release 2013.10.18

It downloads with a m3u8 manifest, requires ffmpeg.

.gitignore

@@ -17,4 +17,12 @@ youtube-dl.tar.gz
 .coverage
 cover/
 updates_key.pem
-*.egg-info
+*.egg-info
+*.srt
+*.sbv
+*.vtt
+*.flv
+*.mp4
+*.part
+test/testdata
+.tox

.travis.yml

@@ -9,6 +9,7 @@ notifications:
     - filippo.valsorda@gmail.com
     - phihag@phihag.de
     - jaime.marquinez.ferrandiz+travis@gmail.com
+    - yasoob.khld@gmail.com
 #  irc:
 #    channels:
 #      - "irc.freenode.org#youtube-dl"

Makefile

@@ -13,13 +13,13 @@ PYTHON=/usr/bin/env python
 
 # set SYSCONFDIR to /etc if PREFIX=/usr or PREFIX=/usr/local
 ifeq ($(PREFIX),/usr)
-    SYSCONFDIR=/etc
+	SYSCONFDIR=/etc
 else
-    ifeq ($(PREFIX),/usr/local)
-        SYSCONFDIR=/etc
-    else
-        SYSCONFDIR=$(PREFIX)/etc
-    endif
+	ifeq ($(PREFIX),/usr/local)
+		SYSCONFDIR=/etc
+	else
+		SYSCONFDIR=$(PREFIX)/etc
+	endif
 endif
 
 install: youtube-dl youtube-dl.1 youtube-dl.bash-completion
@@ -71,6 +71,7 @@ youtube-dl.tar.gz: youtube-dl README.md README.txt youtube-dl.1 youtube-dl.bash-
 		--exclude '*~' \
 		--exclude '__pycache' \
 		--exclude '.git' \
+		--exclude 'testdata' \
 		-- \
 		bin devscripts test youtube_dl \
 		CHANGELOG LICENSE README.md README.txt \

README.md

@@ -16,8 +16,11 @@ which means you can modify it, redistribute it or use it however you like.
 # OPTIONS
     -h, --help                 print this help text and exit
     --version                  print program version and exit
-    -U, --update               update this program to latest version
-    -i, --ignore-errors        continue on download errors
+    -U, --update               update this program to latest version. Make sure
+                               that you have sufficient permissions (run with
+                               sudo if needed)
+    -i, --ignore-errors        continue on download errors, for example to to
+                               skip unavailable videos in a playlist
     --dump-user-agent          display the current browser identification
     --user-agent UA            specify a custom user agent
     --referer REF              specify a custom referer, use if the video access
@@ -27,6 +30,11 @@ which means you can modify it, redistribute it or use it however you like.
     --extractor-descriptions   Output descriptions of all supported extractors
     --proxy URL                Use the specified HTTP/HTTPS proxy
     --no-check-certificate     Suppress HTTPS certificate validation.
+    --cache-dir None           Location in the filesystem where youtube-dl can
+                               store downloaded information permanently. By
+                               default $XDG_CACHE_HOME/youtube-dl or ~/.cache
+                               /youtube-dl .
+    --no-cache-dir             Disable filesystem caching
 
 ## Video Selection:
     --playlist-start NUMBER    playlist video to start at (default is 1)
@@ -43,11 +51,16 @@ which means you can modify it, redistribute it or use it however you like.
     --date DATE                download only videos uploaded in this date
     --datebefore DATE          download only videos uploaded before this date
     --dateafter DATE           download only videos uploaded after this date
+    --no-playlist              download only the currently playing video
+    --age-limit YEARS          download only videos suitable for the given age
+    --download-archive FILE    Download only videos not present in the archive
+                               file. Record all downloaded videos in it.
 
 ## Download Options:
-    -r, --rate-limit LIMIT     maximum download rate (e.g. 50k or 44.6m)
+    -r, --rate-limit LIMIT     maximum download rate in bytes per second (e.g.
+                               50K or 4.2M)
     -R, --retries RETRIES      number of retries (default is 10)
-    --buffer-size SIZE         size of download buffer (e.g. 1024 or 16k)
+    --buffer-size SIZE         size of download buffer (e.g. 1024 or 16K)
                                (default is 1024)
     --no-resize-buffer         do not automatically adjust the buffer size. By
                                default, the buffer size is automatically resized
@@ -88,6 +101,7 @@ which means you can modify it, redistribute it or use it however you like.
                                file modification time
     --write-description        write video description to a .description file
     --write-info-json          write video metadata to a .info.json file
+    --write-annotations        write video annotations to a .annotation file
     --write-thumbnail          write thumbnail image to disk
 
 ## Verbosity / Simulation Options:
@@ -111,25 +125,26 @@ which means you can modify it, redistribute it or use it however you like.
 
 ## Video Format Options:
     -f, --format FORMAT        video format code, specifiy the order of
-                               preference using slashes: "-f 22/17/18"
+                               preference using slashes: "-f 22/17/18". "-f mp4"
+                               and "-f flv" are also supported
     --all-formats              download all available video formats
     --prefer-free-formats      prefer free video formats unless a specific one
                                is requested
     --max-quality FORMAT       highest quality format to download
     -F, --list-formats         list all available formats (currently youtube
                                only)
-    --write-sub                write subtitle file (currently youtube only)
-    --write-auto-sub           write automatic subtitle file (currently youtube
-                               only)
-    --only-sub                 [deprecated] alias of --skip-download
+
+## Subtitle Options:
+    --write-sub                write subtitle file
+    --write-auto-sub           write automatic subtitle file (youtube only)
     --all-subs                 downloads all the available subtitles of the
-                               video (currently youtube only)
+                               video
     --list-subs                lists all available subtitles for the video
-                               (currently youtube only)
-    --sub-format FORMAT        subtitle format [srt/sbv/vtt] (default=srt)
-                               (currently youtube only)
-    --sub-lang LANG            language of the subtitles to download (optional)
-                               use IETF language tags like 'en'
+    --sub-format FORMAT        subtitle format (default=srt) ([sbv/vtt] youtube
+                               only)
+    --sub-lang LANGS           languages of the subtitles to download (optional)
+                               separated by commas, use IETF language tags like
+                               'en,pt'
 
 ## Authentication Options:
     -u, --username USERNAME    account username
@@ -151,6 +166,9 @@ which means you can modify it, redistribute it or use it however you like.
                                processing; the video is erased by default
     --no-post-overwrites       do not overwrite post-processed files; the post-
                                processed files are overwritten by default
+    --embed-subs               embed subtitles in the video (only for mp4
+                               videos)
+    --add-metadata             add metadata to the files
 
 # CONFIGURATION
 

devscripts/bash-completion.in

@@ -4,8 +4,12 @@ __youtube-dl()
     COMPREPLY=()
     cur="${COMP_WORDS[COMP_CWORD]}"
     opts="{{flags}}"
+    keywords=":ytfavorites :ytrecommended :ytsubscriptions :ytwatchlater"
 
-    if [[ ${cur} == * ]] ; then
+    if [[ ${cur} =~ : ]]; then
+        COMPREPLY=( $(compgen -W "${keywords}" -- ${cur}) )
+        return 0
+    elif [[ ${cur} == * ]] ; then
         COMPREPLY=( $(compgen -W "${opts}" -- ${cur}) )
         return 0
     fi

devscripts/buildserver.py

@@ -0,0 +1,405 @@
+#!/usr/bin/python3
+
+from http.server import HTTPServer, BaseHTTPRequestHandler
+from socketserver import ThreadingMixIn
+import argparse
+import ctypes
+import functools
+import sys
+import threading
+import traceback
+import os.path
+
+
+class BuildHTTPServer(ThreadingMixIn, HTTPServer):
+    allow_reuse_address = True
+
+
+advapi32 = ctypes.windll.advapi32
+
+SC_MANAGER_ALL_ACCESS = 0xf003f
+SC_MANAGER_CREATE_SERVICE = 0x02
+SERVICE_WIN32_OWN_PROCESS = 0x10
+SERVICE_AUTO_START = 0x2
+SERVICE_ERROR_NORMAL = 0x1
+DELETE = 0x00010000
+SERVICE_STATUS_START_PENDING = 0x00000002
+SERVICE_STATUS_RUNNING = 0x00000004
+SERVICE_ACCEPT_STOP = 0x1
+
+SVCNAME = 'youtubedl_builder'
+
+LPTSTR = ctypes.c_wchar_p
+START_CALLBACK = ctypes.WINFUNCTYPE(None, ctypes.c_int, ctypes.POINTER(LPTSTR))
+
+
+class SERVICE_TABLE_ENTRY(ctypes.Structure):
+    _fields_ = [
+        ('lpServiceName', LPTSTR),
+        ('lpServiceProc', START_CALLBACK)
+    ]
+
+
+HandlerEx = ctypes.WINFUNCTYPE(
+    ctypes.c_int,     # return
+    ctypes.c_int,     # dwControl
+    ctypes.c_int,     # dwEventType
+    ctypes.c_void_p,  # lpEventData,
+    ctypes.c_void_p,  # lpContext,
+)
+
+
+def _ctypes_array(c_type, py_array):
+    ar = (c_type * len(py_array))()
+    ar[:] = py_array
+    return ar
+
+
+def win_OpenSCManager():
+    res = advapi32.OpenSCManagerW(None, None, SC_MANAGER_ALL_ACCESS)
+    if not res:
+        raise Exception('Opening service manager failed - '
+                        'are you running this as administrator?')
+    return res
+
+
+def win_install_service(service_name, cmdline):
+    manager = win_OpenSCManager()
+    try:
+        h = advapi32.CreateServiceW(
+            manager, service_name, None,
+            SC_MANAGER_CREATE_SERVICE, SERVICE_WIN32_OWN_PROCESS,
+            SERVICE_AUTO_START, SERVICE_ERROR_NORMAL,
+            cmdline, None, None, None, None, None)
+        if not h:
+            raise OSError('Service creation failed: %s' % ctypes.FormatError())
+
+        advapi32.CloseServiceHandle(h)
+    finally:
+        advapi32.CloseServiceHandle(manager)
+
+
+def win_uninstall_service(service_name):
+    manager = win_OpenSCManager()
+    try:
+        h = advapi32.OpenServiceW(manager, service_name, DELETE)
+        if not h:
+            raise OSError('Could not find service %s: %s' % (
+                service_name, ctypes.FormatError()))
+
+        try:
+            if not advapi32.DeleteService(h):
+                raise OSError('Deletion failed: %s' % ctypes.FormatError())
+        finally:
+            advapi32.CloseServiceHandle(h)
+    finally:
+        advapi32.CloseServiceHandle(manager)
+
+
+def win_service_report_event(service_name, msg, is_error=True):
+    with open('C:/sshkeys/log', 'a', encoding='utf-8') as f:
+        f.write(msg + '\n')
+
+    event_log = advapi32.RegisterEventSourceW(None, service_name)
+    if not event_log:
+        raise OSError('Could not report event: %s' % ctypes.FormatError())
+
+    try:
+        type_id = 0x0001 if is_error else 0x0004
+        event_id = 0xc0000000 if is_error else 0x40000000
+        lines = _ctypes_array(LPTSTR, [msg])
+
+        if not advapi32.ReportEventW(
+                event_log, type_id, 0, event_id, None, len(lines), 0,
+                lines, None):
+            raise OSError('Event reporting failed: %s' % ctypes.FormatError())
+    finally:
+        advapi32.DeregisterEventSource(event_log)
+
+
+def win_service_handler(stop_event, *args):
+    try:
+        raise ValueError('Handler called with args ' + repr(args))
+        TODO
+    except Exception as e:
+        tb = traceback.format_exc()
+        msg = str(e) + '\n' + tb
+        win_service_report_event(service_name, msg, is_error=True)
+        raise
+
+
+def win_service_set_status(handle, status_code):
+    svcStatus = SERVICE_STATUS()
+    svcStatus.dwServiceType = SERVICE_WIN32_OWN_PROCESS
+    svcStatus.dwCurrentState = status_code
+    svcStatus.dwControlsAccepted = SERVICE_ACCEPT_STOP
+
+    svcStatus.dwServiceSpecificExitCode = 0
+
+    if not advapi32.SetServiceStatus(handle, ctypes.byref(svcStatus)):
+        raise OSError('SetServiceStatus failed: %r' % ctypes.FormatError())
+
+
+def win_service_main(service_name, real_main, argc, argv_raw):
+    try:
+        #args = [argv_raw[i].value for i in range(argc)]
+        stop_event = threading.Event()
+        handler = HandlerEx(functools.partial(stop_event, win_service_handler))
+        h = advapi32.RegisterServiceCtrlHandlerExW(service_name, handler, None)
+        if not h:
+            raise OSError('Handler registration failed: %s' %
+                          ctypes.FormatError())
+
+        TODO
+    except Exception as e:
+        tb = traceback.format_exc()
+        msg = str(e) + '\n' + tb
+        win_service_report_event(service_name, msg, is_error=True)
+        raise
+
+
+def win_service_start(service_name, real_main):
+    try:
+        cb = START_CALLBACK(
+            functools.partial(win_service_main, service_name, real_main))
+        dispatch_table = _ctypes_array(SERVICE_TABLE_ENTRY, [
+            SERVICE_TABLE_ENTRY(
+                service_name,
+                cb
+            ),
+            SERVICE_TABLE_ENTRY(None, ctypes.cast(None, START_CALLBACK))
+        ])
+
+        if not advapi32.StartServiceCtrlDispatcherW(dispatch_table):
+            raise OSError('ctypes start failed: %s' % ctypes.FormatError())
+    except Exception as e:
+        tb = traceback.format_exc()
+        msg = str(e) + '\n' + tb
+        win_service_report_event(service_name, msg, is_error=True)
+        raise
+
+
+def main(args=None):
+    parser = argparse.ArgumentParser()
+    parser.add_argument('-i', '--install',
+                        action='store_const', dest='action', const='install',
+                        help='Launch at Windows startup')
+    parser.add_argument('-u', '--uninstall',
+                        action='store_const', dest='action', const='uninstall',
+                        help='Remove Windows service')
+    parser.add_argument('-s', '--service',
+                        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',
+                        help='Bind to host:port (default %default)')
+    options = parser.parse_args(args=args)
+
+    if options.action == 'install':
+        fn = os.path.abspath(__file__).replace('v:', '\\\\vboxsrv\\vbox')
+        cmdline = '%s %s -s -b %s' % (sys.executable, fn, options.bind)
+        win_install_service(SVCNAME, cmdline)
+        return
+
+    if options.action == 'uninstall':
+        win_uninstall_service(SVCNAME)
+        return
+
+    if options.action == 'service':
+        win_service_start(SVCNAME, main)
+        return
+
+    host, port_str = options.bind.split(':')
+    port = int(port_str)
+
+    print('Listening on %s:%d' % (host, port))
+    srv = BuildHTTPServer((host, port), BuildHTTPRequestHandler)
+    thr = threading.Thread(target=srv.serve_forever)
+    thr.start()
+    input('Press ENTER to shut down')
+    srv.shutdown()
+    thr.join()
+
+
+def rmtree(path):
+    for name in os.listdir(path):
+        fname = os.path.join(path, name)
+        if os.path.isdir(fname):
+            rmtree(fname)
+        else:
+            os.chmod(fname, 0o666)
+            os.remove(fname)
+    os.rmdir(path)
+
+#==============================================================================
+
+class BuildError(Exception):
+    def __init__(self, output, code=500):
+        self.output = output
+        self.code = code
+
+    def __str__(self):
+        return self.output
+
+
+class HTTPError(BuildError):
+    pass
+
+
+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)
+            try:
+                self.pythonPath, _ = _winreg.QueryValueEx(key, '')
+            finally:
+                _winreg.CloseKey(key)
+        except Exception:
+            raise BuildError('No such Python version: %s' % pythonVersion)
+
+        super(PythonBuilder, self).__init__(**kwargs)
+
+
+class GITInfoBuilder(object):
+    def __init__(self, **kwargs):
+        try:
+            self.user, self.repoName = kwargs['path'][:2]
+            self.rev = kwargs.pop('rev')
+        except ValueError:
+            raise BuildError('Invalid path')
+        except KeyError as e:
+            raise BuildError('Missing mandatory parameter "%s"' % e.args[0])
+
+        path = os.path.join(os.environ['APPDATA'], 'Build archive', self.repoName, self.user)
+        if not os.path.exists(path):
+            os.makedirs(path)
+        self.basePath = tempfile.mkdtemp(dir=path)
+        self.buildPath = os.path.join(self.basePath, 'build')
+
+        super(GITInfoBuilder, self).__init__(**kwargs)
+
+
+class GITBuilder(GITInfoBuilder):
+    def build(self):
+        try:
+            subprocess.check_output(['git', 'clone', 'git://github.com/%s/%s.git' % (self.user, self.repoName), self.buildPath])
+            subprocess.check_output(['git', 'checkout', self.rev], cwd=self.buildPath)
+        except subprocess.CalledProcessError as e:
+            raise BuildError(e.output)
+
+        super(GITBuilder, self).build()
+
+
+class YoutubeDLBuilder(object):
+    authorizedUsers = ['fraca7', 'phihag', 'rg3', 'FiloSottile']
+
+    def __init__(self, **kwargs):
+        if self.repoName != 'youtube-dl':
+            raise BuildError('Invalid repository "%s"' % self.repoName)
+        if self.user not in self.authorizedUsers:
+            raise HTTPError('Unauthorized user "%s"' % self.user, 401)
+
+        super(YoutubeDLBuilder, self).__init__(**kwargs)
+
+    def build(self):
+        try:
+            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)
+
+        super(YoutubeDLBuilder, self).build()
+
+
+class DownloadBuilder(object):
+    def __init__(self, **kwargs):
+        self.handler = kwargs.pop('handler')
+        self.srcPath = os.path.join(self.buildPath, *tuple(kwargs['path'][2:]))
+        self.srcPath = os.path.abspath(os.path.normpath(self.srcPath))
+        if not self.srcPath.startswith(self.buildPath):
+            raise HTTPError(self.srcPath, 401)
+
+        super(DownloadBuilder, self).__init__(**kwargs)
+
+    def build(self):
+        if not os.path.exists(self.srcPath):
+            raise HTTPError('No such file', 404)
+        if os.path.isdir(self.srcPath):
+            raise HTTPError('Is a directory: %s' % self.srcPath, 401)
+
+        self.handler.send_response(200)
+        self.handler.send_header('Content-Type', 'application/octet-stream')
+        self.handler.send_header('Content-Disposition', 'attachment; filename=%s' % os.path.split(self.srcPath)[-1])
+        self.handler.send_header('Content-Length', str(os.stat(self.srcPath).st_size))
+        self.handler.end_headers()
+
+        with open(self.srcPath, 'rb') as src:
+            shutil.copyfileobj(src, self.handler.wfile)
+
+        super(DownloadBuilder, self).build()
+
+
+class CleanupTempDir(object):
+    def build(self):
+        try:
+            rmtree(self.basePath)
+        except Exception as e:
+            print('WARNING deleting "%s": %s' % (self.basePath, e))
+
+        super(CleanupTempDir, self).build()
+
+
+class Null(object):
+    def __init__(self, **kwargs):
+        pass
+
+    def start(self):
+        pass
+
+    def close(self):
+        pass
+
+    def build(self):
+        pass
+
+
+class Builder(PythonBuilder, GITBuilder, YoutubeDLBuilder, DownloadBuilder, CleanupTempDir, Null):
+    pass
+
+
+class BuildHTTPRequestHandler(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()])
+        action, _, path = path.path.strip('/').partition('/')
+        if path:
+            path = path.split('/')
+            if action in self.actionDict:
+                try:
+                    builder = self.actionDict[action](path=path, handler=self, **paramDict)
+                    builder.start()
+                    try:
+                        builder.build()
+                    finally:
+                        builder.close()
+                except BuildError as e:
+                    self.send_response(e.code)
+                    msg = unicode(e).encode('UTF-8')
+                    self.send_header('Content-Type', 'text/plain; charset=UTF-8')
+                    self.send_header('Content-Length', len(msg))
+                    self.end_headers()
+                    self.wfile.write(msg)
+                except HTTPError as e:
+                    self.send_response(e.code, str(e))
+            else:
+                self.send_response(500, 'Unknown build method "%s"' % action)
+        else:
+            self.send_response(500, 'Malformed URL')
+
+#==============================================================================
+
+if __name__ == '__main__':
+    main()

devscripts/gh-pages/add-version.py

@@ -3,31 +3,40 @@
 import json
 import sys
 import hashlib
-import urllib.request
+import os.path
+
 
 if len(sys.argv) <= 1:
-	print('Specify the version number as parameter')
-	sys.exit()
+    print('Specify the version number as parameter')
+    sys.exit()
 version = sys.argv[1]
 
 with open('update/LATEST_VERSION', 'w') as f:
-	f.write(version)
+    f.write(version)
 
 versions_info = json.load(open('update/versions.json'))
 if 'signature' in versions_info:
-	del versions_info['signature']
+    del versions_info['signature']
 
 new_version = {}
 
-filenames = {'bin': 'youtube-dl', 'exe': 'youtube-dl.exe', 'tar': 'youtube-dl-%s.tar.gz' % version}
+filenames = {
+    'bin': 'youtube-dl',
+    'exe': 'youtube-dl.exe',
+    'tar': 'youtube-dl-%s.tar.gz' % version}
+build_dir = os.path.join('..', '..', 'build', version)
 for key, filename in filenames.items():
-	print('Downloading and checksumming %s...' %filename)
-	url = 'http://youtube-dl.org/downloads/%s/%s' % (version, filename)
-	data = urllib.request.urlopen(url).read()
-	sha256sum = hashlib.sha256(data).hexdigest()
-	new_version[key] = (url, sha256sum)
+    url = 'https://yt-dl.org/downloads/%s/%s' % (version, filename)
+    fn = os.path.join(build_dir, filename)
+    with open(fn, 'rb') as f:
+        data = f.read()
+    if not data:
+        raise ValueError('File %s is empty!' % fn)
+    sha256sum = hashlib.sha256(data).hexdigest()
+    new_version[key] = (url, sha256sum)
 
 versions_info['versions'][version] = new_version
 versions_info['latest'] = version
 
-json.dump(versions_info, open('update/versions.json', 'w'), indent=4, sort_keys=True)
+with open('update/versions.json', 'w') as jsonf:
+    json.dump(versions_info, jsonf, indent=4, sort_keys=True)

devscripts/gh-pages/update-feed.py

@@ -22,7 +22,7 @@ entry_template=textwrap.dedent("""
 									<atom:link href="http://rg3.github.io/youtube-dl" />
 									<atom:content type="xhtml">
 										<div xmlns="http://www.w3.org/1999/xhtml">
-											Downloads available at <a href="http://youtube-dl.org/downloads/@VERSION@/">http://youtube-dl.org/downloads/@VERSION@/</a>
+											Downloads available at <a href="https://yt-dl.org/downloads/@VERSION@/">https://yt-dl.org/downloads/@VERSION@/</a>
 										</div>
 									</atom:content>
 									<atom:author>
@@ -54,4 +54,3 @@ atom_template = atom_template.replace('@ENTRIES@', entries_str)
 with open('update/releases.atom','w',encoding='utf-8') as atom_file:
 	atom_file.write(atom_template)
 
-

devscripts/gh-pages/update-sites.py

@@ -0,0 +1,34 @@
+#!/usr/bin/env python3
+
+import sys
+import os
+import textwrap
+
+# We must be able to import youtube_dl
+sys.path.append(os.path.dirname(os.path.dirname(os.path.dirname(os.path.abspath(__file__)))))
+
+import youtube_dl
+
+def main():
+    with open('supportedsites.html.in', 'r', encoding='utf-8') as tmplf:
+        template = tmplf.read()
+
+    ie_htmls = []
+    for ie in sorted(youtube_dl.gen_extractors(), key=lambda i: i.IE_NAME.lower()):
+        ie_html = '<b>{}</b>'.format(ie.IE_NAME)
+        ie_desc = getattr(ie, 'IE_DESC', None)
+        if ie_desc is False:
+            continue
+        elif ie_desc is not None:
+            ie_html += ': {}'.format(ie.IE_DESC)
+        if ie.working() == False:
+            ie_html += ' (Currently broken)'
+        ie_htmls.append('<li>{}</li>'.format(ie_html))
+
+    template = template.replace('@SITES@', textwrap.indent('\n'.join(ie_htmls), '\t'))
+
+    with open('supportedsites.html', 'w', encoding='utf-8') as sitesf:
+        sitesf.write(template)
+
+if __name__ == '__main__':
+    main()

devscripts/release.sh

@@ -55,8 +55,8 @@ git push origin "$version"
 /bin/echo -e "\n### OK, now it is time to build the binaries..."
 REV=$(git rev-parse HEAD)
 make youtube-dl youtube-dl.tar.gz
-wget "http://jeromelaheurte.net:8142/download/rg3/youtube-dl/youtube-dl.exe?rev=$REV" -O youtube-dl.exe || \
-	wget "http://jeromelaheurte.net:8142/build/rg3/youtube-dl/youtube-dl.exe?rev=$REV" -O youtube-dl.exe
+read -p "VM running? (y/n) " -n 1
+wget "http://localhost:8142/build/rg3/youtube-dl/youtube-dl.exe?rev=$REV" -O youtube-dl.exe
 mkdir -p "build/$version"
 mv youtube-dl youtube-dl.exe "build/$version"
 mv youtube-dl.tar.gz "build/$version/youtube-dl-$version.tar.gz"
@@ -67,7 +67,7 @@ RELEASE_FILES="youtube-dl youtube-dl.exe youtube-dl-$version.tar.gz"
 (cd build/$version/ && sha512sum $RELEASE_FILES > SHA2-512SUMS)
 git checkout HEAD -- youtube-dl youtube-dl.exe
 
-/bin/echo -e "\n### Signing and uploading the new binaries to youtube-dl.org..."
+/bin/echo -e "\n### Signing and uploading the new binaries to yt-dl.org ..."
 for f in $RELEASE_FILES; do gpg --detach-sig "build/$version/$f"; done
 scp -r "build/$version" ytdl@yt-dl.org:html/tmp/
 ssh ytdl@yt-dl.org "mv html/tmp/$version html/downloads/"
@@ -85,6 +85,7 @@ ROOT=$(pwd)
     "$ROOT/devscripts/gh-pages/sign-versions.py" < "$ROOT/updates_key.pem"
     "$ROOT/devscripts/gh-pages/generate-download.py"
     "$ROOT/devscripts/gh-pages/update-copyright.py"
+    "$ROOT/devscripts/gh-pages/update-sites.py"
     git add *.html *.html.in update
     git commit -m "release $version"
     git show HEAD

devscripts/youtube_genalgo.py

@@ -1,83 +0,0 @@
-#!/usr/bin/env python
-
-# Generate youtube signature algorithm from test cases
-
-import sys
-
-tests = [
-    # 88
-    ("qwertyuioplkjhgfdsazxcvbnm1234567890QWERTYUIOPLKJHGFDSAZXCVBNM!@#$%^&*()_-+={[]}|:;?/>.<",
-     "J:|}][{=+-_)(*&;%$#@>MNBVCXZASDFGH^KLPOIUYTREWQ0987654321mnbvcxzasdfghrklpoiuytej"),
-    # 87
-    ("qwertyuioplkjhgfdsazxcvbnm1234567890QWERTYUIOPLKJHGFDSAZXCVBNM!@#$^&*()_-+={[]}|:;?/>.<",
-     "!?;:|}][{=+-_)(*&^$#@/MNBVCXZASqFGHJKLPOIUYTREWQ0987654321mnbvcxzasdfghjklpoiuytr"),
-    # 86 - vfl_ymO4Z 2013/06/27
-    ("qwertyuioplkjhgfdsazxcvbnm1234567890QWERTYUIOPLKJHGFDSAZXCVBNM!@#$%^&*()_-+={[|};?/>.<",
-     "ertyuioplkjhgfdsazxcvbnm1234567890QWERTYUIOPLKJHGFDSAZXCVBNM!/#$%^&*()_-+={[|};?@"),
-    # 85
-    ("qwertyuioplkjhgfdsazxcvbnm1234567890QWERTYUIOPLKJHGFDSAZXCVBNM!@#$%^&*()_-+={[};?/>.<",
-     "{>/?;}[.=+-_)(*&^%$#@!MqBVCXZASDFwHJKLPOIUYTREWQ0987654321mnbvcxzasdfghjklpoiuytr"),
-    # 84
-    ("qwertyuioplkjhgfdsazxcvbnm1234567890QWERTYUIOPLKJHGFDSAZXCVBNM!@#$%^&*()_-+={[};?>.<",
-     "<.>?;}[{=+-_)(*&^%$#@!MNBVCXZASDFGHJKLPOIUYTREWe098765432rmnbvcxzasdfghjklpoiuyt1"),
-    # 83
-    ("qwertyuioplkjhgfdsazxcvbnm1234567890QWERTYUIOPLKJHGFDSAZXCVBNM!#$%^&*()_+={[};?/>.<",
-     "D.>/?;}[{=+_)(*&^%$#!MNBVCXeAS<FGHJKLPOIUYTREWZ0987654321mnbvcxzasdfghjklpoiuytrQ"),
-    # 82
-    ("qwertyuioplkjhgfdsazxcvbnm1234567890QWERTYUIOPLKHGFDSAZXCVBNM!@#$%^&*(-+={[};?/>.<",
-     "Q>/?;}[{=+-(*<^%$#@!MNBVCXZASDFGHKLPOIUY8REWT0q&7654321mnbvcxzasdfghjklpoiuytrew9"),
-]
-
-def find_matching(wrong, right):
-    idxs = [wrong.index(c) for c in right]
-    return compress(idxs)
-    return ('s[%d]' % i for i in idxs)
-
-def compress(idxs):
-    def _genslice(start, end, step):
-        starts = '' if start == 0 else str(start)
-        ends = ':%d' % (end+step)
-        steps = '' if step == 1 else (':%d' % step)
-        return 's[%s%s%s]' % (starts, ends, steps)
-
-    step = None
-    for i, prev in zip(idxs[1:], idxs[:-1]):
-        if step is not None:
-            if i - prev == step:
-                continue
-            yield _genslice(start, prev, step)
-            step = None
-            continue
-        if i - prev in [-1, 1]:
-            step = i - prev
-            start = prev
-            continue
-        else:
-            yield 's[%d]' % prev
-    if step is None:
-        yield 's[%d]' % i
-    else:
-        yield _genslice(start, i, step)
-
-def _assert_compress(inp, exp):
-    res = list(compress(inp))
-    if res != exp:
-        print('Got %r, expected %r' % (res, exp))
-        assert res == exp
-_assert_compress([0,2,4,6], ['s[0]', 's[2]', 's[4]', 's[6]'])
-_assert_compress([0,1,2,4,6,7], ['s[:3]', 's[4]', 's[6:8]'])
-_assert_compress([8,0,1,2,4,7,6,9], ['s[8]', 's[:3]', 's[4]', 's[7:5:-1]', 's[9]'])
-
-def gen(wrong, right, indent):
-    code = ' + '.join(find_matching(wrong, right))
-    return 'if len(s) == %d:\n%s    return %s\n' % (len(wrong), indent, code)
-
-def genall(tests):
-    indent = ' ' * 8
-    return indent + (indent + 'el').join(gen(wrong, right, indent) for wrong,right in tests)
-
-def main():
-    print(genall(tests))
-
-if __name__ == '__main__':
-    main()

test/helper.py

@@ -1,38 +1,63 @@
+import errno
 import io
+import hashlib
 import json
 import os.path
+import re
+import types
 
 import youtube_dl.extractor
-from youtube_dl import YoutubeDL, YoutubeDLHandler
-from youtube_dl.utils import (
-    compat_cookiejar,
-    compat_urllib_request,
-)
+from youtube_dl import YoutubeDL
 
-# General configuration (from __init__, not very elegant...)
-jar = compat_cookiejar.CookieJar()
-cookie_processor = compat_urllib_request.HTTPCookieProcessor(jar)
-proxy_handler = compat_urllib_request.ProxyHandler()
-opener = compat_urllib_request.build_opener(proxy_handler, cookie_processor, YoutubeDLHandler())
-compat_urllib_request.install_opener(opener)
 
-PARAMETERS_FILE = os.path.join(os.path.dirname(os.path.abspath(__file__)), "parameters.json")
-with io.open(PARAMETERS_FILE, encoding='utf-8') as pf:
-    parameters = json.load(pf)
+def global_setup():
+    youtube_dl._setup_opener(timeout=10)
+
+
+def get_params(override=None):
+    PARAMETERS_FILE = os.path.join(os.path.dirname(os.path.abspath(__file__)),
+                                   "parameters.json")
+    with io.open(PARAMETERS_FILE, encoding='utf-8') as pf:
+        parameters = json.load(pf)
+    if override:
+        parameters.update(override)
+    return parameters
+
+
+def try_rm(filename):
+    """ Remove a file if it exists """
+    try:
+        os.remove(filename)
+    except OSError as ose:
+        if ose.errno != errno.ENOENT:
+            raise
+
 
 class FakeYDL(YoutubeDL):
-    def __init__(self):
-        self.result = []
+    def __init__(self, override=None):
         # Different instances of the downloader can't share the same dictionary
         # some test set the "sublang" parameter, which would break the md5 checks.
-        self.params = dict(parameters)
-    def to_screen(self, s):
+        params = get_params(override=override)
+        super(FakeYDL, self).__init__(params)
+        self.result = []
+        
+    def to_screen(self, s, skip_eol=None):
         print(s)
+
     def trouble(self, s, tb=None):
         raise Exception(s)
+
     def download(self, x):
         self.result.append(x)
 
