URL = r'https?://(?:www\.)?arte\.tv/guide/(?P<lang>fr|de|en|es)/[^#]*#collection/(?P<id>PL-\d+)'
+
+    _TESTS = [{
+        'url': 'http://www.arte.tv/guide/de/plus7/?country=DE#collection/PL-013263/ARTETV',
+        'info_dict': {
+            'id': 'PL-013263',
+            'title': 'Areva & Uramin',
+        },
+        'playlist_mincount': 6,
+    }, {
+        'url': 'http://www.arte.tv/guide/de/playlists?country=DE#collection/PL-013190/ARTETV',
+        'only_matching': True,
+    }]
+
+    def _real_extract(self, url):
+        playlist_id, lang = self._extract_url_info(url)
+        collection = self._download_json(
+            'https://api.arte.tv/api/player/v1/collectionData/%s/%s?source=videos'
+            % (lang, playlist_id), playlist_id)
+        title = collection.get('title')
+        description = collection.get('shortDescription') or collection.get('teaserText')
+        entries = [
+            self._extract_from_json_url(
+                video['jsonUrl'], video.get('programId') or playlist_id, lang)
+            for video in collection['videos'] if video.get('jsonUrl')]
+        return self.playlist_result(entries, playlist_id, title, description)

youtube_dl/extractor/bandcamp.py

@@ -29,7 +29,7 @@ class BandcampIE(InfoExtractor):
         '_skip': 'There is a limit of 200 free downloads / month for the test song'
     }, {
         'url': 'http://benprunty.bandcamp.com/track/lanius-battle',
-        'md5': '2b68e5851514c20efdff2afc5603b8b4',
+        'md5': '73d0b3171568232574e45652f8720b5c',
         'info_dict': {
             'id': '2650410135',
             'ext': 'mp3',
@@ -48,6 +48,10 @@ class BandcampIE(InfoExtractor):
             if m_trackinfo:
                 json_code = m_trackinfo.group(1)
                 data = json.loads(json_code)[0]
+                track_id = compat_str(data['id'])
+
+                if not data.get('file'):
+                    raise ExtractorError('Not streamable', video_id=track_id, expected=True)
 
                 formats = []
                 for format_id, format_url in data['file'].items():
@@ -64,7 +68,7 @@ class BandcampIE(InfoExtractor):
                 self._sort_formats(formats)
 
                 return {
-                    'id': compat_str(data['id']),
+                    'id': track_id,
                     'title': data['title'],
                     'formats': formats,
                     'duration': float_or_none(data.get('duration')),

youtube_dl/extractor/bilibili.py

@@ -1,34 +1,42 @@
 # coding: utf-8
 from __future__ import unicode_literals
 
+import calendar
+import datetime
 import re
 
 from .common import InfoExtractor
-from ..compat import compat_str
+from ..compat import (
+    compat_etree_fromstring,
+    compat_str,
+    compat_parse_qs,
+    compat_xml_parse_error,
+)
 from ..utils import (
-    int_or_none,
-    unescapeHTML,
     ExtractorError,
+    int_or_none,
+    float_or_none,
     xpath_text,
 )
 
 
 class BiliBiliIE(InfoExtractor):
-    _VALID_URL = r'https?://www\.bilibili\.(?:tv|com)/video/av(?P<id>\d+)(?:/index_(?P<page_num>\d+).html)?'
+    _VALID_URL = r'https?://www\.bilibili\.(?:tv|com)/video/av(?P<id>\d+)'
 
     _TESTS = [{
         'url': 'http://www.bilibili.tv/video/av1074402/',
-        'md5': '2c301e4dab317596e837c3e7633e7d86',
+        'md5': '5f7d29e1a2872f3df0cf76b1f87d3788',
         'info_dict': {
             'id': '1554319',
             'ext': 'flv',
             'title': '【金坷垃】金泡沫',
-            'duration': 308313,
+            'description': 'md5:ce18c2a2d2193f0df2917d270f2e5923',
+            'duration': 308.067,
+            'timestamp': 1398012660,
             'upload_date': '20140420',
             'thumbnail': 're:^https?://.+\.jpg',
-            'description': 'md5:ce18c2a2d2193f0df2917d270f2e5923',
-            'timestamp': 1397983878,
             'uploader': '菊子桑',
+            'uploader_id': '156160',
         },
     }, {
         'url': 'http://www.bilibili.com/video/av1041170/',
@@ -36,75 +44,169 @@ class BiliBiliIE(InfoExtractor):
             'id': '1041170',
             'title': '【BD1080P】刀语【诸神&异域】',
             'description': '这是个神奇的故事~每个人不留弹幕不给走哦~切利哦！~',
-            'uploader': '枫叶逝去',
-            'timestamp': 1396501299,
         },
         'playlist_count': 9,
+    }, {
+        'url': 'http://www.bilibili.com/video/av4808130/',
+        'info_dict': {
+            'id': '4808130',
+            'title': '【长篇】哆啦A梦443【钉铛】',
+            'description': '(2016.05.27)来组合客人的脸吧&amp;amp;寻母六千里锭 抱歉，又轮到周日上班现在才到家 封面www.pixiv.net/member_illust.php?mode=medium&amp;amp;illust_id=56912929',
+        },
+        'playlist': [{
+            'md5': '55cdadedf3254caaa0d5d27cf20a8f9c',
+            'info_dict': {
+                'id': '4808130_part1',
+                'ext': 'flv',
+                'title': '【长篇】哆啦A梦443【钉铛】',
+                'description': '(2016.05.27)来组合客人的脸吧&amp;amp;寻母六千里锭 抱歉，又轮到周日上班现在才到家 封面www.pixiv.net/member_illust.php?mode=medium&amp;amp;illust_id=56912929',
+                'timestamp': 1464564180,
+                'upload_date': '20160529',
+                'uploader': '喜欢拉面',
+                'uploader_id': '151066',
+            },
+        }, {
+            'md5': '926f9f67d0c482091872fbd8eca7ea3d',
+            'info_dict': {
+                'id': '4808130_part2',
+                'ext': 'flv',
+                'title': '【长篇】哆啦A梦443【钉铛】',
+                'description': '(2016.05.27)来组合客人的脸吧&amp;amp;寻母六千里锭 抱歉，又轮到周日上班现在才到家 封面www.pixiv.net/member_illust.php?mode=medium&amp;amp;illust_id=56912929',
+                'timestamp': 1464564180,
+                'upload_date': '20160529',
+                'uploader': '喜欢拉面',
+                'uploader_id': '151066',
+            },
+        }, {
+            'md5': '4b7b225b968402d7c32348c646f1fd83',
+            'info_dict': {
+                'id': '4808130_part3',
+                'ext': 'flv',
+                'title': '【长篇】哆啦A梦443【钉铛】',
+                'description': '(2016.05.27)来组合客人的脸吧&amp;amp;寻母六千里锭 抱歉，又轮到周日上班现在才到家 封面www.pixiv.net/member_illust.php?mode=medium&amp;amp;illust_id=56912929',
+                'timestamp': 1464564180,
+                'upload_date': '20160529',
+                'uploader': '喜欢拉面',
+                'uploader_id': '151066',
+            },
+        }, {
+            'md5': '7b795e214166501e9141139eea236e91',
+            'info_dict': {
+                'id': '4808130_part4',
+                'ext': 'flv',
+                'title': '【长篇】哆啦A梦443【钉铛】',
+                'description': '(2016.05.27)来组合客人的脸吧&amp;amp;寻母六千里锭 抱歉，又轮到周日上班现在才到家 封面www.pixiv.net/member_illust.php?mode=medium&amp;amp;illust_id=56912929',
+                'timestamp': 1464564180,
+                'upload_date': '20160529',
+                'uploader': '喜欢拉面',
+                'uploader_id': '151066',
+            },
+        }],
     }]
 
+    # BiliBili blocks keys from time to time. The current key is extracted from
+    # the Android client
+    # TODO: find the sign algorithm used in the flash player
+    _APP_KEY = '86385cdc024c0f6c'
+
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
         video_id = mobj.group('id')
-        page_num = mobj.group('page_num') or '1'
 
-        view_data = self._download_json(
-            'http://api.bilibili.com/view?type=json&appkey=8e9fc618fbd41e28&id=%s&page=%s' % (video_id, page_num),
-            video_id)
-        if 'error' in view_data:
-            raise ExtractorError('%s said: %s' % (self.IE_NAME, view_data['error']), expected=True)
+        webpage = self._download_webpage(url, video_id)
 
-        cid = view_data['cid']
-        title = unescapeHTML(view_data['title'])
+        params = compat_parse_qs(self._search_regex(
+            [r'EmbedPlayer\([^)]+,\s*"([^"]+)"\)',
+             r'<iframe[^>]+src="https://secure\.bilibili\.com/secure,([^"]+)"'],
+            webpage, 'player parameters'))
+        cid = params['cid'][0]
 
-        doc = self._download_xml(
-            'http://interface.bilibili.com/v_cdn_play?appkey=8e9fc618fbd41e28&cid=%s' % cid,
-            cid,
-            'Downloading page %s/%s' % (page_num, view_data['pages'])
-        )
+        info_xml_str = self._download_webpage(
+            'http://interface.bilibili.com/v_cdn_play',
+            cid, query={'appkey': self._APP_KEY, 'cid': cid},
+            note='Downloading video info page')
 
-        if xpath_text(doc, './result') == 'error':
-            raise ExtractorError('%s said: %s' % (self.IE_NAME, xpath_text(doc, './message')), expected=True)
+        err_msg = None
+        durls = None
+        info_xml = None
+        try:
+            info_xml = compat_etree_fromstring(info_xml_str.encode('utf-8'))
+        except compat_xml_parse_error:
+            info_json = self._parse_json(info_xml_str, video_id, fatal=False)
+            err_msg = (info_json or {}).get('error_text')
+        else:
+            err_msg = xpath_text(info_xml, './message')
+
+        if info_xml is not None:
+            durls = info_xml.findall('./durl')
+        if not durls:
+            if err_msg:
+                raise ExtractorError('%s said: %s' % (self.IE_NAME, err_msg), expected=True)
+            else:
+                raise ExtractorError('No videos found!')
 
         entries = []
 
-        for durl in doc.findall('./durl'):
+        for durl in durls:
             size = xpath_text(durl, ['./filesize', './size'])
             formats = [{
                 'url': durl.find('./url').text,
                 'filesize': int_or_none(size),
-                'ext': 'flv',
             }]
-            backup_urls = durl.find('./backup_url')
-            if backup_urls is not None:
-                for backup_url in backup_urls.findall('./url'):
-                    formats.append({'url': backup_url.text})
-            formats.reverse()
+            for backup_url in durl.findall('./backup_url/url'):
+                formats.append({
+                    'url': backup_url.text,
+                    # backup URLs have lower priorities
+                    'preference': -2 if 'hd.mp4' in backup_url.text else -3,
+                })
+
+            self._sort_formats(formats)
 
             entries.append({
                 'id': '%s_part%s' % (cid, xpath_text(durl, './order')),
-                'title': title,
                 'duration': int_or_none(xpath_text(durl, './length'), 1000),
                 'formats': formats,
             })
 
+        title = self._html_search_regex('<h1[^>]+title="([^"]+)">', webpage, 'title')
+        description = self._html_search_meta('description', webpage)
+        datetime_str = self._html_search_regex(
+            r'<time[^>]+datetime="([^"]+)"', webpage, 'upload time', fatal=False)
+        if datetime_str:
+            timestamp = calendar.timegm(datetime.datetime.strptime(datetime_str, '%Y-%m-%dT%H:%M').timetuple())
+
+        # TODO 'view_count' requires deobfuscating Javascript
         info = {
             'id': compat_str(cid),
             'title': title,
-            'description': view_data.get('description'),
-            'thumbnail': view_data.get('pic'),
-            'uploader': view_data.get('author'),
-            'timestamp': int_or_none(view_data.get('created')),
-            'view_count': int_or_none(view_data.get('play')),
-            'duration': int_or_none(xpath_text(doc, './timelength')),
+            'description': description,
+            'timestamp': timestamp,
+            'thumbnail': self._html_search_meta('thumbnailUrl', webpage),
+            'duration': float_or_none(xpath_text(info_xml, './timelength'), scale=1000),
         }
 
+        uploader_mobj = re.search(
+            r'<a[^>]+href="https?://space\.bilibili\.com/(?P<id>\d+)"[^>]+title="(?P<name>[^"]+)"',
+            webpage)
+        if uploader_mobj:
+            info.update({
+                'uploader': uploader_mobj.group('name'),
+                'uploader_id': uploader_mobj.group('id'),
+            })
+
+        for entry in entries:
+            entry.update(info)
+
         if len(entries) == 1:
-            entries[0].update(info)
             return entries[0]
         else:
-            info.update({
+            for idx, entry in enumerate(entries):
+                entry['id'] = '%s_part%d' % (video_id, (idx + 1))
+
+            return {
                 '_type': 'multi_video',
                 'id': video_id,
+                'title': title,
+                'description': description,
                 'entries': entries,
-            })
-            return info
+            }

youtube_dl/extractor/byutv.py

@@ -11,6 +11,7 @@ class BYUtvIE(InfoExtractor):
     _VALID_URL = r'^https?://(?:www\.)?byutv.org/watch/[0-9a-f-]+/(?P<video_id>[^/?#]+)'
     _TEST = {
         'url': 'http://www.byutv.org/watch/6587b9a3-89d2-42a6-a7f7-fd2f81840a7d/studio-c-season-5-episode-5',
+        'md5': '05850eb8c749e2ee05ad5a1c34668493',
         'info_dict': {
             'id': 'studio-c-season-5-episode-5',
             'ext': 'mp4',
@@ -21,7 +22,8 @@ class BYUtvIE(InfoExtractor):
         },
         'params': {
             'skip_download': True,
-        }
+        },
+        'add_ie': ['Ooyala'],
     }
 
     def _real_extract(self, url):

youtube_dl/extractor/cbs.py

@@ -1,5 +1,7 @@
 from __future__ import unicode_literals
 
+import re
+
 from .theplatform import ThePlatformIE
 from ..utils import (
     xpath_text,
@@ -21,7 +23,7 @@ class CBSBaseIE(ThePlatformIE):
 
 
 class CBSIE(CBSBaseIE):
-    _VALID_URL = r'https?://(?:www\.)?(?:cbs\.com/shows/[^/]+/(?:video|artist)|colbertlateshow\.com/(?:video|podcasts))/[^/]+/(?P<id>[^/]+)'
+    _VALID_URL = r'(?:cbs:(?P<content_id>\w+)|https?://(?:www\.)?(?:cbs\.com/shows/[^/]+/(?:video|artist)|colbertlateshow\.com/(?:video|podcasts))/[^/]+/(?P<display_id>[^/]+))'
 
     _TESTS = [{
         'url': 'http://www.cbs.com/shows/garth-brooks/video/_u7W953k6la293J7EPTd9oHkSPs6Xn6_/connect-chat-feat-garth-brooks/',
@@ -66,11 +68,12 @@ class CBSIE(CBSBaseIE):
     TP_RELEASE_URL_TEMPLATE = 'http://link.theplatform.com/s/dJ5BDC/%s?mbr=true'
 
     def _real_extract(self, url):
-        display_id = self._match_id(url)
-        webpage = self._download_webpage(url, display_id)
-        content_id = self._search_regex(
-            [r"video\.settings\.content_id\s*=\s*'([^']+)';", r"cbsplayer\.contentId\s*=\s*'([^']+)';"],
-            webpage, 'content id')
+        content_id, display_id = re.match(self._VALID_URL, url).groups()
+        if not content_id:
+            webpage = self._download_webpage(url, display_id)
+            content_id = self._search_regex(
+                [r"video\.settings\.content_id\s*=\s*'([^']+)';", r"cbsplayer\.contentId\s*=\s*'([^']+)';"],
+                webpage, 'content id')
         items_data = self._download_xml(
             'http://can.cbs.com/thunder/player/videoPlayerService.php',
             content_id, query={'partner': 'cbs', 'contentId': content_id})

youtube_dl/extractor/channel9.py

@@ -20,54 +20,64 @@ class Channel9IE(InfoExtractor):
     '''
     IE_DESC = 'Channel 9'
     IE_NAME = 'channel9'
-    _VALID_URL = r'https?://(?:www\.)?channel9\.msdn\.com/(?P<contentpath>.+)/?'
+    _VALID_URL = r'https?://(?:www\.)?channel9\.msdn\.com/(?P<contentpath>.+?)(?P<rss>/RSS)?/?(?:[?#&]|$)'
 
-    _TESTS = [
-        {
-            'url': 'http://channel9.msdn.com/Events/TechEd/Australia/2013/KOS002',
-            'md5': 'bbd75296ba47916b754e73c3a4bbdf10',
-            'info_dict': {
-                'id': 'Events/TechEd/Australia/2013/KOS002',
-                'ext': 'mp4',
-                'title': 'Developer Kick-Off Session: Stuff We Love',
-                'description': 'md5:c08d72240b7c87fcecafe2692f80e35f',
-                'duration': 4576,
-                'thumbnail': 're:http://.*\.jpg',
-                'session_code': 'KOS002',
-                'session_day': 'Day 1',
-                'session_room': 'Arena 1A',
-                'session_speakers': ['Ed Blankenship', 'Andrew Coates', 'Brady Gaster', 'Patrick Klug', 'Mads Kristensen'],
-            },
+    _TESTS = [{
+        'url': 'http://channel9.msdn.com/Events/TechEd/Australia/2013/KOS002',
+        'md5': 'bbd75296ba47916b754e73c3a4bbdf10',
+        'info_dict': {
+            'id': 'Events/TechEd/Australia/2013/KOS002',
+            'ext': 'mp4',
+            'title': 'Developer Kick-Off Session: Stuff We Love',
+            'description': 'md5:c08d72240b7c87fcecafe2692f80e35f',
+            'duration': 4576,
+            'thumbnail': 're:http://.*\.jpg',
+            'session_code': 'KOS002',
+            'session_day': 'Day 1',
+            'session_room': 'Arena 1A',
+            'session_speakers': ['Ed Blankenship', 'Andrew Coates', 'Brady Gaster', 'Patrick Klug',
+                                 'Mads Kristensen'],
         },
-        {
-            'url': 'http://channel9.msdn.com/posts/Self-service-BI-with-Power-BI-nuclear-testing',
-            'md5': 'b43ee4529d111bc37ba7ee4f34813e68',
-            'info_dict': {
-                'id': 'posts/Self-service-BI-with-Power-BI-nuclear-testing',
-                'ext': 'mp4',
-                'title': 'Self-service BI with Power BI - nuclear testing',
-                'description': 'md5:d1e6ecaafa7fb52a2cacdf9599829f5b',
-                'duration': 1540,
-                'thumbnail': 're:http://.*\.jpg',
-                'authors': ['Mike Wilmot'],
-            },
+    }, {
+        'url': 'http://channel9.msdn.com/posts/Self-service-BI-with-Power-BI-nuclear-testing',
+        'md5': 'b43ee4529d111bc37ba7ee4f34813e68',
+        'info_dict': {
+            'id': 'posts/Self-service-BI-with-Power-BI-nuclear-testing',
+            'ext': 'mp4',
+            'title': 'Self-service BI with Power BI - nuclear testing',
+            'description': 'md5:d1e6ecaafa7fb52a2cacdf9599829f5b',
+            'duration': 1540,
+            'thumbnail': 're:http://.*\.jpg',
+            'authors': ['Mike Wilmot'],
         },
-        {
-            # low quality mp4 is best
-            'url': 'https://channel9.msdn.com/Events/CPP/CppCon-2015/Ranges-for-the-Standard-Library',
-            'info_dict': {
-                'id': 'Events/CPP/CppCon-2015/Ranges-for-the-Standard-Library',
-                'ext': 'mp4',
-                'title': 'Ranges for the Standard Library',
-                'description': 'md5:2e6b4917677af3728c5f6d63784c4c5d',
-                'duration': 5646,
-                'thumbnail': 're:http://.*\.jpg',
-            },
-            'params': {
-                'skip_download': True,
-            },
-        }
-    ]
+    }, {
+        # low quality mp4 is best
+        'url': 'https://channel9.msdn.com/Events/CPP/CppCon-2015/Ranges-for-the-Standard-Library',
+        'info_dict': {
+            'id': 'Events/CPP/CppCon-2015/Ranges-for-the-Standard-Library',
+            'ext': 'mp4',
+            'title': 'Ranges for the Standard Library',
+            'description': 'md5:2e6b4917677af3728c5f6d63784c4c5d',
+            'duration': 5646,
+            'thumbnail': 're:http://.*\.jpg',
+        },
+        'params': {
+            'skip_download': True,
+        },
+    }, {
+        'url': 'https://channel9.msdn.com/Niners/Splendid22/Queue/76acff796e8f411184b008028e0d492b/RSS',
+        'info_dict': {
+            'id': 'Niners/Splendid22/Queue/76acff796e8f411184b008028e0d492b',
+            'title': 'Channel 9',
+        },
+        'playlist_count': 2,
+    }, {
+        'url': 'https://channel9.msdn.com/Events/DEVintersection/DEVintersection-2016/RSS',
+        'only_matching': True,
+    }, {
+        'url': 'https://channel9.msdn.com/Events/Speakers/scott-hanselman/RSS?UrlSafeName=scott-hanselman',
+        'only_matching': True,
+    }]
 
     _RSS_URL = 'http://channel9.msdn.com/%s/RSS'
 
@@ -254,22 +264,30 @@ class Channel9IE(InfoExtractor):
 
         return self.playlist_result(contents)
 
-    def _extract_list(self, content_path):
-        rss = self._download_xml(self._RSS_URL % content_path, content_path, 'Downloading RSS')
+    def _extract_list(self, video_id, rss_url=None):
+        if not rss_url:
+            rss_url = self._RSS_URL % video_id
+        rss = self._download_xml(rss_url, video_id, 'Downloading RSS')
         entries = [self.url_result(session_url.text, 'Channel9')
                    for session_url in rss.findall('./channel/item/link')]
         title_text = rss.find('./channel/title').text
-        return self.playlist_result(entries, content_path, title_text)
+        return self.playlist_result(entries, video_id, title_text)
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
         content_path = mobj.group('contentpath')
+        rss = mobj.group('rss')
 
-        webpage = self._download_webpage(url, content_path, 'Downloading web page')
+        if rss:
+            return self._extract_list(content_path, url)
 
-        page_type_m = re.search(r'<meta name="WT.entryid" content="(?P<pagetype>[^:]+)[^"]+"/>', webpage)
-        if page_type_m is not None:
-            page_type = page_type_m.group('pagetype')
+        webpage = self._download_webpage(
+            url, content_path, 'Downloading web page')
+
+        page_type = self._search_regex(
+            r'<meta[^>]+name=(["\'])WT\.entryid\1[^>]+content=(["\'])(?P<pagetype>[^:]+).+?\2',
+            webpage, 'page type', default=None, group='pagetype')
+        if page_type:
             if page_type == 'Entry':      # Any 'item'-like page, may contain downloadable content
                 return self._extract_entry_item(webpage, content_path)
             elif page_type == 'Session':  # Event session page, may contain downloadable content
@@ -278,6 +296,5 @@ class Channel9IE(InfoExtractor):
                 return self._extract_list(content_path)
             else:
                 raise ExtractorError('Unexpected WT.entryid %s' % page_type, expected=True)
-
         else:  # Assuming list
             return self._extract_list(content_path)

youtube_dl/extractor/comedycentral.py

@@ -44,10 +44,10 @@ class ComedyCentralShowsIE(MTVServicesInfoExtractor):
     #                     or: http://www.colbertnation.com/the-colbert-report-collections/422008/festival-of-lights/79524
     _VALID_URL = r'''(?x)^(:(?P<shortname>tds|thedailyshow)
                       |https?://(:www\.)?
-                          (?P<showname>thedailyshow|thecolbertreport)\.(?:cc\.)?com/
+                          (?P<showname>thedailyshow|thecolbertreport|tosh)\.(?:cc\.)?com/
                          ((?:full-)?episodes/(?:[0-9a-z]{6}/)?(?P<episode>.*)|
                           (?P<clip>
-                              (?:(?:guests/[^/]+|videos|video-playlists|special-editions|news-team/[^/]+)/[^/]+/(?P<videotitle>[^/?#]+))
+                              (?:(?:guests/[^/]+|videos|video-(?:clips|playlists)|special-editions|news-team/[^/]+)/[^/]+/(?P<videotitle>[^/?#]+))
                               |(the-colbert-report-(videos|collections)/(?P<clipID>[0-9]+)/[^/]*/(?P<cntitle>.*?))
                               |(watch/(?P<date>[^/]*)/(?P<tdstitle>.*))
                           )|
@@ -129,6 +129,9 @@ class ComedyCentralShowsIE(MTVServicesInfoExtractor):
     }, {
         'url': 'http://thedailyshow.cc.com/news-team/michael-che/7wnfel/we-need-to-talk-about-israel',
         'only_matching': True,
+    }, {
+        'url': 'http://tosh.cc.com/video-clips/68g93d/twitter-users-share-summer-plans',
+        'only_matching': True,
     }]
 
     _available_formats = ['3500', '2200', '1700', '1200', '750', '400']

youtube_dl/extractor/common.py

@@ -987,7 +987,7 @@ class InfoExtractor(object):
 
     def _extract_f4m_formats(self, manifest_url, video_id, preference=None, f4m_id=None,
                              transform_source=lambda s: fix_xml_ampersands(s).strip(),
-                             fatal=True):
+                             fatal=True, m3u8_id=None):
         manifest = self._download_xml(
             manifest_url, video_id, 'Downloading f4m manifest',
             'Unable to download f4m manifest',
@@ -1001,11 +1001,11 @@ class InfoExtractor(object):
 
         return self._parse_f4m_formats(
             manifest, manifest_url, video_id, preference=preference, f4m_id=f4m_id,
-            transform_source=transform_source, fatal=fatal)
+            transform_source=transform_source, fatal=fatal, m3u8_id=m3u8_id)
 
     def _parse_f4m_formats(self, manifest, manifest_url, video_id, preference=None, f4m_id=None,
                            transform_source=lambda s: fix_xml_ampersands(s).strip(),
-                           fatal=True):
+                           fatal=True, m3u8_id=None):
         # currently youtube-dl cannot decode the playerVerificationChallenge as Akamai uses Adobe Alchemy
         akamai_pv = manifest.find('{http://ns.adobe.com/f4m/1.0}pv-2.0')
         if akamai_pv is not None and ';' in akamai_pv.text:
@@ -1029,9 +1029,26 @@ class InfoExtractor(object):
             'base URL', default=None)
         if base_url:
             base_url = base_url.strip()
+
+        bootstrap_info = xpath_text(
+            manifest, ['{http://ns.adobe.com/f4m/1.0}bootstrapInfo', '{http://ns.adobe.com/f4m/2.0}bootstrapInfo'],
+            'bootstrap info', default=None)
+
         for i, media_el in enumerate(media_nodes):
-            if manifest_version == '2.0':
-                media_url = media_el.attrib.get('href') or media_el.attrib.get('url')
+            tbr = int_or_none(media_el.attrib.get('bitrate'))
+            width = int_or_none(media_el.attrib.get('width'))
+            height = int_or_none(media_el.attrib.get('height'))
+            format_id = '-'.join(filter(None, [f4m_id, compat_str(i if tbr is None else tbr)]))
+            # If <bootstrapInfo> is present, the specified f4m is a
+            # stream-level manifest, and only set-level manifests may refer to
+            # external resources.  See section 11.4 and section 4 of F4M spec
+            if bootstrap_info is None:
+                media_url = None
+                # @href is introduced in 2.0, see section 11.6 of F4M spec
+                if manifest_version == '2.0':
+                    media_url = media_el.attrib.get('href')
+                if media_url is None:
+                    media_url = media_el.attrib.get('url')
                 if not media_url:
                     continue
                 manifest_url = (
@@ -1041,19 +1058,37 @@ class InfoExtractor(object):
                 # since bitrates in parent manifest (this one) and media_url manifest
                 # may differ leading to inability to resolve the format by requested
                 # bitrate in f4m downloader
-                if determine_ext(manifest_url) == 'f4m':
-                    formats.extend(self._extract_f4m_formats(
+                ext = determine_ext(manifest_url)
+                if ext == 'f4m':
+                    f4m_formats = self._extract_f4m_formats(
                         manifest_url, video_id, preference=preference, f4m_id=f4m_id,
-                        transform_source=transform_source, fatal=fatal))
+                        transform_source=transform_source, fatal=fatal)
+                    # Sometimes stream-level manifest contains single media entry that
+                    # does not contain any quality metadata (e.g. http://matchtv.ru/#live-player).
+                    # At the same time parent's media entry in set-level manifest may
+                    # contain it. We will copy it from parent in such cases.
+                    if len(f4m_formats) == 1:
+                        f = f4m_formats[0]
+                        f.update({
+                            'tbr': f.get('tbr') or tbr,
+                            'width': f.get('width') or width,
+                            'height': f.get('height') or height,
+                            'format_id': f.get('format_id') if not tbr else format_id,
+                        })
+                    formats.extend(f4m_formats)
+                    continue
+                elif ext == 'm3u8':
+                    formats.extend(self._extract_m3u8_formats(
+                        manifest_url, video_id, 'mp4', preference=preference,
+                        m3u8_id=m3u8_id, fatal=fatal))
                     continue
-            tbr = int_or_none(media_el.attrib.get('bitrate'))
             formats.append({
-                'format_id': '-'.join(filter(None, [f4m_id, compat_str(i if tbr is None else tbr)])),
+                'format_id': format_id,
                 'url': manifest_url,
-                'ext': 'flv',
+                'ext': 'flv' if bootstrap_info else None,
                 'tbr': tbr,
-                'width': int_or_none(media_el.attrib.get('width')),
-                'height': int_or_none(media_el.attrib.get('height')),
+                'width': width,
+                'height': height,
                 'preference': preference,
             })
         return formats

youtube_dl/extractor/coub.py

@@ -0,0 +1,143 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+from .common import InfoExtractor
+from ..utils import (
+    ExtractorError,
+    float_or_none,
+    int_or_none,
+    parse_iso8601,
+    qualities,
+)
+
+
+class CoubIE(InfoExtractor):
+    _VALID_URL = r'(?:coub:|https?://(?:coub\.com/(?:view|embed|coubs)/|c-cdn\.coub\.com/fb-player\.swf\?.*\bcoub(?:ID|id)=))(?P<id>[\da-z]+)'
+
+    _TESTS = [{
+        'url': 'http://coub.com/view/5u5n1',
+        'info_dict': {
+            'id': '5u5n1',
+            'ext': 'mp4',
+            'title': 'The Matrix Moonwalk',
+            'thumbnail': 're:^https?://.*\.jpg$',
+            'duration': 4.6,
+            'timestamp': 1428527772,
+            'upload_date': '20150408',
+            'uploader': 'Артём Лоскутников',
+            'uploader_id': 'artyom.loskutnikov',
+            'view_count': int,
+            'like_count': int,
+            'repost_count': int,
+            'comment_count': int,
+            'age_limit': 0,
+        },
+    }, {
+        'url': 'http://c-cdn.coub.com/fb-player.swf?bot_type=vk&coubID=7w5a4',
+        'only_matching': True,
+    }, {
+        'url': 'coub:5u5n1',
+        'only_matching': True,
+    }, {
+        # longer video id
+        'url': 'http://coub.com/view/237d5l5h',
+        'only_matching': True,
+    }]
+
+    def _real_extract(self, url):
+        video_id = self._match_id(url)
+
+        coub = self._download_json(
+            'http://coub.com/api/v2/coubs/%s.json' % video_id, video_id)
+
+        if coub.get('error'):
+            raise ExtractorError(
+                '%s said: %s' % (self.IE_NAME, coub['error']), expected=True)
+
+        title = coub['title']
+
+        file_versions = coub['file_versions']
+
+        QUALITIES = ('low', 'med', 'high')
+
+        MOBILE = 'mobile'
+        IPHONE = 'iphone'
+        HTML5 = 'html5'
+
+        SOURCE_PREFERENCE = (MOBILE, IPHONE, HTML5)
+
+        quality_key = qualities(QUALITIES)
+        preference_key = qualities(SOURCE_PREFERENCE)
+
+        formats = []
+
+        for kind, items in file_versions.get(HTML5, {}).items():
+            if kind not in ('video', 'audio'):
+                continue
+            if not isinstance(items, dict):
+                continue
+            for quality, item in items.items():
+                if not isinstance(item, dict):
+                    continue
+                item_url = item.get('url')
+                if not item_url:
+                    continue
+                formats.append({
+                    'url': item_url,
+                    'format_id': '%s-%s-%s' % (HTML5, kind, quality),
+                    'filesize': int_or_none(item.get('size')),
+                    'vcodec': 'none' if kind == 'audio' else None,
+                    'quality': quality_key(quality),
+                    'preference': preference_key(HTML5),
+                })
+
+        iphone_url = file_versions.get(IPHONE, {}).get('url')
+        if iphone_url:
+            formats.append({
+                'url': iphone_url,
+                'format_id': IPHONE,
+                'preference': preference_key(IPHONE),
+            })
+
+        mobile_url = file_versions.get(MOBILE, {}).get('audio_url')
+        if mobile_url:
+            formats.append({
+                'url': mobile_url,
+                'format_id': '%s-audio' % MOBILE,
+                'preference': preference_key(MOBILE),
+            })
+
+        self._sort_formats(formats)
+
+        thumbnail = coub.get('picture')
+        duration = float_or_none(coub.get('duration'))
+        timestamp = parse_iso8601(coub.get('published_at') or coub.get('created_at'))
+        uploader = coub.get('channel', {}).get('title')
+        uploader_id = coub.get('channel', {}).get('permalink')
+
+        view_count = int_or_none(coub.get('views_count') or coub.get('views_increase_count'))
+        like_count = int_or_none(coub.get('likes_count'))
+        repost_count = int_or_none(coub.get('recoubs_count'))
+        comment_count = int_or_none(coub.get('comments_count'))
+
+        age_restricted = coub.get('age_restricted', coub.get('age_restricted_by_admin'))
+        if age_restricted is not None:
+            age_limit = 18 if age_restricted is True else 0
+        else:
+            age_limit = None
+
+        return {
+            'id': video_id,
+            'title': title,
+            'thumbnail': thumbnail,
+            'duration': duration,
+            'timestamp': timestamp,
+            'uploader': uploader,
+            'uploader_id': uploader_id,
+            'view_count': view_count,
+            'like_count': like_count,
+            'repost_count': repost_count,
+            'comment_count': comment_count,
+            'age_limit': age_limit,
+            'formats': formats,
+        }

youtube_dl/extractor/dw.py

@@ -2,13 +2,16 @@
 from __future__ import unicode_literals
 
 from .common import InfoExtractor
-from ..utils import int_or_none
+from ..utils import (
+    int_or_none,
+    unified_strdate,
+)
 from ..compat import compat_urlparse
 
 
 class DWIE(InfoExtractor):
     IE_NAME = 'dw'
-    _VALID_URL = r'https?://(?:www\.)?dw\.com/(?:[^/]+/)+av-(?P<id>\d+)'
+    _VALID_URL = r'https?://(?:www\.)?dw\.com/(?:[^/]+/)+(?:av|e)-(?P<id>\d+)'
     _TESTS = [{
         # video
         'url': 'http://www.dw.com/en/intelligent-light/av-19112290',
@@ -31,6 +34,16 @@ class DWIE(InfoExtractor):
             'description': 'md5:bc9ca6e4e063361e21c920c53af12405',
             'upload_date': '20160311',
         }
+    }, {
+        'url': 'http://www.dw.com/en/documentaries-welcome-to-the-90s-2016-05-21/e-19220158-9798',
+        'md5': '56b6214ef463bfb9a3b71aeb886f3cf1',
+        'info_dict': {
+            'id': '19274438',
+            'ext': 'mp4',
+            'title': 'Welcome to the 90s – Hip Hop',
+            'description': 'Welcome to the 90s - The Golden Decade of Hip Hop',
+            'upload_date': '20160521',
+        },
     }]
 
     def _real_extract(self, url):
@@ -38,6 +51,7 @@ class DWIE(InfoExtractor):
         webpage = self._download_webpage(url, media_id)
         hidden_inputs = self._hidden_inputs(webpage)
         title = hidden_inputs['media_title']
+        media_id = hidden_inputs.get('media_id') or media_id
 
         if hidden_inputs.get('player_type') == 'video' and hidden_inputs.get('stream_file') == '1':
             formats = self._extract_smil_formats(
@@ -49,13 +63,20 @@ class DWIE(InfoExtractor):
         else:
             formats = [{'url': hidden_inputs['file_name']}]
 