+    def expect_warning(self, regex):
+        # Silence an expected warning matching a regex
+        old_report_warning = self.report_warning
+        def report_warning(self, message):
+            if re.match(regex, message): return
+            old_report_warning(message)
+        self.report_warning = types.MethodType(report_warning, self)
+
 def get_testcases():
     for ie in youtube_dl.extractor.gen_extractors():
         t = getattr(ie, '_TEST', None)
@@ -42,3 +67,6 @@ def get_testcases():
         for t in getattr(ie, '_TESTS', []):
             t['name'] = type(ie).__name__[:-len('IE')]
             yield t
+
+
+md5 = lambda s: hashlib.md5(s.encode('utf-8')).hexdigest()

test/parameters.json

@@ -38,7 +38,6 @@
     "writedescription": false, 
     "writeinfojson": true, 
     "writesubtitles": false,
-    "onlysubtitles": false,
     "allsubtitles": false,
     "listssubtitles": false
 }

test/test_YoutubeDL.py

@@ -0,0 +1,99 @@
+#!/usr/bin/env python
+
+# Allow direct execution
+import os
+import sys
+import unittest
+sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
+
+from test.helper import FakeYDL
+
+
+class YDL(FakeYDL):
+    def __init__(self, *args, **kwargs):
+        super(YDL, self).__init__(*args, **kwargs)
+        self.downloaded_info_dicts = []
+        self.msgs = []
+
+    def process_info(self, info_dict):
+        self.downloaded_info_dicts.append(info_dict)
+
+    def to_screen(self, msg):
+        self.msgs.append(msg)
+
+
+class TestFormatSelection(unittest.TestCase):
+    def test_prefer_free_formats(self):
+        # Same resolution => download webm
+        ydl = YDL()
+        ydl.params['prefer_free_formats'] = True
+        formats = [
+            {u'ext': u'webm', u'height': 460},
+            {u'ext': u'mp4',  u'height': 460},
+        ]
+        info_dict = {u'formats': formats, u'extractor': u'test'}
+        ydl.process_ie_result(info_dict)
+        downloaded = ydl.downloaded_info_dicts[0]
+        self.assertEqual(downloaded[u'ext'], u'webm')
+
+        # Different resolution => download best quality (mp4)
+        ydl = YDL()
+        ydl.params['prefer_free_formats'] = True
+        formats = [
+            {u'ext': u'webm', u'height': 720},
+            {u'ext': u'mp4', u'height': 1080},
+        ]
+        info_dict[u'formats'] = formats
+        ydl.process_ie_result(info_dict)
+        downloaded = ydl.downloaded_info_dicts[0]
+        self.assertEqual(downloaded[u'ext'], u'mp4')
+
+        # No prefer_free_formats => keep original formats order
+        ydl = YDL()
+        ydl.params['prefer_free_formats'] = False
+        formats = [
+            {u'ext': u'webm', u'height': 720},
+            {u'ext': u'flv', u'height': 720},
+        ]
+        info_dict[u'formats'] = formats
+        ydl.process_ie_result(info_dict)
+        downloaded = ydl.downloaded_info_dicts[0]
+        self.assertEqual(downloaded[u'ext'], u'flv')
+
+    def test_format_limit(self):
+        formats = [
+            {u'format_id': u'meh'},
+            {u'format_id': u'good'},
+            {u'format_id': u'great'},
+            {u'format_id': u'excellent'},
+        ]
+        info_dict = {
+            u'formats': formats, u'extractor': u'test', 'id': 'testvid'}
+
+        ydl = YDL()
+        ydl.process_ie_result(info_dict)
+        downloaded = ydl.downloaded_info_dicts[0]
+        self.assertEqual(downloaded[u'format_id'], u'excellent')
+
+        ydl = YDL({'format_limit': 'good'})
+        assert ydl.params['format_limit'] == 'good'
+        ydl.process_ie_result(info_dict)
+        downloaded = ydl.downloaded_info_dicts[0]
+        self.assertEqual(downloaded[u'format_id'], u'good')
+
+        ydl = YDL({'format_limit': 'great', 'format': 'all'})
+        ydl.process_ie_result(info_dict)
+        self.assertEqual(ydl.downloaded_info_dicts[0][u'format_id'], u'meh')
+        self.assertEqual(ydl.downloaded_info_dicts[1][u'format_id'], u'good')
+        self.assertEqual(ydl.downloaded_info_dicts[2][u'format_id'], u'great')
+        self.assertTrue('3' in ydl.msgs[0])
+
+        ydl = YDL()
+        ydl.params['format_limit'] = 'excellent'
+        ydl.process_ie_result(info_dict)
+        downloaded = ydl.downloaded_info_dicts[0]
+        self.assertEqual(downloaded[u'format_id'], u'excellent')
+
+
+if __name__ == '__main__':
+    unittest.main()

test/test_age_restriction.py

@@ -0,0 +1,55 @@
+#!/usr/bin/env python
+
+# Allow direct execution
+import os
+import sys
+import unittest
+sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
+
+from test.helper import global_setup, try_rm
+global_setup()
+
+
+from youtube_dl import YoutubeDL
+
+
+def _download_restricted(url, filename, age):
+    """ Returns true iff the file has been downloaded """
+
+    params = {
+        'age_limit': age,
+        'skip_download': True,
+        'writeinfojson': True,
+        "outtmpl": "%(id)s.%(ext)s",
+    }
+    ydl = YoutubeDL(params)
+    ydl.add_default_info_extractors()
+    json_filename = filename + '.info.json'
+    try_rm(json_filename)
+    ydl.download([url])
+    res = os.path.exists(json_filename)
+    try_rm(json_filename)
+    return res
+
+
+class TestAgeRestriction(unittest.TestCase):
+    def _assert_restricted(self, url, filename, age, old_age=None):
+        self.assertTrue(_download_restricted(url, filename, old_age))
+        self.assertFalse(_download_restricted(url, filename, age))
+
+    def test_youtube(self):
+        self._assert_restricted('07FYdnEawAQ', '07FYdnEawAQ.mp4', 10)
+
+    def test_youporn(self):
+        self._assert_restricted(
+            'http://www.youporn.com/watch/505835/sex-ed-is-it-safe-to-masturbate-daily/',
+            '505835.mp4', 2, old_age=25)
+
+    def test_pornotube(self):
+        self._assert_restricted(
+            'http://pornotube.com/c/173/m/1689755/Marilyn-Monroe-Bathing',
+            '1689755.flv', 13)
+
+
+if __name__ == '__main__':
+    unittest.main()

test/test_all_urls.py

@@ -1,34 +1,66 @@
 #!/usr/bin/env python
 
-import sys
-import unittest
-
 # Allow direct execution
 import os
-sys.path.append(os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
+import sys
+import unittest
+sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
+
+
+from test.helper import get_testcases
+
+from youtube_dl.extractor import (
+    gen_extractors,
+    JustinTVIE,
+    YoutubeIE,
+)
 
-from youtube_dl.extractor import YoutubeIE, YoutubePlaylistIE, YoutubeChannelIE, JustinTVIE, gen_extractors
-from helper import get_testcases
 
 class TestAllURLsMatching(unittest.TestCase):
+    def setUp(self):
+        self.ies = gen_extractors()
+
+    def matching_ies(self, url):
+        return [ie.IE_NAME for ie in self.ies if ie.suitable(url) and ie.IE_NAME != 'generic']
+
+    def assertMatch(self, url, ie_list):
+        self.assertEqual(self.matching_ies(url), ie_list)
+
     def test_youtube_playlist_matching(self):
-        self.assertTrue(YoutubePlaylistIE.suitable(u'ECUl4u3cNGP61MdtwGTqZA0MreSaDybji8'))
-        self.assertTrue(YoutubePlaylistIE.suitable(u'UUBABnxM4Ar9ten8Mdjj1j0Q')) #585
-        self.assertTrue(YoutubePlaylistIE.suitable(u'PL63F0C78739B09958'))
-        self.assertTrue(YoutubePlaylistIE.suitable(u'https://www.youtube.com/playlist?list=UUBABnxM4Ar9ten8Mdjj1j0Q'))
-        self.assertTrue(YoutubePlaylistIE.suitable(u'https://www.youtube.com/course?list=ECUl4u3cNGP61MdtwGTqZA0MreSaDybji8'))
-        self.assertTrue(YoutubePlaylistIE.suitable(u'https://www.youtube.com/playlist?list=PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC'))
-        self.assertTrue(YoutubePlaylistIE.suitable(u'https://www.youtube.com/watch?v=AV6J6_AeFEQ&playnext=1&list=PL4023E734DA416012')) #668
-        self.assertFalse(YoutubePlaylistIE.suitable(u'PLtS2H6bU1M'))
+        assertPlaylist = lambda url: self.assertMatch(url, ['youtube:playlist'])
+        assertPlaylist(u'ECUl4u3cNGP61MdtwGTqZA0MreSaDybji8')
+        assertPlaylist(u'UUBABnxM4Ar9ten8Mdjj1j0Q') #585
+        assertPlaylist(u'PL63F0C78739B09958')
+        assertPlaylist(u'https://www.youtube.com/playlist?list=UUBABnxM4Ar9ten8Mdjj1j0Q')
+        assertPlaylist(u'https://www.youtube.com/course?list=ECUl4u3cNGP61MdtwGTqZA0MreSaDybji8')
+        assertPlaylist(u'https://www.youtube.com/playlist?list=PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC')
+        assertPlaylist(u'https://www.youtube.com/watch?v=AV6J6_AeFEQ&playnext=1&list=PL4023E734DA416012') #668
+        self.assertFalse('youtube:playlist' in self.matching_ies(u'PLtS2H6bU1M'))
 
     def test_youtube_matching(self):
         self.assertTrue(YoutubeIE.suitable(u'PLtS2H6bU1M'))
         self.assertFalse(YoutubeIE.suitable(u'https://www.youtube.com/watch?v=AV6J6_AeFEQ&playnext=1&list=PL4023E734DA416012')) #668
+        self.assertMatch('http://youtu.be/BaW_jenozKc', ['youtube'])
+        self.assertMatch('http://www.youtube.com/v/BaW_jenozKc', ['youtube'])
+        self.assertMatch('https://youtube.googleapis.com/v/BaW_jenozKc', ['youtube'])
 
     def test_youtube_channel_matching(self):
-        self.assertTrue(YoutubeChannelIE.suitable('https://www.youtube.com/channel/HCtnHdj3df7iM'))
-        self.assertTrue(YoutubeChannelIE.suitable('https://www.youtube.com/channel/HCtnHdj3df7iM?feature=gb_ch_rec'))
-        self.assertTrue(YoutubeChannelIE.suitable('https://www.youtube.com/channel/HCtnHdj3df7iM/videos'))
+        assertChannel = lambda url: self.assertMatch(url, ['youtube:channel'])
+        assertChannel('https://www.youtube.com/channel/HCtnHdj3df7iM')
+        assertChannel('https://www.youtube.com/channel/HCtnHdj3df7iM?feature=gb_ch_rec')
+        assertChannel('https://www.youtube.com/channel/HCtnHdj3df7iM/videos')
+
+    def test_youtube_user_matching(self):
+        self.assertMatch('www.youtube.com/NASAgovVideo/videos', ['youtube:user'])
+
+    def test_youtube_feeds(self):
+        self.assertMatch('https://www.youtube.com/feed/watch_later', ['youtube:watch_later'])
+        self.assertMatch('https://www.youtube.com/feed/subscriptions', ['youtube:subscriptions'])
+        self.assertMatch('https://www.youtube.com/feed/recommended', ['youtube:recommended'])
+        self.assertMatch('https://www.youtube.com/my_favorites', ['youtube:favorites'])
+
+    def test_youtube_show_matching(self):
+        self.assertMatch('http://www.youtube.com/show/airdisasters', ['youtube:show'])
 
     def test_justin_tv_channelid_matching(self):
         self.assertTrue(JustinTVIE.suitable(u"justin.tv/vanillatv"))
@@ -47,9 +79,13 @@ class TestAllURLsMatching(unittest.TestCase):
         self.assertTrue(JustinTVIE.suitable(u"http://www.twitch.tv/tsm_theoddone/c/2349361"))
 
     def test_youtube_extract(self):
-        self.assertEqual(YoutubeIE()._extract_id('http://www.youtube.com/watch?&v=BaW_jenozKc'), 'BaW_jenozKc')
-        self.assertEqual(YoutubeIE()._extract_id('https://www.youtube.com/watch?&v=BaW_jenozKc'), 'BaW_jenozKc')
-        self.assertEqual(YoutubeIE()._extract_id('https://www.youtube.com/watch?feature=player_embedded&v=BaW_jenozKc'), 'BaW_jenozKc')
+        assertExtractId = lambda url, id: self.assertEqual(YoutubeIE()._extract_id(url), id)
+        assertExtractId('http://www.youtube.com/watch?&v=BaW_jenozKc', 'BaW_jenozKc')
+        assertExtractId('https://www.youtube.com/watch?&v=BaW_jenozKc', 'BaW_jenozKc')
+        assertExtractId('https://www.youtube.com/watch?feature=player_embedded&v=BaW_jenozKc', 'BaW_jenozKc')
+        assertExtractId('https://www.youtube.com/watch_popup?v=BaW_jenozKc', 'BaW_jenozKc')
+        assertExtractId('http://www.youtube.com/watch?v=BaW_jenozKcsharePLED17F32AD9753930', 'BaW_jenozKc')
+        assertExtractId('BaW_jenozKc', 'BaW_jenozKc')
 
     def test_no_duplicates(self):
         ies = gen_extractors()
@@ -61,6 +97,14 @@ class TestAllURLsMatching(unittest.TestCase):
                 else:
                     self.assertFalse(ie.suitable(url), '%s should not match URL %r' % (type(ie).__name__, url))
 
+    def test_keywords(self):
+        self.assertMatch(':ytsubs', ['youtube:subscriptions'])
+        self.assertMatch(':ytsubscriptions', ['youtube:subscriptions'])
+        self.assertMatch(':thedailyshow', ['ComedyCentral'])
+        self.assertMatch(':tds', ['ComedyCentral'])
+        self.assertMatch(':colbertreport', ['ComedyCentral'])
+        self.assertMatch(':cr', ['ComedyCentral'])
+
 
 if __name__ == '__main__':
     unittest.main()

test/test_dailymotion_subtitles.py

@@ -0,0 +1,70 @@
+#!/usr/bin/env python
+
+# Allow direct execution
+import os
+import sys
+import unittest
+sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
+
+from test.helper import FakeYDL, global_setup, md5
+global_setup()
+
+
+from youtube_dl.extractor import DailymotionIE
+
+class TestDailymotionSubtitles(unittest.TestCase):
+    def setUp(self):
+        self.DL = FakeYDL()
+        self.url = 'http://www.dailymotion.com/video/xczg00'
+    def getInfoDict(self):
+        IE = DailymotionIE(self.DL)
+        info_dict = IE.extract(self.url)
+        return info_dict
+    def getSubtitles(self):
+        info_dict = self.getInfoDict()
+        return info_dict[0]['subtitles']
+    def test_no_writesubtitles(self):
+        subtitles = self.getSubtitles()
+        self.assertEqual(subtitles, None)
+    def test_subtitles(self):
+        self.DL.params['writesubtitles'] = True
+        subtitles = self.getSubtitles()
+        self.assertEqual(md5(subtitles['en']), '976553874490cba125086bbfea3ff76f')
+    def test_subtitles_lang(self):
+        self.DL.params['writesubtitles'] = True
+        self.DL.params['subtitleslangs'] = ['fr']
+        subtitles = self.getSubtitles()
+        self.assertEqual(md5(subtitles['fr']), '594564ec7d588942e384e920e5341792')
+    def test_allsubtitles(self):
+        self.DL.params['writesubtitles'] = True
+        self.DL.params['allsubtitles'] = True
+        subtitles = self.getSubtitles()
+        self.assertEqual(len(subtitles.keys()), 5)
+    def test_list_subtitles(self):
+        self.DL.expect_warning(u'Automatic Captions not supported by this server')
+        self.DL.params['listsubtitles'] = True
+        info_dict = self.getInfoDict()
+        self.assertEqual(info_dict, None)
+    def test_automatic_captions(self):
+        self.DL.expect_warning(u'Automatic Captions not supported by this server')
+        self.DL.params['writeautomaticsub'] = True
+        self.DL.params['subtitleslang'] = ['en']
+        subtitles = self.getSubtitles()
+        self.assertTrue(len(subtitles.keys()) == 0)
+    def test_nosubtitles(self):
+        self.DL.expect_warning(u'video doesn\'t have subtitles')
+        self.url = 'http://www.dailymotion.com/video/x12u166_le-zapping-tele-star-du-08-aout-2013_tv'
+        self.DL.params['writesubtitles'] = True
+        self.DL.params['allsubtitles'] = True
+        subtitles = self.getSubtitles()
+        self.assertEqual(len(subtitles), 0)
+    def test_multiple_langs(self):
+        self.DL.params['writesubtitles'] = True
+        langs = ['es', 'fr', 'de']
+        self.DL.params['subtitleslangs'] = langs
+        subtitles = self.getSubtitles()
+        for lang in langs:
+            self.assertTrue(subtitles.get(lang) is not None, u'Subtitles for \'%s\' not extracted' % lang)
+
+if __name__ == '__main__':
+    unittest.main()

test/test_download.py

@@ -1,43 +1,31 @@
 #!/usr/bin/env python
 
-import errno
+# Allow direct execution
+import os
+import sys
+import unittest
+sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
+
+from test.helper import get_params, get_testcases, global_setup, try_rm, md5
+global_setup()
+
+
 import hashlib
 import io
-import os
 import json
-import unittest
-import sys
 import socket
-import binascii
-
-# Allow direct execution
-sys.path.append(os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
 
 import youtube_dl.YoutubeDL
-from youtube_dl.utils import *
-
-PARAMETERS_FILE = os.path.join(os.path.dirname(os.path.abspath(__file__)), "parameters.json")
+from youtube_dl.utils import (
+    compat_str,
+    compat_urllib_error,
+    DownloadError,
+    ExtractorError,
+    UnavailableVideoError,
+)
 
 RETRIES = 3
 
-# General configuration (from __init__, not very elegant...)
-jar = compat_cookiejar.CookieJar()
-cookie_processor = compat_urllib_request.HTTPCookieProcessor(jar)
-proxy_handler = compat_urllib_request.ProxyHandler()
-opener = compat_urllib_request.build_opener(proxy_handler, cookie_processor, YoutubeDLHandler())
-compat_urllib_request.install_opener(opener)
-socket.setdefaulttimeout(10)
-
-def _try_rm(filename):
-    """ Remove a file if it exists """
-    try:
-        os.remove(filename)
-    except OSError as ose:
-        if ose.errno != errno.ENOENT:
-            raise
-
-md5 = lambda s: hashlib.md5(s.encode('utf-8')).hexdigest()
-
 class YoutubeDL(youtube_dl.YoutubeDL):
     def __init__(self, *args, **kwargs):
         self.to_stderr = self.to_screen
@@ -54,17 +42,12 @@ def _file_md5(fn):
     with open(fn, 'rb') as f:
         return hashlib.md5(f.read()).hexdigest()
 
-from helper import get_testcases
 defs = get_testcases()
 
-with io.open(PARAMETERS_FILE, encoding='utf-8') as pf:
-    parameters = json.load(pf)
-
 
 class TestDownload(unittest.TestCase):
     maxDiff = None
     def setUp(self):
-        self.parameters = parameters
         self.defs = defs
 
 ### Dynamically generate tests
@@ -84,8 +67,7 @@ def generator(test_case):
             print_skipping(test_case['skip'])
             return
 
-        params = self.parameters.copy()
-        params.update(test_case.get('params', {}))
+        params = get_params(test_case.get('params', {}))
 
         ydl = YoutubeDL(params)
         ydl.add_default_info_extractors()
@@ -97,9 +79,9 @@ def generator(test_case):
 
         test_cases = test_case.get('playlist', [test_case])
         for tc in test_cases:
-            _try_rm(tc['file'])
-            _try_rm(tc['file'] + '.part')
-            _try_rm(tc['file'] + '.info.json')
+            try_rm(tc['file'])
+            try_rm(tc['file'] + '.part')
+            try_rm(tc['file'] + '.info.json')
         try:
             for retry in range(1, RETRIES + 1):
                 try:
@@ -127,12 +109,11 @@ def generator(test_case):
                     info_dict = json.load(infof)
                 for (info_field, expected) in tc.get('info_dict', {}).items():
                     if isinstance(expected, compat_str) and expected.startswith('md5:'):
-                        self.assertEqual(expected, 'md5:' + md5(info_dict.get(info_field)))
+                        got = 'md5:' + md5(info_dict.get(info_field))
                     else:
                         got = info_dict.get(info_field)
-                        self.assertEqual(
-                            expected, got,
-                            u'invalid value for field %s, expected %r, got %r' % (info_field, expected, got))
+                    self.assertEqual(expected, got,
+                        u'invalid value for field %s, expected %r, got %r' % (info_field, expected, got))
 
                 # If checkable fields are missing from the test case, print the info_dict
                 test_info_dict = dict((key, value if not isinstance(value, compat_str) or len(value) < 250 else 'md5:' + md5(value))
@@ -146,9 +127,9 @@ def generator(test_case):
                     self.assertTrue(key in info_dict.keys() and info_dict[key])
         finally:
             for tc in test_cases:
-                _try_rm(tc['file'])
-                _try_rm(tc['file'] + '.part')
-                _try_rm(tc['file'] + '.info.json')
+                try_rm(tc['file'])
+                try_rm(tc['file'] + '.part')
+                try_rm(tc['file'] + '.info.json')
 
     return test_template
 

test/test_playlists.py

@@ -0,0 +1,89 @@
+#!/usr/bin/env python
+# encoding: utf-8
+
+
+# Allow direct execution
+import os
+import sys
+import unittest
+sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
+
+from test.helper import FakeYDL, global_setup
+global_setup()
+
+
+from youtube_dl.extractor import (
+    DailymotionPlaylistIE,
+    DailymotionUserIE,
+    VimeoChannelIE,
+    UstreamChannelIE,
+    SoundcloudUserIE,
+    LivestreamIE,
+    NHLVideocenterIE,
+)
+
+
+class TestPlaylists(unittest.TestCase):
+    def assertIsPlaylist(self, info):
+        """Make sure the info has '_type' set to 'playlist'"""
+        self.assertEqual(info['_type'], 'playlist')
+
+    def test_dailymotion_playlist(self):
+        dl = FakeYDL()
+        ie = DailymotionPlaylistIE(dl)
+        result = ie.extract('http://www.dailymotion.com/playlist/xv4bw_nqtv_sport/1#video=xl8v3q')
+        self.assertIsPlaylist(result)
+        self.assertEqual(result['title'], u'SPORT')
+        self.assertTrue(len(result['entries']) > 20)
+
+    def test_dailymotion_user(self):
+        dl = FakeYDL()
+        ie = DailymotionUserIE(dl)
+        result = ie.extract('http://www.dailymotion.com/user/generation-quoi/')
+        self.assertIsPlaylist(result)
+        self.assertEqual(result['title'], u'Génération Quoi')
+        self.assertTrue(len(result['entries']) >= 26)
+
+    def test_vimeo_channel(self):
+        dl = FakeYDL()
+        ie = VimeoChannelIE(dl)
+        result = ie.extract('http://vimeo.com/channels/tributes')
+        self.assertIsPlaylist(result)
+        self.assertEqual(result['title'], u'Vimeo Tributes')
+        self.assertTrue(len(result['entries']) > 24)
+
+    def test_ustream_channel(self):
+        dl = FakeYDL()
+        ie = UstreamChannelIE(dl)
+        result = ie.extract('http://www.ustream.tv/channel/young-americans-for-liberty')
+        self.assertIsPlaylist(result)
+        self.assertEqual(result['id'], u'5124905')
+        self.assertTrue(len(result['entries']) >= 11)
+
+    def test_soundcloud_user(self):
+        dl = FakeYDL()
+        ie = SoundcloudUserIE(dl)
+        result = ie.extract('https://soundcloud.com/the-concept-band')
+        self.assertIsPlaylist(result)
+        self.assertEqual(result['id'], u'9615865')
+        self.assertTrue(len(result['entries']) >= 12)
+
+    def test_livestream_event(self):
+        dl = FakeYDL()
+        ie = LivestreamIE(dl)
+        result = ie.extract('http://new.livestream.com/tedx/cityenglish')
+        self.assertIsPlaylist(result)
+        self.assertEqual(result['title'], u'TEDCity2.0 (English)')
+        self.assertTrue(len(result['entries']) >= 4)
+
+    def test_nhl_videocenter(self):
+        dl = FakeYDL()
+        ie = NHLVideocenterIE(dl)
+        result = ie.extract('http://video.canucks.nhl.com/videocenter/console?catid=999')
+        self.assertIsPlaylist(result)
+        self.assertEqual(result['id'], u'999')
+        self.assertEqual(result['title'], u'Highlights')
+        self.assertEqual(len(result['entries']), 12)
+
+if __name__ == '__main__':
+    unittest.main()

test/test_utils.py

@@ -1,21 +1,30 @@
 #!/usr/bin/env python
-
-# Various small unit tests
-
-import sys
-import unittest
+# coding: utf-8
 
 # Allow direct execution
 import os
-sys.path.append(os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
+import sys
+import unittest
+sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
+
+
+# Various small unit tests
+import xml.etree.ElementTree
 
 #from youtube_dl.utils import htmlentity_transform
-from youtube_dl.utils import timeconvert
-from youtube_dl.utils import sanitize_filename
-from youtube_dl.utils import unescapeHTML
-from youtube_dl.utils import orderedSet
-from youtube_dl.utils import DateRange
-from youtube_dl.utils import unified_strdate
+from youtube_dl.utils import (
+    timeconvert,
+    sanitize_filename,
+    unescapeHTML,
+    orderedSet,
+    DateRange,
+    unified_strdate,
+    find_xpath_attr,
+    get_meta_content,
+    xpath_with_ns,
+    smuggle_url,
+    unsmuggle_url,
+)
 
 if sys.version_info < (3, 0):
     _compat_str = lambda b: b.decode('unicode-escape')
@@ -112,5 +121,55 @@ class TestUtil(unittest.TestCase):
         self.assertEqual(unified_strdate('Dec 14, 2012'), '20121214')
         self.assertEqual(unified_strdate('2012/10/11 01:56:38 +0000'), '20121011')
 
+    def test_find_xpath_attr(self):
+        testxml = u'''<root>
+            <node/>
+            <node x="a"/>
+            <node x="a" y="c" />
+            <node x="b" y="d" />
+        </root>'''
+        doc = xml.etree.ElementTree.fromstring(testxml)
+
+        self.assertEqual(find_xpath_attr(doc, './/fourohfour', 'n', 'v'), None)
+        self.assertEqual(find_xpath_attr(doc, './/node', 'x', 'a'), doc[1])
+        self.assertEqual(find_xpath_attr(doc, './/node', 'y', 'c'), doc[2])
+
+    def test_meta_parser(self):
+        testhtml = u'''
+        <head>
+            <meta name="description" content="foo &amp; bar">
+            <meta content='Plato' name='author'/>
+        </head>
+        '''
+        get_meta = lambda name: get_meta_content(name, testhtml)
+        self.assertEqual(get_meta('description'), u'foo & bar')
+        self.assertEqual(get_meta('author'), 'Plato')
+
+    def test_xpath_with_ns(self):
+        testxml = u'''<root xmlns:media="http://example.com/">
+            <media:song>
+                <media:author>The Author</media:author>
+                <url>http://server.com/download.mp3</url>
+            </media:song>
+        </root>'''
+        doc = xml.etree.ElementTree.fromstring(testxml)
+        find = lambda p: doc.find(xpath_with_ns(p, {'media': 'http://example.com/'}))
+        self.assertTrue(find('media:song') is not None)
+        self.assertEqual(find('media:song/media:author').text, u'The Author')
+        self.assertEqual(find('media:song/url').text, u'http://server.com/download.mp3')
+
+    def test_smuggle_url(self):
+        data = {u"ö": u"ö", u"abc": [3]}
+        url = 'https://foo.bar/baz?x=y#a'
+        smug_url = smuggle_url(url, data)
+        unsmug_url, unsmug_data = unsmuggle_url(smug_url)
+        self.assertEqual(url, unsmug_url)
+        self.assertEqual(data, unsmug_data)
+
+        res_url, res_data = unsmuggle_url(url)
+        self.assertEqual(res_url, url)
+        self.assertEqual(res_data, None)
+
+
 if __name__ == '__main__':
     unittest.main()

test/test_write_annotations.py

@@ -0,0 +1,80 @@
+#!/usr/bin/env python
+# coding: utf-8
+
+# Allow direct execution
+import os
+import sys
+import unittest
+sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
+
+from test.helper import get_params, global_setup, try_rm
+global_setup()
+
+
+import io
+
+import xml.etree.ElementTree
+
+import youtube_dl.YoutubeDL
+import youtube_dl.extractor
+
+
+class YoutubeDL(youtube_dl.YoutubeDL):
+    def __init__(self, *args, **kwargs):
+        super(YoutubeDL, self).__init__(*args, **kwargs)
+        self.to_stderr = self.to_screen
+
+params = get_params({
+    'writeannotations': True,
+    'skip_download': True,
+    'writeinfojson': False,
+    'format': 'flv',
+})
+
+
+
+TEST_ID = 'gr51aVj-mLg'
+ANNOTATIONS_FILE = TEST_ID + '.flv.annotations.xml'
+EXPECTED_ANNOTATIONS = ['Speech bubble', 'Note', 'Title', 'Spotlight', 'Label']
+
+class TestAnnotations(unittest.TestCase):
+    def setUp(self):
+        # Clear old files
+        self.tearDown()
+
+
+    def test_info_json(self):
+        expected = list(EXPECTED_ANNOTATIONS) #Two annotations could have the same text.
+        ie = youtube_dl.extractor.YoutubeIE()
+        ydl = YoutubeDL(params)
+        ydl.add_info_extractor(ie)
+        ydl.download([TEST_ID])
+        self.assertTrue(os.path.exists(ANNOTATIONS_FILE))
+        annoxml = None
+        with io.open(ANNOTATIONS_FILE, 'r', encoding='utf-8') as annof:
+                annoxml = xml.etree.ElementTree.parse(annof)
+        self.assertTrue(annoxml is not None, 'Failed to parse annotations XML')
+        root = annoxml.getroot()
+        self.assertEqual(root.tag, 'document')
+        annotationsTag = root.find('annotations')
+        self.assertEqual(annotationsTag.tag, 'annotations')
+        annotations = annotationsTag.findall('annotation')
+
+        #Not all the annotations have TEXT children and the annotations are returned unsorted.
+        for a in annotations:
+                self.assertEqual(a.tag, 'annotation')
+                if a.get('type') == 'text':
+                        textTag = a.find('TEXT')
+                        text = textTag.text
+                        self.assertTrue(text in expected) #assertIn only added in python 2.7
+                        #remove the first occurance, there could be more than one annotation with the same text
+                        expected.remove(text)
+        #We should have seen (and removed) all the expected annotation texts.
+        self.assertEqual(len(expected), 0, 'Not all expected annotations were found.')
+        
+
+    def tearDown(self):
+        try_rm(ANNOTATIONS_FILE)
+
+if __name__ == '__main__':
+    unittest.main()

test/test_write_info_json.py

@@ -1,37 +1,34 @@
 #!/usr/bin/env python
 # coding: utf-8
 
-import json
+# Allow direct execution
 import os
 import sys
 import unittest
+sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
 
-# Allow direct execution
-sys.path.append(os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
+from test.helper import get_params, global_setup
+global_setup()
+
+
+import io
+import json
 
 import youtube_dl.YoutubeDL
 import youtube_dl.extractor
-from youtube_dl.utils import *
 
-PARAMETERS_FILE = os.path.join(os.path.dirname(os.path.abspath(__file__)), "parameters.json")
-
-# General configuration (from __init__, not very elegant...)
-jar = compat_cookiejar.CookieJar()
-cookie_processor = compat_urllib_request.HTTPCookieProcessor(jar)
-proxy_handler = compat_urllib_request.ProxyHandler()
-opener = compat_urllib_request.build_opener(proxy_handler, cookie_processor, YoutubeDLHandler())
-compat_urllib_request.install_opener(opener)
 
 class YoutubeDL(youtube_dl.YoutubeDL):
     def __init__(self, *args, **kwargs):
         super(YoutubeDL, self).__init__(*args, **kwargs)
         self.to_stderr = self.to_screen
 
-with io.open(PARAMETERS_FILE, encoding='utf-8') as pf:
-    params = json.load(pf)
-params['writeinfojson'] = True
-params['skip_download'] = True
-params['writedescription'] = True
+params = get_params({
+    'writeinfojson': True,
+    'skip_download': True,
+    'writedescription': True,
+})
+
 
 TEST_ID = 'BaW_jenozKc'
 INFO_JSON_FILE = TEST_ID + '.mp4.info.json'
@@ -42,6 +39,7 @@ This is a test video for youtube-dl.
 
 For more information, contact phihag@phihag.de .'''
 
+
 class TestInfoJSON(unittest.TestCase):
     def setUp(self):
         # Clear old files

test/test_youtube_lists.py

@@ -1,20 +1,26 @@
 #!/usr/bin/env python
 
-import sys
-import unittest
-import json
-
 # Allow direct execution
 import os
-sys.path.append(os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
+import sys
+import unittest
+sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
 
-from youtube_dl.extractor import YoutubeUserIE, YoutubePlaylistIE, YoutubeIE, YoutubeChannelIE, YoutubeShowIE
-from youtube_dl.utils import *
+from test.helper import FakeYDL, global_setup
+global_setup()
+
+
+from youtube_dl.extractor import (
+    YoutubeUserIE,
+    YoutubePlaylistIE,
+    YoutubeIE,
+    YoutubeChannelIE,
+    YoutubeShowIE,
+)
 
-from helper import FakeYDL
 
 class TestYoutubeLists(unittest.TestCase):
-    def assertIsPlaylist(self,info):
+    def assertIsPlaylist(self, info):
         """Make sure the info has '_type' set to 'playlist'"""
         self.assertEqual(info['_type'], 'playlist')
 
@@ -27,6 +33,14 @@ class TestYoutubeLists(unittest.TestCase):
         ytie_results = [YoutubeIE()._extract_id(url['url']) for url in result['entries']]
         self.assertEqual(ytie_results, [ 'bV9L5Ht9LgY', 'FXxLjLQi3Fg', 'tU3Bgo5qJZE'])
 
+    def test_youtube_playlist_noplaylist(self):
+        dl = FakeYDL()
+        dl.params['noplaylist'] = True
+        ie = YoutubePlaylistIE(dl)
+        result = ie.extract('https://www.youtube.com/watch?v=FXxLjLQi3Fg&list=PLwiyx1dc3P2JR9N8gQaQN_BCvlSlap7re')
+        self.assertEqual(result['_type'], 'url')
+        self.assertEqual(YoutubeIE()._extract_id(result['url']), 'FXxLjLQi3Fg')
+
     def test_issue_673(self):
         dl = FakeYDL()
         ie = YoutubePlaylistIE(dl)
@@ -92,7 +106,7 @@ class TestYoutubeLists(unittest.TestCase):
         dl = FakeYDL()
         ie = YoutubeShowIE(dl)
         result = ie.extract('http://www.youtube.com/show/airdisasters')
-        self.assertTrue(len(result) >= 4)
+        self.assertTrue(len(result) >= 3)
 
 if __name__ == '__main__':
     unittest.main()

test/test_youtube_sig.py

@@ -1,57 +0,0 @@
-#!/usr/bin/env python
-
-import unittest
-import sys
-
-# Allow direct execution
-import os
-sys.path.append(os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
-
-from youtube_dl.extractor.youtube import YoutubeIE
-from helper import FakeYDL
-
-sig = YoutubeIE(FakeYDL())._decrypt_signature
-
-class TestYoutubeSig(unittest.TestCase):
-    def test_43_43(self):
-        wrong = '5AEEAE0EC39677BC65FD9021CCD115F1F2DBD5A59E4.C0B243A3E2DED6769199AF3461781E75122AE135135'
-        right = '931EA22157E1871643FA9519676DED253A342B0C.4E95A5DBD2F1F511DCC1209DF56CB77693CE0EAE'
-        self.assertEqual(sig(wrong), right)
-
-    def test_88(self):
-        wrong = "qwertyuioplkjhgfdsazxcvbnm1234567890QWERTYUIOPLKJHGFDSAZXCVBNM!@#$%^&*()_-+={[]}|:;?/>.<"
-        right = "J:|}][{=+-_)(*&;%$#@>MNBVCXZASDFGH^KLPOIUYTREWQ0987654321mnbvcxzasdfghrklpoiuytej"
-        self.assertEqual(sig(wrong), right)
-
-    def test_87(self):
-        wrong = "qwertyuioplkjhgfdsazxcvbnm1234567890QWERTYUIOPLKJHGFDSAZXCVBNM!@#$^&*()_-+={[]}|:;?/>.<"
-        right = "!?;:|}][{=+-_)(*&^$#@/MNBVCXZASqFGHJKLPOIUYTREWQ0987654321mnbvcxzasdfghjklpoiuytr"
-        self.assertEqual(sig(wrong), right)
-
-    def test_86(self):
-        wrong = "qwertyuioplkjhgfdsazxcvbnm1234567890QWERTYUIOPLKJHGFDSAZXCVBNM!@#$%^&*()_-+={[|};?/>.<"
-        right = "ertyuioplkjhgfdsazxcvbnm1234567890QWERTYUIOPLKJHGFDSAZXCVBNM!/#$%^&*()_-+={[|};?@"
-        self.assertEqual(sig(wrong), right)
-
-    def test_85(self):
-        wrong = "qwertyuioplkjhgfdsazxcvbnm1234567890QWERTYUIOPLKJHGFDSAZXCVBNM!@#$%^&*()_-+={[};?/>.<"
-        right = "{>/?;}[.=+-_)(*&^%$#@!MqBVCXZASDFwHJKLPOIUYTREWQ0987654321mnbvcxzasdfghjklpoiuytr"
-        self.assertEqual(sig(wrong), right)
-
-    def test_84(self):
-        wrong = "qwertyuioplkjhgfdsazxcvbnm1234567890QWERTYUIOPLKJHGFDSAZXCVBNM!@#$%^&*()_-+={[};?>.<"
-        right = "<.>?;}[{=+-_)(*&^%$#@!MNBVCXZASDFGHJKLPOIUYTREWe098765432rmnbvcxzasdfghjklpoiuyt1"
-        self.assertEqual(sig(wrong), right)
-
-    def test_83(self):
-        wrong = "qwertyuioplkjhgfdsazxcvbnm1234567890QWERTYUIOPLKJHGFDSAZXCVBNM!#$%^&*()_+={[};?/>.<"
-        right = "D.>/?;}[{=+_)(*&^%$#!MNBVCXeAS<FGHJKLPOIUYTREWZ0987654321mnbvcxzasdfghjklpoiuytrQ"
-        self.assertEqual(sig(wrong), right)
-
-    def test_82(self):
-        wrong = "qwertyuioplkjhgfdsazxcvbnm1234567890QWERTYUIOPLKHGFDSAZXCVBNM!@#$%^&*(-+={[};?/>.<"
-        right = "Q>/?;}[{=+-(*<^%$#@!MNBVCXZASDFGHKLPOIUY8REWT0q&7654321mnbvcxzasdfghjklpoiuytrew9"
-        self.assertEqual(sig(wrong), right)
-
-if __name__ == '__main__':
-    unittest.main()

test/test_youtube_signature.py

@@ -0,0 +1,84 @@
+#!/usr/bin/env python
+
+# Allow direct execution
+import os
+import sys
+import unittest
+sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
+
+from test.helper import global_setup
+global_setup()
+
+
+import io
+import re
+import string
+
+from youtube_dl.extractor import YoutubeIE
+from youtube_dl.utils import compat_str, compat_urlretrieve
+
+_TESTS = [
+    (
+        u'https://s.ytimg.com/yts/jsbin/html5player-vflHOr_nV.js',
+        u'js',
+        86,
+        u'>=<;:/.-[+*)(\'&%$#"!ZYX0VUTSRQPONMLKJIHGFEDCBA\\yxwvutsrqponmlkjihgfedcba987654321',
+    ),
+    (
+        u'https://s.ytimg.com/yts/jsbin/html5player-vfldJ8xgI.js',
+        u'js',
+        85,
+        u'3456789a0cdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRS[UVWXYZ!"#$%&\'()*+,-./:;<=>?@',
+    ),
+    (
+        u'https://s.ytimg.com/yts/swfbin/watch_as3-vflg5GhxU.swf',
+        u'swf',
+        82,
+        u':/.-,+*)=\'&%$#"!ZYX0VUTSRQPONMLKJIHGFEDCBAzyxw>utsrqponmlkjihgfedcba987654321'
+    ),
+]
+
+
+class TestSignature(unittest.TestCase):
+    def setUp(self):
+        TEST_DIR = os.path.dirname(os.path.abspath(__file__))
+        self.TESTDATA_DIR = os.path.join(TEST_DIR, 'testdata')
+        if not os.path.exists(self.TESTDATA_DIR):
+            os.mkdir(self.TESTDATA_DIR)
+
+
+def make_tfunc(url, stype, sig_length, expected_sig):
+    basename = url.rpartition('/')[2]
+    m = re.match(r'.*-([a-zA-Z0-9_-]+)\.[a-z]+$', basename)
+    assert m, '%r should follow URL format' % basename
+    test_id = m.group(1)
+
+    def test_func(self):
+        fn = os.path.join(self.TESTDATA_DIR, basename)
+
+        if not os.path.exists(fn):
+            compat_urlretrieve(url, fn)
+
+        ie = YoutubeIE()
+        if stype == 'js':
+            with io.open(fn, encoding='utf-8') as testf:
+                jscode = testf.read()
+            func = ie._parse_sig_js(jscode)
+        else:
+            assert stype == 'swf'
+            with open(fn, 'rb') as testf:
+                swfcode = testf.read()
+            func = ie._parse_sig_swf(swfcode)
+        src_sig = compat_str(string.printable[:sig_length])
+        got_sig = func(src_sig)
+        self.assertEqual(got_sig, expected_sig)
+
+    test_func.__name__ = str('test_signature_' + stype + '_' + test_id)
+    setattr(TestSignature, test_func.__name__, test_func)
+
+for test_spec in _TESTS:
+    make_tfunc(*test_spec)
+
+
+if __name__ == '__main__':
+    unittest.main()

test/test_youtube_subtitles.py

@@ -1,95 +1,95 @@
 #!/usr/bin/env python
 
-import sys
-import unittest
-import json
-import io
-import hashlib
-
 # Allow direct execution
 import os
-sys.path.append(os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
+import sys
+import unittest
+sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
+
+from test.helper import FakeYDL, global_setup, md5
+global_setup()
+
 
 from youtube_dl.extractor import YoutubeIE
-from youtube_dl.utils import *
-from helper import FakeYDL
 
-md5 = lambda s: hashlib.md5(s.encode('utf-8')).hexdigest()
 
 class TestYoutubeSubtitles(unittest.TestCase):
     def setUp(self):
-        DL = FakeYDL()
-        DL.params['allsubtitles'] = False
-        DL.params['writesubtitles'] = False
-        DL.params['subtitlesformat'] = 'srt'
-        DL.params['listsubtitles'] = False
-    def test_youtube_no_subtitles(self):
-        DL = FakeYDL()
-        DL.params['writesubtitles'] = False
-        IE = YoutubeIE(DL)
-        info_dict = IE.extract('QRS8MkLhQmM')
-        subtitles = info_dict[0]['subtitles']
+        self.DL = FakeYDL()
+        self.url = 'QRS8MkLhQmM'
+
+    def getInfoDict(self):
+        IE = YoutubeIE(self.DL)
+        info_dict = IE.extract(self.url)
+        return info_dict
+
+    def getSubtitles(self):
+        info_dict = self.getInfoDict()
+        return info_dict[0]['subtitles']
+
+    def test_youtube_no_writesubtitles(self):
+        self.DL.params['writesubtitles'] = False
+        subtitles = self.getSubtitles()
         self.assertEqual(subtitles, None)
+
     def test_youtube_subtitles(self):
-        DL = FakeYDL()
-        DL.params['writesubtitles'] = True
-        IE = YoutubeIE(DL)
-        info_dict = IE.extract('QRS8MkLhQmM')
-        sub = info_dict[0]['subtitles'][0]
-        self.assertEqual(md5(sub[2]), '4cd9278a35ba2305f47354ee13472260')
-    def test_youtube_subtitles_it(self):
-        DL = FakeYDL()
-        DL.params['writesubtitles'] = True
-        DL.params['subtitleslang'] = 'it'
-        IE = YoutubeIE(DL)
-        info_dict = IE.extract('QRS8MkLhQmM')
-        sub = info_dict[0]['subtitles'][0]
-        self.assertEqual(md5(sub[2]), '164a51f16f260476a05b50fe4c2f161d')
-    def test_youtube_onlysubtitles(self):
-        DL = FakeYDL()
-        DL.params['writesubtitles'] = True
-        DL.params['onlysubtitles'] = True
-        IE = YoutubeIE(DL)
-        info_dict = IE.extract('QRS8MkLhQmM')
-        sub = info_dict[0]['subtitles'][0]
-        self.assertEqual(md5(sub[2]), '4cd9278a35ba2305f47354ee13472260')
+        self.DL.params['writesubtitles'] = True
+        subtitles = self.getSubtitles()
+        self.assertEqual(md5(subtitles['en']), '4cd9278a35ba2305f47354ee13472260')
+
+    def test_youtube_subtitles_lang(self):
+        self.DL.params['writesubtitles'] = True
+        self.DL.params['subtitleslangs'] = ['it']
+        subtitles = self.getSubtitles()
+        self.assertEqual(md5(subtitles['it']), '164a51f16f260476a05b50fe4c2f161d')
+
     def test_youtube_allsubtitles(self):
-        DL = FakeYDL()
-        DL.params['allsubtitles'] = True
-        IE = YoutubeIE(DL)
-        info_dict = IE.extract('QRS8MkLhQmM')
-        subtitles = info_dict[0]['subtitles']
-        self.assertEqual(len(subtitles), 13)
+        self.DL.params['writesubtitles'] = True
+        self.DL.params['allsubtitles'] = True
+        subtitles = self.getSubtitles()
+        self.assertEqual(len(subtitles.keys()), 13)
+
     def test_youtube_subtitles_sbv_format(self):
-        DL = FakeYDL()
-        DL.params['writesubtitles'] = True
-        DL.params['subtitlesformat'] = 'sbv'
-        IE = YoutubeIE(DL)
-        info_dict = IE.extract('QRS8MkLhQmM')
-        sub = info_dict[0]['subtitles'][0]
-        self.assertEqual(md5(sub[2]), '13aeaa0c245a8bed9a451cb643e3ad8b')
+        self.DL.params['writesubtitles'] = True
+        self.DL.params['subtitlesformat'] = 'sbv'
+        subtitles = self.getSubtitles()
+        self.assertEqual(md5(subtitles['en']), '13aeaa0c245a8bed9a451cb643e3ad8b')
+
     def test_youtube_subtitles_vtt_format(self):
-        DL = FakeYDL()
-        DL.params['writesubtitles'] = True
-        DL.params['subtitlesformat'] = 'vtt'
-        IE = YoutubeIE(DL)
-        info_dict = IE.extract('QRS8MkLhQmM')
-        sub = info_dict[0]['subtitles'][0]
-        self.assertEqual(md5(sub[2]), '356cdc577fde0c6783b9b822e7206ff7')
+        self.DL.params['writesubtitles'] = True
+        self.DL.params['subtitlesformat'] = 'vtt'
+        subtitles = self.getSubtitles()
+        self.assertEqual(md5(subtitles['en']), '356cdc577fde0c6783b9b822e7206ff7')
+
     def test_youtube_list_subtitles(self):
-        DL = FakeYDL()
-        DL.params['listsubtitles'] = True
-        IE = YoutubeIE(DL)
-        info_dict = IE.extract('QRS8MkLhQmM')
+        self.DL.expect_warning(u'Video doesn\'t have automatic captions')
+        self.DL.params['listsubtitles'] = True
+        info_dict = self.getInfoDict()
         self.assertEqual(info_dict, None)
+
     def test_youtube_automatic_captions(self):
-        DL = FakeYDL()
-        DL.params['writeautomaticsub'] = True
-        DL.params['subtitleslang'] = 'it'
-        IE = YoutubeIE(DL)
-        info_dict = IE.extract('8YoUxe5ncPo')
-        sub = info_dict[0]['subtitles'][0]
-        self.assertTrue(sub[2] is not None)
+        self.url = '8YoUxe5ncPo'
+        self.DL.params['writeautomaticsub'] = True
+        self.DL.params['subtitleslangs'] = ['it']
+        subtitles = self.getSubtitles()
+        self.assertTrue(subtitles['it'] is not None)
+
+    def test_youtube_nosubtitles(self):
+        self.DL.expect_warning(u'video doesn\'t have subtitles')
+        self.url = 'sAjKT8FhjI8'
+        self.DL.params['writesubtitles'] = True
+        self.DL.params['allsubtitles'] = True
+        subtitles = self.getSubtitles()
+        self.assertEqual(len(subtitles), 0)
+
+    def test_youtube_multiple_langs(self):
+        self.url = 'QRS8MkLhQmM'
+        self.DL.params['writesubtitles'] = True
+        langs = ['it', 'fr', 'de']
+        self.DL.params['subtitleslangs'] = langs
+        subtitles = self.getSubtitles()
+        for lang in langs:
+            self.assertTrue(subtitles.get(lang) is not None, u'Subtitles for \'%s\' not extracted' % lang)
 
 if __name__ == '__main__':
     unittest.main()

tox.ini

@@ -0,0 +1,8 @@
+[tox]
+envlist = py26,py27,py33
+[testenv]
+deps =
+   nose
+   coverage
+commands = nosetests --verbose {posargs:test}  # --with-coverage --cover-package=youtube_dl --cover-html
+                                               # test.test_download:TestDownload.test_NowVideo

youtube_dl/FileDownloader.py

@@ -63,32 +63,57 @@ class FileDownloader(object):
         converted = float(bytes) / float(1024 ** exponent)
         return '%.2f%s' % (converted, suffix)
 
+    @staticmethod
+    def format_seconds(seconds):
+        (mins, secs) = divmod(seconds, 60)
+        (hours, mins) = divmod(mins, 60)
+        if hours > 99:
+            return '--:--:--'
+        if hours == 0:
+            return '%02d:%02d' % (mins, secs)
+        else:
+            return '%02d:%02d:%02d' % (hours, mins, secs)
+
     @staticmethod
     def calc_percent(byte_counter, data_len):
         if data_len is None:
+            return None
+        return float(byte_counter) / float(data_len) * 100.0
+
+    @staticmethod
+    def format_percent(percent):
+        if percent is None:
             return '---.-%'
-        return '%6s' % ('%3.1f%%' % (float(byte_counter) / float(data_len) * 100.0))
+        return '%6s' % ('%3.1f%%' % percent)
 
     @staticmethod
     def calc_eta(start, now, total, current):
         if total is None:
-            return '--:--'
+            return None
         dif = now - start
         if current == 0 or dif < 0.001: # One millisecond
-            return '--:--'
+            return None
         rate = float(current) / dif
-        eta = int((float(total) - float(current)) / rate)
-        (eta_mins, eta_secs) = divmod(eta, 60)
-        if eta_mins > 99:
+        return int((float(total) - float(current)) / rate)
+
+    @staticmethod
+    def format_eta(eta):
+        if eta is None:
             return '--:--'
-        return '%02d:%02d' % (eta_mins, eta_secs)
+        return FileDownloader.format_seconds(eta)
 
     @staticmethod
     def calc_speed(start, now, bytes):
         dif = now - start
         if bytes == 0 or dif < 0.001: # One millisecond
+            return None
+        return float(bytes) / dif
+
+    @staticmethod
+    def format_speed(speed):
+        if speed is None:
             return '%10s' % '---b/s'
-        return '%10s' % ('%s/s' % FileDownloader.format_bytes(float(bytes) / dif))
+        return '%10s' % ('%s/s' % FileDownloader.format_bytes(speed))
 
     @staticmethod
     def best_block_size(elapsed_time, bytes):
@@ -197,11 +222,14 @@ class FileDownloader(object):
         """Report destination filename."""
         self.to_screen(u'[download] Destination: ' + filename)
 
-    def report_progress(self, percent_str, data_len_str, speed_str, eta_str):
+    def report_progress(self, percent, data_len_str, speed, eta):
         """Report download progress."""
         if self.params.get('noprogress', False):
             return
         clear_line = (u'\x1b[K' if sys.stderr.isatty() and os.name != 'nt' else u'')
+        eta_str = self.format_eta(eta)
+        percent_str = self.format_percent(percent)
+        speed_str = self.format_speed(speed)
         if self.params.get('progress_with_newline', False):
             self.to_screen(u'[download] %s of %s at %s ETA %s' %
                 (percent_str, data_len_str, speed_str, eta_str))
@@ -230,16 +258,19 @@ class FileDownloader(object):
         """Report it was impossible to resume download."""
         self.to_screen(u'[download] Unable to resume')
 
-    def report_finish(self):
+    def report_finish(self, data_len_str, tot_time):
         """Report download finished."""
         if self.params.get('noprogress', False):
             self.to_screen(u'[download] Download completed')
         else:
-            self.to_screen(u'')
+            clear_line = (u'\x1b[K' if sys.stderr.isatty() and os.name != 'nt' else u'')
+            self.to_screen(u'\r%s[download] 100%% of %s in %s' %
+                (clear_line, data_len_str, self.format_seconds(tot_time)))
 
     def _download_with_rtmpdump(self, filename, url, player_url, page_url, play_path, tc_url):
         self.report_destination(filename)
         tmpfilename = self.temp_name(filename)
+        test = self.params.get('test', False)
 
         # Check for rtmpdump first
         try:
@@ -261,6 +292,8 @@ class FileDownloader(object):
             basic_args += ['--playpath', play_path]
         if tc_url is not None:
             basic_args += ['--tcUrl', url]
+        if test:
+            basic_args += ['--stop', '1']
         args = basic_args + [[], ['--resume', '--skip', '1']][self.params.get('continuedl', False)]
         if self.params.get('verbose', False):
             try:
@@ -270,7 +303,7 @@ class FileDownloader(object):
                 shell_quote = repr
             self.to_screen(u'[debug] rtmpdump command line: ' + shell_quote(args))
         retval = subprocess.call(args)
-        while retval == 2 or retval == 1:
+        while (retval == 2 or retval == 1) and not test:
             prevsize = os.path.getsize(encodeFilename(tmpfilename))
             self.to_screen(u'\r[rtmpdump] %s bytes' % prevsize, skip_eol=True)
             time.sleep(5.0) # This seems to be needed
@@ -283,7 +316,7 @@ class FileDownloader(object):
                 self.to_screen(u'\r[rtmpdump] Could not download the whole video. This can happen for some advertisements.')
                 retval = 0
                 break
-        if retval == 0:
+        if retval == 0 or (test and retval == 2):
             fsize = os.path.getsize(encodeFilename(tmpfilename))
             self.to_screen(u'\r[rtmpdump] %s bytes' % fsize)
             self.try_rename(tmpfilename, filename)
@@ -329,6 +362,35 @@ class FileDownloader(object):
             self.report_error(u'mplayer exited with code %d' % retval)
             return False
 
+    def _download_m3u8_with_ffmpeg(self, filename, url):
+        self.report_destination(filename)
+        tmpfilename = self.temp_name(filename)
+
+        args = ['ffmpeg', '-y', '-i', url, '-f', 'mp4', tmpfilename]
+        # Check for ffmpeg first
+        try:
+            subprocess.call(['ffmpeg', '-h'], stdout=(open(os.path.devnull, 'w')), stderr=subprocess.STDOUT)
+        except (OSError, IOError):
+            self.report_error(u'm3u8 download detected but "%s" could not be run' % args[0] )
+            return False
+
+        retval = subprocess.call(args)
+        if retval == 0:
+            fsize = os.path.getsize(encodeFilename(tmpfilename))
+            self.to_screen(u'\r[%s] %s bytes' % (args[0], fsize))
+            self.try_rename(tmpfilename, filename)
+            self._hook_progress({
+                'downloaded_bytes': fsize,
+                'total_bytes': fsize,
+                'filename': filename,
+                'status': 'finished',
+            })
+            return True
+        else:
+            self.to_stderr(u"\n")
+            self.report_error(u'ffmpeg exited with code %d' % retval)
+            return False
+
 
     def _do_download(self, filename, info_dict):
         url = info_dict['url']
@@ -339,6 +401,7 @@ class FileDownloader(object):
             self._hook_progress({
                 'filename': filename,
                 'status': 'finished',
+                'total_bytes': os.path.getsize(encodeFilename(filename)),
             })
             return True
 
@@ -354,6 +417,10 @@ class FileDownloader(object):
         if url.startswith('mms') or url.startswith('rtsp'):
             return self._download_with_mplayer(filename, url)
 
+        # m3u8 manifest are downloaded with ffmpeg
+        if determine_ext(url) == u'm3u8':
+            return self._download_m3u8_with_ffmpeg(filename, url)
+
         tmpfilename = self.temp_name(filename)
         stream = None
 
@@ -481,13 +548,14 @@ class FileDownloader(object):
                 block_size = self.best_block_size(after - before, len(data_block))
 
             # Progress message
-            speed_str = self.calc_speed(start, time.time(), byte_counter - resume_len)
+            speed = self.calc_speed(start, time.time(), byte_counter - resume_len)
             if data_len is None:
                 self.report_progress('Unknown %', data_len_str, speed_str, 'Unknown ETA')
+                eta = None
             else:
-                percent_str = self.calc_percent(byte_counter, data_len)
-                eta_str = self.calc_eta(start, time.time(), data_len - resume_len, byte_counter - resume_len)
-                self.report_progress(percent_str, data_len_str, speed_str, eta_str)
+                percent = self.calc_percent(byte_counter, data_len)
+                eta = self.calc_eta(start, time.time(), data_len - resume_len, byte_counter - resume_len)
+                self.report_progress(percent, data_len_str, speed, eta)
 
             self._hook_progress({
                 'downloaded_bytes': byte_counter,
@@ -495,6 +563,8 @@ class FileDownloader(object):
                 'tmpfilename': tmpfilename,
                 'filename': filename,
                 'status': 'downloading',
+                'eta': eta,
+                'speed': speed,
             })
 
             # Apply rate limit
@@ -505,7 +575,7 @@ class FileDownloader(object):
             self.report_error(u'Did not get any data blocks')
             return False
         stream.close()
-        self.report_finish()
+        self.report_finish(data_len_str, (time.time() - start))
         if data_len is not None and byte_counter != data_len:
             raise ContentTooShortError(byte_counter, int(data_len))
         self.try_rename(tmpfilename, filename)
@@ -537,6 +607,8 @@ class FileDownloader(object):
         * downloaded_bytes: Bytes on disks
         * total_bytes: Total bytes, None if unknown
         * tmpfilename: The filename we're currently writing to
+        * eta: The estimated time in seconds, None if unknown
+        * speed: The download speed in bytes/second, None if unknown
 
         Hooks are guaranteed to be called at least once (with status "finished")
         if the download is successful.

youtube_dl/PostProcessor.py

@@ -3,7 +3,14 @@ import subprocess
 import sys
 import time
 
-from .utils import *
+
+from .utils import (
+    compat_subprocess_get_DEVNULL,
+    encodeFilename,
+    PostProcessingError,
+    shell_quote,
+    subtitles_filename,
+)
 
 
 class PostProcessor(object):
@@ -71,12 +78,19 @@ class FFmpegPostProcessor(PostProcessor):
         programs = ['avprobe', 'avconv', 'ffmpeg', 'ffprobe']
         return dict((program, executable(program)) for program in programs)
 
-    def run_ffmpeg(self, path, out_path, opts):
+    def run_ffmpeg_multiple_files(self, input_paths, out_path, opts):
         if not self._exes['ffmpeg'] and not self._exes['avconv']:
             raise FFmpegPostProcessorError(u'ffmpeg or avconv not found. Please install one.')
-        cmd = ([self._exes['avconv'] or self._exes['ffmpeg'], '-y', '-i', encodeFilename(path)]
+
+        files_cmd = []
+        for path in input_paths:
+            files_cmd.extend(['-i', encodeFilename(path)])
+        cmd = ([self._exes['avconv'] or self._exes['ffmpeg'], '-y'] + files_cmd
                + opts +
                [encodeFilename(self._ffmpeg_filename_argument(out_path))])
+
+        if self._downloader.params.get('verbose', False):
+            self._downloader.to_screen(u'[debug] ffmpeg command line: %s' % shell_quote(cmd))
         p = subprocess.Popen(cmd, stdout=subprocess.PIPE, stderr=subprocess.PIPE)
         stdout,stderr = p.communicate()
         if p.returncode != 0:
@@ -84,6 +98,9 @@ class FFmpegPostProcessor(PostProcessor):
             msg = stderr.strip().split('\n')[-1]
             raise FFmpegPostProcessorError(msg)
 
+    def run_ffmpeg(self, path, out_path, opts):
+        self.run_ffmpeg_multiple_files([path], out_path, opts)
+
     def _ffmpeg_filename_argument(self, fn):
         # ffmpeg broke --, see https://ffmpeg.org/trac/ffmpeg/ticket/2127 for details
         if fn.startswith(u'-'):
@@ -100,7 +117,8 @@ class FFmpegExtractAudioPP(FFmpegPostProcessor):
         self._nopostoverwrites = nopostoverwrites
 
     def get_audio_codec(self, path):
-        if not self._exes['ffprobe'] and not self._exes['avprobe']: return None
+        if not self._exes['ffprobe'] and not self._exes['avprobe']:
+            raise PostProcessingError(u'ffprobe or avprobe not found. Please install one.')
         try:
             cmd = [self._exes['avprobe'] or self._exes['ffprobe'], '-show_streams', encodeFilename(self._ffmpeg_filename_argument(path))]
             handle = subprocess.Popen(cmd, stderr=compat_subprocess_get_DEVNULL(), stdout=subprocess.PIPE)
@@ -128,7 +146,7 @@ class FFmpegExtractAudioPP(FFmpegPostProcessor):
         try:
             FFmpegPostProcessor.run_ffmpeg(self, path, out_path, opts)
         except FFmpegPostProcessorError as err:
-            raise AudioConversionError(err.message)
+            raise AudioConversionError(err.msg)
 
     def run(self, information):
         path = information['filepath']
@@ -168,7 +186,8 @@ class FFmpegExtractAudioPP(FFmpegPostProcessor):
             extension = self._preferredcodec
             more_opts = []
             if self._preferredquality is not None:
-                if int(self._preferredquality) < 10:
+                # The opus codec doesn't support the -aq option
+                if int(self._preferredquality) < 10 and extension != 'opus':
                     more_opts += [self._exes['avconv'] and '-q:a' or '-aq', self._preferredquality]
                 else:
                     more_opts += [self._exes['avconv'] and '-b:a' or '-ab', self._preferredquality + 'k']
@@ -198,7 +217,7 @@ class FFmpegExtractAudioPP(FFmpegPostProcessor):
         except:
             etype,e,tb = sys.exc_info()
             if isinstance(e, AudioConversionError):
-                msg = u'audio conversion failed: ' + e.message
+                msg = u'audio conversion failed: ' + e.msg
             else:
                 msg = u'error running ' + (self._exes['avconv'] and 'avconv' or 'ffmpeg')
             raise PostProcessingError(msg)
@@ -208,7 +227,7 @@ class FFmpegExtractAudioPP(FFmpegPostProcessor):
             try:
                 os.utime(encodeFilename(new_path), (time.time(), information['filetime']))
             except:
-                self._downloader.to_stderr(u'WARNING: Cannot update utime of audio file')
+                self._downloader.report_warning(u'Cannot update utime of audio file')
 