+        upload_date = hidden_inputs.get('display_date')
+        if not upload_date:
+            upload_date = self._html_search_regex(
+                r'<span[^>]+class="date">([0-9.]+)\s*\|', webpage,
+                'upload date', default=None)
+            upload_date = unified_strdate(upload_date)
+
         return {
             'id': media_id,
             'title': title,
             'description': self._og_search_description(webpage),
             'thumbnail': hidden_inputs.get('preview_image'),
             'duration': int_or_none(hidden_inputs.get('file_duration')),
-            'upload_date': hidden_inputs.get('display_date'),
+            'upload_date': upload_date,
             'formats': formats,
         }
 

youtube_dl/extractor/eporner.py

@@ -11,8 +11,8 @@ from ..utils import (
 
 
 class EpornerIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:www\.)?eporner\.com/hd-porn/(?P<id>\d+)/(?P<display_id>[\w-]+)'
-    _TEST = {
+    _VALID_URL = r'https?://(?:www\.)?eporner\.com/hd-porn/(?P<id>\w+)/(?P<display_id>[\w-]+)'
+    _TESTS = [{
         'url': 'http://www.eporner.com/hd-porn/95008/Infamous-Tiffany-Teen-Strip-Tease-Video/',
         'md5': '39d486f046212d8e1b911c52ab4691f8',
         'info_dict': {
@@ -23,8 +23,12 @@ class EpornerIE(InfoExtractor):
             'duration': 1838,
             'view_count': int,
             'age_limit': 18,
-        }
-    }
+        },
+    }, {
+        # New (May 2016) URL layout
+        'url': 'http://www.eporner.com/hd-porn/3YRUtzMcWn0/Star-Wars-XXX-Parody/',
+        'only_matching': True,
+    }]
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)

youtube_dl/extractor/espn.py

@@ -8,6 +8,7 @@ class ESPNIE(InfoExtractor):
     _VALID_URL = r'https?://espn\.go\.com/(?:[^/]+/)*(?P<id>[^/]+)'
     _TESTS = [{
         'url': 'http://espn.go.com/video/clip?id=10365079',
+        'md5': '60e5d097a523e767d06479335d1bdc58',
         'info_dict': {
             'id': 'FkYWtmazr6Ed8xmvILvKLWjd4QvYZpzG',
             'ext': 'mp4',
@@ -15,21 +16,22 @@ class ESPNIE(InfoExtractor):
             'description': None,
         },
         'params': {
-            # m3u8 download
             'skip_download': True,
         },
+        'add_ie': ['OoyalaExternal'],
     }, {
         # intl video, from http://www.espnfc.us/video/mls-highlights/150/video/2743663/must-see-moments-best-of-the-mls-season
         'url': 'http://espn.go.com/video/clip?id=2743663',
+        'md5': 'f4ac89b59afc7e2d7dbb049523df6768',
         'info_dict': {
             'id': '50NDFkeTqRHB0nXBOK-RGdSG5YQPuxHg',
             'ext': 'mp4',
             'title': 'Must-See Moments: Best of the MLS season',
         },
         'params': {
-            # m3u8 download
             'skip_download': True,
         },
+        'add_ie': ['OoyalaExternal'],
     }, {
         'url': 'https://espn.go.com/video/iframe/twitter/?cms=espn&id=10365079',
         'only_matching': True,

youtube_dl/extractor/extractors.py

@@ -56,6 +56,7 @@ from .arte import (
     ArteTVDDCIE,
     ArteTVMagazineIE,
     ArteTVEmbedIE,
+    ArteTVPlaylistIE,
 )
 from .atresplayer import AtresPlayerIE
 from .atttechchannel import ATTTechChannelIE
@@ -143,6 +144,7 @@ from .cnn import (
     CNNBlogsIE,
     CNNArticleIE,
 )
+from .coub import CoubIE
 from .collegerama import CollegeRamaIE
 from .comedycentral import ComedyCentralIE, ComedyCentralShowsIE
 from .comcarcoff import ComCarCoffIE
@@ -231,6 +233,7 @@ from .everyonesmixtape import EveryonesMixtapeIE
 from .exfm import ExfmIE
 from .expotv import ExpoTVIE
 from .extremetube import ExtremeTubeIE
+from .eyedotv import EyedoTVIE
 from .facebook import FacebookIE
 from .faz import FazIE
 from .fc2 import FC2IE
@@ -379,6 +382,7 @@ from .leeco import (
     LePlaylistIE,
     LetvCloudIE,
 )
+from .libraryofcongress import LibraryOfCongressIE
 from .libsyn import LibsynIE
 from .lifenews import (
     LifeNewsIE,
@@ -617,6 +621,10 @@ from .qqmusic import (
     QQMusicPlaylistIE,
 )
 from .r7 import R7IE
+from .radiocanada import (
+    RadioCanadaIE,
+    RadioCanadaAudioVideoIE,
+)
 from .radiode import RadioDeIE
 from .radiojavan import RadioJavanIE
 from .radiobremen import RadioBremenIE
@@ -630,8 +638,12 @@ from .rds import RDSIE
 from .redtube import RedTubeIE
 from .regiotv import RegioTVIE
 from .restudy import RestudyIE
+from .reuters import ReutersIE
 from .reverbnation import ReverbNationIE
-from .revision3 import Revision3IE
+from .revision3 import (
+    Revision3EmbedIE,
+    Revision3IE,
+)
 from .rice import RICEIE
 from .ringtv import RingTVIE
 from .ro220 import Ro220IE
@@ -670,6 +682,7 @@ from .screencast import ScreencastIE
 from .screencastomatic import ScreencastOMaticIE
 from .screenjunkies import ScreenJunkiesIE
 from .screenwavemedia import ScreenwaveMediaIE, TeamFourIE
+from .seeker import SeekerIE
 from .senateisvp import SenateISVPIE
 from .sendtonews import SendtoNewsIE
 from .servingsys import ServingSysIE
@@ -827,7 +840,10 @@ from .tvc import (
 )
 from .tvigle import TvigleIE
 from .tvland import TVLandIE
-from .tvp import TvpIE, TvpSeriesIE
+from .tvp import (
+    TVPIE,
+    TVPSeriesIE,
+)
 from .tvplay import TVPlayIE
 from .tweakers import TweakersIE
 from .twentyfourvideo import TwentyFourVideoIE
@@ -941,7 +957,10 @@ from .vube import VubeIE
 from .vuclip import VuClipIE
 from .vulture import VultureIE
 from .walla import WallaIE
-from .washingtonpost import WashingtonPostIE
+from .washingtonpost import (
+    WashingtonPostIE,
+    WashingtonPostArticleIE,
+)
 from .wat import WatIE
 from .watchindianporn import WatchIndianPornIE
 from .wdr import (

youtube_dl/extractor/eyedotv.py

@@ -0,0 +1,64 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+from .common import InfoExtractor
+from ..utils import (
+    xpath_text,
+    parse_duration,
+    ExtractorError,
+)
+
+
+class EyedoTVIE(InfoExtractor):
+    _VALID_URL = r'https?://(?:www\.)?eyedo\.tv/[^/]+/(?:#!/)?Live/Detail/(?P<id>[0-9]+)'
+    _TEST = {
+        'url': 'https://www.eyedo.tv/en-US/#!/Live/Detail/16301',
+        'md5': 'ba14f17995cdfc20c36ba40e21bf73f7',
+        'info_dict': {
+            'id': '16301',
+            'ext': 'mp4',
+            'title': 'Journée du conseil scientifique de l\'Afnic 2015',
+            'description': 'md5:4abe07293b2f73efc6e1c37028d58c98',
+            'uploader': 'Afnic Live',
+            'uploader_id': '8023',
+        }
+    }
+    _ROOT_URL = 'http://live.eyedo.net:1935/'
+
+    def _real_extract(self, url):
+        video_id = self._match_id(url)
+        video_data = self._download_xml('http://eyedo.tv/api/live/GetLive/%s' % video_id, video_id)
+
+        def _add_ns(path):
+            return self._xpath_ns(path, 'http://schemas.datacontract.org/2004/07/EyeDo.Core.Implementation.Web.ViewModels.Api')
+
+        title = xpath_text(video_data, _add_ns('Titre'), 'title', True)
+        state_live_code = xpath_text(video_data, _add_ns('StateLiveCode'), 'title', True)
+        if state_live_code == 'avenir':
+            raise ExtractorError(
+                '%s said: We\'re sorry, but this video is not yet available.' % self.IE_NAME,
+                expected=True)
+
+        is_live = state_live_code == 'live'
+        m3u8_url = None
+        # http://eyedo.tv/Content/Html5/Scripts/html5view.js
+        if is_live:
+            if xpath_text(video_data, 'Cdn') == 'true':
+                m3u8_url = 'http://rrr.sz.xlcdn.com/?account=eyedo&file=A%s&type=live&service=wowza&protocol=http&output=playlist.m3u8' % video_id
+            else:
+                m3u8_url = self._ROOT_URL + 'w/%s/eyedo_720p/playlist.m3u8' % video_id
+        else:
+            m3u8_url = self._ROOT_URL + 'replay-w/%s/mp4:%s.mp4/playlist.m3u8' % (video_id, video_id)
+
+        return {
+            'id': video_id,
+            'title': title,
+            'formats': self._extract_m3u8_formats(
+                m3u8_url, video_id, 'mp4', 'm3u8' if is_live else 'm3u8_native'),
+            'description': xpath_text(video_data, _add_ns('Description')),
+            'duration': parse_duration(xpath_text(video_data, _add_ns('Duration'))),
+            'uploader': xpath_text(video_data, _add_ns('Createur')),
+            'uploader_id': xpath_text(video_data, _add_ns('CreateurId')),
+            'chapter': xpath_text(video_data, _add_ns('ChapitreTitre')),
+            'chapter_id': xpath_text(video_data, _add_ns('ChapitreId')),
+        }

youtube_dl/extractor/formula1.py

@@ -13,7 +13,8 @@ class Formula1IE(InfoExtractor):
             'id': 'JvYXJpMzE6pArfHWm5ARp5AiUmD-gibV',
             'ext': 'flv',
             'title': 'Race highlights - Spain 2016',
-        }
+        },
+        'add_ie': ['Ooyala'],
     }
 
     def _real_extract(self, url):

youtube_dl/extractor/generic.py

@@ -62,6 +62,7 @@ from .digiteka import DigitekaIE
 from .instagram import InstagramIE
 from .liveleak import LiveLeakIE
 from .threeqsdn import ThreeQSDNIE
+from .theplatform import ThePlatformIE
 
 
 class GenericIE(InfoExtractor):
@@ -783,6 +784,19 @@ class GenericIE(InfoExtractor):
                 'title': 'Rosetta #CometLanding webcast HL 10',
             }
         },
+        # Another Livestream embed, without 'new.' in URL
+        {
+            'url': 'https://www.freespeech.org/',
+            'info_dict': {
+                'id': '123537347',
+                'ext': 'mp4',
+                'title': 're:^FSTV [0-9]{4}-[0-9]{2}-[0-9]{2} [0-9]{2}:[0-9]{2}$',
+            },
+            'params': {
+                # Live stream
+                'skip_download': True,
+            },
+        },
         # LazyYT
         {
             'url': 'http://discourse.ubuntu.com/t/unity-8-desktop-mode-windows-on-mir/1986',
@@ -867,18 +881,6 @@ class GenericIE(InfoExtractor):
                 'title': 'EP3S5 - Bon Appétit - Baqueira Mi Corazon !',
             }
         },
-        # Kaltura embed
-        {
-            'url': 'http://www.monumentalnetwork.com/videos/john-carlson-postgame-2-25-15',
-            'info_dict': {
-                'id': '1_eergr3h1',
-                'ext': 'mp4',
-                'upload_date': '20150226',
-                'uploader_id': 'MonumentalSports-Kaltura@perfectsensedigital.com',
-                'timestamp': int,
-                'title': 'John Carlson Postgame 2/25/15',
-            },
-        },
         # Kaltura embed (different embed code)
         {
             'url': 'http://www.premierchristianradio.com/Shows/Saturday/Unbelievable/Conference-Videos/Os-Guinness-Is-It-Fools-Talk-Unbelievable-Conference-2014',
@@ -904,6 +906,19 @@ class GenericIE(InfoExtractor):
                 'uploader_id': 'echojecka',
             },
         },
+        # Kaltura embed with single quotes
+        {
+            'url': 'http://fod.infobase.com/p_ViewPlaylist.aspx?AssignmentID=NUN8ZY',
+            'info_dict': {
+                'id': '0_izeg5utt',
+                'ext': 'mp4',
+                'title': '35871',
+                'timestamp': 1355743100,
+                'upload_date': '20121217',
+                'uploader_id': 'batchUser',
+            },
+            'add_ie': ['Kaltura'],
+        },
         # Eagle.Platform embed (generic URL)
         {
             'url': 'http://lenta.ru/news/2015/03/06/navalny/',
@@ -1018,14 +1033,18 @@ class GenericIE(InfoExtractor):
         },
         # UDN embed
         {
-            'url': 'http://www.udn.com/news/story/7314/822787',
+            'url': 'https://video.udn.com/news/300346',
             'md5': 'fd2060e988c326991037b9aff9df21a6',
             'info_dict': {
                 'id': '300346',
                 'ext': 'mp4',
                 'title': '中一中男師變性 全校師生力挺',
                 'thumbnail': 're:^https?://.*\.jpg$',
-            }
+            },
+            'params': {
+                # m3u8 download
+                'skip_download': True,
+            },
         },
         # Ooyala embed
         {
@@ -1193,6 +1212,16 @@ class GenericIE(InfoExtractor):
                 'uploader': 'Lake8737',
             }
         },
+        # Duplicated embedded video URLs
+        {
+            'url': 'http://www.hudl.com/athlete/2538180/highlights/149298443',
+            'info_dict': {
+                'id': '149298443_480_16c25b74_2',
+                'ext': 'mp4',
+                'title': 'vs. Blue Orange Spring Game',
+                'uploader': 'www.hudl.com',
+            },
+        },
     ]
 
     def report_following_redirect(self, new_url):
@@ -1499,6 +1528,11 @@ class GenericIE(InfoExtractor):
         if bc_urls:
             return _playlist_from_matches(bc_urls, ie='BrightcoveNew')
 
+        # Look for ThePlatform embeds
+        tp_urls = ThePlatformIE._extract_urls(webpage)
+        if tp_urls:
+            return _playlist_from_matches(tp_urls, ie='ThePlatform')
+
         # Look for embedded rtl.nl player
         matches = re.findall(
             r'<iframe[^>]+?src="((?:https?:)?//(?:www\.)?rtl\.nl/system/videoplayer/[^"]+(?:video_)?embed[^"]+)"',
@@ -1862,7 +1896,7 @@ class GenericIE(InfoExtractor):
             return self.url_result(self._proto_relative_url(mobj.group('url'), scheme='http:'), 'CondeNast')
 
         mobj = re.search(
-            r'<iframe[^>]+src="(?P<url>https?://new\.livestream\.com/[^"]+/player[^"]+)"',
+            r'<iframe[^>]+src="(?P<url>https?://(?:new\.)?livestream\.com/[^"]+/player[^"]+)"',
             webpage)
         if mobj is not None:
             return self.url_result(mobj.group('url'), 'Livestream')
@@ -1874,7 +1908,7 @@ class GenericIE(InfoExtractor):
             return self.url_result(mobj.group('url'), 'Zapiks')
 
         # Look for Kaltura embeds
-        mobj = (re.search(r"(?s)kWidget\.(?:thumb)?[Ee]mbed\(\{.*?'wid'\s*:\s*'_?(?P<partner_id>[^']+)',.*?'entry_?[Ii]d'\s*:\s*'(?P<id>[^']+)',", webpage) or
+        mobj = (re.search(r"(?s)kWidget\.(?:thumb)?[Ee]mbed\(\{.*?(?P<q1>['\"])wid(?P=q1)\s*:\s*(?P<q2>['\"])_?(?P<partner_id>[^'\"]+)(?P=q2),.*?(?P<q3>['\"])entry_?[Ii]d(?P=q3)\s*:\s*(?P<q4>['\"])(?P<id>[^'\"]+)(?P=q4),", webpage) or
                 re.search(r'(?s)(?P<q1>["\'])(?:https?:)?//cdnapi(?:sec)?\.kaltura\.com/.*?(?:p|partner_id)/(?P<partner_id>\d+).*?(?P=q1).*?entry_?[Ii]d\s*:\s*(?P<q2>["\'])(?P<id>.+?)(?P=q2)', webpage))
         if mobj is not None:
             return self.url_result(smuggle_url(
@@ -2105,7 +2139,7 @@ class GenericIE(InfoExtractor):
             raise UnsupportedError(url)
 
         entries = []
-        for video_url in found:
+        for video_url in orderedSet(found):
             video_url = unescapeHTML(video_url)
             video_url = video_url.replace('\\/', '/')
             video_url = compat_urlparse.urljoin(url, video_url)

youtube_dl/extractor/groupon.py

@@ -14,6 +14,7 @@ class GrouponIE(InfoExtractor):
             'description': 'Studio kept at 105 degrees and 40% humidity with anti-microbial and anti-slip Flotex flooring; certified instructors',
         },
         'playlist': [{
+            'md5': '42428ce8a00585f9bc36e49226eae7a1',
             'info_dict': {
                 'id': 'fk6OhWpXgIQ',
                 'ext': 'mp4',
@@ -24,10 +25,11 @@ class GrouponIE(InfoExtractor):
                 'uploader_id': 'groupon',
                 'uploader': 'Groupon',
             },
+            'add_ie': ['Youtube'],
         }],
         'params': {
             'skip_download': True,
-        }
+        },
     }
 
     _PROVIDERS = {

youtube_dl/extractor/howcast.py

@@ -8,7 +8,7 @@ class HowcastIE(InfoExtractor):
     _VALID_URL = r'https?://(?:www\.)?howcast\.com/videos/(?P<id>\d+)'
     _TEST = {
         'url': 'http://www.howcast.com/videos/390161-How-to-Tie-a-Square-Knot-Properly',
-        'md5': '8b743df908c42f60cf6496586c7f12c3',
+        'md5': '7d45932269a288149483144f01b99789',
         'info_dict': {
             'id': '390161',
             'ext': 'mp4',
@@ -19,9 +19,9 @@ class HowcastIE(InfoExtractor):
             'duration': 56.823,
         },
         'params': {
-            # m3u8 download
             'skip_download': True,
         },
+        'add_ie': ['Ooyala'],
     }
 
     def _real_extract(self, url):

youtube_dl/extractor/libraryofcongress.py

@@ -0,0 +1,143 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+import re
+
+from .common import InfoExtractor
+
+from ..utils import (
+    determine_ext,
+    float_or_none,
+    int_or_none,
+    parse_filesize,
+)
+
+
+class LibraryOfCongressIE(InfoExtractor):
+    IE_NAME = 'loc'
+    IE_DESC = 'Library of Congress'
+    _VALID_URL = r'https?://(?:www\.)?loc\.gov/(?:item/|today/cyberlc/feature_wdesc\.php\?.*\brec=)(?P<id>[0-9]+)'
+    _TESTS = [{
+        # embedded via <div class="media-player"
+        'url': 'http://loc.gov/item/90716351/',
+        'md5': '353917ff7f0255aa6d4b80a034833de8',
+        'info_dict': {
+            'id': '90716351',
+            'ext': 'mp4',
+            'title': "Pa's trip to Mars",
+            'thumbnail': 're:^https?://.*\.jpg$',
+            'duration': 0,
+            'view_count': int,
+        },
+    }, {
+        # webcast embedded via mediaObjectId
+        'url': 'https://www.loc.gov/today/cyberlc/feature_wdesc.php?rec=5578',
+        'info_dict': {
+            'id': '5578',
+            'ext': 'mp4',
+            'title': 'Help! Preservation Training Needs Here, There & Everywhere',
+            'duration': 3765,
+            'view_count': int,
+            'subtitles': 'mincount:1',
+        },
+        'params': {
+            'skip_download': True,
+        },
+    }, {
+        # with direct download links
+        'url': 'https://www.loc.gov/item/78710669/',
+        'info_dict': {
+            'id': '78710669',
+            'ext': 'mp4',
+            'title': 'La vie et la passion de Jesus-Christ',
+            'duration': 0,
+            'view_count': int,
+            'formats': 'mincount:4',
+        },
+        'params': {
+            'skip_download': True,
+        },
+    }]
+
+    def _real_extract(self, url):
+        video_id = self._match_id(url)
+        webpage = self._download_webpage(url, video_id)
+
+        media_id = self._search_regex(
+            (r'id=(["\'])media-player-(?P<id>.+?)\1',
+             r'<video[^>]+id=(["\'])uuid-(?P<id>.+?)\1',
+             r'<video[^>]+data-uuid=(["\'])(?P<id>.+?)\1',
+             r'mediaObjectId\s*:\s*(["\'])(?P<id>.+?)\1'),
+            webpage, 'media id', group='id')
+
+        data = self._download_json(
+            'https://media.loc.gov/services/v1/media?id=%s&context=json' % media_id,
+            video_id)['mediaObject']
+
+        derivative = data['derivatives'][0]
+        media_url = derivative['derivativeUrl']
+
+        title = derivative.get('shortName') or data.get('shortName') or self._og_search_title(
+            webpage)
+
+        # Following algorithm was extracted from setAVSource js function
+        # found in webpage
+        media_url = media_url.replace('rtmp', 'https')
+
+        is_video = data.get('mediaType', 'v').lower() == 'v'
+        ext = determine_ext(media_url)
+        if ext not in ('mp4', 'mp3'):
+            media_url += '.mp4' if is_video else '.mp3'
+
+        if 'vod/mp4:' in media_url:
+            formats = [{
+                'url': media_url.replace('vod/mp4:', 'hls-vod/media/') + '.m3u8',
+                'format_id': 'hls',
+                'ext': 'mp4',
+                'protocol': 'm3u8_native',
+                'quality': 1,
+            }]
+        elif 'vod/mp3:' in media_url:
+            formats = [{
+                'url': media_url.replace('vod/mp3:', ''),
+                'vcodec': 'none',
+            }]
+
+        download_urls = set()
+        for m in re.finditer(
+                r'<option[^>]+value=(["\'])(?P<url>.+?)\1[^>]+data-file-download=[^>]+>\s*(?P<id>.+?)(?:(?:&nbsp;|\s+)\((?P<size>.+?)\))?\s*<', webpage):
+            format_id = m.group('id').lower()
+            if format_id == 'gif':
+                continue
+            download_url = m.group('url')
+            if download_url in download_urls:
+                continue
+            download_urls.add(download_url)
+            formats.append({
+                'url': download_url,
+                'format_id': format_id,
+                'filesize_approx': parse_filesize(m.group('size')),
+            })
+
+        self._sort_formats(formats)
+
+        duration = float_or_none(data.get('duration'))
+        view_count = int_or_none(data.get('viewCount'))
+
+        subtitles = {}
+        cc_url = data.get('ccUrl')
+        if cc_url:
+            subtitles.setdefault('en', []).append({
+                'url': cc_url,
+                'ext': 'ttml',
+            })
+
+        return {
+            'id': video_id,
+            'title': title,
+            'thumbnail': self._og_search_thumbnail(webpage, default=None),
+            'duration': duration,
+            'view_count': view_count,
+            'formats': formats,
+            'subtitles': subtitles,
+        }

youtube_dl/extractor/lifenews.py

@@ -7,48 +7,53 @@ from .common import InfoExtractor
 from ..compat import compat_urlparse
 from ..utils import (
     determine_ext,
-    int_or_none,
-    remove_end,
-    unified_strdate,
     ExtractorError,
+    int_or_none,
+    parse_iso8601,
+    remove_end,
 )
 
 
 class LifeNewsIE(InfoExtractor):
-    IE_NAME = 'lifenews'
-    IE_DESC = 'LIFE | NEWS'
-    _VALID_URL = r'https?://lifenews\.ru/(?:mobile/)?(?P<section>news|video)/(?P<id>\d+)'
+    IE_NAME = 'life'
+    IE_DESC = 'Life.ru'
+    _VALID_URL = r'https?://life\.ru/t/[^/]+/(?P<id>\d+)'
 
     _TESTS = [{
         # single video embedded via video/source
-        'url': 'http://lifenews.ru/news/98736',
+        'url': 'https://life.ru/t/новости/98736',
         'md5': '77c95eaefaca216e32a76a343ad89d23',
         'info_dict': {
             'id': '98736',
             'ext': 'mp4',
             'title': 'Мужчина нашел дома архив оборонного завода',
             'description': 'md5:3b06b1b39b5e2bea548e403d99b8bf26',
+            'timestamp': 1344154740,
             'upload_date': '20120805',
+            'view_count': int,
         }
     }, {
         # single video embedded via iframe
-        'url': 'http://lifenews.ru/news/152125',
+        'url': 'https://life.ru/t/новости/152125',
         'md5': '77d19a6f0886cd76bdbf44b4d971a273',
         'info_dict': {
             'id': '152125',
             'ext': 'mp4',
             'title': 'В Сети появилось видео захвата «Правым сектором» колхозных полей ',
             'description': 'Жители двух поселков Днепропетровской области не простили радикалам угрозу лишения плодородных земель и пошли в лобовую. ',
+            'timestamp': 1427961840,
             'upload_date': '20150402',
+            'view_count': int,
         }
     }, {
         # two videos embedded via iframe
-        'url': 'http://lifenews.ru/news/153461',
+        'url': 'https://life.ru/t/новости/153461',
         'info_dict': {
             'id': '153461',
             'title': 'В Москве спасли потерявшегося медвежонка, который спрятался на дереве',
             'description': 'Маленький хищник не смог найти дорогу домой и обрел временное убежище на тополе недалеко от жилого массива, пока его не нашла соседская собака.',
-            'upload_date': '20150505',
+            'timestamp': 1430825520,
+            'view_count': int,
         },
         'playlist': [{
             'md5': '9b6ef8bc0ffa25aebc8bdb40d89ab795',
@@ -57,6 +62,7 @@ class LifeNewsIE(InfoExtractor):
                 'ext': 'mp4',
                 'title': 'В Москве спасли потерявшегося медвежонка, который спрятался на дереве (Видео 1)',
                 'description': 'Маленький хищник не смог найти дорогу домой и обрел временное убежище на тополе недалеко от жилого массива, пока его не нашла соседская собака.',
+                'timestamp': 1430825520,
                 'upload_date': '20150505',
             },
         }, {
@@ -66,22 +72,25 @@ class LifeNewsIE(InfoExtractor):
                 'ext': 'mp4',
                 'title': 'В Москве спасли потерявшегося медвежонка, который спрятался на дереве (Видео 2)',
                 'description': 'Маленький хищник не смог найти дорогу домой и обрел временное убежище на тополе недалеко от жилого массива, пока его не нашла соседская собака.',
+                'timestamp': 1430825520,
                 'upload_date': '20150505',
             },
         }],
     }, {
-        'url': 'http://lifenews.ru/video/13035',
+        'url': 'https://life.ru/t/новости/213035',
+        'only_matching': True,
+    }, {
+        'url': 'https://life.ru/t/%D0%BD%D0%BE%D0%B2%D0%BE%D1%81%D1%82%D0%B8/153461',
+        'only_matching': True,
+    }, {
+        'url': 'https://life.ru/t/новости/411489/manuel_vals_nazval_frantsiiu_tsieliu_nomier_odin_dlia_ighil',
         'only_matching': True,
     }]
 
     def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
-        video_id = mobj.group('id')
-        section = mobj.group('section')
+        video_id = self._match_id(url)
 
-        webpage = self._download_webpage(
-            'http://lifenews.ru/%s/%s' % (section, video_id),
-            video_id, 'Downloading page')
+        webpage = self._download_webpage(url, video_id)
 
         video_urls = re.findall(
             r'<video[^>]+><source[^>]+src=["\'](.+?)["\']', webpage)
@@ -95,26 +104,22 @@ class LifeNewsIE(InfoExtractor):
 
         title = remove_end(
             self._og_search_title(webpage),
-            ' - Первый по срочным новостям — LIFE | NEWS')
+            ' - Life.ru')
 
         description = self._og_search_description(webpage)
 
         view_count = self._html_search_regex(
-            r'<div class=\'views\'>\s*(\d+)\s*</div>', webpage, 'view count', fatal=False)
-        comment_count = self._html_search_regex(
-            r'=\'commentCount\'[^>]*>\s*(\d+)\s*<',
-            webpage, 'comment count', fatal=False)
+            r'<div[^>]+class=(["\']).*?\bhits-count\b.*?\1[^>]*>\s*(?P<value>\d+)\s*</div>',
+            webpage, 'view count', fatal=False, group='value')
 
-        upload_date = self._html_search_regex(
-            r'<time[^>]*datetime=\'([^\']+)\'', webpage, 'upload date', fatal=False)
-        if upload_date is not None:
-            upload_date = unified_strdate(upload_date)
+        timestamp = parse_iso8601(self._search_regex(
+            r'<time[^>]+datetime=(["\'])(?P<value>.+?)\1',
+            webpage, 'upload date', fatal=False, group='value'))
 
         common_info = {
             'description': description,
             'view_count': int_or_none(view_count),
-            'comment_count': int_or_none(comment_count),
-            'upload_date': upload_date,
+            'timestamp': timestamp,
         }
 
         def make_entry(video_id, video_url, index=None):
@@ -183,7 +188,8 @@ class LifeEmbedIE(InfoExtractor):
             ext = determine_ext(video_url)
             if ext == 'm3u8':
                 formats.extend(self._extract_m3u8_formats(
-                    video_url, video_id, 'mp4', m3u8_id='m3u8'))
+                    video_url, video_id, 'mp4',
+                    entry_protocol='m3u8_native', m3u8_id='m3u8'))
             else:
                 formats.append({
                     'url': video_url,

youtube_dl/extractor/livestream.py

@@ -150,7 +150,7 @@ class LivestreamIE(InfoExtractor):
         }
 
     def _extract_stream_info(self, stream_info):
-        broadcast_id = stream_info['broadcast_id']
+        broadcast_id = compat_str(stream_info['broadcast_id'])
         is_live = stream_info.get('is_live')
 
         formats = []

youtube_dl/extractor/ooyala.py

@@ -8,6 +8,7 @@ from ..utils import (
     float_or_none,
     ExtractorError,
     unsmuggle_url,
+    determine_ext,
 )
 from ..compat import compat_urllib_parse_urlencode
 
@@ -15,71 +16,80 @@ from ..compat import compat_urllib_parse_urlencode
 class OoyalaBaseIE(InfoExtractor):
     _PLAYER_BASE = 'http://player.ooyala.com/'
     _CONTENT_TREE_BASE = _PLAYER_BASE + 'player_api/v1/content_tree/'
-    _AUTHORIZATION_URL_TEMPLATE = _PLAYER_BASE + 'sas/player_api/v1/authorization/embed_code/%s/%s?'
+    _AUTHORIZATION_URL_TEMPLATE = _PLAYER_BASE + 'sas/player_api/v2/authorization/embed_code/%s/%s?'
 
     def _extract(self, content_tree_url, video_id, domain='example.org'):
         content_tree = self._download_json(content_tree_url, video_id)['content_tree']
         metadata = content_tree[list(content_tree)[0]]
         embed_code = metadata['embed_code']
         pcode = metadata.get('asset_pcode') or embed_code
-        video_info = {
-            'id': embed_code,
-            'title': metadata['title'],
-            'description': metadata.get('description'),
-            'thumbnail': metadata.get('thumbnail_image') or metadata.get('promo_image'),
-            'duration': float_or_none(metadata.get('duration'), 1000),
-        }
+        title = metadata['title']
+
+        auth_data = self._download_json(
+            self._AUTHORIZATION_URL_TEMPLATE % (pcode, embed_code) +
+            compat_urllib_parse_urlencode({
+                'domain': domain,
+                'supportedFormats': 'mp4,rtmp,m3u8,hds',
+            }), video_id)
+
+        cur_auth_data = auth_data['authorization_data'][embed_code]
 
         urls = []
         formats = []
-        for supported_format in ('mp4', 'm3u8', 'hds', 'rtmp'):
-            auth_data = self._download_json(
-                self._AUTHORIZATION_URL_TEMPLATE % (pcode, embed_code) +
-                compat_urllib_parse_urlencode({
-                    'domain': domain,
-                    'supportedFormats': supported_format
-                }),
-                video_id, 'Downloading %s JSON' % supported_format)
-
-            cur_auth_data = auth_data['authorization_data'][embed_code]
-
-            if cur_auth_data['authorized']:
-                for stream in cur_auth_data['streams']:
-                    url = base64.b64decode(
-                        stream['url']['data'].encode('ascii')).decode('utf-8')
-                    if url in urls:
-                        continue
-                    urls.append(url)
-                    delivery_type = stream['delivery_type']
-                    if delivery_type == 'hls' or '.m3u8' in url:
-                        formats.extend(self._extract_m3u8_formats(
-                            url, embed_code, 'mp4', 'm3u8_native',
-                            m3u8_id='hls', fatal=False))
-                    elif delivery_type == 'hds' or '.f4m' in url:
-                        formats.extend(self._extract_f4m_formats(
-                            url + '?hdcore=3.7.0', embed_code, f4m_id='hds', fatal=False))
-                    elif '.smil' in url:
-                        formats.extend(self._extract_smil_formats(
-                            url, embed_code, fatal=False))
-                    else:
-                        formats.append({
-                            'url': url,
-                            'ext': stream.get('delivery_type'),
-                            'vcodec': stream.get('video_codec'),
-                            'format_id': delivery_type,
-                            'width': int_or_none(stream.get('width')),
-                            'height': int_or_none(stream.get('height')),
-                            'abr': int_or_none(stream.get('audio_bitrate')),
-                            'vbr': int_or_none(stream.get('video_bitrate')),
-                            'fps': float_or_none(stream.get('framerate')),
-                        })
-            else:
-                raise ExtractorError('%s said: %s' % (
-                    self.IE_NAME, cur_auth_data['message']), expected=True)
+        if cur_auth_data['authorized']:
+            for stream in cur_auth_data['streams']:
+                s_url = base64.b64decode(
+                    stream['url']['data'].encode('ascii')).decode('utf-8')
+                if s_url in urls:
+                    continue
+                urls.append(s_url)
+                ext = determine_ext(s_url, None)
+                delivery_type = stream['delivery_type']
+                if delivery_type == 'hls' or ext == 'm3u8':
+                    formats.extend(self._extract_m3u8_formats(
+                        s_url, embed_code, 'mp4', 'm3u8_native',
+                        m3u8_id='hls', fatal=False))
+                elif delivery_type == 'hds' or ext == 'f4m':
+                    formats.extend(self._extract_f4m_formats(
+                        s_url + '?hdcore=3.7.0', embed_code, f4m_id='hds', fatal=False))
+                elif ext == 'smil':
+                    formats.extend(self._extract_smil_formats(
+                        s_url, embed_code, fatal=False))
+                else:
+                    formats.append({
+                        'url': s_url,
+                        'ext': ext or stream.get('delivery_type'),
+                        'vcodec': stream.get('video_codec'),
+                        'format_id': delivery_type,
+                        'width': int_or_none(stream.get('width')),
+                        'height': int_or_none(stream.get('height')),
+                        'abr': int_or_none(stream.get('audio_bitrate')),
+                        'vbr': int_or_none(stream.get('video_bitrate')),
+                        'fps': float_or_none(stream.get('framerate')),
+                    })
+        else:
+            raise ExtractorError('%s said: %s' % (
+                self.IE_NAME, cur_auth_data['message']), expected=True)
         self._sort_formats(formats)
 