         information['filepath'] = new_path
         return self._nopostoverwrites,information
@@ -231,3 +250,262 @@ class FFmpegVideoConvertor(FFmpegPostProcessor):
         information['format'] = self._preferedformat
         information['ext'] = self._preferedformat
         return False,information
+
+
+class FFmpegEmbedSubtitlePP(FFmpegPostProcessor):
+    # See http://www.loc.gov/standards/iso639-2/ISO-639-2_utf-8.txt
+    _lang_map = {
+        'aa': 'aar',
+        'ab': 'abk',
+        'ae': 'ave',
+        'af': 'afr',
+        'ak': 'aka',
+        'am': 'amh',
+        'an': 'arg',
+        'ar': 'ara',
+        'as': 'asm',
+        'av': 'ava',
+        'ay': 'aym',
+        'az': 'aze',
+        'ba': 'bak',
+        'be': 'bel',
+        'bg': 'bul',
+        'bh': 'bih',
+        'bi': 'bis',
+        'bm': 'bam',
+        'bn': 'ben',
+        'bo': 'bod',
+        'br': 'bre',
+        'bs': 'bos',
+        'ca': 'cat',
+        'ce': 'che',
+        'ch': 'cha',
+        'co': 'cos',
+        'cr': 'cre',
+        'cs': 'ces',
+        'cu': 'chu',
+        'cv': 'chv',
+        'cy': 'cym',
+        'da': 'dan',
+        'de': 'deu',
+        'dv': 'div',
+        'dz': 'dzo',
+        'ee': 'ewe',
+        'el': 'ell',
+        'en': 'eng',
+        'eo': 'epo',
+        'es': 'spa',
+        'et': 'est',
+        'eu': 'eus',
+        'fa': 'fas',
+        'ff': 'ful',
+        'fi': 'fin',
+        'fj': 'fij',
+        'fo': 'fao',
+        'fr': 'fra',
+        'fy': 'fry',
+        'ga': 'gle',
+        'gd': 'gla',
+        'gl': 'glg',
+        'gn': 'grn',
+        'gu': 'guj',
+        'gv': 'glv',
+        'ha': 'hau',
+        'he': 'heb',
+        'hi': 'hin',
+        'ho': 'hmo',
+        'hr': 'hrv',
+        'ht': 'hat',
+        'hu': 'hun',
+        'hy': 'hye',
+        'hz': 'her',
+        'ia': 'ina',
+        'id': 'ind',
+        'ie': 'ile',
+        'ig': 'ibo',
+        'ii': 'iii',
+        'ik': 'ipk',
+        'io': 'ido',
+        'is': 'isl',
+        'it': 'ita',
+        'iu': 'iku',
+        'ja': 'jpn',
+        'jv': 'jav',
+        'ka': 'kat',
+        'kg': 'kon',
+        'ki': 'kik',
+        'kj': 'kua',
+        'kk': 'kaz',
+        'kl': 'kal',
+        'km': 'khm',
+        'kn': 'kan',
+        'ko': 'kor',
+        'kr': 'kau',
+        'ks': 'kas',
+        'ku': 'kur',
+        'kv': 'kom',
+        'kw': 'cor',
+        'ky': 'kir',
+        'la': 'lat',
+        'lb': 'ltz',
+        'lg': 'lug',
+        'li': 'lim',
+        'ln': 'lin',
+        'lo': 'lao',
+        'lt': 'lit',
+        'lu': 'lub',
+        'lv': 'lav',
+        'mg': 'mlg',
+        'mh': 'mah',
+        'mi': 'mri',
+        'mk': 'mkd',
+        'ml': 'mal',
+        'mn': 'mon',
+        'mr': 'mar',
+        'ms': 'msa',
+        'mt': 'mlt',
+        'my': 'mya',
+        'na': 'nau',
+        'nb': 'nob',
+        'nd': 'nde',
+        'ne': 'nep',
+        'ng': 'ndo',
+        'nl': 'nld',
+        'nn': 'nno',
+        'no': 'nor',
+        'nr': 'nbl',
+        'nv': 'nav',
+        'ny': 'nya',
+        'oc': 'oci',
+        'oj': 'oji',
+        'om': 'orm',
+        'or': 'ori',
+        'os': 'oss',
+        'pa': 'pan',
+        'pi': 'pli',
+        'pl': 'pol',
+        'ps': 'pus',
+        'pt': 'por',
+        'qu': 'que',
+        'rm': 'roh',
+        'rn': 'run',
+        'ro': 'ron',
+        'ru': 'rus',
+        'rw': 'kin',
+        'sa': 'san',
+        'sc': 'srd',
+        'sd': 'snd',
+        'se': 'sme',
+        'sg': 'sag',
+        'si': 'sin',
+        'sk': 'slk',
+        'sl': 'slv',
+        'sm': 'smo',
+        'sn': 'sna',
+        'so': 'som',
+        'sq': 'sqi',
+        'sr': 'srp',
+        'ss': 'ssw',
+        'st': 'sot',
+        'su': 'sun',
+        'sv': 'swe',
+        'sw': 'swa',
+        'ta': 'tam',
+        'te': 'tel',
+        'tg': 'tgk',
+        'th': 'tha',
+        'ti': 'tir',
+        'tk': 'tuk',
+        'tl': 'tgl',
+        'tn': 'tsn',
+        'to': 'ton',
+        'tr': 'tur',
+        'ts': 'tso',
+        'tt': 'tat',
+        'tw': 'twi',
+        'ty': 'tah',
+        'ug': 'uig',
+        'uk': 'ukr',
+        'ur': 'urd',
+        'uz': 'uzb',
+        've': 'ven',
+        'vi': 'vie',
+        'vo': 'vol',
+        'wa': 'wln',
+        'wo': 'wol',
+        'xh': 'xho',
+        'yi': 'yid',
+        'yo': 'yor',
+        'za': 'zha',
+        'zh': 'zho',
+        'zu': 'zul',
+    }
+
+    def __init__(self, downloader=None, subtitlesformat='srt'):
+        super(FFmpegEmbedSubtitlePP, self).__init__(downloader)
+        self._subformat = subtitlesformat
+
+    @classmethod
+    def _conver_lang_code(cls, code):
+        """Convert language code from ISO 639-1 to ISO 639-2/T"""
+        return cls._lang_map.get(code[:2])
+
+    def run(self, information):
+        if information['ext'] != u'mp4':
+            self._downloader.to_screen(u'[ffmpeg] Subtitles can only be embedded in mp4 files')
+            return True, information
+        if not information.get('subtitles'):
+            self._downloader.to_screen(u'[ffmpeg] There aren\'t any subtitles to embed') 
+            return True, information
+
+        sub_langs = [key for key in information['subtitles']]
+        filename = information['filepath']
+        input_files = [filename] + [subtitles_filename(filename, lang, self._subformat) for lang in sub_langs]
+
+        opts = ['-map', '0:0', '-map', '0:1', '-c:v', 'copy', '-c:a', 'copy']
+        for (i, lang) in enumerate(sub_langs):
+            opts.extend(['-map', '%d:0' % (i+1), '-c:s:%d' % i, 'mov_text'])
+            lang_code = self._conver_lang_code(lang)
+            if lang_code is not None:
+                opts.extend(['-metadata:s:s:%d' % i, 'language=%s' % lang_code])
+        opts.extend(['-f', 'mp4'])
+
+        temp_filename = filename + u'.temp'
+        self._downloader.to_screen(u'[ffmpeg] Embedding subtitles in \'%s\'' % filename)
+        self.run_ffmpeg_multiple_files(input_files, temp_filename, opts)
+        os.remove(encodeFilename(filename))
+        os.rename(encodeFilename(temp_filename), encodeFilename(filename))
+
+        return True, information
+
+
+class FFmpegMetadataPP(FFmpegPostProcessor):
+    def run(self, info):
+        metadata = {}
+        if info.get('title') is not None:
+            metadata['title'] = info['title']
+        if info.get('upload_date') is not None:
+            metadata['date'] = info['upload_date']
+        if info.get('uploader') is not None:
+            metadata['artist'] = info['uploader']
+        elif info.get('uploader_id') is not None:
+            metadata['artist'] = info['uploader_id']
+
+        if not metadata:
+            self._downloader.to_screen(u'[ffmpeg] There isn\'t any metadata to add')
+            return True, info
+
+        filename = info['filepath']
+        ext = os.path.splitext(filename)[1][1:]
+        temp_filename = filename + u'.temp'
+
+        options = ['-c', 'copy']
+        for (name, value) in metadata.items():
+            options.extend(['-metadata', '%s="%s"' % (name, value)])
+        options.extend(['-f', ext])
+
+        self._downloader.to_screen(u'[ffmpeg] Adding metadata to \'%s\'' % filename)
+        self.run_ffmpeg(filename, temp_filename, options)
+        os.remove(encodeFilename(filename))
+        os.rename(encodeFilename(temp_filename), encodeFilename(filename))
+        return True, info

youtube_dl/YoutubeDL.py

@@ -3,6 +3,7 @@
 
 from __future__ import absolute_import
 
+import errno
 import io
 import os
 import re
@@ -70,16 +71,26 @@ class YoutubeDL(object):
     logtostderr:       Log messages to stderr instead of stdout.
     writedescription:  Write the video description to a .description file
     writeinfojson:     Write the video description to a .info.json file
+    writeannotations:  Write the video annotations to a .annotations.xml file
     writethumbnail:    Write the thumbnail image to a file
     writesubtitles:    Write the video subtitles to a file
     writeautomaticsub: Write the automatic subtitles to a file
     allsubtitles:      Downloads all the subtitles of the video
+                       (requires writesubtitles or writeautomaticsub)
     listsubtitles:     Lists all available subtitles for the video
     subtitlesformat:   Subtitle format [srt/sbv/vtt] (default=srt)
-    subtitleslang:     Language of the subtitles to download
+    subtitleslangs:    List of languages of the subtitles to download
     keepvideo:         Keep the video file after post-processing
     daterange:         A DateRange object, download only if the upload_date is in the range.
     skip_download:     Skip the actual download of the video file
+    cachedir:          Location of the cache files in the filesystem.
+                       None to disable filesystem cache.
+    noplaylist:        Download single video instead of a playlist if in doubt.
+    age_limit:         An integer representing the user's age in years.
+                       Unsuitable videos for the given age are skipped.
+    downloadarchive:   File name of a file where all downloads are recorded.
+                       Videos already present in the file are not downloaded
+                       again.
     
     The following parameters are not used by YoutubeDL itself, they are used by
     the FileDownloader:
@@ -97,11 +108,23 @@ class YoutubeDL(object):
     def __init__(self, params):
         """Create a FileDownloader object with the given options."""
         self._ies = []
+        self._ies_instances = {}
         self._pps = []
         self._progress_hooks = []
         self._download_retcode = 0
         self._num_downloads = 0
         self._screen_file = [sys.stdout, sys.stderr][params.get('logtostderr', False)]
+
+        if (sys.version_info >= (3,) and sys.platform != 'win32' and
+                sys.getfilesystemencoding() in ['ascii', 'ANSI_X3.4-1968']
+                and not params['restrictfilenames']):
+            # On Python 3, the Unicode filesystem API will throw errors (#1474)
+            self.report_warning(
+                u'Assuming --restrict-filenames since file system encoding '
+                u'cannot encode all charactes. '
+                u'Set the LC_ALL environment variable to fix this.')
+            params['restrictfilenames'] = True
+
         self.params = params
         self.fd = FileDownloader(self, self.params)
 
@@ -111,8 +134,21 @@ class YoutubeDL(object):
     def add_info_extractor(self, ie):
         """Add an InfoExtractor object to the end of the list."""
         self._ies.append(ie)
+        self._ies_instances[ie.ie_key()] = ie
         ie.set_downloader(self)
 
+    def get_info_extractor(self, ie_key):
+        """
+        Get an instance of an IE with name ie_key, it will try to get one from
+        the _ies list, if there's no instance it will create a new one and add
+        it to the extractor list.
+        """
+        ie = self._ies_instances.get(ie_key)
+        if ie is None:
+            ie = get_info_extractor(ie_key)()
+            self.add_info_extractor(ie)
+        return ie
+
     def add_default_info_extractors(self):
         """
         Add the InfoExtractors returned by gen_extractors to the end of the list
@@ -127,14 +163,10 @@ class YoutubeDL(object):
 
     def to_screen(self, message, skip_eol=False):
         """Print message to stdout if not in quiet mode."""
-        assert type(message) == type(u'')
         if not self.params.get('quiet', False):
             terminator = [u'\n', u''][skip_eol]
             output = message + terminator
-            if 'b' in getattr(self._screen_file, 'mode', '') or sys.version_info[0] < 3: # Python 2 lies about the mode of sys.stdout/sys.stderr
-                output = output.encode(preferredencoding(), 'ignore')
-            self._screen_file.write(output)
-            self._screen_file.flush()
+            write_string(output, self._screen_file)
 
     def to_stderr(self, message):
         """Print message to stderr."""
@@ -227,6 +259,10 @@ class YoutubeDL(object):
         """ Report that the metadata file has been written """
         self.to_screen(u'[info] Video description metadata as JSON to: ' + infofn)
 
+    def report_writeannotations(self, annofn):
+        """ Report that the annotations file has been written. """
+        self.to_screen(u'[info] Writing video annotations to: ' + annofn)
+
     def report_file_already_downloaded(self, file_name):
         """Report file has already been fully downloaded."""
         try:
@@ -264,7 +300,7 @@ class YoutubeDL(object):
             self.report_error(u'Erroneous output template')
             return None
         except ValueError as err:
-            self.report_error(u'Insufficient system charset ' + repr(preferredencoding()))
+            self.report_error(u'Error in output template: ' + str(err) + u' (encoding: ' + repr(preferredencoding()) + ')')
             return None
 
     def _match_entry(self, info_dict):
@@ -284,6 +320,13 @@ class YoutubeDL(object):
             dateRange = self.params.get('daterange', DateRange())
             if date not in dateRange:
                 return u'[download] %s upload date is not in range %s' % (date_from_str(date).isoformat(), dateRange)
+        age_limit = self.params.get('age_limit')
+        if age_limit is not None:
+            if age_limit < info_dict.get('age_limit', 0):
+                return u'Skipping "' + title + '" because it is age restricted'
+        if self.in_download_archive(info_dict):
+            return (u'%(title)s has already been recorded in archive'
+                    % info_dict)
         return None
         
     def extract_info(self, url, download=True, ie_key=None, extra_info={}):
@@ -294,9 +337,7 @@ class YoutubeDL(object):
          '''
         
         if ie_key:
-            ie = get_info_extractor(ie_key)()
-            ie.set_downloader(self)
-            ies = [ie]
+            ies = [self.get_info_extractor(ie_key)]
         else:
             ies = self._ies
 
@@ -348,13 +389,8 @@ class YoutubeDL(object):
 
         result_type = ie_result.get('_type', 'video') # If not given we suppose it's a video, support the default old system
         if result_type == 'video':
-            if 'playlist' not in ie_result:
-                # It isn't part of a playlist
-                ie_result['playlist'] = None
-                ie_result['playlist_index'] = None
-            if download:
-                self.process_info(ie_result)
-            return ie_result
+            ie_result.update(extra_info)
+            return self.process_video_result(ie_result)
         elif result_type == 'url':
             # We have to add extra_info to the results because it may be
             # contained in a playlist
@@ -412,6 +448,91 @@ class YoutubeDL(object):
         else:
             raise Exception('Invalid result type: %s' % result_type)
 
+    def process_video_result(self, info_dict, download=True):
+        assert info_dict.get('_type', 'video') == 'video'
+
+        if 'playlist' not in info_dict:
+            # It isn't part of a playlist
+            info_dict['playlist'] = None
+            info_dict['playlist_index'] = None
+
+        # This extractors handle format selection themselves
+        if info_dict['extractor'] in [u'youtube', u'Youku', u'YouPorn', u'mixcloud']:
+            self.process_info(info_dict)
+            return info_dict
+
+        # We now pick which formats have to be downloaded
+        if info_dict.get('formats') is None:
+            # There's only one format available
+            formats = [info_dict]
+        else:
+            formats = info_dict['formats']
+
+        # We check that all the formats have the format and format_id fields
+        for (i, format) in enumerate(formats):
+            if format.get('format') is None:
+                if format.get('height') is not None:
+                    if format.get('width') is not None:
+                        format_desc = u'%sx%s' % (format['width'], format['height'])
+                    else:
+                        format_desc = u'%sp' % format['height']
+                else:
+                    format_desc = '???'
+                format['format'] = format_desc
+            if format.get('format_id') is None:
+                format['format_id'] = compat_str(i)
+
+        if self.params.get('listformats', None):
+            self.list_formats(info_dict)
+            return
+
+        format_limit = self.params.get('format_limit', None)
+        if format_limit:
+            formats = list(takewhile_inclusive(
+                lambda f: f['format_id'] != format_limit, formats
+            ))
+        if self.params.get('prefer_free_formats'):
+            def _free_formats_key(f):
+                try:
+                    ext_ord = [u'flv', u'mp4', u'webm'].index(f['ext'])
+                except ValueError:
+                    ext_ord = -1
+                # We only compare the extension if they have the same height and width
+                return (f.get('height'), f.get('width'), ext_ord)
+            formats = sorted(formats, key=_free_formats_key)
+
+        req_format = self.params.get('format', 'best')
+        formats_to_download = []
+        if req_format == 'best' or req_format is None:
+            formats_to_download = [formats[-1]]
+        elif req_format == 'worst':
+            formats_to_download = [formats[0]]
+        # The -1 is for supporting YoutubeIE
+        elif req_format in ('-1', 'all'):
+            formats_to_download = formats
+        else:
+            # We can accept formats requestd in the format: 34/10/5, we pick
+            # the first that is available, starting from left
+            req_formats = req_format.split('/')
+            for rf in req_formats:
+                matches = filter(lambda f:f['format_id'] == rf ,formats)
+                if matches:
+                    formats_to_download = [matches[0]]
+                    break
+        if not formats_to_download:
+            raise ExtractorError(u'requested format not available')
+
+        if download:
+            if len(formats_to_download) > 1:
+                self.to_screen(u'[info] %s: downloading video in %s formats' % (info_dict['id'], len(formats_to_download)))
+            for format in formats_to_download:
+                new_info = dict(info_dict)
+                new_info.update(format)
+                self.process_info(new_info)
+        # We update the info dict with the best quality format (backwards compatibility)
+        info_dict.update(formats_to_download[-1])
+        return info_dict
+
     def process_info(self, info_dict):
         """Process a single resolved IE result."""
 
@@ -447,7 +568,8 @@ class YoutubeDL(object):
         if self.params.get('forceid', False):
             compat_print(info_dict['id'])
         if self.params.get('forceurl', False):
-            compat_print(info_dict['url'])
+            # For RTMP URLs, also include the playpath
+            compat_print(info_dict['url'] + info_dict.get('play_path', u''))
         if self.params.get('forcethumbnail', False) and 'thumbnail' in info_dict:
             compat_print(info_dict['thumbnail'])
         if self.params.get('forcedescription', False) and 'description' in info_dict:
@@ -478,45 +600,45 @@ class YoutubeDL(object):
                 self.report_writedescription(descfn)
                 with io.open(encodeFilename(descfn), 'w', encoding='utf-8') as descfile:
                     descfile.write(info_dict['description'])
+            except (KeyError, TypeError):
+                self.report_warning(u'There\'s no description to write.')
             except (OSError, IOError):
                 self.report_error(u'Cannot write description file ' + descfn)
                 return
 
-        if (self.params.get('writesubtitles', False) or self.params.get('writeautomaticsub')) and 'subtitles' in info_dict and info_dict['subtitles']:
+        if self.params.get('writeannotations', False):
+            try:
+               annofn = filename + u'.annotations.xml'
+               self.report_writeannotations(annofn)
+               with io.open(encodeFilename(annofn), 'w', encoding='utf-8') as annofile:
+                   annofile.write(info_dict['annotations'])
+            except (KeyError, TypeError):
+                self.report_warning(u'There are no annotations to write.')
+            except (OSError, IOError):
+                 self.report_error(u'Cannot write annotations file: ' + annofn)
+                 return
+
+        subtitles_are_requested = any([self.params.get('writesubtitles', False),
+                                       self.params.get('writeautomaticsub')])
+
+        if  subtitles_are_requested and 'subtitles' in info_dict and info_dict['subtitles']:
             # subtitles download errors are already managed as troubles in relevant IE
             # that way it will silently go on when used with unsupporting IE
-            subtitle = info_dict['subtitles'][0]
-            (sub_error, sub_lang, sub) = subtitle
+            subtitles = info_dict['subtitles']
             sub_format = self.params.get('subtitlesformat')
-            if sub_error:
-                self.report_warning("Some error while getting the subtitles")
-            else:
+            for sub_lang in subtitles.keys():
+                sub = subtitles[sub_lang]
+                if sub is None:
+                    continue
                 try:
-                    sub_filename = filename.rsplit('.', 1)[0] + u'.' + sub_lang + u'.' + sub_format
+                    sub_filename = subtitles_filename(filename, sub_lang, sub_format)
                     self.report_writesubtitles(sub_filename)
                     with io.open(encodeFilename(sub_filename), 'w', encoding='utf-8') as subfile:
-                        subfile.write(sub)
+                            subfile.write(sub)
                 except (OSError, IOError):
                     self.report_error(u'Cannot write subtitles file ' + descfn)
                     return
 
-        if self.params.get('allsubtitles', False) and 'subtitles' in info_dict and info_dict['subtitles']:
-            subtitles = info_dict['subtitles']
-            sub_format = self.params.get('subtitlesformat')
-            for subtitle in subtitles:
-                (sub_error, sub_lang, sub) = subtitle
-                if sub_error:
-                    self.report_warning("Some error while getting the subtitles")
-                else:
-                    try:
-                        sub_filename = filename.rsplit('.', 1)[0] + u'.' + sub_lang + u'.' + sub_format
-                        self.report_writesubtitles(sub_filename)
-                        with io.open(encodeFilename(sub_filename), 'w', encoding='utf-8') as subfile:
-                                subfile.write(sub)
-                    except (OSError, IOError):
-                        self.report_error(u'Cannot write subtitles file ' + descfn)
-                        return
-
         if self.params.get('writeinfojson', False):
             infofn = filename + u'.info.json'
             self.report_writeinfojson(infofn)
@@ -528,18 +650,20 @@ class YoutubeDL(object):
                 return
 
         if self.params.get('writethumbnail', False):
-            if 'thumbnail' in info_dict:
-                thumb_format = info_dict['thumbnail'].rpartition(u'/')[2].rpartition(u'.')[2]
-                if not thumb_format:
-                    thumb_format = 'jpg'
+            if info_dict.get('thumbnail') is not None:
+                thumb_format = determine_ext(info_dict['thumbnail'], u'jpg')
                 thumb_filename = filename.rpartition('.')[0] + u'.' + thumb_format
                 self.to_screen(u'[%s] %s: Downloading thumbnail ...' %
                                (info_dict['extractor'], info_dict['id']))
-                uf = compat_urllib_request.urlopen(info_dict['thumbnail'])
-                with open(thumb_filename, 'wb') as thumbf:
-                    shutil.copyfileobj(uf, thumbf)
-                self.to_screen(u'[%s] %s: Writing thumbnail to: %s' %
-                               (info_dict['extractor'], info_dict['id'], thumb_filename))
+                try:
+                    uf = compat_urllib_request.urlopen(info_dict['thumbnail'])
+                    with open(thumb_filename, 'wb') as thumbf:
+                        shutil.copyfileobj(uf, thumbf)
+                    self.to_screen(u'[%s] %s: Writing thumbnail to: %s' %
+                        (info_dict['extractor'], info_dict['id'], thumb_filename))
+                except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
+                    self.report_warning(u'Unable to download thumbnail "%s": %s' %
+                        (info_dict['thumbnail'], compat_str(err)))
 
         if not self.params.get('skip_download', False):
             if self.params.get('nooverwrites', False) and os.path.exists(encodeFilename(filename)):
@@ -547,11 +671,11 @@ class YoutubeDL(object):
             else:
                 try:
                     success = self.fd._do_download(filename, info_dict)
-                except (OSError, IOError) as err:
-                    raise UnavailableVideoError()
                 except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
                     self.report_error(u'unable to download video data: %s' % str(err))
                     return
+                except (OSError, IOError) as err:
+                    raise UnavailableVideoError(err)
                 except (ContentTooShortError, ) as err:
                     self.report_error(u'content too short (expected %s bytes and served %s)' % (err.expected, err.downloaded))
                     return
@@ -563,6 +687,8 @@ class YoutubeDL(object):
                     self.report_error(u'postprocessing: %s' % str(err))
                     return
 
+        self.record_download_archive(info_dict)
+
     def download(self, url_list):
         """Download a given list of URLs."""
         if len(url_list) > 1 and self.fixed_template():
@@ -595,10 +721,47 @@ class YoutubeDL(object):
                         # No clear decision yet, let IE decide
                         keep_video = keep_video_wish
             except PostProcessingError as e:
-                self.to_stderr(u'ERROR: ' + e.msg)
+                self.report_error(e.msg)
         if keep_video is False and not self.params.get('keepvideo', False):
             try:
                 self.to_screen(u'Deleting original file %s (pass -k to keep)' % filename)
                 os.remove(encodeFilename(filename))
             except (IOError, OSError):
                 self.report_warning(u'Unable to remove downloaded video file')
+
+    def in_download_archive(self, info_dict):
+        fn = self.params.get('download_archive')
+        if fn is None:
+            return False
+        vid_id = info_dict['extractor'] + u' ' + info_dict['id']
+        try:
+            with locked_file(fn, 'r', encoding='utf-8') as archive_file:
+                for line in archive_file:
+                    if line.strip() == vid_id:
+                        return True
+        except IOError as ioe:
+            if ioe.errno != errno.ENOENT:
+                raise
+        return False
+
+    def record_download_archive(self, info_dict):
+        fn = self.params.get('download_archive')
+        if fn is None:
+            return
+        vid_id = info_dict['extractor'] + u' ' + info_dict['id']
+        with locked_file(fn, 'a', encoding='utf-8') as archive_file:
+            archive_file.write(vid_id + u'\n')
+
+    def list_formats(self, info_dict):
+        formats_s = []
+        for format in info_dict.get('formats', [info_dict]):
+            formats_s.append("%s\t:\t%s\t[%s]" % (format['format_id'],
+                                                format['ext'],
+                                                format.get('format', '???'),
+                                                )
+                            )
+        if len(formats_s) != 1:
+            formats_s[0]  += ' (worst)'
+            formats_s[-1] += ' (best)'
+        formats_s = "\n".join(formats_s)
+        self.to_screen(u"[info] Available formats for %s:\nformat code\textension\n%s" % (info_dict['id'], formats_s)) 

youtube_dl/__init__.py

@@ -27,11 +27,17 @@ __authors__  = (
     'Johny Mo Swag',
     'Axel Noack',
     'Albert Kim',
+    'Pierre Rudloff',
+    'Huarong Huo',
+    'Ismael Mejía',
+    'Steffan \'Ruirize\' James',
+    'Andras Elso',
 )
 
 __license__ = 'Public Domain'
 
 import codecs
+import collections
 import getpass
 import optparse
 import os
@@ -41,16 +47,43 @@ import shlex
 import socket
 import subprocess
 import sys
-import warnings
+import traceback
 import platform
 
-from .utils import *
+
+from .utils import (
+    compat_cookiejar,
+    compat_print,
+    compat_str,
+    compat_urllib_request,
+    DateRange,
+    decodeOption,
+    determine_ext,
+    DownloadError,
+    get_cachedir,
+    make_HTTPS_handler,
+    MaxDownloadsReached,
+    platform_name,
+    preferredencoding,
+    SameFileError,
+    std_headers,
+    write_string,
+    YoutubeDLHandler,
+)
 from .update import update_self
 from .version import __version__
-from .FileDownloader import *
+from .FileDownloader import (
+    FileDownloader,
+)
 from .extractor import gen_extractors
 from .YoutubeDL import YoutubeDL
-from .PostProcessor import *
+from .PostProcessor import (
+    FFmpegMetadataPP,
+    FFmpegVideoConvertor,
+    FFmpegExtractAudioPP,
+    FFmpegEmbedSubtitlePP,
+)
+
 
 def parseOpts(overrideArguments=None):
     def _readOptions(filename_bytes):
@@ -82,6 +115,9 @@ def parseOpts(overrideArguments=None):
 
         return "".join(opts)
 
+    def _comma_separated_values_options_callback(option, opt_str, value, parser):
+        setattr(parser.values, option.dest, value.split(','))
+
     def _find_term_columns():
         columns = os.environ.get('COLUMNS', None)
         if columns:
@@ -95,6 +131,16 @@ def parseOpts(overrideArguments=None):
             pass
         return None
 
+    def _hide_login_info(opts):
+        opts = list(opts)
+        for private_opt in ['-p', '--password', '-u', '--username']:
+            try:
+                i = opts.index(private_opt)
+                opts[i+1] = '<PRIVATE>'
+            except ValueError:
+                pass
+        return opts
+
     max_width = 80
     max_help_position = 80
 
@@ -119,6 +165,7 @@ def parseOpts(overrideArguments=None):
     selection      = optparse.OptionGroup(parser, 'Video Selection')
     authentication = optparse.OptionGroup(parser, 'Authentication Options')
     video_format   = optparse.OptionGroup(parser, 'Video Format Options')
+    subtitles      = optparse.OptionGroup(parser, 'Subtitle Options')
     downloader     = optparse.OptionGroup(parser, 'Download Options')
     postproc       = optparse.OptionGroup(parser, 'Post-processing Options')
     filesystem     = optparse.OptionGroup(parser, 'Filesystem Options')
@@ -129,9 +176,9 @@ def parseOpts(overrideArguments=None):
     general.add_option('-v', '--version',
             action='version', help='print program version and exit')
     general.add_option('-U', '--update',
-            action='store_true', dest='update_self', help='update this program to latest version')
+            action='store_true', dest='update_self', help='update this program to latest version. Make sure that you have sufficient permissions (run with sudo if needed)')
     general.add_option('-i', '--ignore-errors',
-            action='store_true', dest='ignoreerrors', help='continue on download errors', default=False)
+            action='store_true', dest='ignoreerrors', help='continue on download errors, for example to to skip unavailable videos in a playlist', default=False)
     general.add_option('--dump-user-agent',
             action='store_true', dest='dump_user_agent',
             help='display the current browser identification', default=False)
@@ -148,6 +195,12 @@ def parseOpts(overrideArguments=None):
             help='Output descriptions of all supported extractors', default=False)
     general.add_option('--proxy', dest='proxy', default=None, help='Use the specified HTTP/HTTPS proxy', metavar='URL')
     general.add_option('--no-check-certificate', action='store_true', dest='no_check_certificate', default=False, help='Suppress HTTPS certificate validation.')
+    general.add_option(
+        '--cache-dir', dest='cachedir', default=get_cachedir(),
+        help='Location in the filesystem where youtube-dl can store downloaded information permanently. By default $XDG_CACHE_HOME/youtube-dl or ~/.cache/youtube-dl .')
+    general.add_option(
+        '--no-cache-dir', action='store_const', const=None, dest='cachedir',
+        help='Disable filesystem caching')
 
 
     selection.add_option('--playlist-start',
@@ -162,6 +215,13 @@ def parseOpts(overrideArguments=None):
     selection.add_option('--date', metavar='DATE', dest='date', help='download only videos uploaded in this date', default=None)
     selection.add_option('--datebefore', metavar='DATE', dest='datebefore', help='download only videos uploaded before this date', default=None)
     selection.add_option('--dateafter', metavar='DATE', dest='dateafter', help='download only videos uploaded after this date', default=None)
+    selection.add_option('--no-playlist', action='store_true', dest='noplaylist', help='download only the currently playing video', default=False)
+    selection.add_option('--age-limit', metavar='YEARS', dest='age_limit',
+                         help='download only videos suitable for the given age',
+                         default=None, type=int)
+    selection.add_option('--download-archive', metavar='FILE',
+                         dest='download_archive',
+                         help='Download only videos not present in the archive file. Record all downloaded videos in it.')
 
 
     authentication.add_option('-u', '--username',
@@ -175,8 +235,8 @@ def parseOpts(overrideArguments=None):
 
 
     video_format.add_option('-f', '--format',
-            action='store', dest='format', metavar='FORMAT',
-            help='video format code, specifiy the order of preference using slashes: "-f 22/17/18"')
+            action='store', dest='format', metavar='FORMAT', default='best',
+            help='video format code, specifiy the order of preference using slashes: "-f 22/17/18". "-f mp4" and "-f flv" are also supported')
     video_format.add_option('--all-formats',
             action='store_const', dest='format', help='download all available video formats', const='all')
     video_format.add_option('--prefer-free-formats',
@@ -185,34 +245,33 @@ def parseOpts(overrideArguments=None):
             action='store', dest='format_limit', metavar='FORMAT', help='highest quality format to download')
     video_format.add_option('-F', '--list-formats',
             action='store_true', dest='listformats', help='list all available formats (currently youtube only)')
-    video_format.add_option('--write-sub', '--write-srt',
+
+    subtitles.add_option('--write-sub', '--write-srt',
             action='store_true', dest='writesubtitles',
-            help='write subtitle file (currently youtube only)', default=False)
-    video_format.add_option('--write-auto-sub', '--write-automatic-sub',
+            help='write subtitle file', default=False)
+    subtitles.add_option('--write-auto-sub', '--write-automatic-sub',
             action='store_true', dest='writeautomaticsub',
-            help='write automatic subtitle file (currently youtube only)', default=False)
-    video_format.add_option('--only-sub',
-            action='store_true', dest='skip_download',
-            help='[deprecated] alias of --skip-download', default=False)
-    video_format.add_option('--all-subs',
+            help='write automatic subtitle file (youtube only)', default=False)
+    subtitles.add_option('--all-subs',
             action='store_true', dest='allsubtitles',
-            help='downloads all the available subtitles of the video (currently youtube only)', default=False)
-    video_format.add_option('--list-subs',
+            help='downloads all the available subtitles of the video', default=False)
+    subtitles.add_option('--list-subs',
             action='store_true', dest='listsubtitles',
-            help='lists all available subtitles for the video (currently youtube only)', default=False)
-    video_format.add_option('--sub-format',
+            help='lists all available subtitles for the video', default=False)
+    subtitles.add_option('--sub-format',
             action='store', dest='subtitlesformat', metavar='FORMAT',
-            help='subtitle format [srt/sbv/vtt] (default=srt) (currently youtube only)', default='srt')
-    video_format.add_option('--sub-lang', '--srt-lang',
-            action='store', dest='subtitleslang', metavar='LANG',
-            help='language of the subtitles to download (optional) use IETF language tags like \'en\'')
+            help='subtitle format (default=srt) ([sbv/vtt] youtube only)', default='srt')
+    subtitles.add_option('--sub-lang', '--sub-langs', '--srt-lang',
+            action='callback', dest='subtitleslangs', metavar='LANGS', type='str',
+            default=[], callback=_comma_separated_values_options_callback,
+            help='languages of the subtitles to download (optional) separated by commas, use IETF language tags like \'en,pt\'')
 
     downloader.add_option('-r', '--rate-limit',
-            dest='ratelimit', metavar='LIMIT', help='maximum download rate (e.g. 50k or 44.6m)')
+            dest='ratelimit', metavar='LIMIT', help='maximum download rate in bytes per second (e.g. 50K or 4.2M)')
     downloader.add_option('-R', '--retries',
             dest='retries', metavar='RETRIES', help='number of retries (default is %default)', default=10)
     downloader.add_option('--buffer-size',
-            dest='buffersize', metavar='SIZE', help='size of download buffer (e.g. 1024 or 16k) (default is %default)', default="1024")
+            dest='buffersize', metavar='SIZE', help='size of download buffer (e.g. 1024 or 16K) (default is %default)', default="1024")
     downloader.add_option('--no-resize-buffer',
             action='store_true', dest='noresizebuffer',
             help='do not automatically adjust the buffer size. By default, the buffer size is automatically resized from an initial value of SIZE.', default=False)
@@ -254,6 +313,10 @@ def parseOpts(overrideArguments=None):
     verbosity.add_option('--dump-intermediate-pages',
             action='store_true', dest='dump_intermediate_pages', default=False,
             help='print downloaded pages to debug problems(very verbose)')
+    verbosity.add_option('--youtube-print-sig-code',
+            action='store_true', dest='youtube_print_sig_code', default=False,
+            help=optparse.SUPPRESS_HELP)
+
 
     filesystem.add_option('-t', '--title',
             action='store_true', dest='usetitle', help='use title in file name (default)', default=False)
@@ -303,6 +366,9 @@ def parseOpts(overrideArguments=None):
     filesystem.add_option('--write-info-json',
             action='store_true', dest='writeinfojson',
             help='write video metadata to a .info.json file', default=False)
+    filesystem.add_option('--write-annotations',
+            action='store_true', dest='writeannotations',
+            help='write video annotations to a .annotation file', default=False)
     filesystem.add_option('--write-thumbnail',
             action='store_true', dest='writethumbnail',
             help='write thumbnail image to disk', default=False)
@@ -320,6 +386,10 @@ def parseOpts(overrideArguments=None):
             help='keeps the video file on disk after the post-processing; the video is erased by default')
     postproc.add_option('--no-post-overwrites', action='store_true', dest='nopostoverwrites', default=False,
             help='do not overwrite post-processed files; the post-processed files are overwritten by default')
+    postproc.add_option('--embed-subs', action='store_true', dest='embedsubtitles', default=False,
+            help='embed subtitles in the video (only for mp4 videos)')
+    postproc.add_option('--add-metadata', action='store_true', dest='addmetadata', default=False,
+            help='add metadata to the files')
 
 
     parser.add_option_group(general)
@@ -328,28 +398,33 @@ def parseOpts(overrideArguments=None):
     parser.add_option_group(filesystem)
     parser.add_option_group(verbosity)
     parser.add_option_group(video_format)
+    parser.add_option_group(subtitles)
     parser.add_option_group(authentication)
     parser.add_option_group(postproc)
 
     if overrideArguments is not None:
         opts, args = parser.parse_args(overrideArguments)
         if opts.verbose:
-            sys.stderr.write(u'[debug] Override config: ' + repr(overrideArguments) + '\n')
+            write_string(u'[debug] Override config: ' + repr(overrideArguments) + '\n')
     else:
         xdg_config_home = os.environ.get('XDG_CONFIG_HOME')
         if xdg_config_home:
-            userConfFile = os.path.join(xdg_config_home, 'youtube-dl.conf')
+            userConfFile = os.path.join(xdg_config_home, 'youtube-dl', 'config')
+            if not os.path.isfile(userConfFile):
+                userConfFile = os.path.join(xdg_config_home, 'youtube-dl.conf')
         else:
-            userConfFile = os.path.join(os.path.expanduser('~'), '.config', 'youtube-dl.conf')
+            userConfFile = os.path.join(os.path.expanduser('~'), '.config', 'youtube-dl', 'config')
+            if not os.path.isfile(userConfFile):
+                userConfFile = os.path.join(os.path.expanduser('~'), '.config', 'youtube-dl.conf')
         systemConf = _readOptions('/etc/youtube-dl.conf')
         userConf = _readOptions(userConfFile)
-        commandLineConf = sys.argv[1:] 
+        commandLineConf = sys.argv[1:]
         argv = systemConf + userConf + commandLineConf
         opts, args = parser.parse_args(argv)
         if opts.verbose:
-            sys.stderr.write(u'[debug] System config: ' + repr(systemConf) + '\n')
-            sys.stderr.write(u'[debug] User config: ' + repr(userConf) + '\n')
-            sys.stderr.write(u'[debug] Command-line args: ' + repr(commandLineConf) + '\n')
+            write_string(u'[debug] System config: ' + repr(_hide_login_info(systemConf)) + '\n')
+            write_string(u'[debug] User config: ' + repr(_hide_login_info(userConf)) + '\n')
+            write_string(u'[debug] Command-line args: ' + repr(_hide_login_info(commandLineConf)) + '\n')
 
     return parser, opts, args
 
@@ -372,12 +447,12 @@ def _real_main(argv=None):
         except (IOError, OSError) as err:
             if opts.verbose:
                 traceback.print_exc()
-            sys.stderr.write(u'ERROR: unable to open cookie file\n')
+            write_string(u'ERROR: unable to open cookie file\n')
             sys.exit(101)
     # Set user agent
     if opts.user_agent is not None:
         std_headers['User-Agent'] = opts.user_agent
-    
+
     # Set referer
     if opts.referer is not None:
         std_headers['Referer'] = opts.referer
@@ -398,28 +473,14 @@ def _real_main(argv=None):
             batchurls = batchfd.readlines()
             batchurls = [x.strip() for x in batchurls]
             batchurls = [x for x in batchurls if len(x) > 0 and not re.search(r'^[#/;]', x)]
+            if opts.verbose:
+                write_string(u'[debug] Batch file urls: ' + repr(batchurls) + u'\n')
         except IOError:
             sys.exit(u'ERROR: batch file could not be read')
     all_urls = batchurls + args
     all_urls = [url.strip() for url in all_urls]
 