-        video_info['formats'] = formats
-        return video_info
+        subtitles = {}
+        for lang, sub in metadata.get('closed_captions_vtt', {}).get('captions', {}).items():
+            sub_url = sub.get('url')
+            if not sub_url:
+                continue
+            subtitles[lang] = [{
+                'url': sub_url,
+            }]
+
+        return {
+            'id': embed_code,
+            'title': title,
+            'description': metadata.get('description'),
+            'thumbnail': metadata.get('thumbnail_image') or metadata.get('promo_image'),
+            'duration': float_or_none(metadata.get('duration'), 1000),
+            'subtitles': subtitles,
+            'formats': formats,
+        }
 
 
 class OoyalaIE(OoyalaBaseIE):

youtube_dl/extractor/periscope.py

@@ -2,7 +2,10 @@
 from __future__ import unicode_literals
 
 from .common import InfoExtractor
-from ..utils import parse_iso8601
+from ..utils import (
+    parse_iso8601,
+    unescapeHTML,
+)
 
 
 class PeriscopeIE(InfoExtractor):
@@ -42,8 +45,11 @@ class PeriscopeIE(InfoExtractor):
         broadcast = broadcast_data['broadcast']
         status = broadcast['status']
 
-        uploader = broadcast.get('user_display_name') or broadcast_data.get('user', {}).get('display_name')
-        uploader_id = broadcast.get('user_id') or broadcast_data.get('user', {}).get('id')
+        user = broadcast_data.get('user', {})
+
+        uploader = broadcast.get('user_display_name') or user.get('display_name')
+        uploader_id = (broadcast.get('username') or user.get('username') or
+                       broadcast.get('user_id') or user.get('id'))
 
         title = '%s - %s' % (uploader, status) if uploader else status
         state = broadcast.get('state').lower()
@@ -92,6 +98,7 @@ class PeriscopeUserIE(InfoExtractor):
         'info_dict': {
             'id': 'LularoeHusbandMike',
             'title': 'LULAROE HUSBAND MIKE',
+            'description': 'md5:6cf4ec8047768098da58e446e82c82f0',
         },
         # Periscope only shows videos in the last 24 hours, so it's possible to
         # get 0 videos
@@ -103,16 +110,19 @@ class PeriscopeUserIE(InfoExtractor):
 
         webpage = self._download_webpage(url, user_id)
 
-        broadcast_data = self._parse_json(self._html_search_meta(
-            'broadcast-data', webpage, default='{}'), user_id)
-        username = broadcast_data.get('user', {}).get('display_name')
-        user_broadcasts = self._parse_json(
-            self._html_search_meta('user-broadcasts', webpage, default='{}'),
+        data_store = self._parse_json(
+            unescapeHTML(self._search_regex(
+                r'data-store=(["\'])(?P<data>.+?)\1',
+                webpage, 'data store', default='{}', group='data')),
             user_id)
 
+        user = data_store.get('User', {}).get('user', {})
+        title = user.get('display_name') or user.get('username')
+        description = user.get('description')
+
         entries = [
             self.url_result(
                 'https://www.periscope.tv/%s/%s' % (user_id, broadcast['id']))
-            for broadcast in user_broadcasts.get('broadcasts', [])]
+            for broadcast in data_store.get('UserBroadcastHistory', {}).get('broadcasts', [])]
 
-        return self.playlist_result(entries, user_id, username)
+        return self.playlist_result(entries, user_id, title, description)

youtube_dl/extractor/playwire.py

@@ -4,9 +4,8 @@ import re
 
 from .common import InfoExtractor
 from ..utils import (
-    xpath_text,
+    dict_get,
     float_or_none,
-    int_or_none,
 )
 
 
@@ -23,6 +22,19 @@ class PlaywireIE(InfoExtractor):
             'duration': 145.94,
         },
     }, {
+        # m3u8 in f4m
+        'url': 'http://config.playwire.com/21772/videos/v2/4840492/zeus.json',
+        'info_dict': {
+            'id': '4840492',
+            'ext': 'mp4',
+            'title': 'ITV EL SHOW FULL',
+        },
+        'params': {
+            # m3u8 download
+            'skip_download': True,
+        },
+    }, {
+        # Multiple resolutions while bitrates missing
         'url': 'http://cdn.playwire.com/11625/embed/85228.html',
         'only_matching': True,
     }, {
@@ -48,25 +60,10 @@ class PlaywireIE(InfoExtractor):
         thumbnail = content.get('poster')
         src = content['media']['f4m']
 
-        f4m = self._download_xml(src, video_id)
-        base_url = xpath_text(f4m, './{http://ns.adobe.com/f4m/1.0}baseURL', 'base url', fatal=True)
-        formats = []
-        for media in f4m.findall('./{http://ns.adobe.com/f4m/1.0}media'):
-            media_url = media.get('url')
-            if not media_url:
-                continue
-            tbr = int_or_none(media.get('bitrate'))
-            width = int_or_none(media.get('width'))
-            height = int_or_none(media.get('height'))
-            f = {
-                'url': '%s/%s' % (base_url, media.attrib['url']),
-                'tbr': tbr,
-                'width': width,
-                'height': height,
-            }
-            if not (tbr or width or height):
-                f['quality'] = 1 if '-hd.' in media_url else 0
-            formats.append(f)
+        formats = self._extract_f4m_formats(src, video_id, m3u8_id='hls')
+        for a_format in formats:
+            if not dict_get(a_format, ['tbr', 'width', 'height']):
+                a_format['quality'] = 1 if '-hd.' in a_format['url'] else 0
         self._sort_formats(formats)
 
         return {

youtube_dl/extractor/radiocanada.py

@@ -0,0 +1,130 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+import re
+
+from .common import InfoExtractor
+from ..utils import (
+    xpath_text,
+    find_xpath_attr,
+    determine_ext,
+    int_or_none,
+    unified_strdate,
+    xpath_element,
+    ExtractorError,
+)
+
+
+class RadioCanadaIE(InfoExtractor):
+    IE_NAME = 'radiocanada'
+    _VALID_URL = r'(?:radiocanada:|https?://ici\.radio-canada\.ca/widgets/mediaconsole/)(?P<app_code>[^:/]+)[:/](?P<id>[0-9]+)'
+    _TEST = {
+        'url': 'http://ici.radio-canada.ca/widgets/mediaconsole/medianet/7184272',
+        'info_dict': {
+            'id': '7184272',
+            'ext': 'flv',
+            'title': 'Le parcours du tireur capté sur vidéo',
+            'description': 'Images des caméras de surveillance fournies par la GRC montrant le parcours du tireur d\'Ottawa',
+            'upload_date': '20141023',
+        },
+        'params': {
+            # rtmp download
+            'skip_download': True,
+        },
+    }
+
+    def _real_extract(self, url):
+        app_code, video_id = re.match(self._VALID_URL, url).groups()
+
+        formats = []
+        # TODO: extract m3u8 and f4m formats
+        # m3u8 formats can be extracted using ipad device_type return 403 error code when ffmpeg try to download segements
+        # f4m formats can be extracted using flashhd device_type but they produce unplayable file
+        for device_type in ('flash',):
+            v_data = self._download_xml(
+                'http://api.radio-canada.ca/validationMedia/v1/Validation.ashx',
+                video_id, note='Downloading %s XML' % device_type, query={
+                    'appCode': app_code,
+                    'idMedia': video_id,
+                    'connectionType': 'broadband',
+                    'multibitrate': 'true',
+                    'deviceType': device_type,
+                    # paysJ391wsHjbOJwvCs26toz and bypasslock are used to bypass geo-restriction
+                    'paysJ391wsHjbOJwvCs26toz': 'CA',
+                    'bypasslock': 'NZt5K62gRqfc',
+                })
+            v_url = xpath_text(v_data, 'url')
+            if not v_url:
+                continue
+            if v_url == 'null':
+                raise ExtractorError('%s said: %s' % (
+                    self.IE_NAME, xpath_text(v_data, 'message')), expected=True)
+            ext = determine_ext(v_url)
+            if ext == 'm3u8':
+                formats.extend(self._extract_m3u8_formats(
+                    v_url, video_id, 'mp4', m3u8_id='hls', fatal=False))
+            elif ext == 'f4m':
+                formats.extend(self._extract_f4m_formats(v_url, video_id, f4m_id='hds', fatal=False))
+            else:
+                ext = determine_ext(v_url)
+                bitrates = xpath_element(v_data, 'bitrates')
+                for url_e in bitrates.findall('url'):
+                    tbr = int_or_none(url_e.get('bitrate'))
+                    if not tbr:
+                        continue
+                    formats.append({
+                        'format_id': 'rtmp-%d' % tbr,
+                        'url': re.sub(r'\d+\.%s' % ext, '%d.%s' % (tbr, ext), v_url),
+                        'ext': 'flv',
+                        'protocol': 'rtmp',
+                        'width': int_or_none(url_e.get('width')),
+                        'height': int_or_none(url_e.get('height')),
+                        'tbr': tbr,
+                    })
+        self._sort_formats(formats)
+
+        metadata = self._download_xml(
+            'http://api.radio-canada.ca/metaMedia/v1/index.ashx',
+            video_id, note='Downloading metadata XML', query={
+                'appCode': app_code,
+                'idMedia': video_id,
+            })
+
+        def get_meta(name):
+            el = find_xpath_attr(metadata, './/Meta', 'name', name)
+            return el.text if el is not None else None
+
+        return {
+            'id': video_id,
+            'title': get_meta('Title'),
+            'description': get_meta('Description') or get_meta('ShortDescription'),
+            'thumbnail': get_meta('imageHR') or get_meta('imageMR') or get_meta('imageBR'),
+            'duration': int_or_none(get_meta('length')),
+            'series': get_meta('Emission'),
+            'season_number': int_or_none('SrcSaison'),
+            'episode_number': int_or_none('SrcEpisode'),
+            'upload_date': unified_strdate(get_meta('Date')),
+            'formats': formats,
+        }
+
+
+class RadioCanadaAudioVideoIE(InfoExtractor):
+    'radiocanada:audiovideo'
+    _VALID_URL = r'https?://ici\.radio-canada\.ca/audio-video/media-(?P<id>[0-9]+)'
+    _TEST = {
+        'url': 'http://ici.radio-canada.ca/audio-video/media-7527184/barack-obama-au-vietnam',
+        'info_dict': {
+            'id': '7527184',
+            'ext': 'flv',
+            'title': 'Barack Obama au Vietnam',
+            'description': 'Les États-Unis lèvent l\'embargo sur la vente d\'armes qui datait de la guerre du Vietnam',
+            'upload_date': '20160523',
+        },
+        'params': {
+            # rtmp download
+            'skip_download': True,
+        },
+    }
+
+    def _real_extract(self, url):
+        return self.url_result('radiocanada:medianet:%s' % self._match_id(url))

youtube_dl/extractor/reuters.py

@@ -0,0 +1,69 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+import re
+
+from .common import InfoExtractor
+from ..utils import (
+    js_to_json,
+    int_or_none,
+    unescapeHTML,
+)
+
+
+class ReutersIE(InfoExtractor):
+    _VALID_URL = r'https?://(?:www\.)?reuters\.com/.*?\?.*?videoId=(?P<id>[0-9]+)'
+    _TEST = {
+        'url': 'http://www.reuters.com/video/2016/05/20/san-francisco-police-chief-resigns?videoId=368575562',
+        'md5': '8015113643a0b12838f160b0b81cc2ee',
+        'info_dict': {
+            'id': '368575562',
+            'ext': 'mp4',
+            'title': 'San Francisco police chief resigns',
+        }
+    }
+
+    def _real_extract(self, url):
+        video_id = self._match_id(url)
+        webpage = self._download_webpage(
+            'http://www.reuters.com/assets/iframe/yovideo?videoId=%s' % video_id, video_id)
+        video_data = js_to_json(self._search_regex(
+            r'(?s)Reuters\.yovideo\.drawPlayer\(({.*?})\);',
+            webpage, 'video data'))
+
+        def get_json_value(key, fatal=False):
+            return self._search_regex('"%s"\s*:\s*"([^"]+)"' % key, video_data, key, fatal=fatal)
+
+        title = unescapeHTML(get_json_value('title', fatal=True))
+        mmid, fid = re.search(r',/(\d+)\?f=(\d+)', get_json_value('flv', fatal=True)).groups()
+
+        mas_data = self._download_json(
+            'http://mas-e.cds1.yospace.com/mas/%s/%s?trans=json' % (mmid, fid),
+            video_id, transform_source=js_to_json)
+        formats = []
+        for f in mas_data:
+            f_url = f.get('url')
+            if not f_url:
+                continue
+            method = f.get('method')
+            if method == 'hls':
+                formats.extend(self._extract_m3u8_formats(
+                    f_url, video_id, 'mp4', 'm3u8_native', m3u8_id='hls', fatal=False))
+            else:
+                container = f.get('container')
+                ext = '3gp' if method == 'mobile' else container
+                formats.append({
+                    'format_id': ext,
+                    'url': f_url,
+                    'ext': ext,
+                    'container': container if method != 'mobile' else None,
+                })
+        self._sort_formats(formats)
+
+        return {
+            'id': video_id,
+            'title': title,
+            'thumbnail': get_json_value('thumb'),
+            'duration': int_or_none(get_json_value('seconds')),
+            'formats': formats,
+        }

youtube_dl/extractor/revision3.py

@@ -13,8 +13,64 @@ from ..utils import (
 )
 