-    # General configuration
-    cookie_processor = compat_urllib_request.HTTPCookieProcessor(jar)
-    if opts.proxy is not None:
-        if opts.proxy == '':
-            proxies = {}
-        else:
-            proxies = {'http': opts.proxy, 'https': opts.proxy}
-    else:
-        proxies = compat_urllib_request.getproxies()
-        # Set HTTPS proxy to HTTP one if given (https://github.com/rg3/youtube-dl/issues/805)
-        if 'http' in proxies and 'https' not in proxies:
-            proxies['https'] = proxies['http']
-    proxy_handler = compat_urllib_request.ProxyHandler(proxies)
-    https_handler = make_HTTPS_handler(opts)
-    opener = compat_urllib_request.build_opener(https_handler, proxy_handler, cookie_processor, YoutubeDLHandler())
-    compat_urllib_request.install_opener(opener)
-    socket.setdefaulttimeout(300) # 5 minutes should be enough (famous last words)
+    opener = _setup_opener(jar=jar, opts=opts)
 
     extractors = gen_extractors()
 
@@ -436,6 +497,8 @@ def _real_main(argv=None):
             if not ie._WORKING:
                 continue
             desc = getattr(ie, 'IE_DESC', ie.IE_NAME)
+            if desc is False:
+                continue
             if hasattr(ie, 'SEARCH_KEY'):
                 _SEARCHES = (u'cute kittens', u'slithering pythons', u'falling cat', u'angry poodle', u'purple fish', u'running tortoise')
                 _COUNTS = (u'', u'5', u'10', u'all')
@@ -507,6 +570,11 @@ def _real_main(argv=None):
     else:
         date = DateRange(opts.dateafter, opts.datebefore)
 
+    # --all-sub automatically sets --write-sub if --write-auto-sub is not given
+    # this was the old behaviour if only --all-sub was given.
+    if opts.allsubtitles and (opts.writeautomaticsub == False):
+        opts.writesubtitles = True
+
     if sys.version_info < (3,):
         # In Python 2, sys.argv is a bytestring (also note http://bugs.python.org/issue2128 for Windows systems)
         if opts.outtmpl is not None:
@@ -519,6 +587,10 @@ def _real_main(argv=None):
             or (opts.useid and u'%(id)s.%(ext)s')
             or (opts.autonumber and u'%(autonumber)s-%(id)s.%(ext)s')
             or u'%(title)s-%(id)s.%(ext)s')
+    if '%(ext)s' not in outtmpl and opts.extractaudio:
+        parser.error(u'Cannot download a video and extract audio into the same'
+                     u' file! Use "%%(ext)s" instead of %r' %
+                     determine_ext(outtmpl, u''))
 
     # YoutubeDL
     ydl = YoutubeDL({
@@ -553,11 +625,13 @@ def _real_main(argv=None):
         'progress_with_newline': opts.progress_with_newline,
         'playliststart': opts.playliststart,
         'playlistend': opts.playlistend,
+        'noplaylist': opts.noplaylist,
         'logtostderr': opts.outtmpl == '-',
         'consoletitle': opts.consoletitle,
         'nopart': opts.nopart,
         'updatetime': opts.updatetime,
         'writedescription': opts.writedescription,
+        'writeannotations': opts.writeannotations,
         'writeinfojson': opts.writeinfojson,
         'writethumbnail': opts.writethumbnail,
         'writesubtitles': opts.writesubtitles,
@@ -565,7 +639,7 @@ def _real_main(argv=None):
         'allsubtitles': opts.allsubtitles,
         'listsubtitles': opts.listsubtitles,
         'subtitlesformat': opts.subtitlesformat,
-        'subtitleslang': opts.subtitleslang,
+        'subtitleslangs': opts.subtitleslangs,
         'matchtitle': decodeOption(opts.matchtitle),
         'rejecttitle': decodeOption(opts.rejecttitle),
         'max_downloads': opts.max_downloads,
@@ -577,10 +651,14 @@ def _real_main(argv=None):
         'min_filesize': opts.min_filesize,
         'max_filesize': opts.max_filesize,
         'daterange': date,
+        'cachedir': opts.cachedir,
+        'youtube_print_sig_code': opts.youtube_print_sig_code,
+        'age_limit': opts.age_limit,
+        'download_archive': opts.download_archive,
         })
 
     if opts.verbose:
-        ydl.to_screen(u'[debug] youtube-dl version ' + __version__)
+        write_string(u'[debug] youtube-dl version ' + __version__ + u'\n')
         try:
             sp = subprocess.Popen(
                 ['git', 'rev-parse', '--short', 'HEAD'],
@@ -589,23 +667,36 @@ def _real_main(argv=None):
             out, err = sp.communicate()
             out = out.decode().strip()
             if re.match('[0-9a-f]+', out):
-                ydl.to_screen(u'[debug] Git HEAD: ' + out)
+                write_string(u'[debug] Git HEAD: ' + out + u'\n')
         except:
-            sys.exc_clear()
-        ydl.to_screen(u'[debug] Python version %s - %s' %(platform.python_version(), platform.platform()))
-        ydl.to_screen(u'[debug] Proxy map: ' + str(proxy_handler.proxies))
+            try:
+                sys.exc_clear()
+            except:
+                pass
+        write_string(u'[debug] Python version %s - %s' %(platform.python_version(), platform_name()) + u'\n')
+
+        proxy_map = {}
+        for handler in opener.handlers:
+            if hasattr(handler, 'proxies'):
+                proxy_map.update(handler.proxies)
+        write_string(u'[debug] Proxy map: ' + compat_str(proxy_map) + u'\n')
 
     ydl.add_default_info_extractors()
 
     # PostProcessors
+    # Add the metadata pp first, the other pps will copy it
+    if opts.addmetadata:
+        ydl.add_post_processor(FFmpegMetadataPP())
     if opts.extractaudio:
         ydl.add_post_processor(FFmpegExtractAudioPP(preferredcodec=opts.audioformat, preferredquality=opts.audioquality, nopostoverwrites=opts.nopostoverwrites))
     if opts.recodevideo:
         ydl.add_post_processor(FFmpegVideoConvertor(preferedformat=opts.recodevideo))
+    if opts.embedsubtitles:
+        ydl.add_post_processor(FFmpegEmbedSubtitlePP(subtitlesformat=opts.subtitlesformat))
 
     # Update version
     if opts.update_self:
-        update_self(ydl.to_screen, opts.verbose, sys.argv[0])
+        update_self(ydl.to_screen, opts.verbose)
 
     # Maybe do nothing
     if len(all_urls) < 1:
@@ -624,11 +715,42 @@ def _real_main(argv=None):
     if opts.cookiefile is not None:
         try:
             jar.save()
-        except (IOError, OSError) as err:
+        except (IOError, OSError):
             sys.exit(u'ERROR: unable to save cookie jar')
 
     sys.exit(retcode)
 
+
+def _setup_opener(jar=None, opts=None, timeout=300):
+    if opts is None:
+        FakeOptions = collections.namedtuple(
+            'FakeOptions', ['proxy', 'no_check_certificate'])
+        opts = FakeOptions(proxy=None, no_check_certificate=False)
+
+    cookie_processor = compat_urllib_request.HTTPCookieProcessor(jar)
+    if opts.proxy is not None:
+        if opts.proxy == '':
+            proxies = {}
+        else:
+            proxies = {'http': opts.proxy, 'https': opts.proxy}
+    else:
+        proxies = compat_urllib_request.getproxies()
+        # Set HTTPS proxy to HTTP one if given (https://github.com/rg3/youtube-dl/issues/805)
+        if 'http' in proxies and 'https' not in proxies:
+            proxies['https'] = proxies['http']
+    proxy_handler = compat_urllib_request.ProxyHandler(proxies)
+    https_handler = make_HTTPS_handler(opts)
+    opener = compat_urllib_request.build_opener(
+        https_handler, proxy_handler, cookie_processor, YoutubeDLHandler())
+    # Delete the default user-agent header, which would otherwise apply in
+    # cases where our custom HTTP handler doesn't come into play
+    # (See https://github.com/rg3/youtube-dl/issues/1309 for details)
+    opener.addheaders = []
+    compat_urllib_request.install_opener(opener)
+    socket.setdefaulttimeout(timeout)
+    return opener
+
+
 def main(argv=None):
     try:
         _real_main(argv)

youtube_dl/aes.py

@@ -0,0 +1,202 @@
+__all__ = ['aes_encrypt', 'key_expansion', 'aes_ctr_decrypt', 'aes_decrypt_text']
+
+import base64
+from math import ceil
+
+from .utils import bytes_to_intlist, intlist_to_bytes
+
+BLOCK_SIZE_BYTES = 16
+
+def aes_ctr_decrypt(data, key, counter):
+    """
+    Decrypt with aes in counter mode
+    
+    @param {int[]} data        cipher
+    @param {int[]} key         16/24/32-Byte cipher key
+    @param {instance} counter  Instance whose next_value function (@returns {int[]}  16-Byte block)
+                               returns the next counter block
+    @returns {int[]}           decrypted data
+    """
+    expanded_key = key_expansion(key)
+    block_count = int(ceil(float(len(data)) / BLOCK_SIZE_BYTES))
+    
+    decrypted_data=[]
+    for i in range(block_count):
+        counter_block = counter.next_value()
+        block = data[i*BLOCK_SIZE_BYTES : (i+1)*BLOCK_SIZE_BYTES]
+        block += [0]*(BLOCK_SIZE_BYTES - len(block))
+        
+        cipher_counter_block = aes_encrypt(counter_block, expanded_key)
+        decrypted_data += xor(block, cipher_counter_block)
+    decrypted_data = decrypted_data[:len(data)]
+    
+    return decrypted_data
+
+def key_expansion(data):
+    """
+    Generate key schedule
+    
+    @param {int[]} data  16/24/32-Byte cipher key
+    @returns {int[]}     176/208/240-Byte expanded key 
+    """
+    data = data[:] # copy
+    rcon_iteration = 1
+    key_size_bytes = len(data)
+    expanded_key_size_bytes = (key_size_bytes // 4 + 7) * BLOCK_SIZE_BYTES
+    
+    while len(data) < expanded_key_size_bytes:
+        temp = data[-4:]
+        temp = key_schedule_core(temp, rcon_iteration)
+        rcon_iteration += 1
+        data += xor(temp, data[-key_size_bytes : 4-key_size_bytes])
+        
+        for _ in range(3):
+            temp = data[-4:]
+            data += xor(temp, data[-key_size_bytes : 4-key_size_bytes])
+        
+        if key_size_bytes == 32:
+            temp = data[-4:]
+            temp = sub_bytes(temp)
+            data += xor(temp, data[-key_size_bytes : 4-key_size_bytes])
+        
+        for _ in range(3 if key_size_bytes == 32  else 2 if key_size_bytes == 24 else 0):
+            temp = data[-4:]
+            data += xor(temp, data[-key_size_bytes : 4-key_size_bytes])
+    data = data[:expanded_key_size_bytes]
+    
+    return data
+
+def aes_encrypt(data, expanded_key):
+    """
+    Encrypt one block with aes
+    
+    @param {int[]} data          16-Byte state
+    @param {int[]} expanded_key  176/208/240-Byte expanded key 
+    @returns {int[]}             16-Byte cipher
+    """
+    rounds = len(expanded_key) // BLOCK_SIZE_BYTES - 1
+    
+    data = xor(data, expanded_key[:BLOCK_SIZE_BYTES])
+    for i in range(1, rounds+1):
+        data = sub_bytes(data)
+        data = shift_rows(data)
+        if i != rounds:
+            data = mix_columns(data)
+        data = xor(data, expanded_key[i*BLOCK_SIZE_BYTES : (i+1)*BLOCK_SIZE_BYTES])
+    
+    return data
+
+def aes_decrypt_text(data, password, key_size_bytes):
+    """
+    Decrypt text
+    - The first 8 Bytes of decoded 'data' are the 8 high Bytes of the counter
+    - The cipher key is retrieved by encrypting the first 16 Byte of 'password'
+      with the first 'key_size_bytes' Bytes from 'password' (if necessary filled with 0's)
+    - Mode of operation is 'counter'
+    
+    @param {str} data                    Base64 encoded string
+    @param {str,unicode} password        Password (will be encoded with utf-8)
+    @param {int} key_size_bytes          Possible values: 16 for 128-Bit, 24 for 192-Bit or 32 for 256-Bit
+    @returns {str}                       Decrypted data
+    """
+    NONCE_LENGTH_BYTES = 8
+    
+    data = bytes_to_intlist(base64.b64decode(data))
+    password = bytes_to_intlist(password.encode('utf-8'))
+    
+    key = password[:key_size_bytes] + [0]*(key_size_bytes - len(password))
+    key = aes_encrypt(key[:BLOCK_SIZE_BYTES], key_expansion(key)) * (key_size_bytes // BLOCK_SIZE_BYTES)
+    
+    nonce = data[:NONCE_LENGTH_BYTES]
+    cipher = data[NONCE_LENGTH_BYTES:]
+    
+    class Counter:
+        __value = nonce + [0]*(BLOCK_SIZE_BYTES - NONCE_LENGTH_BYTES)
+        def next_value(self):
+            temp = self.__value
+            self.__value = inc(self.__value)
+            return temp
+    
+    decrypted_data = aes_ctr_decrypt(cipher, key, Counter())
+    plaintext = intlist_to_bytes(decrypted_data)
+    
+    return plaintext
+
+RCON = (0x8d, 0x01, 0x02, 0x04, 0x08, 0x10, 0x20, 0x40, 0x80, 0x1b, 0x36)
+SBOX = (0x63, 0x7C, 0x77, 0x7B, 0xF2, 0x6B, 0x6F, 0xC5, 0x30, 0x01, 0x67, 0x2B, 0xFE, 0xD7, 0xAB, 0x76,
+        0xCA, 0x82, 0xC9, 0x7D, 0xFA, 0x59, 0x47, 0xF0, 0xAD, 0xD4, 0xA2, 0xAF, 0x9C, 0xA4, 0x72, 0xC0,
+        0xB7, 0xFD, 0x93, 0x26, 0x36, 0x3F, 0xF7, 0xCC, 0x34, 0xA5, 0xE5, 0xF1, 0x71, 0xD8, 0x31, 0x15,
+        0x04, 0xC7, 0x23, 0xC3, 0x18, 0x96, 0x05, 0x9A, 0x07, 0x12, 0x80, 0xE2, 0xEB, 0x27, 0xB2, 0x75,
+        0x09, 0x83, 0x2C, 0x1A, 0x1B, 0x6E, 0x5A, 0xA0, 0x52, 0x3B, 0xD6, 0xB3, 0x29, 0xE3, 0x2F, 0x84,
+        0x53, 0xD1, 0x00, 0xED, 0x20, 0xFC, 0xB1, 0x5B, 0x6A, 0xCB, 0xBE, 0x39, 0x4A, 0x4C, 0x58, 0xCF,
+        0xD0, 0xEF, 0xAA, 0xFB, 0x43, 0x4D, 0x33, 0x85, 0x45, 0xF9, 0x02, 0x7F, 0x50, 0x3C, 0x9F, 0xA8,
+        0x51, 0xA3, 0x40, 0x8F, 0x92, 0x9D, 0x38, 0xF5, 0xBC, 0xB6, 0xDA, 0x21, 0x10, 0xFF, 0xF3, 0xD2,
+        0xCD, 0x0C, 0x13, 0xEC, 0x5F, 0x97, 0x44, 0x17, 0xC4, 0xA7, 0x7E, 0x3D, 0x64, 0x5D, 0x19, 0x73,
+        0x60, 0x81, 0x4F, 0xDC, 0x22, 0x2A, 0x90, 0x88, 0x46, 0xEE, 0xB8, 0x14, 0xDE, 0x5E, 0x0B, 0xDB,
+        0xE0, 0x32, 0x3A, 0x0A, 0x49, 0x06, 0x24, 0x5C, 0xC2, 0xD3, 0xAC, 0x62, 0x91, 0x95, 0xE4, 0x79,
+        0xE7, 0xC8, 0x37, 0x6D, 0x8D, 0xD5, 0x4E, 0xA9, 0x6C, 0x56, 0xF4, 0xEA, 0x65, 0x7A, 0xAE, 0x08,
+        0xBA, 0x78, 0x25, 0x2E, 0x1C, 0xA6, 0xB4, 0xC6, 0xE8, 0xDD, 0x74, 0x1F, 0x4B, 0xBD, 0x8B, 0x8A,
+        0x70, 0x3E, 0xB5, 0x66, 0x48, 0x03, 0xF6, 0x0E, 0x61, 0x35, 0x57, 0xB9, 0x86, 0xC1, 0x1D, 0x9E,
+        0xE1, 0xF8, 0x98, 0x11, 0x69, 0xD9, 0x8E, 0x94, 0x9B, 0x1E, 0x87, 0xE9, 0xCE, 0x55, 0x28, 0xDF,
+        0x8C, 0xA1, 0x89, 0x0D, 0xBF, 0xE6, 0x42, 0x68, 0x41, 0x99, 0x2D, 0x0F, 0xB0, 0x54, 0xBB, 0x16)
+MIX_COLUMN_MATRIX = ((2,3,1,1),
+                     (1,2,3,1),
+                     (1,1,2,3),
+                     (3,1,1,2))
+
+def sub_bytes(data):
+    return [SBOX[x] for x in data]
+
+def rotate(data):
+    return data[1:] + [data[0]]
+
+def key_schedule_core(data, rcon_iteration):
+    data = rotate(data)
+    data = sub_bytes(data)
+    data[0] = data[0] ^ RCON[rcon_iteration]
+    
+    return data
+
+def xor(data1, data2):
+    return [x^y for x, y in zip(data1, data2)]
+
+def mix_column(data):
+    data_mixed = []
+    for row in range(4):
+        mixed = 0
+        for column in range(4):
+            addend = data[column]
+            if MIX_COLUMN_MATRIX[row][column] in (2,3):
+                addend <<= 1
+                if addend > 0xff:
+                    addend &= 0xff
+                    addend ^= 0x1b
+                if MIX_COLUMN_MATRIX[row][column] == 3:
+                    addend ^= data[column]
+            mixed ^= addend & 0xff
+        data_mixed.append(mixed)
+    return data_mixed
+
+def mix_columns(data):
+    data_mixed = []
+    for i in range(4):
+        column = data[i*4 : (i+1)*4]
+        data_mixed += mix_column(column)
+    return data_mixed
+
+def shift_rows(data):
+    data_shifted = []
+    for column in range(4):
+        for row in range(4):
+            data_shifted.append( data[((column + row) & 0b11) * 4 + row] )
+    return data_shifted
+
+def inc(data):
+    data = data[:] # copy
+    for i in range(len(data)-1,-1,-1):
+        if data[i] == 255:
+            data[i] = 0
+        else:
+            data[i] = data[i] + 1
+            break
+    return data

youtube_dl/extractor/__init__.py

@@ -1,64 +1,145 @@
-
+from .appletrailers import AppleTrailersIE
+from .addanime import AddAnimeIE
+from .archiveorg import ArchiveOrgIE
 from .ard import ARDIE
-from .arte import ArteTvIE
+from .arte import (
+    ArteTvIE,
+    ArteTVPlus7IE,
+    ArteTVCreativeIE,
+    ArteTVFutureIE,
+)
 from .auengine import AUEngineIE
 from .bandcamp import BandcampIE
 from .bliptv import BlipTVIE, BlipTVUserIE
+from .bloomberg import BloombergIE
 from .breakcom import BreakIE
+from .brightcove import BrightcoveIE
+from .c56 import C56IE
+from .canalplus import CanalplusIE
+from .canalc2 import Canalc2IE
+from .cinemassacre import CinemassacreIE
+from .cnn import CNNIE
 from .collegehumor import CollegeHumorIE
 from .comedycentral import ComedyCentralIE
+from .condenast import CondeNastIE
+from .criterion import CriterionIE
 from .cspan import CSpanIE
-from .dailymotion import DailymotionIE
+from .dailymotion import (
+    DailymotionIE,
+    DailymotionPlaylistIE,
+    DailymotionUserIE,
+)
+from .daum import DaumIE
 from .depositfiles import DepositFilesIE
+from .dotsub import DotsubIE
+from .dreisat import DreiSatIE
+from .defense import DefenseGouvFrIE
+from .ebaumsworld import EbaumsWorldIE
+from .ehow import EHowIE
 from .eighttracks import EightTracksIE
 from .escapist import EscapistIE
+from .exfm import ExfmIE
 from .facebook import FacebookIE
+from .faz import FazIE
+from .fktv import (
+    FKTVIE,
+    FKTVPosteckeIE,
+)
 from .flickr import FlickrIE
+from .francetv import (
+    PluzzIE,
+    FranceTvInfoIE,
+    France2IE,
+    GenerationQuoiIE
+)
+from .freesound import FreesoundIE
 from .funnyordie import FunnyOrDieIE
 from .gamespot import GameSpotIE
 from .gametrailers import GametrailersIE
 from .generic import GenericIE
 from .googleplus import GooglePlusIE
 from .googlesearch import GoogleSearchIE
+from .hark import HarkIE
 from .hotnewhiphop import HotNewHipHopIE
 from .howcast import HowcastIE
 from .hypem import HypemIE
+from .ign import IGNIE, OneUPIE
 from .ina import InaIE
 from .infoq import InfoQIE
 from .instagram import InstagramIE
+from .internetvideoarchive import InternetVideoArchiveIE
+from .jeuxvideo import JeuxVideoIE
 from .jukebox import JukeboxIE
 from .justintv import JustinTVIE
+from .kankan import KankanIE
+from .kickstarter import KickStarterIE
 from .keek import KeekIE
 from .liveleak import LiveLeakIE
+from .livestream import LivestreamIE
 from .metacafe import MetacafeIE
+from .metacritic import MetacriticIE
+from .mit import TechTVMITIE, MITIE
 from .mixcloud import MixcloudIE
 from .mtv import MTVIE
+from .muzu import MuzuTVIE
 from .myspass import MySpassIE
 from .myvideo import MyVideoIE
+from .naver import NaverIE
 from .nba import NBAIE
+from .nbc import NBCNewsIE
+from .newgrounds import NewgroundsIE
+from .nhl import NHLIE, NHLVideocenterIE
+from .nowvideo import NowVideoIE
+from .ooyala import OoyalaIE
+from .orf import ORFIE
+from .pbs import PBSIE
 from .photobucket import PhotobucketIE
 from .pornotube import PornotubeIE
 from .rbmaradio import RBMARadioIE
 from .redtube import RedTubeIE
 from .ringtv import RingTVIE
-from .soundcloud import SoundcloudIE, SoundcloudSetIE
+from .ro220 import Ro220IE
+from .rottentomatoes import RottenTomatoesIE
+from .roxwel import RoxwelIE
+from .rtlnow import RTLnowIE
+from .rutube import RutubeIE
+from .sina import SinaIE
+from .slashdot import SlashdotIE
+from .slideshare import SlideshareIE
+from .sohu import SohuIE
+from .soundcloud import SoundcloudIE, SoundcloudSetIE, SoundcloudUserIE
+from .southparkstudios import SouthParkStudiosIE
 from .spiegel import SpiegelIE
 from .stanfordoc import StanfordOpenClassroomIE
 from .statigram import StatigramIE
 from .steam import SteamIE
+from .sztvhu import SztvHuIE
 from .teamcoco import TeamcocoIE
+from .techtalks import TechTalksIE
 from .ted import TEDIE
 from .tf1 import TF1IE
+from .thisav import ThisAVIE
 from .traileraddict import TrailerAddictIE
+from .trilulilu import TriluliluIE
 from .tudou import TudouIE
 from .tumblr import TumblrIE
 from .tutv import TutvIE
-from .ustream import UstreamIE
+from .unistra import UnistraIE
+from .ustream import UstreamIE, UstreamChannelIE
 from .vbox7 import Vbox7IE
+from .veehd import VeeHDIE
+from .veoh import VeohIE
 from .vevo import VevoIE
-from .vimeo import VimeoIE
+from .vice import ViceIE
+from .viddler import ViddlerIE
+from .videodetective import VideoDetectiveIE
+from .videofyme import VideofyMeIE
+from .videopremium import VideoPremiumIE
+from .vimeo import VimeoIE, VimeoChannelIE
 from .vine import VineIE
 from .wat import WatIE
+from .websurg import WeBSurgIE
+from .weibo import WeiboIE
 from .wimp import WimpIE
 from .worldstarhiphop import WorldStarHipHopIE
 from .xhamster import XHamsterIE
@@ -68,7 +149,19 @@ from .yahoo import YahooIE, YahooSearchIE
 from .youjizz import YouJizzIE
 from .youku import YoukuIE
 from .youporn import YouPornIE
-from .youtube import YoutubeIE, YoutubePlaylistIE, YoutubeSearchIE, YoutubeUserIE, YoutubeChannelIE, YoutubeShowIE
+from .youtube import (
+    YoutubeIE,
+    YoutubePlaylistIE,
+    YoutubeSearchIE,
+    YoutubeUserIE,
+    YoutubeChannelIE,
+    YoutubeShowIE,
+    YoutubeSubscriptionsIE,
+    YoutubeRecommendedIE,
+    YoutubeTruncatedURLIE,
+    YoutubeWatchLaterIE,
+    YoutubeFavouritesIE,
+)
 from .zdf import ZDFIE
 
 
@@ -79,12 +172,14 @@ _ALL_CLASSES = [
 ]
 _ALL_CLASSES.append(GenericIE)
 
+
 def gen_extractors():
     """ Return a list of an instance of every supported extractor.
     The order does matter; the first extractor matched is the one handling the URL.
     """
     return [klass() for klass in _ALL_CLASSES]
 