 
+class Revision3EmbedIE(InfoExtractor):
+    IE_NAME = 'revision3:embed'
+    _VALID_URL = r'(?:revision3:(?:(?P<playlist_type>[^:]+):)?|https?://(?:(?:(?:www|embed)\.)?(?:revision3|animalist)|(?:(?:api|embed)\.)?seekernetwork)\.com/player/embed\?videoId=)(?P<playlist_id>\d+)'
+    _TEST = {
+        'url': 'http://api.seekernetwork.com/player/embed?videoId=67558',
+        'md5': '83bcd157cab89ad7318dd7b8c9cf1306',
+        'info_dict': {
+            'id': '67558',
+            'ext': 'mp4',
+            'title': 'The Pros & Cons Of Zoos',
+            'description': 'Zoos are often depicted as a terrible place for animals to live, but is there any truth to this?',
+            'uploader_id': 'dnews',
+            'uploader': 'DNews',
+        }
+    }
+    _API_KEY = 'ba9c741bce1b9d8e3defcc22193f3651b8867e62'
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        playlist_id = mobj.group('playlist_id')
+        playlist_type = mobj.group('playlist_type') or 'video_id'
+        video_data = self._download_json(
+            'http://revision3.com/api/getPlaylist.json', playlist_id, query={
+                'api_key': self._API_KEY,
+                'codecs': 'h264,vp8,theora',
+                playlist_type: playlist_id,
+            })['items'][0]
+
+        formats = []
+        for vcodec, media in video_data['media'].items():
+            for quality_id, quality in media.items():
+                if quality_id == 'hls':
+                    formats.extend(self._extract_m3u8_formats(
+                        quality['url'], playlist_id, 'mp4',
+                        'm3u8_native', m3u8_id='hls', fatal=False))
+                else:
+                    formats.append({
+                        'url': quality['url'],
+                        'format_id': '%s-%s' % (vcodec, quality_id),
+                        'tbr': int_or_none(quality.get('bitrate')),
+                        'vcodec': vcodec,
+                    })
+        self._sort_formats(formats)
+
+        return {
+            'id': playlist_id,
+            'title': unescapeHTML(video_data['title']),
+            'description': unescapeHTML(video_data.get('summary')),
+            'uploader': video_data.get('show', {}).get('name'),
+            'uploader_id': video_data.get('show', {}).get('slug'),
+            'duration': int_or_none(video_data.get('duration')),
+            'formats': formats,
+        }
+
+
 class Revision3IE(InfoExtractor):
-    _VALID_URL = r'https?://(?:www\.)?(?P<domain>(?:revision3|testtube|animalist)\.com)/(?P<id>[^/]+(?:/[^/?#]+)?)'
+    IE_NAME = 'revision'
+    _VALID_URL = r'https?://(?:www\.)?(?P<domain>(?:revision3|animalist)\.com)/(?P<id>[^/]+(?:/[^/?#]+)?)'
     _TESTS = [{
         'url': 'http://www.revision3.com/technobuffalo/5-google-predictions-for-2016',
         'md5': 'd94a72d85d0a829766de4deb8daaf7df',
@@ -32,52 +88,14 @@ class Revision3IE(InfoExtractor):
         }
     }, {
         # Show
-        'url': 'http://testtube.com/brainstuff',
-        'info_dict': {
-            'id': '251',
-            'title': 'BrainStuff',
-            'description': 'Whether the topic is popcorn or particle physics, you can count on the HowStuffWorks team to explore-and explain-the everyday science in the world around us on BrainStuff.',
-        },
-        'playlist_mincount': 93,
-    }, {
-        'url': 'https://testtube.com/dnews/5-weird-ways-plants-can-eat-animals?utm_source=FB&utm_medium=DNews&utm_campaign=DNewsSocial',
-        'info_dict': {
-            'id': '58227',
-            'display_id': 'dnews/5-weird-ways-plants-can-eat-animals',
-            'duration': 275,
-            'ext': 'webm',
-            'title': '5 Weird Ways Plants Can Eat Animals',
-            'description': 'Why have some plants evolved to eat meat?',
-            'upload_date': '20150120',
-            'timestamp': 1421763300,
-            'uploader': 'DNews',
-            'uploader_id': 'dnews',
-        },
-    }, {
-        'url': 'http://testtube.com/tt-editors-picks/the-israel-palestine-conflict-explained-in-ten-min',
-        'info_dict': {
-            'id': '71618',
-            'ext': 'mp4',
-            'display_id': 'tt-editors-picks/the-israel-palestine-conflict-explained-in-ten-min',
-            'title': 'The Israel-Palestine Conflict Explained in Ten Minutes',
-            'description': 'If you\'d like to learn about the struggle between Israelis and Palestinians, this video is a great place to start',
-            'uploader': 'Editors\' Picks',
-            'uploader_id': 'tt-editors-picks',
-            'timestamp': 1453309200,
-            'upload_date': '20160120',
-        },
-        'add_ie': ['Youtube'],
+        'url': 'http://revision3.com/variant',
+        'only_matching': True,
     }, {
         # Tag
-        'url': 'http://testtube.com/tech-news',
-        'info_dict': {
-            'id': '21018',
-            'title': 'tech news',
-        },
-        'playlist_mincount': 9,
+        'url': 'http://revision3.com/vr',
+        'only_matching': True,
     }]
     _PAGE_DATA_TEMPLATE = 'http://www.%s/apiProxy/ddn/%s?domain=%s'
-    _API_KEY = 'ba9c741bce1b9d8e3defcc22193f3651b8867e62'
 
     def _real_extract(self, url):
         domain, display_id = re.match(self._VALID_URL, url).groups()
@@ -119,33 +137,9 @@ class Revision3IE(InfoExtractor):
                 })
                 return info
 
-            video_data = self._download_json(
-                'http://revision3.com/api/getPlaylist.json?api_key=%s&codecs=h264,vp8,theora&video_id=%s' % (self._API_KEY, video_id),
-                video_id)['items'][0]
-
-            formats = []
-            for vcodec, media in video_data['media'].items():
-                for quality_id, quality in media.items():
-                    if quality_id == 'hls':
-                        formats.extend(self._extract_m3u8_formats(
-                            quality['url'], video_id, 'mp4',
-                            'm3u8_native', m3u8_id='hls', fatal=False))
-                    else:
-                        formats.append({
-                            'url': quality['url'],
-                            'format_id': '%s-%s' % (vcodec, quality_id),
-                            'tbr': int_or_none(quality.get('bitrate')),
-                            'vcodec': vcodec,
-                        })
-            self._sort_formats(formats)
-
             info.update({
-                'title': unescapeHTML(video_data['title']),
-                'description': unescapeHTML(video_data.get('summary')),
-                'uploader': video_data.get('show', {}).get('name'),
-                'uploader_id': video_data.get('show', {}).get('slug'),
-                'duration': int_or_none(video_data.get('duration')),
-                'formats': formats,
+                '_type': 'url_transparent',
+                'url': 'revision3:%s' % video_id,
             })
             return info
         else:

youtube_dl/extractor/seeker.py

@@ -0,0 +1,57 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+import re
+
+from .common import InfoExtractor
+
+
+class SeekerIE(InfoExtractor):
+    _VALID_URL = r'https?://(?:www\.)?seeker\.com/(?P<display_id>.*)-(?P<article_id>\d+)\.html'
+    _TESTS = [{
+        # player.loadRevision3Item
+        'url': 'http://www.seeker.com/should-trump-be-required-to-release-his-tax-returns-1833805621.html',
+        'md5': '30c1dc4030cc715cf05b423d0947ac18',
+        'info_dict': {
+            'id': '76243',
+            'ext': 'webm',
+            'title': 'Should Trump Be Required To Release His Tax Returns?',
+            'description': 'Donald Trump has been secretive about his "big," "beautiful" tax returns. So what can we learn if he decides to release them?',
+            'uploader': 'Seeker Daily',
+            'uploader_id': 'seekerdaily',
+        }
+    }, {
+        'url': 'http://www.seeker.com/changes-expected-at-zoos-following-recent-gorilla-lion-shootings-1834116536.html',
+        'playlist': [
+            {
+                'md5': '83bcd157cab89ad7318dd7b8c9cf1306',
+                'info_dict': {
+                    'id': '67558',
+                    'ext': 'mp4',
+                    'title': 'The Pros & Cons Of Zoos',
+                    'description': 'Zoos are often depicted as a terrible place for animals to live, but is there any truth to this?',
+                    'uploader': 'DNews',
+                    'uploader_id': 'dnews',
+                },
+            }
+        ],
+        'info_dict': {
+            'id': '1834116536',
+            'title': 'After Gorilla Killing, Changes Ahead for Zoos',
+            'description': 'The largest association of zoos and others are hoping to learn from recent incidents that led to the shooting deaths of a gorilla and two lions.',
+        },
+    }]
+
+    def _real_extract(self, url):
+        display_id, article_id = re.match(self._VALID_URL, url).groups()
+        webpage = self._download_webpage(url, display_id)
+        mobj = re.search(r"player\.loadRevision3Item\('([^']+)'\s*,\s*(\d+)\);", webpage)
+        if mobj:
+            playlist_type, playlist_id = mobj.groups()
+            return self.url_result(
+                'revision3:%s:%s' % (playlist_type, playlist_id), 'Revision3Embed', playlist_id)
+        else:
+            entries = [self.url_result('revision3:video_id:%s' % video_id, 'Revision3Embed', video_id) for video_id in re.findall(
+                r'<iframe[^>]+src=[\'"](?:https?:)?//api\.seekernetwork\.com/player/embed\?videoId=(\d+)', webpage)]
+            return self.playlist_result(
+                entries, article_id, self._og_search_title(webpage), self._og_search_description(webpage))

youtube_dl/extractor/spankwire.py

@@ -96,20 +96,18 @@ class SpankwireIE(InfoExtractor):
         formats = []
         for height, video_url in zip(heights, video_urls):
             path = compat_urllib_parse_urlparse(video_url).path
-            _, quality = path.split('/')[4].split('_')[:2]
-            f = {
-                'url': video_url,
-                'height': height,
-            }
-            tbr = self._search_regex(r'^(\d+)[Kk]$', quality, 'tbr', default=None)
-            if tbr:
-                f.update({
-                    'tbr': int(tbr),
-                    'format_id': '%dp' % height,
-                })
+            m = re.search(r'/(?P<height>\d+)[pP]_(?P<tbr>\d+)[kK]', path)
+            if m:
+                tbr = int(m.group('tbr'))
+                height = int(m.group('height'))
             else:
-                f['format_id'] = quality
-            formats.append(f)
+                tbr = None
+            formats.append({
+                'url': video_url,
+                'format_id': '%dp' % height,
+                'height': height,
+                'tbr': tbr,
+            })
         self._sort_formats(formats)
 
         age_limit = self._rta_search(webpage)

youtube_dl/extractor/teachingchannel.py

@@ -11,6 +11,7 @@ class TeachingChannelIE(InfoExtractor):
 
     _TEST = {
         'url': 'https://www.teachingchannel.org/videos/teacher-teaming-evolution',
+        'md5': '3d6361864d7cac20b57c8784da17166f',
         'info_dict': {
             'id': 'F3bnlzbToeI6pLEfRyrlfooIILUjz4nM',
             'ext': 'mp4',
@@ -19,9 +20,9 @@ class TeachingChannelIE(InfoExtractor):
             'duration': 422.255,
         },
         'params': {
-            # m3u8 download
             'skip_download': True,
         },
+        'add_ie': ['Ooyala'],
     }
 
     def _real_extract(self, url):

youtube_dl/extractor/tf1.py

@@ -6,7 +6,7 @@ from .common import InfoExtractor
 
 class TF1IE(InfoExtractor):
     """TF1 uses the wat.tv player."""
-    _VALID_URL = r'https?://(?:(?:videos|www|lci)\.tf1|www\.tfou)\.fr/(?:[^/]+/)*(?P<id>.+?)\.html'
+    _VALID_URL = r'https?://(?:(?:videos|www|lci)\.tf1|(?:www\.)?(?:tfou|ushuaiatv|histoire|tvbreizh))\.fr/(?:[^/]+/)*(?P<id>[^/?#.]+)'
     _TESTS = [{
         'url': 'http://videos.tf1.fr/auto-moto/citroen-grand-c4-picasso-2013-presentation-officielle-8062060.html',
         'info_dict': {
@@ -48,6 +48,6 @@ class TF1IE(InfoExtractor):
         video_id = self._match_id(url)
         webpage = self._download_webpage(url, video_id)
         wat_id = self._html_search_regex(
-            r'(["\'])(?:https?:)?//www\.wat\.tv/embedframe/.*?(?P<id>\d{8})(?:#.*?)?\1',
+            r'(["\'])(?:https?:)?//www\.wat\.tv/embedframe/.*?(?P<id>\d{8}).*?\1',
             webpage, 'wat id', group='id')
         return self.url_result('wat:%s' % wat_id, 'Wat')

youtube_dl/extractor/theplatform.py

@@ -151,6 +151,22 @@ class ThePlatformIE(ThePlatformBaseIE):
         'only_matching': True,
     }]
 
+    @classmethod
+    def _extract_urls(cls, webpage):
+        m = re.search(
+            r'''(?x)
+                    <meta\s+
+                        property=(["'])(?:og:video(?::(?:secure_)?url)?|twitter:player)\1\s+
+                        content=(["'])(?P<url>https?://player\.theplatform\.com/p/.+?)\2
+            ''', webpage)
+        if m:
+            return [m.group('url')]
+
+        matches = re.findall(
+            r'<(?:iframe|script)[^>]+src=(["\'])((?:https?:)?//player\.theplatform\.com/p/.+?)\1', webpage)
+        if matches:
+            return list(zip(*matches))[1]
+
     @staticmethod
     def _sign_url(url, sig_key, sig_secret, life=600, include_qs=False):
         flags = '10' if include_qs else '00'

youtube_dl/extractor/tvp.py

@@ -1,4 +1,4 @@
-# -*- coding: utf-8 -*-
+# coding: utf-8
 from __future__ import unicode_literals
 
 import re
@@ -6,20 +6,13 @@ import re
 from .common import InfoExtractor
 
 
-class TvpIE(InfoExtractor):
-    IE_NAME = 'tvp.pl'
-    _VALID_URL = r'https?://(?:vod|www)\.tvp\.pl/.*/(?P<id>\d+)$'
+class TVPIE(InfoExtractor):
+    IE_NAME = 'tvp'
+    IE_DESC = 'Telewizja Polska'
+    _VALID_URL = r'https?://[^/]+\.tvp\.(?:pl|info)/(?:(?!\d+/)[^/]+/)*(?P<id>\d+)'
 
     _TESTS = [{
-        'url': 'http://vod.tvp.pl/filmy-fabularne/filmy-za-darmo/ogniem-i-mieczem/wideo/odc-2/4278035',
-        'md5': 'cdd98303338b8a7f7abab5cd14092bf2',
-        'info_dict': {
-            'id': '4278035',
-            'ext': 'wmv',
-            'title': 'Ogniem i mieczem, odc. 2',
-        },
-    }, {
-        'url': 'http://vod.tvp.pl/seriale/obyczajowe/czas-honoru/sezon-1-1-13/i-seria-odc-13/194536',
+        'url': 'http://vod.tvp.pl/194536/i-seria-odc-13',
         'md5': '8aa518c15e5cc32dfe8db400dc921fbb',
         'info_dict': {
             'id': '194536',
@@ -36,12 +29,22 @@ class TvpIE(InfoExtractor):
         },
     }, {
         'url': 'http://vod.tvp.pl/seriale/obyczajowe/na-sygnale/sezon-2-27-/odc-39/17834272',
-        'md5': 'c3b15ed1af288131115ff17a17c19dda',
-        'info_dict': {
-            'id': '17834272',
-            'ext': 'mp4',
-            'title': 'Na sygnale, odc. 39',
-        },
+        'only_matching': True,
+    }, {
+        'url': 'http://wiadomosci.tvp.pl/25169746/24052016-1200',
+        'only_matching': True,
+    }, {
+        'url': 'http://krakow.tvp.pl/25511623/25lecie-mck-wyjatkowe-miejsce-na-mapie-krakowa',
+        'only_matching': True,
+    }, {
+        'url': 'http://teleexpress.tvp.pl/25522307/wierni-wzieli-udzial-w-procesjach',
+        'only_matching': True,
+    }, {
+        'url': 'http://sport.tvp.pl/25522165/krychowiak-uspokaja-w-sprawie-kontuzji-dwa-tygodnie-to-maksimum',
+        'only_matching': True,
+    }, {
+        'url': 'http://www.tvp.info/25511919/trwa-rewolucja-wladza-zdecydowala-sie-na-pogwalcenie-konstytucji',
+        'only_matching': True,
     }]
 
     def _real_extract(self, url):
@@ -92,8 +95,8 @@ class TvpIE(InfoExtractor):
         }
 
 
-class TvpSeriesIE(InfoExtractor):
-    IE_NAME = 'tvp.pl:Series'
+class TVPSeriesIE(InfoExtractor):
+    IE_NAME = 'tvp:series'
     _VALID_URL = r'https?://vod\.tvp\.pl/(?:[^/]+/){2}(?P<id>[^/]+)/?$'
 
     _TESTS = [{
@@ -127,7 +130,7 @@ class TvpSeriesIE(InfoExtractor):
         videos_paths = re.findall(
             '(?s)class="shortTitle">.*?href="(/[^"]+)', playlist)
         entries = [
-            self.url_result('http://vod.tvp.pl%s' % v_path, ie=TvpIE.ie_key())
+            self.url_result('http://vod.tvp.pl%s' % v_path, ie=TVPIE.ie_key())
             for v_path in videos_paths]
 
         return {

youtube_dl/extractor/udemy.py

@@ -142,7 +142,9 @@ class UdemyIE(InfoExtractor):
             self._LOGIN_URL, None, 'Downloading login popup')
 
         def is_logged(webpage):
-            return any(p in webpage for p in ['href="https://www.udemy.com/user/logout/', '>Logout<'])
+            return any(re.search(p, webpage) for p in (
+                r'href=["\'](?:https://www\.udemy\.com)?/user/logout/',
+                r'>Logout<'))
 
         # already logged in
         if is_logged(login_popup):

youtube_dl/extractor/udn.py

@@ -2,10 +2,13 @@
 from __future__ import unicode_literals
 
 import json
+import re
+
 from .common import InfoExtractor
 from ..utils import (
+    determine_ext,
+    int_or_none,
     js_to_json,
-    ExtractorError,
 )
 from ..compat import compat_urlparse
 
@@ -16,13 +19,16 @@ class UDNEmbedIE(InfoExtractor):
     _VALID_URL = r'https?:' + _PROTOCOL_RELATIVE_VALID_URL
     _TESTS = [{
         'url': 'http://video.udn.com/embed/news/300040',
-        'md5': 'de06b4c90b042c128395a88f0384817e',
         'info_dict': {
             'id': '300040',
             'ext': 'mp4',
             'title': '生物老師男變女 全校挺"做自己"',
             'thumbnail': 're:^https?://.*\.jpg$',
-        }
+        },
+        'params': {
+            # m3u8 download
+            'skip_download': True,
+        },
     }, {
         'url': 'https://video.udn.com/embed/news/300040',
         'only_matching': True,
@@ -38,39 +44,53 @@ class UDNEmbedIE(InfoExtractor):
         page = self._download_webpage(url, video_id)
 
         options = json.loads(js_to_json(self._html_search_regex(
-            r'var options\s*=\s*([^;]+);', page, 'video urls dictionary')))
+            r'var\s+options\s*=\s*([^;]+);', page, 'video urls dictionary')))
 
         video_urls = options['video']
 
         if video_urls.get('youtube'):
             return self.url_result(video_urls.get('youtube'), 'Youtube')
 
-        try:
-            del video_urls['youtube']
-        except KeyError:
-            pass
+        formats = []
+        for video_type, api_url in video_urls.items():
+            if not api_url:
+                continue
 
-        formats = [{
-            'url': self._download_webpage(
+            video_url = self._download_webpage(
                 compat_urlparse.urljoin(url, api_url), video_id,
-                'retrieve url for %s video' % video_type),
-            'format_id': video_type,
-            'preference': 0 if video_type == 'mp4' else -1,
-        } for video_type, api_url in video_urls.items() if api_url]
+                note='retrieve url for %s video' % video_type)
 
-        if not formats:
-            raise ExtractorError('No videos found', expected=True)
+            ext = determine_ext(video_url)
+            if ext == 'm3u8':
+                formats.extend(self._extract_m3u8_formats(
+                    video_url, video_id, ext='mp4', m3u8_id='hls'))
+            elif ext == 'f4m':
+                formats.extend(self._extract_f4m_formats(
+                    video_url, video_id, f4m_id='hds'))
+            else:
+                mobj = re.search(r'_(?P<height>\d+)p_(?P<tbr>\d+).mp4', video_url)
+                a_format = {
+                    'url': video_url,
+                    # video_type may be 'mp4', which confuses YoutubeDL
+                    'format_id': 'http-' + video_type,
+                }
+                if mobj:
+                    a_format.update({
+                        'height': int_or_none(mobj.group('height')),
+                        'tbr': int_or_none(mobj.group('tbr')),
+                    })
+                formats.append(a_format)
 
         self._sort_formats(formats)
 
-        thumbnail = None
-
-        if options.get('gallery') and len(options['gallery']):
-            thumbnail = options['gallery'][0].get('original')
+        thumbnails = [{
+            'url': img_url,
+            'id': img_type,
+        } for img_type, img_url in options.get('gallery', [{}])[0].items() if img_url]
 
         return {
             'id': video_id,
             'formats': formats,
             'title': options['title'],
-            'thumbnail': thumbnail
+            'thumbnails': thumbnails,
         }

youtube_dl/extractor/veoh.py

@@ -37,6 +37,7 @@ class VeohIE(InfoExtractor):
                 'uploader': 'afp-news',
                 'duration': 123,
             },
+            'skip': 'This video has been deleted.',
         },
         {
             'url': 'http://www.veoh.com/watch/v69525809F6Nc4frX',

youtube_dl/extractor/vice.py

@@ -11,12 +11,14 @@ class ViceIE(InfoExtractor):
 
     _TESTS = [{
         'url': 'http://www.vice.com/video/cowboy-capitalists-part-1',
+        'md5': 'e9d77741f9e42ba583e683cd170660f7',
         'info_dict': {
             'id': '43cW1mYzpia9IlestBjVpd23Yu3afAfp',
             'ext': 'flv',
             'title': 'VICE_COWBOYCAPITALISTS_PART01_v1_VICE_WM_1080p.mov',
             'duration': 725.983,
         },
+        'add_ie': ['Ooyala'],
     }, {
         'url': 'http://www.vice.com/video/how-to-hack-a-car',
         'md5': '6fb2989a3fed069fb8eab3401fc2d3c9',
@@ -29,6 +31,7 @@ class ViceIE(InfoExtractor):
             'uploader': 'Motherboard',
             'upload_date': '20140529',
         },
+        'add_ie': ['Youtube'],
     }, {
         'url': 'https://news.vice.com/video/experimenting-on-animals-inside-the-monkey-lab',
         'only_matching': True,

youtube_dl/extractor/viewlift.py

@@ -141,6 +141,10 @@ class ViewLiftIE(ViewLiftBaseIE):
     }, {
         'url': 'http://www.kesari.tv/news/video/1461919076414',
         'only_matching': True,
+    }, {
+        # Was once Kaltura embed
+        'url': 'https://www.monumentalsportsnetwork.com/videos/john-carlson-postgame-2-25-15',
+        'only_matching': True,
     }]
 
     def _real_extract(self, url):

youtube_dl/extractor/vk.py

@@ -217,7 +217,6 @@ class VKIE(InfoExtractor):
         mobj = re.match(self._VALID_URL, url)
         video_id = mobj.group('videoid')
 
-        info_url = url
         if video_id:
             info_url = 'https://vk.com/al_video.php?act=show&al=1&module=video&video=%s' % video_id
             # Some videos (removed?) can only be downloaded with list id specified

youtube_dl/extractor/vlive.py

@@ -1,8 +1,7 @@
 # coding: utf-8
-from __future__ import division, unicode_literals
+from __future__ import unicode_literals
 
 import re
-import time
 
 from .common import InfoExtractor
 from ..utils import (
@@ -23,7 +22,7 @@ class VLiveIE(InfoExtractor):
         'info_dict': {
             'id': '1326',
             'ext': 'mp4',
-            'title': "[V] Girl's Day's Broadcast",
+            'title': "[V LIVE] Girl's Day's Broadcast",
             'creator': "Girl's Day",
             'view_count': int,
         },
@@ -35,24 +34,11 @@ class VLiveIE(InfoExtractor):
         webpage = self._download_webpage(
             'http://www.vlive.tv/video/%s' % video_id, video_id)
 
-        # UTC+x - UTC+9 (KST)
-        tz = time.altzone if time.localtime().tm_isdst == 1 else time.timezone
-        tz_offset = -tz // 60 - 9 * 60
-        self._set_cookie('vlive.tv', 'timezoneOffset', '%d' % tz_offset)
-
-        status_params = self._download_json(
-            'http://www.vlive.tv/video/status?videoSeq=%s' % video_id,
-            video_id, 'Downloading JSON status',
-            headers={'Referer': url.encode('utf-8')})
-        status = status_params.get('status')
-        air_start = status_params.get('onAirStartAt', '')
-        is_live = status_params.get('isLive')
-
         video_params = self._search_regex(
-            r'vlive\.tv\.video\.ajax\.request\.handler\.init\((.+)\)',
+            r'\bvlive\.video\.init\(([^)]+)\)',
             webpage, 'video params')
-        live_params, long_video_id, key = re.split(
-            r'"\s*,\s*"', video_params)[1:4]
+        status, _, _, live_params, long_video_id, key = re.split(
+            r'"\s*,\s*"', video_params)[2:8]
 
         if status == 'LIVE_ON_AIR' or status == 'BIG_EVENT_ON_AIR':
             live_params = self._parse_json('"%s"' % live_params, video_id)
@@ -61,8 +47,6 @@ class VLiveIE(InfoExtractor):
         elif status == 'VOD_ON_AIR' or status == 'BIG_EVENT_INTRO':
             if long_video_id and key:
                 return self._replay(video_id, webpage, long_video_id, key)
-            elif is_live:
-                status = 'LIVE_END'
             else:
                 status = 'COMING_SOON'
 
@@ -70,7 +54,7 @@ class VLiveIE(InfoExtractor):
             raise ExtractorError('Uploading for replay. Please wait...',
                                  expected=True)
         elif status == 'COMING_SOON':
-            raise ExtractorError('Coming soon! %s' % air_start, expected=True)
+            raise ExtractorError('Coming soon!', expected=True)
         elif status == 'CANCELED':
             raise ExtractorError('We are sorry, '
                                  'but the live broadcast has been canceled.',

youtube_dl/extractor/voxmedia.py

@@ -15,7 +15,8 @@ class VoxMediaIE(InfoExtractor):
             'ext': 'mp4',
             'title': 'Google\'s new material design direction',
             'description': 'md5:2f44f74c4d14a1f800ea73e1c6832ad2',
-        }
+        },
+        'add_ie': ['Ooyala'],
     }, {
         # data-ooyala-id
         'url': 'http://www.theverge.com/2014/10/21/7025853/google-nexus-6-hands-on-photos-video-android-phablet',
@@ -25,7 +26,8 @@ class VoxMediaIE(InfoExtractor):
             'ext': 'mp4',
             'title': 'The Nexus 6: hands-on with Google\'s phablet',
             'description': 'md5:87a51fe95ff8cea8b5bdb9ac7ae6a6af',
-        }
+        },
+        'add_ie': ['Ooyala'],
     }, {
         # volume embed
         'url': 'http://www.vox.com/2016/3/31/11336640/mississippi-lgbt-religious-freedom-bill',
@@ -35,7 +37,8 @@ class VoxMediaIE(InfoExtractor):
             'ext': 'mp4',
             'title': 'The new frontier of LGBTQ civil rights, explained',
             'description': 'md5:0dc58e94a465cbe91d02950f770eb93f',
-        }
+        },
+        'add_ie': ['Ooyala'],
     }, {
         # youtube embed
         'url': 'http://www.vox.com/2016/3/24/11291692/robot-dance',
@@ -48,7 +51,8 @@ class VoxMediaIE(InfoExtractor):
             'upload_date': '20160324',
             'uploader_id': 'voxdotcom',
             'uploader': 'Vox',
-        }
+        },
+        'add_ie': ['Youtube'],
     }, {
         # SBN.VideoLinkset.entryGroup multiple ooyala embeds
         'url': 'http://www.sbnation.com/college-football-recruiting/2015/2/3/7970291/national-signing-day-rationalizations-itll-be-ok-itll-be-ok',
@@ -117,7 +121,7 @@ class VoxMediaIE(InfoExtractor):
             volume_webpage = self._download_webpage(
                 'http://volume.vox-cdn.com/embed/%s' % volume_uuid, volume_uuid)
             video_data = self._parse_json(self._search_regex(
-                r'Volume\.createVideo\(({.+})\s*,\s*{.*}\);', volume_webpage, 'video data'), volume_uuid)
+                r'Volume\.createVideo\(({.+})\s*,\s*{.*}\s*,\s*\[.*\]\s*,\s*{.*}\);', volume_webpage, 'video data'), volume_uuid)
             for provider_video_type in ('ooyala', 'youtube'):
                 provider_video_id = video_data.get('%s_id' % provider_video_type)
                 if provider_video_id:

youtube_dl/extractor/washingtonpost.py

@@ -11,7 +11,96 @@ from ..utils import (
 
 
 class WashingtonPostIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:www\.)?washingtonpost\.com/.*?/(?P<id>[^/]+)/(?:$|[?#])'
+    IE_NAME = 'washingtonpost'
+    _VALID_URL = r'(?:washingtonpost:|https?://(?:www\.)?washingtonpost\.com/video/(?:[^/]+/)*)(?P<id>[\da-f]{8}-[\da-f]{4}-[\da-f]{4}-[\da-f]{4}-[\da-f]{12})'
+    _TEST = {
+        'url': 'https://www.washingtonpost.com/video/c/video/480ba4ee-1ec7-11e6-82c2-a7dcb313287d',
+        'md5': '6f537e1334b714eb15f9563bd4b9cdfa',
+        'info_dict': {
+            'id': '480ba4ee-1ec7-11e6-82c2-a7dcb313287d',
+            'ext': 'mp4',
+            'title': 'Egypt finds belongings, debris from plane crash',
+            'description': 'md5:a17ceee432f215a5371388c1f680bd86',
+            'upload_date': '20160520',
+            'uploader': 'Reuters',
+            'timestamp': 1463778452,
+        },
+    }
+
+    def _real_extract(self, url):
+        video_id = self._match_id(url)
+        video_data = self._download_json(
+            'http://www.washingtonpost.com/posttv/c/videojson/%s?resType=jsonp' % video_id,
+            video_id, transform_source=strip_jsonp)[0]['contentConfig']
+        title = video_data['title']
+
+        urls = []
+        formats = []
+        for s in video_data.get('streams', []):
+            s_url = s.get('url')
+            if not s_url or s_url in urls:
+                continue
+            urls.append(s_url)
+            video_type = s.get('type')
+            if video_type == 'smil':
+                continue
+            elif video_type in ('ts', 'hls') and ('_master.m3u8' in s_url or '_mobile.m3u8' in s_url):
+                m3u8_formats = self._extract_m3u8_formats(
+                    s_url, video_id, 'mp4', 'm3u8_native', m3u8_id='hls', fatal=False)
+                for m3u8_format in m3u8_formats:
+                    width = m3u8_format.get('width')
+                    if not width:
+                        continue
+                    vbr = self._search_regex(
+                        r'%d_%d_(\d+)' % (width, m3u8_format['height']), m3u8_format['url'], 'vbr', default=None)
+                    if vbr:
+                        m3u8_format.update({
+                            'vbr': int_or_none(vbr),
+                        })
+                formats.extend(m3u8_formats)
+            else:
+                width = int_or_none(s.get('width'))
+                vbr = int_or_none(s.get('bitrate'))
+                has_width = width != 0
+                formats.append({
+                    'format_id': (
+                        '%s-%d-%d' % (video_type, width, vbr)
+                        if width
+                        else video_type),
+                    'vbr': vbr if has_width else None,
+                    'width': width,
+                    'height': int_or_none(s.get('height')),
+                    'acodec': s.get('audioCodec'),
+                    'vcodec': s.get('videoCodec') if has_width else 'none',
+                    'filesize': int_or_none(s.get('fileSize')),
+                    'url': s_url,
+                    'ext': 'mp4',
+                    'protocol': 'm3u8_native' if video_type in ('ts', 'hls') else None,
+                })
+        source_media_url = video_data.get('sourceMediaURL')
+        if source_media_url:
+            formats.append({
+                'format_id': 'source_media',
+                'url': source_media_url,
+            })
+        self._sort_formats(
+            formats, ('width', 'height', 'vbr', 'filesize', 'tbr', 'format_id'))
+
+        return {
+            'id': video_id,
+            'title': title,
+            'description': video_data.get('blurb'),
+            'uploader': video_data.get('credits', {}).get('source'),
+            'formats': formats,
+            'duration': int_or_none(video_data.get('videoDuration'), 100),
+            'timestamp': int_or_none(
+                video_data.get('dateConfig', {}).get('dateFirstPublished'), 1000),
+        }
+
+
+class WashingtonPostArticleIE(InfoExtractor):
+    IE_NAME = 'washingtonpost:article'
+    _VALID_URL = r'https?://(?:www\.)?washingtonpost\.com/(?:[^/]+/)*(?P<id>[^/?#]+)'
     _TESTS = [{
         'url': 'http://www.washingtonpost.com/sf/national/2014/03/22/sinkhole-of-bureaucracy/',
         'info_dict': {
@@ -63,6 +152,10 @@ class WashingtonPostIE(InfoExtractor):
         }]
     }]
 
+    @classmethod
+    def suitable(cls, url):
+        return False if WashingtonPostIE.suitable(url) else super(WashingtonPostArticleIE, cls).suitable(url)
+
     def _real_extract(self, url):
         page_id = self._match_id(url)
         webpage = self._download_webpage(url, page_id)
@@ -74,54 +167,7 @@ class WashingtonPostIE(InfoExtractor):
                 <div\s+class="posttv-video-embed[^>]*?data-uuid=|
                 data-video-uuid=
             )"([^"]+)"''', webpage)
-        entries = []
-        for i, uuid in enumerate(uuids, start=1):
-            vinfo_all = self._download_json(
-                'http://www.washingtonpost.com/posttv/c/videojson/%s?resType=jsonp' % uuid,
-                page_id,
-                transform_source=strip_jsonp,
-                note='Downloading information of video %d/%d' % (i, len(uuids))
-            )
-            vinfo = vinfo_all[0]['contentConfig']
-            uploader = vinfo.get('credits', {}).get('source')
-            timestamp = int_or_none(
-                vinfo.get('dateConfig', {}).get('dateFirstPublished'), 1000)
-
-            formats = [{
-                'format_id': (
-                    '%s-%s-%s' % (s.get('type'), s.get('width'), s.get('bitrate'))
-                    if s.get('width')
-                    else s.get('type')),
-                'vbr': s.get('bitrate') if s.get('width') != 0 else None,
-                'width': s.get('width'),
-                'height': s.get('height'),
-                'acodec': s.get('audioCodec'),
-                'vcodec': s.get('videoCodec') if s.get('width') != 0 else 'none',
-                'filesize': s.get('fileSize'),
-                'url': s.get('url'),
-                'ext': 'mp4',
-                'preference': -100 if s.get('type') == 'smil' else None,
-                'protocol': {
-                    'MP4': 'http',
-                    'F4F': 'f4m',
-                }.get(s.get('type')),
-            } for s in vinfo.get('streams', [])]
-            source_media_url = vinfo.get('sourceMediaURL')
-            if source_media_url:
-                formats.append({
-                    'format_id': 'source_media',
-                    'url': source_media_url,
-                })
-            self._sort_formats(formats)
-            entries.append({
-                'id': uuid,
-                'title': vinfo['title'],
-                'description': vinfo.get('blurb'),
-                'uploader': uploader,
-                'formats': formats,
-                'duration': int_or_none(vinfo.get('videoDuration'), 100),
-                'timestamp': timestamp,
-            })
+        entries = [self.url_result('washingtonpost:%s' % uuid, 'WashingtonPost', uuid) for uuid in uuids]
 
         return {
             '_type': 'playlist',

youtube_dl/extractor/wat.py

@@ -2,25 +2,26 @@
 from __future__ import unicode_literals
 
 import re
-import hashlib
 
 from .common import InfoExtractor
+from ..compat import compat_str
 from ..utils import (
     ExtractorError,
     unified_strdate,
+    HEADRequest,
+    float_or_none,
 )
 
 
 class WatIE(InfoExtractor):
-    _VALID_URL = r'(?:wat:(?P<real_id>\d{8})|https?://www\.wat\.tv/video/(?P<display_id>.*)-(?P<short_id>.*?)_.*?\.html)'
+    _VALID_URL = r'(?:wat:|https?://(?:www\.)?wat\.tv/video/.*-)(?P<id>[0-9a-z]+)'
     IE_NAME = 'wat.tv'
     _TESTS = [
         {
             'url': 'http://www.wat.tv/video/soupe-figues-l-orange-aux-epices-6z1uz_2hvf7_.html',
-            'md5': 'ce70e9223945ed26a8056d413ca55dc9',
+            'md5': '83d882d9de5c9d97f0bb2c6273cde56a',
             'info_dict': {
                 'id': '11713067',
-                'display_id': 'soupe-figues-l-orange-aux-epices',
                 'ext': 'mp4',
                 'title': 'Soupe de figues à l\'orange et aux épices',
                 'description': 'Retrouvez l\'émission "Petits plats en équilibre", diffusée le 18 août 2014.',
@@ -33,7 +34,6 @@ class WatIE(InfoExtractor):
             'md5': 'fbc84e4378165278e743956d9c1bf16b',
             'info_dict': {
                 'id': '11713075',
-                'display_id': 'gregory-lemarchal-voix-ange',
                 'ext': 'mp4',
                 'title': 'Grégory Lemarchal, une voix d\'ange depuis 10 ans (1/3)',
                 'description': 'md5:b7a849cf16a2b733d9cd10c52906dee3',
@@ -44,96 +44,85 @@ class WatIE(InfoExtractor):
         },
     ]
 