+
 def get_info_extractor(ie_name):
     """Returns the info extractor class with the given ie_name"""
     return globals()[ie_name+'IE']

youtube_dl/extractor/addanime.py

@@ -0,0 +1,75 @@
+import re
+
+from .common import InfoExtractor
+from ..utils import (
+    compat_HTTPError,
+    compat_str,
+    compat_urllib_parse,
+    compat_urllib_parse_urlparse,
+
+    ExtractorError,
+)
+
+
+class AddAnimeIE(InfoExtractor):
+
+    _VALID_URL = r'^http://(?:\w+\.)?add-anime\.net/watch_video.php\?(?:.*?)v=(?P<video_id>[\w_]+)(?:.*)'
+    IE_NAME = u'AddAnime'
+    _TEST = {
+        u'url': u'http://www.add-anime.net/watch_video.php?v=24MR3YO5SAS9',
+        u'file': u'24MR3YO5SAS9.flv',
+        u'md5': u'1036a0e0cd307b95bd8a8c3a5c8cfaf1',
+        u'info_dict': {
+            u"description": u"One Piece 606",
+            u"title": u"One Piece 606"
+        }
+    }
+
+    def _real_extract(self, url):
+        try:
+            mobj = re.match(self._VALID_URL, url)
+            video_id = mobj.group('video_id')
+            webpage = self._download_webpage(url, video_id)
+        except ExtractorError as ee:
+            if not isinstance(ee.cause, compat_HTTPError):
+                raise
+
+            redir_webpage = ee.cause.read().decode('utf-8')
+            action = self._search_regex(
+                r'<form id="challenge-form" action="([^"]+)"',
+                redir_webpage, u'Redirect form')
+            vc = self._search_regex(
+                r'<input type="hidden" name="jschl_vc" value="([^"]+)"/>',
+                redir_webpage, u'redirect vc value')
+            av = re.search(
+                r'a\.value = ([0-9]+)[+]([0-9]+)[*]([0-9]+);',
+                redir_webpage)
+            if av is None:
+                raise ExtractorError(u'Cannot find redirect math task')
+            av_res = int(av.group(1)) + int(av.group(2)) * int(av.group(3))
+
+            parsed_url = compat_urllib_parse_urlparse(url)
+            av_val = av_res + len(parsed_url.netloc)
+            confirm_url = (
+                parsed_url.scheme + u'://' + parsed_url.netloc +
+                action + '?' +
+                compat_urllib_parse.urlencode({
+                    'jschl_vc': vc, 'jschl_answer': compat_str(av_val)}))
+            self._download_webpage(
+                confirm_url, video_id,
+                note=u'Confirming after redirect')
+            webpage = self._download_webpage(url, video_id)
+
+        video_url = self._search_regex(r"var normal_video_file = '(.*?)';",
+                                       webpage, u'video file URL')
+        video_title = self._og_search_title(webpage)
+        video_description = self._og_search_description(webpage)
+
+        return {
+            '_type': 'video',
+            'id':  video_id,
+            'url': video_url,
+            'ext': 'flv',
+            'title': video_title,
+            'description': video_description
+        }

youtube_dl/extractor/appletrailers.py

@@ -0,0 +1,138 @@
+import re
+import xml.etree.ElementTree
+import json
+
+from .common import InfoExtractor
+from ..utils import (
+    compat_urlparse,
+    determine_ext,
+)
+
+
+class AppleTrailersIE(InfoExtractor):
+    _VALID_URL = r'https?://(?:www\.)?trailers.apple.com/trailers/(?P<company>[^/]+)/(?P<movie>[^/]+)'
+    _TEST = {
+        u"url": u"http://trailers.apple.com/trailers/wb/manofsteel/",
+        u"playlist": [
+            {
+                u"file": u"manofsteel-trailer4.mov",
+                u"md5": u"d97a8e575432dbcb81b7c3acb741f8a8",
+                u"info_dict": {
+                    u"duration": 111,
+                    u"title": u"Trailer 4",
+                    u"upload_date": u"20130523",
+                    u"uploader_id": u"wb",
+                },
+            },
+            {
+                u"file": u"manofsteel-trailer3.mov",
+                u"md5": u"b8017b7131b721fb4e8d6f49e1df908c",
+                u"info_dict": {
+                    u"duration": 182,
+                    u"title": u"Trailer 3",
+                    u"upload_date": u"20130417",
+                    u"uploader_id": u"wb",
+                },
+            },
+            {
+                u"file": u"manofsteel-trailer.mov",
+                u"md5": u"d0f1e1150989b9924679b441f3404d48",
+                u"info_dict": {
+                    u"duration": 148,
+                    u"title": u"Trailer",
+                    u"upload_date": u"20121212",
+                    u"uploader_id": u"wb",
+                },
+            },
+            {
+                u"file": u"manofsteel-teaser.mov",
+                u"md5": u"5fe08795b943eb2e757fa95cb6def1cb",
+                u"info_dict": {
+                    u"duration": 93,
+                    u"title": u"Teaser",
+                    u"upload_date": u"20120721",
+                    u"uploader_id": u"wb",
+                },
+            }
+        ]
+    }
+
+    _JSON_RE = r'iTunes.playURL\((.*?)\);'
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        movie = mobj.group('movie')
+        uploader_id = mobj.group('company')
+
+        playlist_url = compat_urlparse.urljoin(url, u'includes/playlists/itunes.inc')
+        playlist_snippet = self._download_webpage(playlist_url, movie)
+        playlist_cleaned = re.sub(r'(?s)<script[^<]*?>.*?</script>', u'', playlist_snippet)
+        playlist_cleaned = re.sub(r'<img ([^<]*?)>', r'<img \1/>', playlist_cleaned)
+        # The ' in the onClick attributes are not escaped, it couldn't be parsed
+        # with xml.etree.ElementTree.fromstring
+        # like: http://trailers.apple.com/trailers/wb/gravity/
+        def _clean_json(m):
+            return u'iTunes.playURL(%s);' % m.group(1).replace('\'', '&#39;')
+        playlist_cleaned = re.sub(self._JSON_RE, _clean_json, playlist_cleaned)
+        playlist_html = u'<html>' + playlist_cleaned + u'</html>'
+
+        doc = xml.etree.ElementTree.fromstring(playlist_html)
+        playlist = []
+        for li in doc.findall('./div/ul/li'):
+            on_click = li.find('.//a').attrib['onClick']
+            trailer_info_json = self._search_regex(self._JSON_RE,
+                on_click, u'trailer info')
+            trailer_info = json.loads(trailer_info_json)
+            title = trailer_info['title']
+            video_id = movie + '-' + re.sub(r'[^a-zA-Z0-9]', '', title).lower()
+            thumbnail = li.find('.//img').attrib['src']
+            upload_date = trailer_info['posted'].replace('-', '')
+
+            runtime = trailer_info['runtime']
+            m = re.search(r'(?P<minutes>[0-9]+):(?P<seconds>[0-9]{1,2})', runtime)
+            duration = None
+            if m:
+                duration = 60 * int(m.group('minutes')) + int(m.group('seconds'))
+
+            first_url = trailer_info['url']
+            trailer_id = first_url.split('/')[-1].rpartition('_')[0].lower()
+            settings_json_url = compat_urlparse.urljoin(url, 'includes/settings/%s.json' % trailer_id)
+            settings_json = self._download_webpage(settings_json_url, trailer_id, u'Downloading settings json')
+            settings = json.loads(settings_json)
+
+            formats = []
+            for format in settings['metadata']['sizes']:
+                # The src is a file pointing to the real video file
+                format_url = re.sub(r'_(\d*p.mov)', r'_h\1', format['src'])
+                formats.append({
+                    'url': format_url,
+                    'ext': determine_ext(format_url),
+                    'format': format['type'],
+                    'width': format['width'],
+                    'height': int(format['height']),
+                })
+            formats = sorted(formats, key=lambda f: (f['height'], f['width']))
+
+            info = {
+                '_type': 'video',
+                'id': video_id,
+                'title': title,
+                'formats': formats,
+                'title': title,
+                'duration': duration,
+                'thumbnail': thumbnail,
+                'upload_date': upload_date,
+                'uploader_id': uploader_id,
+                'user_agent': 'QuickTime compatible (youtube-dl)',
+            }
+            # TODO: Remove when #980 has been merged
+            info['url'] = formats[-1]['url']
+            info['ext'] = formats[-1]['ext']
+
+            playlist.append(info)
+
+        return {
+            '_type': 'playlist',
+            'id': movie,
+            'entries': playlist,
+        }

youtube_dl/extractor/archiveorg.py

@@ -0,0 +1,68 @@
+import json
+import re
+
+from .common import InfoExtractor
+from ..utils import (
+    determine_ext,
+    unified_strdate,
+)
+
+
+class ArchiveOrgIE(InfoExtractor):
+    IE_NAME = 'archive.org'
+    IE_DESC = 'archive.org videos'
+    _VALID_URL = r'(?:https?://)?(?:www\.)?archive.org/details/(?P<id>[^?/]+)(?:[?].*)?$'
+    _TEST = {
+        u"url": u"http://archive.org/details/XD300-23_68HighlightsAResearchCntAugHumanIntellect",
+        u'file': u'XD300-23_68HighlightsAResearchCntAugHumanIntellect.ogv',
+        u'md5': u'8af1d4cf447933ed3c7f4871162602db',
+        u'info_dict': {
+            u"title": u"1968 Demo - FJCC Conference Presentation Reel #1",
+            u"description": u"Reel 1 of 3: Also known as the \"Mother of All Demos\", Doug Engelbart's presentation at the Fall Joint Computer Conference in San Francisco, December 9, 1968 titled \"A Research Center for Augmenting Human Intellect.\" For this presentation, Doug and his team astonished the audience by not only relating their research, but demonstrating it live. This was the debut of the mouse, interactive computing, hypermedia, computer supported software engineering, video teleconferencing, etc. See also <a href=\"http://dougengelbart.org/firsts/dougs-1968-demo.html\" rel=\"nofollow\">Doug's 1968 Demo page</a> for more background, highlights, links, and the detailed paper published in this conference proceedings. Filmed on 3 reels: Reel 1 | <a href=\"http://www.archive.org/details/XD300-24_68HighlightsAResearchCntAugHumanIntellect\" rel=\"nofollow\">Reel 2</a> | <a href=\"http://www.archive.org/details/XD300-25_68HighlightsAResearchCntAugHumanIntellect\" rel=\"nofollow\">Reel 3</a>",
+            u"upload_date": u"19681210",
+            u"uploader": u"SRI International"
+        }
+    }
+
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        video_id = mobj.group('id')
+
+        json_url = url + (u'?' if u'?' in url else '&') + u'output=json'
+        json_data = self._download_webpage(json_url, video_id)
+        data = json.loads(json_data)
+
+        title = data['metadata']['title'][0]
+        description = data['metadata']['description'][0]
+        uploader = data['metadata']['creator'][0]
+        upload_date = unified_strdate(data['metadata']['date'][0])
+
+        formats = [{
+                'format': fdata['format'],
+                'url': 'http://' + data['server'] + data['dir'] + fn,
+                'file_size': int(fdata['size']),
+            }
+            for fn,fdata in data['files'].items()
+            if 'Video' in fdata['format']]
+        formats.sort(key=lambda fdata: fdata['file_size'])
+        for f in formats:
+            f['ext'] = determine_ext(f['url'])
+
+        info = {
+            '_type': 'video',
+            'id': video_id,
+            'title': title,
+            'formats': formats,
+            'description': description,
+            'uploader': uploader,
+            'upload_date': upload_date,
+        }
+        thumbnail = data.get('misc', {}).get('image')
+        if thumbnail:
+            info['thumbnail'] = thumbnail
+
+        # TODO: Remove when #980 has been merged
+        info.update(formats[-1])
+
+        return info

youtube_dl/extractor/ard.py

@@ -32,7 +32,7 @@ class ARDIE(InfoExtractor):
         # determine title and media streams from webpage
         html = self._download_webpage(url, video_id)
         title = re.search(self._TITLE, html).group('title')
-        streams = [m.groupdict() for m in re.finditer(self._MEDIA_STREAM, html)]
+        streams = [mo.groupdict() for mo in re.finditer(self._MEDIA_STREAM, html)]
         if not streams:
             assert '"fsk"' in html
             raise ExtractorError(u'This video is only available after 8:00 pm')

youtube_dl/extractor/arte.py

@@ -1,32 +1,34 @@
+# encoding: utf-8
 import re
 import json
+import xml.etree.ElementTree
 
 from .common import InfoExtractor
 from ..utils import (
-    # This is used by the not implemented extractLiveStream method
-    compat_urllib_parse,
-
     ExtractorError,
+    find_xpath_attr,
     unified_strdate,
+    determine_ext,
+    get_element_by_id,
 )
 
+# There are different sources of video in arte.tv, the extraction process 
+# is different for each one. The videos usually expire in 7 days, so we can't
+# add tests.
+
 class ArteTvIE(InfoExtractor):
-    """
-    There are two sources of video in arte.tv: videos.arte.tv and
-    www.arte.tv/guide, the extraction process is different for each one.
-    The videos expire in 7 days, so we can't add tests.
-    """
-    _EMISSION_URL = r'(?:http://)?www\.arte.tv/guide/(?:fr|de)/(?:(?:sendungen|emissions)/)?(?P<id>.*?)/(?P<name>.*?)(\?.*)?'
-    _VIDEOS_URL = r'(?:http://)?videos.arte.tv/(?:fr|de)/.*-(?P<id>.*?).html'
+    _VIDEOS_URL = r'(?:http://)?videos.arte.tv/(?P<lang>fr|de)/.*-(?P<id>.*?).html'
+    _LIVEWEB_URL = r'(?:http://)?liveweb.arte.tv/(?P<lang>fr|de)/(?P<subpage>.+?)/(?P<name>.+)'
     _LIVE_URL = r'index-[0-9]+\.html$'
 
     IE_NAME = u'arte.tv'
 
     @classmethod
     def suitable(cls, url):
-        return any(re.match(regex, url) for regex in (cls._EMISSION_URL, cls._VIDEOS_URL))
+        return any(re.match(regex, url) for regex in (cls._VIDEOS_URL, cls._LIVEWEB_URL))
 
     # TODO implement Live Stream
+    # from ..utils import compat_urllib_parse
     # def extractLiveStream(self, url):
     #     video_lang = url.split('/')[-4]
     #     info = self.grep_webpage(
@@ -54,62 +56,32 @@ class ArteTvIE(InfoExtractor):
     #     video_url = u'%s/%s' % (info.get('url'), info.get('path'))
 
     def _real_extract(self, url):
-        mobj = re.match(self._EMISSION_URL, url)
-        if mobj is not None:
-            name = mobj.group('name')
-            # This is not a real id, it can be for example AJT for the news
-            # http://www.arte.tv/guide/fr/emissions/AJT/arte-journal
-            video_id = mobj.group('id')
-            return self._extract_emission(url, video_id)
-
         mobj = re.match(self._VIDEOS_URL, url)
         if mobj is not None:
             id = mobj.group('id')
-            return self._extract_video(url, id)
+            lang = mobj.group('lang')
+            return self._extract_video(url, id, lang)
+
+        mobj = re.match(self._LIVEWEB_URL, url)
+        if mobj is not None:
+            name = mobj.group('name')
+            lang = mobj.group('lang')
+            return self._extract_liveweb(url, name, lang)
 
         if re.search(self._LIVE_URL, video_id) is not None:
             raise ExtractorError(u'Arte live streams are not yet supported, sorry')
             # self.extractLiveStream(url)
             # return
 
-    def _extract_emission(self, url, video_id):
-        """Extract from www.arte.tv/guide"""
-        webpage = self._download_webpage(url, video_id)
-        json_url = self._html_search_regex(r'arte_vp_url="(.*?)"', webpage, 'json url')
-
-        json_info = self._download_webpage(json_url, video_id, 'Downloading info json')
-        self.report_extraction(video_id)
-        info = json.loads(json_info)
-        player_info = info['videoJsonPlayer']
-
-        info_dict = {'id': player_info['VID'],
-                     'title': player_info['VTI'],
-                     'description': player_info['VDE'],
-                     'upload_date': unified_strdate(player_info['VDA'].split(' ')[0]),
-                     'thumbnail': player_info['programImage'],
-                     'ext': 'flv',
-                     }
-
-        formats = player_info['VSR'].values()
-        # We order the formats by quality
-        formats = sorted(formats, key=lambda f: int(f['height']))
-        # Pick the best quality
-        format_info = formats[-1]
-        if format_info['mediaType'] == u'rtmp':
-            info_dict['url'] = format_info['streamer']
-            info_dict['play_path'] = 'mp4:' + format_info['url']
-        else:
-            info_dict['url'] = format_info['url']
-
-        return info_dict
-
-    def _extract_video(self, url, video_id):
+    def _extract_video(self, url, video_id, lang):
         """Extract from videos.arte.tv"""
-        config_xml_url = url.replace('/videos/', '/do_delegate/videos/')
-        config_xml_url = config_xml_url.replace('.html', ',view,asPlayerXml.xml')
-        config_xml = self._download_webpage(config_xml_url, video_id)
-        config_xml_url = self._html_search_regex(r'<video lang=".*?" ref="(.*?)"', config_xml, 'config xml url')
-        config_xml = self._download_webpage(config_xml_url, video_id)
+        ref_xml_url = url.replace('/videos/', '/do_delegate/videos/')
+        ref_xml_url = ref_xml_url.replace('.html', ',view,asPlayerXml.xml')
+        ref_xml = self._download_webpage(ref_xml_url, video_id, note=u'Downloading metadata')
+        ref_xml_doc = xml.etree.ElementTree.fromstring(ref_xml)
+        config_node = find_xpath_attr(ref_xml_doc, './/video', 'lang', lang)
+        config_xml_url = config_node.attrib['ref']
+        config_xml = self._download_webpage(config_xml_url, video_id, note=u'Downloading configuration')
 
         video_urls = list(re.finditer(r'<url quality="(?P<quality>.*?)">(?P<url>.*?)</url>', config_xml))
         def _key(m):
@@ -131,3 +103,129 @@ class ArteTvIE(InfoExtractor):
                 'url': video_url,
                 'ext': 'flv',
                 }
+
+    def _extract_liveweb(self, url, name, lang):
+        """Extract form http://liveweb.arte.tv/"""
+        webpage = self._download_webpage(url, name)
+        video_id = self._search_regex(r'eventId=(\d+?)("|&)', webpage, u'event id')
+        config_xml = self._download_webpage('http://download.liveweb.arte.tv/o21/liveweb/events/event-%s.xml' % video_id,
+                                            video_id, u'Downloading information')
+        config_doc = xml.etree.ElementTree.fromstring(config_xml.encode('utf-8'))
+        event_doc = config_doc.find('event')
+        url_node = event_doc.find('video').find('urlHd')
+        if url_node is None:
+            url_node = video_doc.find('urlSd')
+
+        return {'id': video_id,
+                'title': event_doc.find('name%s' % lang.capitalize()).text,
+                'url': url_node.text.replace('MP4', 'mp4'),
+                'ext': 'flv',
+                'thumbnail': self._og_search_thumbnail(webpage),
+                }
+
+
+class ArteTVPlus7IE(InfoExtractor):
+    IE_NAME = u'arte.tv:+7'
+    _VALID_URL = r'https?://www\.arte.tv/guide/(?P<lang>fr|de)/(?:(?:sendungen|emissions)/)?(?P<id>.*?)/(?P<name>.*?)(\?.*)?'
+
+    @classmethod
+    def _extract_url_info(cls, url):
+        mobj = re.match(cls._VALID_URL, url)
+        lang = mobj.group('lang')
+        # This is not a real id, it can be for example AJT for the news
+        # http://www.arte.tv/guide/fr/emissions/AJT/arte-journal
+        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):
+        json_url = self._html_search_regex(r'arte_vp_url="(.*?)"', webpage, 'json url')
+
+        json_info = self._download_webpage(json_url, video_id, 'Downloading info json')
+        self.report_extraction(video_id)
+        info = json.loads(json_info)
+        player_info = info['videoJsonPlayer']
+
+        info_dict = {
+            'id': player_info['VID'],
+            'title': player_info['VTI'],
+            'description': player_info.get('VDE'),
+            'upload_date': unified_strdate(player_info.get('VDA', '').split(' ')[0]),
+            'thumbnail': player_info.get('programImage') or player_info.get('VTU', {}).get('IUR'),
+        }
+
+        formats = player_info['VSR'].values()
+        def _match_lang(f):
+            if f.get('versionCode') is None:
+                return True
+            # Return true if that format is in the language of the url
+            if lang == 'fr':
+                l = 'F'
+            elif lang == 'de':
+                l = 'A'
+            regexes = [r'VO?%s' % l, r'VO?.-ST%s' % l]
+            return any(re.match(r, f['versionCode']) for r in regexes)
+        # Some formats may not be in the same language as the url
+        formats = filter(_match_lang, formats)
+        # Some formats use the m3u8 protocol
+        formats = filter(lambda f: f.get('videoFormat') != 'M3U8', formats)
+        # We order the formats by quality
+        formats = sorted(formats, key=lambda f: int(f.get('height',-1)))
+        # Prefer videos without subtitles in the same language
+        formats = sorted(formats, key=lambda f: re.match(r'VO(F|A)-STM\1', f.get('versionCode', '')) is None)
+        # Pick the best quality
+        def _format(format_info):
+            info = {
+                'width': format_info.get('width'),
+                'height': format_info.get('height'),
+            }
+            if format_info['mediaType'] == u'rtmp':
+                info['url'] = format_info['streamer']
+                info['play_path'] = 'mp4:' + format_info['url']
+                info['ext'] = 'flv'
+            else:
+                info['url'] = format_info['url']
+                info['ext'] = determine_ext(info['url'])
+            return info
+        info_dict['formats'] = [_format(f) for f in formats]
+        # TODO: Remove when #980 has been merged 
+        info_dict.update(info_dict['formats'][-1])
+
+        return info_dict
+
+
+# It also uses the arte_vp_url url from the webpage to extract the information
+class ArteTVCreativeIE(ArteTVPlus7IE):
+    IE_NAME = u'arte.tv:creative'
+    _VALID_URL = r'https?://creative\.arte\.tv/(?P<lang>fr|de)/magazine?/(?P<id>.+)'
+
+    _TEST = {
+        u'url': u'http://creative.arte.tv/de/magazin/agentur-amateur-corporate-design',
+        u'file': u'050489-002.mp4',
+        u'info_dict': {
+            u'title': u'Agentur Amateur #2 - Corporate Design',
+        },
+    }
+
+
+class ArteTVFutureIE(ArteTVPlus7IE):
+    IE_NAME = u'arte.tv:future'
+    _VALID_URL = r'https?://future\.arte\.tv/(?P<lang>fr|de)/(thema|sujet)/.*?#article-anchor-(?P<id>\d+)'
+
+    _TEST = {
+        u'url': u'http://future.arte.tv/fr/sujet/info-sciences#article-anchor-7081',
+        u'file': u'050940-003.mp4',
+        u'info_dict': {
+            u'title': u'Les champignons au secours de la planète',
+        },
+    }
+
+    def _real_extract(self, url):
+        anchor_id, lang = self._extract_url_info(url)
+        webpage = self._download_webpage(url, anchor_id)
+        row = get_element_by_id(anchor_id, webpage)
+        return self._extract_from_webpage(row, anchor_id, lang)

youtube_dl/extractor/auengine.py

@@ -8,6 +8,14 @@ from ..utils import (
 )
 
 class AUEngineIE(InfoExtractor):
+    _TEST = {
+        u'url': u'http://auengine.com/embed.php?file=lfvlytY6&w=650&h=370',
+        u'file': u'lfvlytY6.mp4',
+        u'md5': u'48972bdbcf1a3a2f5533e62425b41d4f',
+        u'info_dict': {
+            u"title": u"[Commie]The Legend of the Legendary Heroes - 03 - Replication Eye (Alpha Stigma)[F9410F5A]"
+        }
+    }
     _VALID_URL = r'(?:http://)?(?:www\.)?auengine\.com/embed.php\?.*?file=([^&]+).*?'
 
     def _real_extract(self, url):

youtube_dl/extractor/bliptv.py

@@ -27,7 +27,7 @@ class BlipTVIE(InfoExtractor):
     _TEST = {
         u'url': u'http://blip.tv/cbr/cbr-exclusive-gotham-city-imposters-bats-vs-jokerz-short-3-5796352',
         u'file': u'5779306.m4v',
-        u'md5': u'b2d849efcf7ee18917e4b4d9ff37cafe',
+        u'md5': u'80baf1ec5c3d2019037c1c707d676b9f',
         u'info_dict': {
             u"upload_date": u"20111205", 
             u"description": u"md5:9bc31f227219cde65e47eeec8d2dc596", 
@@ -103,14 +103,19 @@ class BlipTVIE(InfoExtractor):
                     data = json_data
 
                 upload_date = datetime.datetime.strptime(data['datestamp'], '%m-%d-%y %H:%M%p').strftime('%Y%m%d')
-                video_url = data['media']['url']
+                if 'additionalMedia' in data:
+                    formats = sorted(data['additionalMedia'], key=lambda f: int(f['media_height']))
+                    best_format = formats[-1]
+                    video_url = best_format['url']
+                else:
+                    video_url = data['media']['url']
                 umobj = re.match(self._URL_EXT, video_url)
                 if umobj is None:
                     raise ValueError('Can not determine filename extension')
                 ext = umobj.group(1)
 
                 info = {
-                    'id': data['item_id'],
+                    'id': compat_str(data['item_id']),
                     'url': video_url,
                     'uploader': data['display_name'],
                     'upload_date': upload_date,
@@ -184,5 +189,5 @@ class BlipTVUserIE(InfoExtractor):
             pagenum += 1
 
         urls = [u'http://blip.tv/%s' % video_id for video_id in video_ids]
-        url_entries = [self.url_result(url, 'BlipTV') for url in urls]
+        url_entries = [self.url_result(vurl, 'BlipTV') for vurl in urls]
         return [self.playlist_result(url_entries, playlist_title = username)]

youtube_dl/extractor/bloomberg.py

@@ -0,0 +1,27 @@
+import re
+
+from .common import InfoExtractor
+
+
+class BloombergIE(InfoExtractor):
+    _VALID_URL = r'https?://www\.bloomberg\.com/video/(?P<name>.+?).html'
+
+    _TEST = {
+        u'url': u'http://www.bloomberg.com/video/shah-s-presentation-on-foreign-exchange-strategies-qurhIVlJSB6hzkVi229d8g.html',
+        u'file': u'12bzhqZTqQHmmlA8I-i0NpzJgcG5NNYX.mp4',
+        u'info_dict': {
+            u'title': u'Shah\'s Presentation on Foreign-Exchange Strategies',
+            u'description': u'md5:abc86e5236f9f0e4866c59ad36736686',
+        },
+        u'params': {
+            # Requires ffmpeg (m3u8 manifest)
+            u'skip_download': True,
+        },
+    }
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        name = mobj.group('name')
+        webpage = self._download_webpage(url, name)
+        ooyala_url = self._og_search_video_url(webpage)
+        return self.url_result(ooyala_url, ie='Ooyala')

youtube_dl/extractor/breakcom.py

@@ -1,6 +1,8 @@
 import re
+import json
 
 from .common import InfoExtractor
+from ..utils import determine_ext
 
 
 class BreakIE(InfoExtractor):
@@ -17,17 +19,20 @@ class BreakIE(InfoExtractor):
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
         video_id = mobj.group(1).split("-")[-1]
-        webpage = self._download_webpage(url, video_id)
-        video_url = re.search(r"videoPath: '(.+?)',",webpage).group(1)
-        key = re.search(r"icon: '(.+?)',",webpage).group(1)
-        final_url = str(video_url)+"?"+str(key)
-        thumbnail_url = re.search(r"thumbnailURL: '(.+?)'",webpage).group(1)
-        title = re.search(r"sVidTitle: '(.+)',",webpage).group(1)
-        ext = video_url.split('.')[-1]
+        embed_url = 'http://www.break.com/embed/%s' % video_id
+        webpage = self._download_webpage(embed_url, video_id)
+        info_json = self._search_regex(r'var embedVars = ({.*?});', webpage,
+                                       u'info json', flags=re.DOTALL)
+        info = json.loads(info_json)
+        video_url = info['videoUri']
+        m_youtube = re.search(r'(https?://www\.youtube\.com/watch\?v=.*)', video_url)
+        if m_youtube is not None:
+            return self.url_result(m_youtube.group(1), 'Youtube')
+        final_url = video_url + '?' + info['AuthToken']
         return [{
             'id':        video_id,
             'url':       final_url,
-            'ext':       ext,
-            'title':     title,
-            'thumbnail': thumbnail_url,
+            'ext':       determine_ext(final_url),
+            'title':     info['contentName'],
+            'thumbnail': info['thumbUri'],
         }]

youtube_dl/extractor/brightcove.py

@@ -0,0 +1,134 @@
+# encoding: utf-8
+
+import re
+import json
+import xml.etree.ElementTree
+
+from .common import InfoExtractor
+from ..utils import (
+    compat_urllib_parse,
+    find_xpath_attr,
+    compat_urlparse,
+
+    ExtractorError,
+)
+
+class BrightcoveIE(InfoExtractor):
+    _VALID_URL = r'https?://.*brightcove\.com/(services|viewer).*\?(?P<query>.*)'
+    _FEDERATED_URL_TEMPLATE = 'http://c.brightcove.com/services/viewer/htmlFederated?%s'
+    _PLAYLIST_URL_TEMPLATE = 'http://c.brightcove.com/services/json/experience/runtime/?command=get_programming_for_experience&playerKey=%s'
+
+    _TESTS = [
+        {
+            # From http://www.8tv.cat/8aldia/videos/xavier-sala-i-martin-aquesta-tarda-a-8-al-dia/
+            u'url': u'http://c.brightcove.com/services/viewer/htmlFederated?playerID=1654948606001&flashID=myExperience&%40videoPlayer=2371591881001',
+            u'file': u'2371591881001.mp4',
+            u'md5': u'9e80619e0a94663f0bdc849b4566af19',
+            u'note': u'Test Brightcove downloads and detection in GenericIE',
+            u'info_dict': {
+                u'title': u'Xavier Sala i Martín: “Un banc que no presta és un banc zombi que no serveix per a res”',
+                u'uploader': u'8TV',
+                u'description': u'md5:a950cc4285c43e44d763d036710cd9cd',
+            }
+        },
+        {
+            # From http://medianetwork.oracle.com/video/player/1785452137001
+            u'url': u'http://c.brightcove.com/services/viewer/htmlFederated?playerID=1217746023001&flashID=myPlayer&%40videoPlayer=1785452137001',
+            u'file': u'1785452137001.flv',
+            u'info_dict': {
+                u'title': u'JVMLS 2012: Arrays 2.0 - Opportunities and Challenges',
+                u'description': u'John Rose speaks at the JVM Language Summit, August 1, 2012.',
+                u'uploader': u'Oracle',
+            },
+        },
+    ]
+
+    @classmethod
+    def _build_brighcove_url(cls, object_str):
+        """
+        Build a Brightcove url from a xml string containing
+        <object class="BrightcoveExperience">{params}</object>
+        """
+
+        # Fix up some stupid HTML, see https://github.com/rg3/youtube-dl/issues/1553
+        object_str = re.sub(r'(<param name="[^"]+" value="[^"]+")>',
+                            lambda m: m.group(1) + '/>', object_str)
+        # Fix up some stupid XML, see https://github.com/rg3/youtube-dl/issues/1608
+        object_str = object_str.replace(u'<--', u'<!--')
+
+        object_doc = xml.etree.ElementTree.fromstring(object_str)
+        assert u'BrightcoveExperience' in object_doc.attrib['class']
+        params = {'flashID': object_doc.attrib['id'],
+                  'playerID': find_xpath_attr(object_doc, './param', 'name', 'playerID').attrib['value'],
+                  }
+        playerKey = find_xpath_attr(object_doc, './param', 'name', 'playerKey')
+        # Not all pages define this value
+        if playerKey is not None:
+            params['playerKey'] = playerKey.attrib['value']
+        videoPlayer = find_xpath_attr(object_doc, './param', 'name', '@videoPlayer')
+        if videoPlayer is not None:
+            params['@videoPlayer'] = videoPlayer.attrib['value']
+        data = compat_urllib_parse.urlencode(params)
+        return cls._FEDERATED_URL_TEMPLATE % data
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        query_str = mobj.group('query')
+        query = compat_urlparse.parse_qs(query_str)
+
+        videoPlayer = query.get('@videoPlayer')
+        if videoPlayer:
+            return self._get_video_info(videoPlayer[0], query_str)
+        else:
+            player_key = query['playerKey']
+            return self._get_playlist_info(player_key[0])
+
+    def _get_video_info(self, video_id, query):
+        request_url = self._FEDERATED_URL_TEMPLATE % query
+        webpage = self._download_webpage(request_url, video_id)
+
+        self.report_extraction(video_id)
+        info = self._search_regex(r'var experienceJSON = ({.*?});', webpage, 'json')
+        info = json.loads(info)['data']
+        video_info = info['programmedContent']['videoPlayer']['mediaDTO']
+
+        return self._extract_video_info(video_info)
+
+    def _get_playlist_info(self, player_key):
+        playlist_info = self._download_webpage(self._PLAYLIST_URL_TEMPLATE % player_key,
+                                               player_key, u'Downloading playlist information')
+
+        json_data = json.loads(playlist_info)
+        if 'videoList' not in json_data:
+            raise ExtractorError(u'Empty playlist')
+        playlist_info = json_data['videoList']
+        videos = [self._extract_video_info(video_info) for video_info in playlist_info['mediaCollectionDTO']['videoDTOs']]
+
+        return self.playlist_result(videos, playlist_id=playlist_info['id'],
+                                    playlist_title=playlist_info['mediaCollectionDTO']['displayName'])
+
+    def _extract_video_info(self, video_info):
+        info = {
+            'id': video_info['id'],
+            'title': video_info['displayName'],
+            'description': video_info.get('shortDescription'),
+            'thumbnail': video_info.get('videoStillURL') or video_info.get('thumbnailURL'),
+            'uploader': video_info.get('publisherName'),
+        }
+
+        renditions = video_info.get('renditions')
+        if renditions:
+            renditions = sorted(renditions, key=lambda r: r['size'])
+            best_format = renditions[-1]
+            info.update({
+                'url': best_format['defaultURL'],
+                'ext': 'mp4',
+            })
+        elif video_info.get('FLVFullLengthURL') is not None:
+            info.update({
+                'url': video_info['FLVFullLengthURL'],
+                'ext': 'flv',
+            })
+        else:
+            raise ExtractorError(u'Unable to extract video url for %s' % info['id'])
+        return info

youtube_dl/extractor/c56.py

@@ -0,0 +1,36 @@
+# coding: utf-8
+
+import re
+import json
+
+from .common import InfoExtractor
+from ..utils import determine_ext
+
+class C56IE(InfoExtractor):
+    _VALID_URL = r'https?://((www|player)\.)?56\.com/(.+?/)?(v_|(play_album.+-))(?P<textid>.+?)\.(html|swf)'
+    IE_NAME = u'56.com'
+
+    _TEST ={
+        u'url': u'http://www.56.com/u39/v_OTM0NDA3MTY.html',
+        u'file': u'93440716.flv',
+        u'md5': u'e59995ac63d0457783ea05f93f12a866',
+        u'info_dict': {
+            u'title': u'网事知多少 第32期：车怒',
+        },
+    }
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url, flags=re.VERBOSE)
+        text_id = mobj.group('textid')
+        info_page = self._download_webpage('http://vxml.56.com/json/%s/' % text_id,
+                                           text_id, u'Downloading video info')
+        info = json.loads(info_page)['info']
+        best_format = sorted(info['rfiles'], key=lambda f: int(f['filesize']))[-1]
+        video_url = best_format['url']
+
+        return {'id': info['vid'],
+                'title': info['Subject'],
+                'url': video_url,
+                'ext': determine_ext(video_url),
+                'thumbnail': info.get('bimg') or info.get('img'),
+                }

youtube_dl/extractor/canalc2.py

@@ -0,0 +1,35 @@
+# coding: utf-8
+import re
+
+from .common import InfoExtractor
+
+
+class Canalc2IE(InfoExtractor):
+    IE_NAME = 'canalc2.tv'
+    _VALID_URL = r'http://.*?\.canalc2\.tv/video\.asp\?idVideo=(\d+)&voir=oui'
+
+    _TEST = {
+        u'url': u'http://www.canalc2.tv/video.asp?idVideo=12163&voir=oui',
+        u'file': u'12163.mp4',
+        u'md5': u'060158428b650f896c542dfbb3d6487f',
+        u'info_dict': {
+            u'title': u'Terrasses du Numérique'
+        }
+    }
+
+    def _real_extract(self, url):
+        video_id = re.match(self._VALID_URL, url).group(1)
+        webpage = self._download_webpage(url, video_id)
+        file_name = self._search_regex(
+            r"so\.addVariable\('file','(.*?)'\);",
+            webpage, 'file name')
+        video_url = 'http://vod-flash.u-strasbg.fr:8080/' + file_name
+
+        title = self._html_search_regex(
+            r'class="evenement8">(.*?)</a>', webpage, u'title')
+        
+        return {'id': video_id,
+                'ext': 'mp4',
+                'url': video_url,
+                'title': title,
+                }

youtube_dl/extractor/canalplus.py

@@ -0,0 +1,54 @@
+# encoding: utf-8
+import re
+import xml.etree.ElementTree
+
+from .common import InfoExtractor
+from ..utils import unified_strdate
+
+class CanalplusIE(InfoExtractor):
+    _VALID_URL = r'https?://(www\.canalplus\.fr/.*?/(?P<path>.*)|player\.canalplus\.fr/#/(?P<id>\d+))'
+    _VIDEO_INFO_TEMPLATE = 'http://service.canal-plus.com/video/rest/getVideosLiees/cplus/%s'
+    IE_NAME = u'canalplus.fr'
+
+    _TEST = {
+        u'url': u'http://www.canalplus.fr/c-infos-documentaires/pid1830-c-zapping.html?vid=922470',
+        u'file': u'922470.flv',
+        u'info_dict': {
+            u'title': u'Zapping - 26/08/13',
+            u'description': u'Le meilleur de toutes les chaînes, tous les jours.\nEmission du 26 août 2013',
+            u'upload_date': u'20130826',
+        },
+        u'params': {
+            u'skip_download': True,
+        },
+    }
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        video_id = mobj.group('id')
+        if video_id is None:
+            webpage = self._download_webpage(url, mobj.group('path'))
+            video_id = self._search_regex(r'videoId = "(\d+)";', webpage, u'video id')
+        info_url = self._VIDEO_INFO_TEMPLATE % video_id
+        info_page = self._download_webpage(info_url,video_id, 
+                                           u'Downloading video info')
+
+        self.report_extraction(video_id)
+        doc = xml.etree.ElementTree.fromstring(info_page.encode('utf-8'))
+        video_info = [video for video in doc if video.find('ID').text == video_id][0]
+        infos = video_info.find('INFOS')
+        media = video_info.find('MEDIA')
+        formats = [media.find('VIDEOS/%s' % format)
+            for format in ['BAS_DEBIT', 'HAUT_DEBIT', 'HD']]
+        video_url = [format.text for format in formats if format is not None][-1]
+
+        return {'id': video_id,
+                'title': u'%s - %s' % (infos.find('TITRAGE/TITRE').text,
+                                       infos.find('TITRAGE/SOUS_TITRE').text),
+                'url': video_url,
+                'ext': 'flv',
+                'upload_date': unified_strdate(infos.find('PUBLICATION/DATE').text),
+                'thumbnail': media.find('IMAGES/GRAND').text,
+                'description': infos.find('DESCRIPTION').text,
+                'view_count': int(infos.find('NB_VUES').text),
+                }