-    def download_video_info(self, real_id):
+    def _real_extract(self, url):
+        video_id = self._match_id(url)
+        video_id = video_id if video_id.isdigit() and len(video_id) > 6 else compat_str(int(video_id, 36))
+
         # 'contentv4' is used in the website, but it also returns the related
         # videos, we don't need them
-        info = self._download_json('http://www.wat.tv/interface/contentv3/' + real_id, real_id)
-        return info['media']
-
-    def _real_extract(self, url):
-        def real_id_for_chapter(chapter):
-            return chapter['tc_start'].split('-')[0]
-        mobj = re.match(self._VALID_URL, url)
-        display_id = mobj.group('display_id')
-        real_id = mobj.group('real_id')
-        if not real_id:
-            short_id = mobj.group('short_id')
-            webpage = self._download_webpage(url, display_id or short_id)
-            real_id = self._search_regex(r'xtpage = ".*-(.*?)";', webpage, 'real id')
-
-        video_info = self.download_video_info(real_id)
+        video_info = self._download_json(
+            'http://www.wat.tv/interface/contentv3/' + video_id, video_id)['media']
 
         error_desc = video_info.get('error_desc')
         if error_desc:
             raise ExtractorError(
                 '%s returned error: %s' % (self.IE_NAME, error_desc), expected=True)
 
-        geo_list = video_info.get('geoList')
-        country = geo_list[0] if geo_list else ''
-
         chapters = video_info['chapters']
         first_chapter = chapters[0]
-        files = video_info['files']
-        first_file = files[0]
 
-        if real_id_for_chapter(first_chapter) != real_id:
+        def video_id_for_chapter(chapter):
+            return chapter['tc_start'].split('-')[0]
+
+        if video_id_for_chapter(first_chapter) != video_id:
             self.to_screen('Multipart video detected')
-            chapter_urls = []
-            for chapter in chapters:
-                chapter_id = real_id_for_chapter(chapter)
-                # Yes, when we this chapter is processed by WatIE,
-                # it will download the info again
-                chapter_info = self.download_video_info(chapter_id)
-                chapter_urls.append(chapter_info['url'])
-            entries = [self.url_result(chapter_url) for chapter_url in chapter_urls]
-            return self.playlist_result(entries, real_id, video_info['title'])
-
-        upload_date = None
-        if 'date_diffusion' in first_chapter:
-            upload_date = unified_strdate(first_chapter['date_diffusion'])
+            entries = [self.url_result('wat:%s' % video_id_for_chapter(chapter)) for chapter in chapters]
+            return self.playlist_result(entries, video_id, video_info['title'])
         # Otherwise we can continue and extract just one part, we have to use
-        # the short id for getting the video url
+        # the video id for getting the video url
 
-        formats = [{
-            'url': 'http://wat.tv/get/android5/%s.mp4' % real_id,
-            'format_id': 'Mobile',
-        }]
+        date_diffusion = first_chapter.get('date_diffusion')
+        upload_date = unified_strdate(date_diffusion) if date_diffusion else None
 
-        fmts = [('SD', 'web')]
-        if first_file.get('hasHD'):
-            fmts.append(('HD', 'webhd'))
+        def extract_url(path_template, url_type):
+            req_url = 'http://www.wat.tv/get/%s' % (path_template % video_id)
+            head = self._request_webpage(HEADRequest(req_url), video_id, 'Extracting %s url' % url_type)
+            red_url = head.geturl()
+            if req_url == red_url:
+                raise ExtractorError(
+                    '%s said: Sorry, this video is not available from your country.' % self.IE_NAME,
+                    expected=True)
+            return red_url
 
-        def compute_token(param):
-            timestamp = '%08x' % int(self._download_webpage(
-                'http://www.wat.tv/servertime', real_id,
-                'Downloading server time').split('|')[0])
-            magic = '9b673b13fa4682ed14c3cfa5af5310274b514c4133e9b3a81e6e3aba009l2564'
-            return '%s/%s' % (hashlib.md5((magic + param + timestamp).encode('ascii')).hexdigest(), timestamp)
+        m3u8_url = extract_url('ipad/%s.m3u8', 'm3u8')
+        http_url = extract_url('android5/%s.mp4', 'http')
 
-        for fmt in fmts:
-            webid = '/%s/%s' % (fmt[1], real_id)
-            video_url = self._download_webpage(
-                'http://www.wat.tv/get%s?token=%s&getURL=1&country=%s' % (webid, compute_token(webid), country),
-                real_id,
-                'Downloading %s video URL' % fmt[0],
-                'Failed to download %s video URL' % fmt[0],
-                False)
-            if not video_url:
+        formats = []
+        m3u8_formats = self._extract_m3u8_formats(
+            m3u8_url, video_id, 'mp4', 'm3u8_native', m3u8_id='hls')
+        formats.extend(m3u8_formats)
+        formats.extend(self._extract_f4m_formats(
+            m3u8_url.replace('ios.', 'web.').replace('.m3u8', '.f4m'),
+            video_id, f4m_id='hds', fatal=False))
+        for m3u8_format in m3u8_formats:
+            mobj = re.search(
+                r'audio.*?%3D(\d+)(?:-video.*?%3D(\d+))?', m3u8_format['url'])
+            if not mobj:
                 continue
-            formats.append({
-                'url': video_url,
-                'ext': 'mp4',
-                'format_id': fmt[0],
+            abr, vbr = mobj.groups()
+            abr, vbr = float_or_none(abr, 1000), float_or_none(vbr, 1000)
+            m3u8_format.update({
+                'vbr': vbr,
+                'abr': abr,
             })
+            if not vbr or not abr:
+                continue
+            f = m3u8_format.copy()
+            f.update({
+                'url': re.sub(r'%s-\d+00-\d+' % video_id, '%s-%d00-%d' % (video_id, round(vbr / 100), round(abr)), http_url),
+                'format_id': f['format_id'].replace('hls', 'http'),
+                'protocol': 'http',
+            })
+            formats.append(f)
+        self._sort_formats(formats)
 
         return {
-            'id': real_id,
-            'display_id': display_id,
+            'id': video_id,
             'title': first_chapter['title'],
             'thumbnail': first_chapter['preview'],
             'description': first_chapter['description'],
             'view_count': video_info['views'],
             'upload_date': upload_date,
-            'duration': first_file['duration'],
+            'duration': video_info['files'][0]['duration'],
             'formats': formats,
         }

youtube_dl/extractor/xhamster.py

@@ -12,37 +12,52 @@ from ..utils import (
 
 
 class XHamsterIE(InfoExtractor):
-    _VALID_URL = r'(?P<proto>https?)://(?:.+?\.)?xhamster\.com/movies/(?P<id>[0-9]+)/(?P<seo>.+?)\.html(?:\?.*)?'
-    _TESTS = [
-        {
-            'url': 'http://xhamster.com/movies/1509445/femaleagent_shy_beauty_takes_the_bait.html',
-            'info_dict': {
-                'id': '1509445',
-                'ext': 'mp4',
-                'title': 'FemaleAgent Shy beauty takes the bait',
-                'upload_date': '20121014',
-                'uploader': 'Ruseful2011',
-                'duration': 893.52,
-                'age_limit': 18,
-            }
+    _VALID_URL = r'(?P<proto>https?)://(?:.+?\.)?xhamster\.com/movies/(?P<id>[0-9]+)/(?P<seo>.*?)\.html(?:\?.*)?'
+    _TESTS = [{
+        'url': 'http://xhamster.com/movies/1509445/femaleagent_shy_beauty_takes_the_bait.html',
+        'md5': '8281348b8d3c53d39fffb377d24eac4e',
+        'info_dict': {
+            'id': '1509445',
+            'ext': 'mp4',
+            'title': 'FemaleAgent Shy beauty takes the bait',
+            'upload_date': '20121014',
+            'uploader': 'Ruseful2011',
+            'duration': 893.52,
+            'age_limit': 18,
         },
-        {
-            'url': 'http://xhamster.com/movies/2221348/britney_spears_sexy_booty.html?hd',
-            'info_dict': {
-                'id': '2221348',
-                'ext': 'mp4',
-                'title': 'Britney Spears  Sexy Booty',
-                'upload_date': '20130914',
-                'uploader': 'jojo747400',
-                'duration': 200.48,
-                'age_limit': 18,
-            }
+    }, {
+        'url': 'http://xhamster.com/movies/2221348/britney_spears_sexy_booty.html?hd',
+        'info_dict': {
+            'id': '2221348',
+            'ext': 'mp4',
+            'title': 'Britney Spears  Sexy Booty',
+            'upload_date': '20130914',
+            'uploader': 'jojo747400',
+            'duration': 200.48,
+            'age_limit': 18,
         },
-        {
-            'url': 'https://xhamster.com/movies/2272726/amber_slayed_by_the_knight.html',
-            'only_matching': True,
+        'params': {
+            'skip_download': True,
         },
-    ]
+    }, {
+        # empty seo
+        'url': 'http://xhamster.com/movies/5667973/.html',
+        'info_dict': {
+            'id': '5667973',
+            'ext': 'mp4',
+            'title': '....',
+            'upload_date': '20160208',
+            'uploader': 'parejafree',
+            'duration': 72.0,
+            'age_limit': 18,
+        },
+        'params': {
+            'skip_download': True,
+        },
+    }, {
+        'url': 'https://xhamster.com/movies/2272726/amber_slayed_by_the_knight.html',
+        'only_matching': True,
+    }]
 
     def _real_extract(self, url):
         def extract_video_url(webpage, name):
@@ -170,7 +185,7 @@ class XHamsterEmbedIE(InfoExtractor):
         webpage = self._download_webpage(url, video_id)
 
         video_url = self._search_regex(
-            r'href="(https?://xhamster\.com/movies/%s/[^"]+\.html[^"]*)"' % video_id,
+            r'href="(https?://xhamster\.com/movies/%s/[^"]*\.html[^"]*)"' % video_id,
             webpage, 'xhamster url', default=None)
 
         if not video_url:

youtube_dl/extractor/yandexmusic.py

@@ -20,18 +20,24 @@ class YandexMusicBaseIE(InfoExtractor):
             error = response.get('error')
             if error:
                 raise ExtractorError(error, expected=True)
+            if response.get('type') == 'captcha' or 'captcha' in response:
+                YandexMusicBaseIE._raise_captcha()
+
+    @staticmethod
+    def _raise_captcha():
+        raise ExtractorError(
+            'YandexMusic has considered youtube-dl requests automated and '
+            'asks you to solve a CAPTCHA. You can either wait for some '
+            'time until unblocked and optionally use --sleep-interval '
+            'in future or alternatively you can go to https://music.yandex.ru/ '
+            'solve CAPTCHA, then export cookies and pass cookie file to '
+            'youtube-dl with --cookies',
+            expected=True)
 
     def _download_webpage(self, *args, **kwargs):
         webpage = super(YandexMusicBaseIE, self)._download_webpage(*args, **kwargs)
         if 'Нам очень жаль, но запросы, поступившие с вашего IP-адреса, похожи на автоматические.' in webpage:
-            raise ExtractorError(
-                'YandexMusic has considered youtube-dl requests automated and '
-                'asks you to solve a CAPTCHA. You can either wait for some '
-                'time until unblocked and optionally use --sleep-interval '
-                'in future or alternatively you can go to https://music.yandex.ru/ '
-                'solve CAPTCHA, then export cookies and pass cookie file to '
-                'youtube-dl with --cookies',
-                expected=True)
+            self._raise_captcha()
         return webpage
 
     def _download_json(self, *args, **kwargs):

youtube_dl/extractor/youku.py

@@ -275,6 +275,8 @@ class YoukuIE(InfoExtractor):
                     'format_id': self.get_format_name(fm),
                     'ext': self.parse_ext_l(fm),
                     'filesize': int(seg['size']),
+                    'width': stream.get('width'),
+                    'height': stream.get('height'),
                 })
 
         return {

youtube_dl/options.py

@@ -395,8 +395,8 @@ def parseOpts(overrideArguments=None):
 
     downloader = optparse.OptionGroup(parser, 'Download Options')
     downloader.add_option(
-        '-r', '--rate-limit',
-        dest='ratelimit', metavar='LIMIT',
+        '-r', '--limit-rate', '--rate-limit',
+        dest='ratelimit', metavar='RATE',
         help='Maximum download rate in bytes per second (e.g. 50K or 4.2M)')
     downloader.add_option(
         '-R', '--retries',

youtube_dl/update.py

@@ -83,11 +83,8 @@ def update_self(to_screen, verbose, opener):
 
     print_notes(to_screen, versions_info['versions'])
 
-    filename = sys.argv[0]
-    # Py2EXE: Filename could be different
-    if hasattr(sys, 'frozen') and not os.path.isfile(filename):
-        if os.path.isfile(filename + '.exe'):
-            filename += '.exe'
+    # sys.executable is set to the full pathname of the exe-file for py2exe
+    filename = sys.executable if hasattr(sys, 'frozen') else sys.argv[0]
 
     if not os.access(filename, os.W_OK):
         to_screen('ERROR: no write permissions on %s' % filename)
@@ -95,7 +92,7 @@ def update_self(to_screen, verbose, opener):
 
     # Py2EXE
     if hasattr(sys, 'frozen'):
-        exe = os.path.abspath(filename)
+        exe = filename
         directory = os.path.dirname(exe)
         if not os.access(directory, os.W_OK):
             to_screen('ERROR: no write permissions on %s' % directory)

youtube_dl/utils.py

@@ -105,9 +105,9 @@ KNOWN_EXTENSIONS = (
     'f4f', 'f4m', 'm3u8', 'smil')
 
 # needed for sanitizing filenames in restricted mode
-ACCENT_CHARS = dict(zip('ÂÃÄÀÁÅÆÇÈÉÊËÌÍÎÏÐÑÒÓÔÕÖØŒÙÚÛÜÝÞßàáâãäåæçèéêëìíîïðñòóôõöøœùúûüýþÿ',
-                        itertools.chain('AAAAAA', ['AE'], 'CEEEEIIIIDNOOOOOO', ['OE'], 'UUUUYP', ['ss'],
-                                        'aaaaaa', ['ae'], 'ceeeeiiiionoooooo', ['oe'], 'uuuuypy')))
+ACCENT_CHARS = dict(zip('ÂÃÄÀÁÅÆÇÈÉÊËÌÍÎÏÐÑÒÓÔÕÖŐØŒÙÚÛÜŰÝÞßàáâãäåæçèéêëìíîïðñòóôõöőøœùúûüűýþÿ',
+                        itertools.chain('AAAAAA', ['AE'], 'CEEEEIIIIDNOOOOOOO', ['OE'], 'UUUUUYP', ['ss'],
+                                        'aaaaaa', ['ae'], 'ceeeeiiiionooooooo', ['oe'], 'uuuuuypy')))
 
 
 def preferredencoding():
@@ -861,9 +861,13 @@ class YoutubeDLHandler(compat_urllib_request.HTTPHandler):
                 # As of RFC 2616 default charset is iso-8859-1 that is respected by python 3
                 if sys.version_info >= (3, 0):
                     location = location.encode('iso-8859-1').decode('utf-8')
+                else:
+                    location = location.decode('utf-8')
                 location_escaped = escape_url(location)
                 if location != location_escaped:
                     del resp.headers['Location']
+                    if sys.version_info < (3, 0):
+                        location_escaped = location_escaped.encode('utf-8')
                     resp.headers['Location'] = location_escaped
         return resp
 
@@ -1035,6 +1039,7 @@ def unified_strdate(date_str, day_first=True):
         format_expressions.extend([
             '%d-%m-%Y',
             '%d.%m.%Y',
+            '%d.%m.%y',
             '%d/%m/%Y',
             '%d/%m/%y',
             '%d/%m/%Y %H:%M:%S',
@@ -1055,7 +1060,10 @@ def unified_strdate(date_str, day_first=True):
     if upload_date is None:
         timetuple = email.utils.parsedate_tz(date_str)
         if timetuple:
-            upload_date = datetime.datetime(*timetuple[:6]).strftime('%Y%m%d')
+            try:
+                upload_date = datetime.datetime(*timetuple[:6]).strftime('%Y%m%d')
+            except ValueError:
+                pass
     if upload_date is not None:
         return compat_str(upload_date)
 
@@ -1907,7 +1915,7 @@ def parse_age_limit(s):
 
 def strip_jsonp(code):
     return re.sub(
-        r'(?s)^[a-zA-Z0-9_.]+\s*\(\s*(.*)\);?\s*?(?://[^\n]*)*$', r'\1', code)
+        r'(?s)^[a-zA-Z0-9_.$]+\s*\(\s*(.*)\);?\s*?(?://[^\n]*)*$', r'\1', code)
 
 
 def js_to_json(code):

youtube_dl/version.py

@@ -1,3 +1,3 @@
 from __future__ import unicode_literals
 
-__version__ = '2016.05.21.2'
+__version__ = '2016.06.03'