youtube_dl/extractor/cinemassacre.py

@@ -0,0 +1,91 @@
+# encoding: utf-8
+import re
+
+from .common import InfoExtractor
+from ..utils import (
+    ExtractorError,
+)
+
+
+class CinemassacreIE(InfoExtractor):
+    _VALID_URL = r'(?:http://)?(?:www\.)?(?P<url>cinemassacre\.com/(?P<date_Y>[0-9]{4})/(?P<date_m>[0-9]{2})/(?P<date_d>[0-9]{2})/.+?)(?:[/?].*)?'
+    _TESTS = [{
+        u'url': u'http://cinemassacre.com/2012/11/10/avgn-the-movie-trailer/',
+        u'file': u'19911.flv',
+        u'info_dict': {
+            u'upload_date': u'20121110',
+            u'title': u'“Angry Video Game Nerd: The Movie” – Trailer',
+            u'description': u'md5:fb87405fcb42a331742a0dce2708560b',
+        },
+        u'params': {
+            # rtmp download
+            u'skip_download': True,
+        },
+    },
+    {
+        u'url': u'http://cinemassacre.com/2013/10/02/the-mummys-hand-1940',
+        u'file': u'521be8ef82b16.flv',
+        u'info_dict': {
+            u'upload_date': u'20131002',
+            u'title': u'The Mummy’s Hand (1940)',
+        },
+        u'params': {
+            # rtmp download
+            u'skip_download': True,
+        },
+    }]
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+
+        webpage_url = u'http://' + mobj.group('url')
+        webpage = self._download_webpage(webpage_url, None) # Don't know video id yet
+        video_date = mobj.group('date_Y') + mobj.group('date_m') + mobj.group('date_d')
+        mobj = re.search(r'src="(?P<embed_url>http://player\.screenwavemedia\.com/play/(?:embed|player)\.php\?id=(?:Cinemassacre-)?(?P<video_id>.+?))"', webpage)
+        if not mobj:
+            raise ExtractorError(u'Can\'t extract embed url and video id')
+        playerdata_url = mobj.group(u'embed_url')
+        video_id = mobj.group(u'video_id')
+
+        video_title = self._html_search_regex(r'<title>(?P<title>.+?)\|',
+            webpage, u'title')
+        video_description = self._html_search_regex(r'<div class="entry-content">(?P<description>.+?)</div>',
+            webpage, u'description', flags=re.DOTALL, fatal=False)
+        if len(video_description) == 0:
+            video_description = None
+
+        playerdata = self._download_webpage(playerdata_url, video_id)
+        base_url = self._html_search_regex(r'\'streamer\': \'(?P<base_url>rtmp://.*?)/(?:vod|Cinemassacre)\'',
+            playerdata, u'base_url')
+        base_url += '/Cinemassacre/'
+        # Important: The file names in playerdata are not used by the player and even wrong for some videos
+        sd_file = 'Cinemassacre-%s_high.mp4' % video_id
+        hd_file = 'Cinemassacre-%s.mp4' % video_id
+        video_thumbnail = 'http://image.screenwavemedia.com/Cinemassacre/Cinemassacre-%s_thumb_640x360.jpg' % video_id
+
+        formats = [
+            {
+                'url': base_url + sd_file,
+                'ext': 'flv',
+                'format': 'sd',
+                'format_id': 'sd',
+            },
+            {
+                'url': base_url + hd_file,
+                'ext': 'flv',
+                'format': 'hd',
+                'format_id': 'hd',
+            },
+        ]
+
+        info = {
+            'id': video_id,
+            'title': video_title,
+            'formats': formats,
+            'description': video_description,
+            'upload_date': video_date,
+            'thumbnail': video_thumbnail,
+        }
+        # TODO: Remove when #980 has been merged
+        info.update(formats[-1])
+        return info

youtube_dl/extractor/cnn.py

@@ -0,0 +1,58 @@
+import re
+import xml.etree.ElementTree
+
+from .common import InfoExtractor
+from ..utils import determine_ext
+
+
+class CNNIE(InfoExtractor):
+    _VALID_URL = r'''(?x)https?://(edition\.)?cnn\.com/video/(data/.+?|\?)/
+        (?P<path>.+?/(?P<title>[^/]+?)(?:\.cnn|(?=&)))'''
+
+    _TESTS = [{
+        u'url': u'http://edition.cnn.com/video/?/video/sports/2013/06/09/nadal-1-on-1.cnn',
+        u'file': u'sports_2013_06_09_nadal-1-on-1.cnn.mp4',
+        u'md5': u'3e6121ea48df7e2259fe73a0628605c4',
+        u'info_dict': {
+            u'title': u'Nadal wins 8th French Open title',
+            u'description': u'World Sport\'s Amanda Davies chats with 2013 French Open champion Rafael Nadal.',
+        },
+    },
+    {
+        u"url": u"http://edition.cnn.com/video/?/video/us/2013/08/21/sot-student-gives-epic-speech.georgia-institute-of-technology&utm_source=feedburner&utm_medium=feed&utm_campaign=Feed%3A+rss%2Fcnn_topstories+%28RSS%3A+Top+Stories%29",
+        u"file": u"us_2013_08_21_sot-student-gives-epic-speech.georgia-institute-of-technology.mp4",
+        u"md5": u"b5cc60c60a3477d185af8f19a2a26f4e",
+        u"info_dict": {
+            u"title": "Student's epic speech stuns new freshmen",
+            u"description": "A Georgia Tech student welcomes the incoming freshmen with an epic speech backed by music from \"2001: A Space Odyssey.\""
+        }
+    }]
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        path = mobj.group('path')
+        page_title = mobj.group('title')
+        info_url = u'http://cnn.com/video/data/3.0/%s/index.xml' % path
+        info_xml = self._download_webpage(info_url, page_title)
+        info = xml.etree.ElementTree.fromstring(info_xml.encode('utf-8'))
+
+        formats = []
+        for f in info.findall('files/file'):
+            mf = re.match(r'(\d+)x(\d+)(?:_(.*)k)?',f.attrib['bitrate'])
+            if mf is not None:
+                formats.append((int(mf.group(1)), int(mf.group(2)), int(mf.group(3) or 0), f.text))
+        formats = sorted(formats)
+        (_,_,_, video_path) = formats[-1]
+        video_url = 'http://ht.cdn.turner.com/cnn/big%s' % video_path
+
+        thumbnails = sorted([((int(t.attrib['height']),int(t.attrib['width'])), t.text) for t in info.findall('images/image')])
+        thumbs_dict = [{'resolution': res, 'url': t_url} for (res, t_url) in thumbnails]
+
+        return {'id': info.attrib['id'],
+                'title': info.find('headline').text,
+                'url': video_url,
+                'ext': determine_ext(video_url),
+                'thumbnail': thumbnails[-1][1],
+                'thumbnails': thumbs_dict,
+                'description': info.find('description').text,
+                }

youtube_dl/extractor/collegehumor.py

@@ -1,26 +1,36 @@
 import re
-import socket
 import xml.etree.ElementTree
 
 from .common import InfoExtractor
 from ..utils import (
-    compat_http_client,
-    compat_str,
-    compat_urllib_error,
     compat_urllib_parse_urlparse,
-    compat_urllib_request,
+    determine_ext,
 
     ExtractorError,
 )
 
 
 class CollegeHumorIE(InfoExtractor):
-    _WORKING = False
-    _VALID_URL = r'^(?:https?://)?(?:www\.)?collegehumor\.com/video/(?P<videoid>[0-9]+)/(?P<shorttitle>.*)$'
+    _VALID_URL = r'^(?:https?://)?(?:www\.)?collegehumor\.com/(video|embed|e)/(?P<videoid>[0-9]+)/?(?P<shorttitle>.*)$'
 
-    def report_manifest(self, video_id):
-        """Report information extraction."""
-        self.to_screen(u'%s: Downloading XML manifest' % video_id)
+    _TESTS = [{
+        u'url': u'http://www.collegehumor.com/video/6902724/comic-con-cosplay-catastrophe',
+        u'file': u'6902724.mp4',
+        u'md5': u'1264c12ad95dca142a9f0bf7968105a0',
+        u'info_dict': {
+            u'title': u'Comic-Con Cosplay Catastrophe',
+            u'description': u'Fans get creative this year at San Diego.  Too creative.  And yes, that\'s really Joss Whedon.',
+        },
+    },
+    {
+        u'url': u'http://www.collegehumor.com/video/3505939/font-conference',
+        u'file': u'3505939.mp4',
+        u'md5': u'c51ca16b82bb456a4397987791a835f5',
+        u'info_dict': {
+            u'title': u'Font Conference',
+            u'description': u'This video wasn\'t long enough, so we made it double-spaced.',
+        },
+    }]
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
@@ -36,39 +46,42 @@ class CollegeHumorIE(InfoExtractor):
 
         self.report_extraction(video_id)
         xmlUrl = 'http://www.collegehumor.com/moogaloop/video/' + video_id
-        try:
-            metaXml = compat_urllib_request.urlopen(xmlUrl).read()
-        except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
-            raise ExtractorError(u'Unable to download video info XML: %s' % compat_str(err))
+        metaXml = self._download_webpage(xmlUrl, video_id,
+                                         u'Downloading info XML',
+                                         u'Unable to download video info XML')
 
         mdoc = xml.etree.ElementTree.fromstring(metaXml)
         try:
             videoNode = mdoc.findall('./video')[0]
+            youtubeIdNode = videoNode.find('./youtubeID')
+            if youtubeIdNode is not None:
+                return self.url_result(youtubeIdNode.text, 'Youtube')
             info['description'] = videoNode.findall('./description')[0].text
             info['title'] = videoNode.findall('./caption')[0].text
             info['thumbnail'] = videoNode.findall('./thumbnail')[0].text
-            manifest_url = videoNode.findall('./file')[0].text
+            next_url = videoNode.findall('./file')[0].text
         except IndexError:
             raise ExtractorError(u'Invalid metadata XML file')
 
-        manifest_url += '?hdcore=2.10.3'
-        self.report_manifest(video_id)
-        try:
-            manifestXml = compat_urllib_request.urlopen(manifest_url).read()
-        except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
-            raise ExtractorError(u'Unable to download video info XML: %s' % compat_str(err))
+        if next_url.endswith(u'manifest.f4m'):
+            manifest_url = next_url + '?hdcore=2.10.3'
+            manifestXml = self._download_webpage(manifest_url, video_id,
+                                         u'Downloading XML manifest',
+                                         u'Unable to download video info XML')
 
-        adoc = xml.etree.ElementTree.fromstring(manifestXml)
-        try:
-            media_node = adoc.findall('./{http://ns.adobe.com/f4m/1.0}media')[0]
-            node_id = media_node.attrib['url']
-            video_id = adoc.findall('./{http://ns.adobe.com/f4m/1.0}id')[0].text
-        except IndexError as err:
-            raise ExtractorError(u'Invalid manifest file')
+            adoc = xml.etree.ElementTree.fromstring(manifestXml)
+            try:
+                media_node = adoc.findall('./{http://ns.adobe.com/f4m/1.0}media')[0]
+                node_id = media_node.attrib['url']
+                video_id = adoc.findall('./{http://ns.adobe.com/f4m/1.0}id')[0].text
+            except IndexError as err:
+                raise ExtractorError(u'Invalid manifest file')
+            url_pr = compat_urllib_parse_urlparse(info['thumbnail'])
+            info['url'] = url_pr.scheme + '://' + url_pr.netloc + video_id[:-2].replace('.csmil','').replace(',','')
+            info['ext'] = 'mp4'
+        else:
+            # Old-style direct links
+            info['url'] = next_url
+            info['ext'] = determine_ext(info['url'])
 
-        url_pr = compat_urllib_parse_urlparse(manifest_url)
-        url = url_pr.scheme + '://' + url_pr.netloc + '/z' + video_id[:-2] + '/' + node_id + 'Seg1-Frag1'
-
-        info['url'] = url
-        info['ext'] = 'f4f'
-        return [info]
+        return info

youtube_dl/extractor/comedycentral.py

@@ -24,7 +24,9 @@ class ComedyCentralIE(InfoExtractor):
                          (full-episodes/(?P<episode>.*)|
                           (?P<clip>
                               (the-colbert-report-(videos|collections)/(?P<clipID>[0-9]+)/[^/]*/(?P<cntitle>.*?))
-                              |(watch/(?P<date>[^/]*)/(?P<tdstitle>.*)))))
+                              |(watch/(?P<date>[^/]*)/(?P<tdstitle>.*)))|
+                          (?P<interview>
+                              extended-interviews/(?P<interID>[0-9]+)/playlist_tds_extended_(?P<interview_title>.*?)/.*?)))
                      $"""
     _TEST = {
         u'url': u'http://www.thedailyshow.com/watch/thu-december-13-2012/kristen-stewart',
@@ -49,12 +51,12 @@ class ComedyCentralIE(InfoExtractor):
         '400': 'mp4',
     }
     _video_dimensions = {
-        '3500': '1280x720',
-        '2200': '960x540',
-        '1700': '768x432',
-        '1200': '640x360',
-        '750': '512x288',
-        '400': '384x216',
+        '3500': (1280, 720),
+        '2200': (960, 540),
+        '1700': (768, 432),
+        '1200': (640, 360),
+        '750': (512, 288),
+        '400': (384, 216),
     }
 
     @classmethod
@@ -62,11 +64,13 @@ class ComedyCentralIE(InfoExtractor):
         """Receives a URL and returns True if suitable for this IE."""
         return re.match(cls._VALID_URL, url, re.VERBOSE) is not None
 
-    def _print_formats(self, formats):
-        print('Available formats:')
-        for x in formats:
-            print('%s\t:\t%s\t[%s]' %(x, self._video_extensions.get(x, 'mp4'), self._video_dimensions.get(x, '???')))
-
+    @staticmethod
+    def _transform_rtmp_url(rtmp_video_url):
+        m = re.match(r'^rtmpe?://.*?/(?P<finalid>gsp.comedystor/.*)$', rtmp_video_url)
+        if not m:
+            raise ExtractorError(u'Cannot transform RTMP url')
+        base = 'http://mtvnmobile.vo.llnwd.net/kip0/_pxn=1+_pxI0=Ripod-h264+_pxL0=undefined+_pxM0=+_pxK=18639+_pxE=mp4/44620/mtvnorigin/'
+        return base + m.group('finalid')
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url, re.VERBOSE)
@@ -87,6 +91,9 @@ class ComedyCentralIE(InfoExtractor):
             else:
                 epTitle = mobj.group('cntitle')
             dlNewest = False
+        elif mobj.group('interview'):
+            epTitle = mobj.group('interview_title')
+            dlNewest = False
         else:
             dlNewest = not mobj.group('episode')
             if dlNewest:
@@ -150,40 +157,31 @@ class ComedyCentralIE(InfoExtractor):
                 self._downloader.report_error(u'unable to download ' + mediaId + ': No videos found')
                 continue
 
-            if self._downloader.params.get('listformats', None):
-                self._print_formats([i[0] for i in turls])
-                return
-
-            # For now, just pick the highest bitrate
-            format,rtmp_video_url = turls[-1]
-
-            # Get the format arg from the arg stream
-            req_format = self._downloader.params.get('format', None)
-
-            # Select format if we can find one
-            for f,v in turls:
-                if f == req_format:
-                    format, rtmp_video_url = f, v
-                    break
-
-            m = re.match(r'^rtmpe?://.*?/(?P<finalid>gsp.comedystor/.*)$', rtmp_video_url)
-            if not m:
-                raise ExtractorError(u'Cannot transform RTMP url')
-            base = 'http://mtvnmobile.vo.llnwd.net/kip0/_pxn=1+_pxI0=Ripod-h264+_pxL0=undefined+_pxM0=+_pxK=18639+_pxE=mp4/44620/mtvnorigin/'
-            video_url = base + m.group('finalid')
+            formats = []
+            for format, rtmp_video_url in turls:
+                w, h = self._video_dimensions.get(format, (None, None))
+                formats.append({
+                    'url': self._transform_rtmp_url(rtmp_video_url),
+                    'ext': self._video_extensions.get(format, 'mp4'),
+                    'format_id': format,
+                    'height': h,
+                    'width': w,
+                })
 
             effTitle = showId + u'-' + epTitle + u' part ' + compat_str(partNum+1)
             info = {
                 'id': shortMediaId,
-                'url': video_url,
+                'formats': formats,
                 'uploader': showId,
                 'upload_date': officialDate,
                 'title': effTitle,
-                'ext': 'mp4',
-                'format': format,
                 'thumbnail': None,
                 'description': compat_str(officialTitle),
             }
+
+            # TODO: Remove when #980 has been merged
+            info.update(info['formats'][-1])
+
             results.append(info)
 
         return results

youtube_dl/extractor/common.py

@@ -3,6 +3,7 @@ import os
 import re
 import socket
 import sys
+import netrc
 
 from ..utils import (
     compat_http_client,
@@ -13,6 +14,7 @@ from ..utils import (
     clean_html,
     compiled_regex_type,
     ExtractorError,
+    unescapeHTML,
 )
 
 class InfoExtractor(object):
@@ -33,9 +35,13 @@ class InfoExtractor(object):
     title:          Video title, unescaped.
     ext:            Video filename extension.
 
+    Instead of url and ext, formats can also specified.
+
     The following fields are optional:
 
     format:         The video format, defaults to ext (used for --get-format)
+    thumbnails:     A list of dictionaries (with the entries "resolution" and
+                    "url") for the varying thumbnails
     thumbnail:      Full URL to a video thumbnail image.
     description:    One-line video description.
     uploader:       Full name of the video uploader.
@@ -43,12 +49,25 @@ class InfoExtractor(object):
     uploader_id:    Nickname or id of the video uploader.
     location:       Physical location of the video.
     player_url:     SWF Player URL (used for rtmpdump).
-    subtitles:      The subtitle file contents.
+    subtitles:      The subtitle file contents as a dictionary in the format
+                    {language: subtitles}.
     view_count:     How many users have watched the video on the platform.
     urlhandle:      [internal] The urlHandle to be used to download the file,
                     like returned by urllib.request.urlopen
+    age_limit:      Age restriction for the video, as an integer (years)
+    formats:        A list of dictionaries for each format available, it must
+                    be ordered from worst to best quality. Potential fields:
+                    * url       Mandatory. The URL of the video file
+                    * ext       Will be calculated from url if missing
+                    * format    A human-readable description of the format
+                                ("mp4 container with h264/opus").
+                                Calculated from width and height if missing.
+                    * format_id A short description of the format
+                                ("mp4_h264_opus" or "19")
+                    * width     Width of the video, if known
+                    * height    Height of the video, if known
 
-    The fields should all be Unicode strings.
+    Unless mentioned otherwise, the fields should be Unicode strings.
 
     Subclasses of this one should re-define the _real_initialize() and
     _real_extract() methods and define a _VALID_URL regexp.
@@ -73,7 +92,13 @@ class InfoExtractor(object):
     @classmethod
     def suitable(cls, url):
         """Receives a URL and returns True if suitable for this IE."""
-        return re.match(cls._VALID_URL, url) is not None
+
+        # This does not use has/getattr intentionally - we want to know whether
+        # we have cached the regexp for *this* class, whereas getattr would also
+        # match the superclass
+        if '_VALID_URL_RE' not in cls.__dict__:
+            cls._VALID_URL_RE = re.compile(cls._VALID_URL)
+        return cls._VALID_URL_RE.match(url) is not None
 
     @classmethod
     def working(cls):
@@ -103,6 +128,11 @@ class InfoExtractor(object):
         """Real extraction process. Redefine in subclasses."""
         pass
 
+    @classmethod
+    def ie_key(cls):
+        """A string for getting the InfoExtractor with get_info_extractor"""
+        return cls.__name__[:-2]
+
     @property
     def IE_NAME(self):
         return type(self).__name__[:-2]
@@ -118,18 +148,28 @@ class InfoExtractor(object):
         except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
             if errnote is None:
                 errnote = u'Unable to download webpage'
-            raise ExtractorError(u'%s: %s' % (errnote, compat_str(err)), sys.exc_info()[2])
+            raise ExtractorError(u'%s: %s' % (errnote, compat_str(err)), sys.exc_info()[2], cause=err)
 
     def _download_webpage_handle(self, url_or_request, video_id, note=None, errnote=None):
         """ Returns a tuple (page content as string, URL handle) """
+
+        # Strip hashes from the URL (#1038)
+        if isinstance(url_or_request, (compat_str, str)):
+            url_or_request = url_or_request.partition('#')[0]
+
         urlh = self._request_webpage(url_or_request, video_id, note, errnote)
         content_type = urlh.headers.get('Content-Type', '')
+        webpage_bytes = urlh.read()
         m = re.match(r'[a-zA-Z0-9_.-]+/[a-zA-Z0-9_.-]+\s*;\s*charset=(.+)', content_type)
         if m:
             encoding = m.group(1)
         else:
-            encoding = 'utf-8'
-        webpage_bytes = urlh.read()
+            m = re.search(br'<meta[^>]+charset=[\'"]?([^\'")]+)[ /\'">]',
+                          webpage_bytes[:1024])
+            if m:
+                encoding = m.group(1).decode('ascii')
+            else:
+                encoding = 'utf-8'
         if self._downloader.params.get('dump_intermediate_pages', False):
             try:
                 url = url_or_request.get_full_url()
@@ -161,12 +201,11 @@ class InfoExtractor(object):
         """Report attempt to confirm age."""
         self.to_screen(u'Confirming age')
 
+    def report_login(self):
+        """Report attempt to log in."""
+        self.to_screen(u'Logging in')
+
     #Methods for following #608
-    #They set the correct value of the '_type' key
-    def video_result(self, video_info):
-        """Returns a video"""
-        video_info['_type'] = 'video'
-        return video_info
     def url_result(self, url, ie=None):
         """Returns a url that points to a page that should be processed"""
         #TODO: ie should be the class used for getting the info
@@ -225,6 +264,70 @@ class InfoExtractor(object):
         else:
             return res
 
+    def _get_login_info(self):
+        """
+        Get the the login info as (username, password)
+        It will look in the netrc file using the _NETRC_MACHINE value
+        If there's no info available, return (None, None)
+        """
+        if self._downloader is None:
+            return (None, None)
+
+        username = None
+        password = None
+        downloader_params = self._downloader.params
+
+        # Attempt to use provided username and password or .netrc data
+        if downloader_params.get('username', None) is not None:
+            username = downloader_params['username']
+            password = downloader_params['password']
+        elif downloader_params.get('usenetrc', False):
+            try:
+                info = netrc.netrc().authenticators(self._NETRC_MACHINE)
+                if info is not None:
+                    username = info[0]
+                    password = info[2]
+                else:
+                    raise netrc.NetrcParseError('No authenticators for %s' % self._NETRC_MACHINE)
+            except (IOError, netrc.NetrcParseError) as err:
+                self._downloader.report_warning(u'parsing .netrc: %s' % compat_str(err))
+        
+        return (username, password)
+
+    # Helper functions for extracting OpenGraph info
+    @staticmethod
+    def _og_regex(prop):
+        return r'<meta.+?property=[\'"]og:%s[\'"].+?content=(?:"(.+?)"|\'(.+?)\')' % re.escape(prop)
+
+    def _og_search_property(self, prop, html, name=None, **kargs):
+        if name is None:
+            name = 'OpenGraph %s' % prop
+        escaped = self._search_regex(self._og_regex(prop), html, name, flags=re.DOTALL, **kargs)
+        return unescapeHTML(escaped)
+
+    def _og_search_thumbnail(self, html, **kargs):
+        return self._og_search_property('image', html, u'thumbnail url', fatal=False, **kargs)
+
+    def _og_search_description(self, html, **kargs):
+        return self._og_search_property('description', html, fatal=False, **kargs)
+
+    def _og_search_title(self, html, **kargs):
+        return self._og_search_property('title', html, **kargs)
+
+    def _og_search_video_url(self, html, name='video url', **kargs):
+        return self._html_search_regex([self._og_regex('video:secure_url'),
+                                        self._og_regex('video')],
+                                       html, name, **kargs)
+
+    def _rta_search(self, html):
+        # See http://www.rtalabel.org/index.php?content=howtofaq#single
+        if re.search(r'(?ix)<meta\s+name="rating"\s+'
+                     r'     content="RTA-5042-1996-1400-1577-RTA"',
+                     html):
+            return 18
+        return 0
+
+
 class SearchInfoExtractor(InfoExtractor):
     """
     Base class for paged search queries extractors.
@@ -262,7 +365,7 @@ class SearchInfoExtractor(InfoExtractor):
 
     def _get_n_results(self, query, n):
         """Get a specified number of results for a query"""
-        raise NotImplementedError("This method must be implemented by sublclasses")
+        raise NotImplementedError("This method must be implemented by subclasses")
 
     @property
     def SEARCH_KEY(self):

youtube_dl/extractor/condenast.py

@@ -0,0 +1,106 @@
+# coding: utf-8
+
+import re
+import json
+
+from .common import InfoExtractor
+from ..utils import (
+    compat_urllib_parse,
+    orderedSet,
+    compat_urllib_parse_urlparse,
+    compat_urlparse,
+)
+
+
+class CondeNastIE(InfoExtractor):
+    """
+    Condé Nast is a media group, some of its sites use a custom HTML5 player
+    that works the same in all of them.
+    """
+
+    # The keys are the supported sites and the values are the name to be shown
+    # to the user and in the extractor description.
+    _SITES = {'wired': u'WIRED',
+              'gq': u'GQ',
+              'vogue': u'Vogue',
+              'glamour': u'Glamour',
+              'wmagazine': u'W Magazine',
+              'vanityfair': u'Vanity Fair',
+              }
+
+    _VALID_URL = r'http://(video|www).(?P<site>%s).com/(?P<type>watch|series|video)/(?P<id>.+)' % '|'.join(_SITES.keys())
+    IE_DESC = u'Condé Nast media group: %s' % ', '.join(sorted(_SITES.values()))
+
+    _TEST = {
+        u'url': u'http://video.wired.com/watch/3d-printed-speakers-lit-with-led',
+        u'file': u'5171b343c2b4c00dd0c1ccb3.mp4',
+        u'md5': u'1921f713ed48aabd715691f774c451f7',
+        u'info_dict': {
+            u'title': u'3D Printed Speakers Lit With LED',
+            u'description': u'Check out these beautiful 3D printed LED speakers.  You can\'t actually buy them, but LumiGeek is working on a board that will let you make you\'re own.',
+        }
+    }
+
+    def _extract_series(self, url, webpage):
+        title = self._html_search_regex(r'<div class="cne-series-info">.*?<h1>(.+?)</h1>',
+                                        webpage, u'series title', flags=re.DOTALL)
+        url_object = compat_urllib_parse_urlparse(url)
+        base_url = '%s://%s' % (url_object.scheme, url_object.netloc)
+        m_paths = re.finditer(r'<p class="cne-thumb-title">.*?<a href="(/watch/.+?)["\?]',
+                              webpage, flags=re.DOTALL)
+        paths = orderedSet(m.group(1) for m in m_paths)
+        build_url = lambda path: compat_urlparse.urljoin(base_url, path)
+        entries = [self.url_result(build_url(path), 'CondeNast') for path in paths]
+        return self.playlist_result(entries, playlist_title=title)
+
+    def _extract_video(self, webpage):
+        description = self._html_search_regex([r'<div class="cne-video-description">(.+?)</div>',
+                                               r'<div class="video-post-content">(.+?)</div>',
+                                               ],
+                                              webpage, u'description',
+                                              fatal=False, flags=re.DOTALL)
+        params = self._search_regex(r'var params = {(.+?)}[;,]', webpage,
+                                    u'player params', flags=re.DOTALL)
+        video_id = self._search_regex(r'videoId: [\'"](.+?)[\'"]', params, u'video id')
+        player_id = self._search_regex(r'playerId: [\'"](.+?)[\'"]', params, u'player id')
+        target = self._search_regex(r'target: [\'"](.+?)[\'"]', params, u'target')
+        data = compat_urllib_parse.urlencode({'videoId': video_id,
+                                              'playerId': player_id,
+                                              'target': target,
+                                              })
+        base_info_url = self._search_regex(r'url = [\'"](.+?)[\'"][,;]',
+                                           webpage, u'base info url',
+                                           default='http://player.cnevids.com/player/loader.js?')
+        info_url = base_info_url + data
+        info_page = self._download_webpage(info_url, video_id,
+                                           u'Downloading video info')
+        video_info = self._search_regex(r'var video = ({.+?});', info_page, u'video info')
+        video_info = json.loads(video_info)
+
+        def _formats_sort_key(f):
+            type_ord = 1 if f['type'] == 'video/mp4' else 0
+            quality_ord = 1 if f['quality'] == 'high' else 0
+            return (quality_ord, type_ord)
+        best_format = sorted(video_info['sources'][0], key=_formats_sort_key)[-1]
+
+        return {'id': video_id,
+                'url': best_format['src'],
+                'ext': best_format['type'].split('/')[-1],
+                'title': video_info['title'],
+                'thumbnail': video_info['poster_frame'],
+                'description': description,
+                }
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        site = mobj.group('site')
+        url_type = mobj.group('type')
+        id = mobj.group('id')
+
+        self.to_screen(u'Extracting from %s with the Condé Nast extractor' % self._SITES[site])
+        webpage = self._download_webpage(url, id)
+
+        if url_type == 'series':
+            return self._extract_series(url, webpage)
+        else:
+            return self._extract_video(webpage)

youtube_dl/extractor/criterion.py

@@ -0,0 +1,40 @@
+# -*- coding: utf-8 -*-
+
+import re
+
+from .common import InfoExtractor
+from ..utils import determine_ext
+
+class CriterionIE(InfoExtractor):
+    _VALID_URL = r'https?://www\.criterion\.com/films/(\d*)-.+'
+    _TEST = {
+        u'url': u'http://www.criterion.com/films/184-le-samourai',
+        u'file': u'184.mp4',
+        u'md5': u'bc51beba55685509883a9a7830919ec3',
+        u'info_dict': {
+            u"title": u"Le Samouraï",
+            u"description" : u'md5:a2b4b116326558149bef81f76dcbb93f',
+        }
+    }
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        video_id = mobj.group(1)
+        webpage = self._download_webpage(url, video_id)
+
+        final_url = self._search_regex(r'so.addVariable\("videoURL", "(.+?)"\)\;',
+                                webpage, 'video url')
+        title = self._html_search_regex(r'<meta content="(.+?)" property="og:title" />',
+                                webpage, 'video title')
+        description = self._html_search_regex(r'<meta name="description" content="(.+?)" />',
+                                webpage, 'video description')
+        thumbnail = self._search_regex(r'so.addVariable\("thumbnailURL", "(.+?)"\)\;',
+                                webpage, 'thumbnail url')
+
+        return {'id': video_id,
+                'url' : final_url,
+                'title': title,
+                'ext': determine_ext(final_url),
+                'description': description,
+                'thumbnail': thumbnail,
+                }

youtube_dl/extractor/cspan.py

@@ -34,8 +34,6 @@ class CSpanIE(InfoExtractor):
         description = self._html_search_regex(r'<meta (?:property="og:|name=")description" content="(.*?)"',
                                               webpage, 'description',
                                               flags=re.MULTILINE|re.DOTALL)
-        thumbnail = self._html_search_regex(r'<meta property="og:image" content="(.*?)"',
-                                            webpage, 'thumbnail')
 
         url = self._search_regex(r'<string name="URL">(.*?)</string>',
                                  video_info, 'video url')
@@ -49,5 +47,5 @@ class CSpanIE(InfoExtractor):
                 'url': url,
                 'play_path': path,
                 'description': description,
-                'thumbnail': thumbnail,
+                'thumbnail': self._og_search_thumbnail(webpage),
                 }

youtube_dl/extractor/dailymotion.py

@@ -1,28 +1,58 @@
 import re
+import json
+import itertools
 
 from .common import InfoExtractor
+from .subtitles import SubtitlesInfoExtractor
+
 from ..utils import (
     compat_urllib_request,
-    compat_urllib_parse,
+    compat_str,
+    get_element_by_attribute,
+    get_element_by_id,
+    orderedSet,
 
     ExtractorError,
-    unescapeHTML,
 )
 
-class DailymotionIE(InfoExtractor):
+class DailymotionBaseInfoExtractor(InfoExtractor):
+    @staticmethod
+    def _build_request(url):
+        """Build a request with the family filter disabled"""
+        request = compat_urllib_request.Request(url)
+        request.add_header('Cookie', 'family_filter=off')
+        return request
+
+class DailymotionIE(DailymotionBaseInfoExtractor, SubtitlesInfoExtractor):
     """Information Extractor for Dailymotion"""
 
-    _VALID_URL = r'(?i)(?:https?://)?(?:www\.)?dailymotion\.[a-z]{2,3}/video/([^/]+)'
+    _VALID_URL = r'(?i)(?:https?://)?(?:www\.)?dailymotion\.[a-z]{2,3}/(?:embed/)?video/([^/]+)'
     IE_NAME = u'dailymotion'
-    _TEST = {
-        u'url': u'http://www.dailymotion.com/video/x33vw9_tutoriel-de-youtubeur-dl-des-video_tech',
-        u'file': u'x33vw9.mp4',
-        u'md5': u'392c4b85a60a90dc4792da41ce3144eb',
-        u'info_dict': {
-            u"uploader": u"Alex and Van .", 
-            u"title": u"Tutoriel de Youtubeur\"DL DES VIDEO DE YOUTUBE\""
-        }
-    }
+    _TESTS = [
+        {
+            u'url': u'http://www.dailymotion.com/video/x33vw9_tutoriel-de-youtubeur-dl-des-video_tech',
+            u'file': u'x33vw9.mp4',
+            u'md5': u'392c4b85a60a90dc4792da41ce3144eb',
+            u'info_dict': {
+                u"uploader": u"Amphora Alex and Van .", 
+                u"title": u"Tutoriel de Youtubeur\"DL DES VIDEO DE YOUTUBE\""
+            }
+        },
+        # Vevo video
+        {
+            u'url': u'http://www.dailymotion.com/video/x149uew_katy-perry-roar-official_musi',
+            u'file': u'USUV71301934.mp4',
+            u'info_dict': {
+                u'title': u'Roar (Official)',
+                u'uploader': u'Katy Perry',
+                u'upload_date': u'20130905',
+            },
+            u'params': {
+                u'skip_download': True,
+            },
+            u'skip': u'VEVO is only available in some countries',
+        },
+    ]
 
     def _real_extract(self, url):
         # Extract id and simplified title from URL
@@ -31,41 +61,24 @@ class DailymotionIE(InfoExtractor):
         video_id = mobj.group(1).split('_')[0].split('?')[0]
 
         video_extension = 'mp4'
+        url = 'http://www.dailymotion.com/video/%s' % video_id
 
         # Retrieve video webpage to extract further information
-        request = compat_urllib_request.Request(url)
-        request.add_header('Cookie', 'family_filter=off')
+        request = self._build_request(url)
         webpage = self._download_webpage(request, video_id)
 
         # Extract URL, uploader and title from webpage
         self.report_extraction(video_id)
-        mobj = re.search(r'\s*var flashvars = (.*)', webpage)
-        if mobj is None:
-            raise ExtractorError(u'Unable to extract media URL')
-        flashvars = compat_urllib_parse.unquote(mobj.group(1))
 
-        for key in ['hd1080URL', 'hd720URL', 'hqURL', 'sdURL', 'ldURL', 'video_url']:
-            if key in flashvars:
-                max_quality = key
-                self.to_screen(u'Using %s' % key)
-                break
-        else:
-            raise ExtractorError(u'Unable to extract video URL')
+        # It may just embed a vevo video:
+        m_vevo = re.search(
+            r'<link rel="video_src" href="[^"]*?vevo.com[^"]*?videoId=(?P<id>[\w]*)',
+            webpage)
+        if m_vevo is not None:
+            vevo_id = m_vevo.group('id')
+            self.to_screen(u'Vevo video detected: %s' % vevo_id)
+            return self.url_result(u'vevo:%s' % vevo_id, ie='Vevo')
 
-        mobj = re.search(r'"' + max_quality + r'":"(.+?)"', flashvars)
-        if mobj is None:
-            raise ExtractorError(u'Unable to extract video URL')
-
-        video_url = compat_urllib_parse.unquote(mobj.group(1)).replace('\\/', '/')
-
-        # TODO: support choosing qualities
-
-        mobj = re.search(r'<meta property="og:title" content="(?P<title>[^"]*)" />', webpage)
-        if mobj is None:
-            raise ExtractorError(u'Unable to extract title')
-        video_title = unescapeHTML(mobj.group('title'))
-
-        video_uploader = None
         video_uploader = self._search_regex([r'(?im)<span class="owner[^\"]+?">[^<]+?<a [^>]+?>([^<]+?)</a>',
                                              # Looking for official user
                                              r'<(?:span|a) .*?rel="author".*?>([^<]+?)</'],
@@ -76,11 +89,112 @@ class DailymotionIE(InfoExtractor):
         if mobj is not None:
             video_upload_date = mobj.group(3) + mobj.group(2) + mobj.group(1)
 
+        embed_url = 'http://www.dailymotion.com/embed/video/%s' % video_id
+        embed_page = self._download_webpage(embed_url, video_id,
+                                            u'Downloading embed page')
+        info = self._search_regex(r'var info = ({.*?}),$', embed_page,
+            'video info', flags=re.MULTILINE)
+        info = json.loads(info)
+        if info.get('error') is not None:
+            msg = 'Couldn\'t get video, Dailymotion says: %s' % info['error']['title']
+            raise ExtractorError(msg, expected=True)
+
+        # TODO: support choosing qualities
+
+        for key in ['stream_h264_hd1080_url','stream_h264_hd_url',
+                    'stream_h264_hq_url','stream_h264_url',
+                    'stream_h264_ld_url']:
+            if info.get(key):#key in info and info[key]:
+                max_quality = key
+                self.to_screen(u'Using %s' % key)
+                break
+        else:
+            raise ExtractorError(u'Unable to extract video URL')
+        video_url = info[max_quality]
+
+        # subtitles
+        video_subtitles = self.extract_subtitles(video_id)
+        if self._downloader.params.get('listsubtitles', False):
+            self._list_available_subtitles(video_id)
+            return
+
         return [{
             'id':       video_id,
             'url':      video_url,
             'uploader': video_uploader,
             'upload_date':  video_upload_date,
-            'title':    video_title,
+            'title':    self._og_search_title(webpage),
             'ext':      video_extension,
+            'subtitles':    video_subtitles,
+            'thumbnail': info['thumbnail_url']
         }]
+
+    def _get_available_subtitles(self, video_id):
+        try:
+            sub_list = self._download_webpage(
+                'https://api.dailymotion.com/video/%s/subtitles?fields=id,language,url' % video_id,
+                video_id, note=False)
+        except ExtractorError as err:
+            self._downloader.report_warning(u'unable to download video subtitles: %s' % compat_str(err))
+            return {}
+        info = json.loads(sub_list)
+        if (info['total'] > 0):
+            sub_lang_list = dict((l['language'], l['url']) for l in info['list'])
+            return sub_lang_list
+        self._downloader.report_warning(u'video doesn\'t have subtitles')
+        return {}
+
+
+class DailymotionPlaylistIE(DailymotionBaseInfoExtractor):
+    IE_NAME = u'dailymotion:playlist'
+    _VALID_URL = r'(?:https?://)?(?:www\.)?dailymotion\.[a-z]{2,3}/playlist/(?P<id>.+?)/'
+    _MORE_PAGES_INDICATOR = r'<div class="next">.*?<a.*?href="/playlist/.+?".*?>.*?</a>.*?</div>'
+    _PAGE_TEMPLATE = 'https://www.dailymotion.com/playlist/%s/%s'
+
+    def _extract_entries(self, id):
+        video_ids = []
+        for pagenum in itertools.count(1):
+            request = self._build_request(self._PAGE_TEMPLATE % (id, pagenum))
+            webpage = self._download_webpage(request,
+                                             id, u'Downloading page %s' % pagenum)
+
+            playlist_el = get_element_by_attribute(u'class', u'video_list', webpage)
+            video_ids.extend(re.findall(r'data-id="(.+?)"', playlist_el))
+
+            if re.search(self._MORE_PAGES_INDICATOR, webpage, re.DOTALL) is None:
+                break
+        return [self.url_result('http://www.dailymotion.com/video/%s' % video_id, 'Dailymotion')
+                   for video_id in orderedSet(video_ids)]
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        playlist_id = mobj.group('id')
+        webpage = self._download_webpage(url, playlist_id)
+
+        return {'_type': 'playlist',
+                'id': playlist_id,
+                'title': get_element_by_id(u'playlist_name', webpage),
+                'entries': self._extract_entries(playlist_id),
+                }
+
+
+class DailymotionUserIE(DailymotionPlaylistIE):
+    IE_NAME = u'dailymotion:user'
+    _VALID_URL = r'(?:https?://)?(?:www\.)?dailymotion\.[a-z]{2,3}/user/(?P<user>[^/]+)'
+    _MORE_PAGES_INDICATOR = r'<div class="next">.*?<a.*?href="/user/.+?".*?>.*?</a>.*?</div>'
+    _PAGE_TEMPLATE = 'http://www.dailymotion.com/user/%s/%s'
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        user = mobj.group('user')
+        webpage = self._download_webpage(url, user)
+        full_user = self._html_search_regex(
+            r'<a class="label" href="/%s".*?>(.*?)</' % re.escape(user),
+            webpage, u'user', flags=re.DOTALL)
+
+        return {
+            '_type': 'playlist',
+            'id': user,
+            'title': full_user,
+            'entries': self._extract_entries(user),
+        }

youtube_dl/extractor/daum.py

@@ -0,0 +1,74 @@
+# encoding: utf-8
+import re
+import xml.etree.ElementTree
+
+from .common import InfoExtractor
+from ..utils import (
+    compat_urllib_parse,
+    determine_ext,
+)
+
+
+class DaumIE(InfoExtractor):
+    _VALID_URL = r'https?://tvpot\.daum\.net/.*?clipid=(?P<id>\d+)'
+    IE_NAME = u'daum.net'
+
+    _TEST = {
+        u'url': u'http://tvpot.daum.net/clip/ClipView.do?clipid=52554690',
+        u'file': u'52554690.mp4',
+        u'info_dict': {
+            u'title': u'DOTA 2GETHER 시즌2 6회 - 2부',
+            u'description': u'DOTA 2GETHER 시즌2 6회 - 2부',
+            u'upload_date': u'20130831',
+            u'duration': 3868,
+        },
+    }
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        video_id = mobj.group(1)
+        canonical_url = 'http://tvpot.daum.net/v/%s' % video_id
+        webpage = self._download_webpage(canonical_url, video_id)
+        full_id = self._search_regex(r'<link rel="video_src" href=".+?vid=(.+?)"',
+            webpage, u'full id')
+        query = compat_urllib_parse.urlencode({'vid': full_id})
+        info_xml = self._download_webpage(
+            'http://tvpot.daum.net/clip/ClipInfoXml.do?' + query, video_id,
+            u'Downloading video info')
+        urls_xml = self._download_webpage(
+            'http://videofarm.daum.net/controller/api/open/v1_2/MovieData.apixml?' + query,
+            video_id, u'Downloading video formats info')
+        info = xml.etree.ElementTree.fromstring(info_xml.encode('utf-8'))
+        urls = xml.etree.ElementTree.fromstring(urls_xml.encode('utf-8'))
+
+        self.to_screen(u'%s: Getting video urls' % video_id)
+        formats = []
+        for format_el in urls.findall('result/output_list/output_list'):
+            profile = format_el.attrib['profile']
+            format_query = compat_urllib_parse.urlencode({
+                'vid': full_id,
+                'profile': profile,
+            })
+            url_xml = self._download_webpage(
+                'http://videofarm.daum.net/controller/api/open/v1_2/MovieLocation.apixml?' + format_query,
+                video_id, note=False)
+            url_doc = xml.etree.ElementTree.fromstring(url_xml.encode('utf-8'))
+            format_url = url_doc.find('result/url').text
+            formats.append({
+                'url': format_url,
+                'ext': determine_ext(format_url),
+                'format_id': profile,
+            })
+
+        info = {
+            'id': video_id,
+            'title': info.find('TITLE').text,
+            'formats': formats,
+            'thumbnail': self._og_search_thumbnail(webpage),
+            'description': info.find('CONTENTS').text,
+            'duration': int(info.find('DURATION').text),
+            'upload_date': info.find('REGDTTM').text[:8],
+        }
+        # TODO: Remove when #980 has been merged
+        info.update(formats[-1])
+        return info

youtube_dl/extractor/defense.py

@@ -0,0 +1,39 @@
+import re
+import json
+
+from .common import InfoExtractor
+
+
+class DefenseGouvFrIE(InfoExtractor):
+    _IE_NAME = 'defense.gouv.fr'
+    _VALID_URL = (r'http://.*?\.defense\.gouv\.fr/layout/set/'
+        r'ligthboxvideo/base-de-medias/webtv/(.*)')
+
+    _TEST = {
+        u'url': (u'http://www.defense.gouv.fr/layout/set/ligthboxvideo/'
+        u'base-de-medias/webtv/attaque-chimique-syrienne-du-21-aout-2013-1'),
+        u'file': u'11213.mp4',
+        u'md5': u'75bba6124da7e63d2d60b5244ec9430c',
+        "info_dict": {
+            "title": "attaque-chimique-syrienne-du-21-aout-2013-1"
+        }
+    }
+
+    def _real_extract(self, url):
+        title = re.match(self._VALID_URL, url).group(1)
+        webpage = self._download_webpage(url, title)
+        video_id = self._search_regex(
+            r"flashvars.pvg_id=\"(\d+)\";",
+            webpage, 'ID')
+        
+        json_url = ('http://static.videos.gouv.fr/brightcovehub/export/json/'
+            + video_id)
+        info = self._download_webpage(json_url, title,
+                                                  'Downloading JSON config')
+        video_url = json.loads(info)['renditions'][0]['url']
+        
+        return {'id': video_id,
+                'ext': 'mp4',
+                'url': video_url,
+                'title': title,
+                }

youtube_dl/extractor/dotsub.py

@@ -0,0 +1,41 @@
+import re
+import json
+import time
+
+from .common import InfoExtractor
+
+
+class DotsubIE(InfoExtractor):
+    _VALID_URL = r'(?:http://)?(?:www\.)?dotsub\.com/view/([^/]+)'
+    _TEST = {
+        u'url': u'http://dotsub.com/view/aed3b8b2-1889-4df5-ae63-ad85f5572f27',
+        u'file': u'aed3b8b2-1889-4df5-ae63-ad85f5572f27.flv',
+        u'md5': u'0914d4d69605090f623b7ac329fea66e',
+        u'info_dict': {
+            u"title": u"Pyramids of Waste (2010), AKA The Lightbulb Conspiracy - Planned obsolescence documentary",
+            u"uploader": u"4v4l0n42",
+            u'description': u'Pyramids of Waste (2010) also known as "The lightbulb conspiracy" is a documentary about how our economic system based on consumerism  and planned obsolescence is breaking our planet down.\r\n\r\nSolutions to this can be found at:\r\nhttp://robotswillstealyourjob.com\r\nhttp://www.federicopistono.org\r\n\r\nhttp://opensourceecology.org\r\nhttp://thezeitgeistmovement.com',
+            u'thumbnail': u'http://dotsub.com/media/aed3b8b2-1889-4df5-ae63-ad85f5572f27/p',
+            u'upload_date': u'20101213',
+        }
+    }
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        video_id = mobj.group(1)
+        info_url = "https://dotsub.com/api/media/%s/metadata" %(video_id)
+        webpage = self._download_webpage(info_url, video_id)
+        info = json.loads(webpage)
+        date = time.gmtime(info['dateCreated']/1000) # The timestamp is in miliseconds
+
+        return [{
+            'id':          video_id,
+            'url':         info['mediaURI'],
+            'ext':         'flv',
+            'title':       info['title'],
+            'thumbnail':   info['screenshotURI'],
+            'description': info['description'],
+            'uploader':    info['user'],
+            'view_count':  info['numberOfViews'],
+            'upload_date': u'%04i%02i%02i' % (date.tm_year, date.tm_mon, date.tm_mday),
+        }]

youtube_dl/extractor/dreisat.py

@@ -0,0 +1,85 @@
+# coding: utf-8
+
+import re
+import xml.etree.ElementTree
+
+from .common import InfoExtractor
+from ..utils import (
+    determine_ext,
+    unified_strdate,
+)
+
+
+class DreiSatIE(InfoExtractor):
+    IE_NAME = '3sat'
+    _VALID_URL = r'(?:http://)?(?:www\.)?3sat.de/mediathek/index.php\?(?:(?:mode|display)=[^&]+&)*obj=(?P<id>[0-9]+)$'
+    _TEST = {
+        u"url": u"http://www.3sat.de/mediathek/index.php?obj=36983",
+        u'file': u'36983.webm',
+        u'md5': u'57c97d0469d71cf874f6815aa2b7c944',
+        u'info_dict': {
+            u"title": u"Kaffeeland Schweiz",
+            u"description": u"Über 80 Kaffeeröstereien liefern in der Schweiz das Getränk, in das das Land so vernarrt ist: Mehr als 1000 Tassen trinkt ein Schweizer pro Jahr. SCHWEIZWEIT nimmt die Kaffeekultur unter die...", 
+            u"uploader": u"3sat",
+            u"upload_date": u"20130622"
+        }
+    }
+
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        video_id = mobj.group('id')
+        details_url = 'http://www.3sat.de/mediathek/xmlservice/web/beitragsDetails?ak=web&id=%s' % video_id
+        details_xml = self._download_webpage(details_url, video_id, note=u'Downloading video details')
+        details_doc = xml.etree.ElementTree.fromstring(details_xml.encode('utf-8'))
+
+        thumbnail_els = details_doc.findall('.//teaserimage')
+        thumbnails = [{
+            'width': te.attrib['key'].partition('x')[0],
+            'height': te.attrib['key'].partition('x')[2],
+            'url': te.text,
+        } for te in thumbnail_els]
+
+        information_el = details_doc.find('.//information')
+        video_title = information_el.find('./title').text
+        video_description = information_el.find('./detail').text
+
+        details_el = details_doc.find('.//details')
+        video_uploader = details_el.find('./channel').text
+        upload_date = unified_strdate(details_el.find('./airtime').text)
+
+        format_els = details_doc.findall('.//formitaet')
+        formats = [{
+            'format_id': fe.attrib['basetype'],
+            'width': int(fe.find('./width').text),
+            'height': int(fe.find('./height').text),
+            'url': fe.find('./url').text,
+            'ext': determine_ext(fe.find('./url').text),
+            'filesize': int(fe.find('./filesize').text),
+            'video_bitrate': int(fe.find('./videoBitrate').text),
+            '3sat_qualityname': fe.find('./quality').text,
+        } for fe in format_els
+            if not fe.find('./url').text.startswith('http://www.metafilegenerator.de/')]
+
+        def _sortkey(format):
+            qidx = ['low', 'med', 'high', 'veryhigh'].index(format['3sat_qualityname'])
+            prefer_http = 1 if 'rtmp' in format['url'] else 0
+            return (qidx, prefer_http, format['video_bitrate'])
+        formats.sort(key=_sortkey)
+
+        info = {
+            '_type': 'video',
+            'id': video_id,
+            'title': video_title,
+            'formats': formats,
+            'description': video_description,
+            'thumbnails': thumbnails,
+            'thumbnail': thumbnails[-1]['url'],
+            'uploader': video_uploader,
+            'upload_date': upload_date,
+        }
+
+        # TODO: Remove when #980 has been merged
+        info.update(formats[-1])
+
+        return info

youtube_dl/extractor/ebaumsworld.py

@@ -0,0 +1,37 @@
+import re
+import xml.etree.ElementTree
+
+from .common import InfoExtractor
+from ..utils import determine_ext
+
+
+class EbaumsWorldIE(InfoExtractor):
+    _VALID_URL = r'https?://www\.ebaumsworld\.com/video/watch/(?P<id>\d+)'
+
+    _TEST = {
+        u'url': u'http://www.ebaumsworld.com/video/watch/83367677/',
+        u'file': u'83367677.mp4',
+        u'info_dict': {
+            u'title': u'A Giant Python Opens The Door',
+            u'description': u'This is how nightmares start...',
+            u'uploader': u'jihadpizza',
+        },
+    }
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        video_id = mobj.group('id')
+        config_xml = self._download_webpage(
+            'http://www.ebaumsworld.com/video/player/%s' % video_id, video_id)
+        config = xml.etree.ElementTree.fromstring(config_xml.encode('utf-8'))
+        video_url = config.find('file').text
+
+        return {
+            'id': video_id,
+            'title': config.find('title').text,
+            'url': video_url,
+            'ext': determine_ext(video_url),
+            'description': config.find('description').text,
+            'thumbnail': config.find('image').text,
+            'uploader': config.find('username').text,
+        }

youtube_dl/extractor/ehow.py

@@ -0,0 +1,46 @@
+import re
+
+from ..utils import (
+    compat_urllib_parse,
+    determine_ext
+)
+from .common import InfoExtractor
+
+
+class EHowIE(InfoExtractor):
+    IE_NAME = u'eHow'
+    _VALID_URL = r'(?:https?://)?(?:www\.)?ehow\.com/[^/_?]*_(?P<id>[0-9]+)'
+    _TEST = {
+        u'url': u'http://www.ehow.com/video_12245069_hardwood-flooring-basics.html',
+        u'file': u'12245069.flv',
+        u'md5': u'9809b4e3f115ae2088440bcb4efbf371',
+        u'info_dict': {
+            u"title": u"Hardwood Flooring Basics",
+            u"description": u"Hardwood flooring may be time consuming, but its ultimately a pretty straightforward concept. Learn about hardwood flooring basics with help from a hardware flooring business owner in this free video...",
+   			u"uploader": u"Erick Nathan"
+        }
+    }
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        video_id = mobj.group('id')
+        webpage = self._download_webpage(url, video_id)
+        video_url = self._search_regex(r'(?:file|source)=(http[^\'"&]*)',
+            webpage, u'video URL')
+        final_url = compat_urllib_parse.unquote(video_url)        
+        uploader = self._search_regex(r'<meta name="uploader" content="(.+?)" />',
+            webpage, u'uploader')
+        title = self._og_search_title(webpage).replace(' | eHow', '')
+        ext = determine_ext(final_url)
+
+        return {
+            '_type':       'video',
+            'id':          video_id,
+            'url':         final_url,
+            'ext':         ext,
+            'title':       title,
+            'thumbnail':   self._og_search_thumbnail(webpage),
+            'description': self._og_search_description(webpage),
+            'uploader':    uploader,
+        }
+

youtube_dl/extractor/escapist.py

@@ -36,11 +36,7 @@ class EscapistIE(InfoExtractor):
         videoDesc = self._html_search_regex('<meta name="description" content="([^"]*)"',
             webpage, u'description', fatal=False)
 
-        imgUrl = self._html_search_regex('<meta property="og:image" content="([^"]*)"',
-            webpage, u'thumbnail', fatal=False)
-
-        playerUrl = self._html_search_regex('<meta property="og:video" content="([^"]*)"',
-            webpage, u'player url')
+        playerUrl = self._og_search_video_url(webpage, name='player url')
 
         title = self._html_search_regex('<meta name="title" content="([^"]*)"',
             webpage, u'player url').split(' : ')[-1]
@@ -70,7 +66,7 @@ class EscapistIE(InfoExtractor):
             'upload_date': None,
             'title': title,
             'ext': 'mp4',
-            'thumbnail': imgUrl,
+            'thumbnail': self._og_search_thumbnail(webpage),
             'description': videoDesc,
             'player_url': playerUrl,
         }

youtube_dl/extractor/exfm.py

@@ -0,0 +1,54 @@
+import re
+import json
+
+from .common import InfoExtractor
+
+
+class ExfmIE(InfoExtractor):
+    IE_NAME = u'exfm'
+    IE_DESC = u'ex.fm'
+    _VALID_URL = r'(?:http://)?(?:www\.)?ex\.fm/song/([^/]+)'
+    _SOUNDCLOUD_URL = r'(?:http://)?(?:www\.)?api\.soundcloud.com/tracks/([^/]+)/stream'
+    _TESTS = [
+        {
+            u'url': u'http://ex.fm/song/1bgtzg',
+            u'file': u'95223130.mp3',
+            u'md5': u'8a7967a3fef10e59a1d6f86240fd41cf',
+            u'info_dict': {
+                u"title": u"We Can't Stop - Miley Cyrus",
+                u"uploader": u"Miley Cyrus",
+                u'upload_date': u'20130603',
+                u'description': u'Download "We Can\'t Stop" \r\niTunes: http://smarturl.it/WeCantStop?IQid=SC\r\nAmazon: http://smarturl.it/WeCantStopAMZ?IQid=SC',
+            },
+            u'note': u'Soundcloud song',
+        },
+        {
+            u'url': u'http://ex.fm/song/wddt8',
+            u'file': u'wddt8.mp3',
+            u'md5': u'966bd70741ac5b8570d8e45bfaed3643',
+            u'info_dict': {
+                u'title': u'Safe and Sound',
+                u'uploader': u'Capital Cities',
+            },
+        },
+    ]
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        song_id = mobj.group(1)
+        info_url = "http://ex.fm/api/v3/song/%s" %(song_id)
+        webpage = self._download_webpage(info_url, song_id)
+        info = json.loads(webpage)
+        song_url = info['song']['url']
+        if re.match(self._SOUNDCLOUD_URL, song_url) is not None:
+            self.to_screen('Soundcloud song detected')
+            return self.url_result(song_url.replace('/stream',''), 'Soundcloud')
+        return [{
+            'id':          song_id,
+            'url':         song_url,
+            'ext':         'mp3',
+            'title':       info['song']['title'],
+            'thumbnail':   info['song']['image']['large'],
+            'uploader':    info['song']['artist'],
+            'view_count':  info['song']['loved_count'],
+        }]

youtube_dl/extractor/facebook.py

@@ -106,8 +106,8 @@ class FacebookIE(InfoExtractor):
         video_duration = int(video_data['video_duration'])
         thumbnail = video_data['thumbnail_src']
 
-        video_title = self._html_search_regex('<h2 class="uiHeaderTitle">([^<]+)</h2>',
-            webpage, u'title')
+        video_title = self._html_search_regex(
+            r'<h2 class="uiHeaderTitle">([^<]*)</h2>', webpage, u'title')
 
         info = {
             'id': video_id,

youtube_dl/extractor/faz.py

@@ -0,0 +1,60 @@
+# encoding: utf-8
+import re
+import xml.etree.ElementTree
+
+from .common import InfoExtractor
+from ..utils import (
+    determine_ext,
+    clean_html,
+    get_element_by_attribute,
+)
+
+
+class FazIE(InfoExtractor):
+    IE_NAME = u'faz.net'
+    _VALID_URL = r'https?://www\.faz\.net/multimedia/videos/.*?-(?P<id>\d+).html'
+
+    _TEST = {
+        u'url': u'http://www.faz.net/multimedia/videos/stockholm-chemie-nobelpreis-fuer-drei-amerikanische-forscher-12610585.html',
+        u'file': u'12610585.mp4',
+        u'info_dict': {
+            u'title': u'Stockholm: Chemie-Nobelpreis für drei amerikanische Forscher',
+            u'description': u'md5:1453fbf9a0d041d985a47306192ea253',
+        },
+    }
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        video_id = mobj.group('id')
+        self.to_screen(video_id)
+        webpage = self._download_webpage(url, video_id)
+        config_xml_url = self._search_regex(r'writeFLV\(\'(.+?)\',', webpage,
+            u'config xml url')
+        config_xml = self._download_webpage(config_xml_url, video_id,
+            u'Downloading config xml')
+        config = xml.etree.ElementTree.fromstring(config_xml.encode('utf-8'))
+
+        encodings = config.find('ENCODINGS')
+        formats = []
+        for code in ['LOW', 'HIGH', 'HQ']:
+            encoding = encodings.find(code)
+            if encoding is None:
+                continue
+            encoding_url = encoding.find('FILENAME').text
+            formats.append({
+                'url': encoding_url,
+                'ext': determine_ext(encoding_url),
+                'format_id': code.lower(),
+            })
+
+        descr_html = get_element_by_attribute('class', 'Content Copy', webpage)
+        info = {
+            'id': video_id,
+            'title': self._og_search_title(webpage),
+            'formats': formats,
+            'description': clean_html(descr_html),
+            'thumbnail': config.find('STILL/STILL_BIG').text,
+        }
+        # TODO: Remove when #980 has been merged
+        info.update(formats[-1])
+        return info

youtube_dl/extractor/fktv.py

@@ -0,0 +1,79 @@
+import re
+import random
+import json
+
+from .common import InfoExtractor
+from ..utils import (
+    determine_ext,
+    get_element_by_id,
+    clean_html,
+)
+
+
+class FKTVIE(InfoExtractor):
+    IE_NAME = u'fernsehkritik.tv'
+    _VALID_URL = r'(?:http://)?(?:www\.)?fernsehkritik.tv/folge-(?P<ep>[0-9]+)(?:/.*)?'
+
+    _TEST = {
+        u'url': u'http://fernsehkritik.tv/folge-1',
+        u'file': u'00011.flv',
+        u'info_dict': {
+            u'title': u'Folge 1 vom 10. April 2007',
+            u'description': u'md5:fb4818139c7cfe6907d4b83412a6864f',
+        },
+    }
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        episode = int(mobj.group('ep'))
+
+        server = random.randint(2, 4)
+        video_thumbnail = 'http://fernsehkritik.tv/images/magazin/folge%d.jpg' % episode
+        start_webpage = self._download_webpage('http://fernsehkritik.tv/folge-%d/Start' % episode,
+            episode)
+        playlist = self._search_regex(r'playlist = (\[.*?\]);', start_webpage,
+            u'playlist', flags=re.DOTALL)
+        files = json.loads(re.sub('{[^{}]*?}', '{}', playlist))
+        # TODO: return a single multipart video
+        videos = []
+        for i, _ in enumerate(files, 1):
+            video_id = '%04d%d' % (episode, i)
+            video_url = 'http://dl%d.fernsehkritik.tv/fernsehkritik%d%s.flv' % (server, episode, '' if i == 1 else '-%d' % i)
+            video_title = 'Fernsehkritik %d.%d' % (episode, i)
+            videos.append({
+                'id': video_id,
+                'url': video_url,
+                'ext': determine_ext(video_url),
+                'title': clean_html(get_element_by_id('eptitle', start_webpage)),
+                'description': clean_html(get_element_by_id('contentlist', start_webpage)),
+                'thumbnail': video_thumbnail
+            })
+        return videos
+
+
+class FKTVPosteckeIE(InfoExtractor):
+    IE_NAME = u'fernsehkritik.tv:postecke'
+    _VALID_URL = r'(?:http://)?(?:www\.)?fernsehkritik.tv/inline-video/postecke.php\?(.*&)?ep=(?P<ep>[0-9]+)(&|$)'
+    _TEST = {
+        u'url': u'http://fernsehkritik.tv/inline-video/postecke.php?iframe=true&width=625&height=440&ep=120',
+        u'file': u'0120.flv',
+        u'md5': u'262f0adbac80317412f7e57b4808e5c4',
+        u'info_dict': {
+            u"title": u"Postecke 120"
+        }
+    }
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        episode = int(mobj.group('ep'))
+
+        server = random.randint(2, 4)
+        video_id = '%04d' % episode
+        video_url = 'http://dl%d.fernsehkritik.tv/postecke/postecke%d.flv' % (server, episode)
+        video_title = 'Postecke %d' % episode
+        return {
+            'id':       video_id,
+            'url':      video_url,
+            'ext':      determine_ext(video_url),
+            'title':    video_title,
+        }

youtube_dl/extractor/flickr.py

@@ -9,7 +9,7 @@ from ..utils import (
 
 class FlickrIE(InfoExtractor):
     """Information Extractor for Flickr videos"""
-    _VALID_URL = r'(?:https?://)?(?:www\.)?flickr\.com/photos/(?P<uploader_id>[\w\-_@]+)/(?P<id>\d+).*'
+    _VALID_URL = r'(?:https?://)?(?:www\.|secure\.)?flickr\.com/photos/(?P<uploader_id>[\w\-_@]+)/(?P<id>\d+).*'
     _TEST = {
         u'url': u'http://www.flickr.com/photos/forestwander-nature-pictures/5645318632/in/photostream/',
         u'file': u'5645318632.mp4',
@@ -47,21 +47,12 @@ class FlickrIE(InfoExtractor):
             raise ExtractorError(u'Unable to extract video url')
         video_url = mobj.group(1) + unescapeHTML(mobj.group(2))
 
-        video_title = self._html_search_regex(r'<meta property="og:title" content=(?:"([^"]+)"|\'([^\']+)\')',
-            webpage, u'video title')
-
-        video_description = self._html_search_regex(r'<meta property="og:description" content=(?:"([^"]+)"|\'([^\']+)\')',
-            webpage, u'description', fatal=False)
-
-        thumbnail = self._html_search_regex(r'<meta property="og:image" content=(?:"([^"]+)"|\'([^\']+)\')',
-            webpage, u'thumbnail', fatal=False)
-
         return [{
             'id':          video_id,
             'url':         video_url,
             'ext':         'mp4',
-            'title':       video_title,
-            'description': video_description,
-            'thumbnail':   thumbnail,
+            'title':       self._og_search_title(webpage),
+            'description': self._og_search_description(webpage),
+            'thumbnail':   self._og_search_thumbnail(webpage),
             'uploader_id': video_uploader_id,
         }]

youtube_dl/extractor/francetv.py

@@ -0,0 +1,129 @@
+# encoding: utf-8
+import re
+import xml.etree.ElementTree
+import json
+
+from .common import InfoExtractor
+from ..utils import (
+    compat_urlparse,
+)
+
+
+class FranceTVBaseInfoExtractor(InfoExtractor):
+    def _extract_video(self, video_id):
+        xml_desc = self._download_webpage(
+            'http://www.francetvinfo.fr/appftv/webservices/video/'
+            'getInfosOeuvre.php?id-diffusion='
+            + video_id, video_id, 'Downloading XML config')
+        info = xml.etree.ElementTree.fromstring(xml_desc.encode('utf-8'))
+
+        manifest_url = info.find('videos/video/url').text
+        video_url = manifest_url.replace('manifest.f4m', 'index_2_av.m3u8')
+        video_url = video_url.replace('/z/', '/i/')
+        thumbnail_path = info.find('image').text
+
+        return {'id': video_id,
+                'ext': 'mp4',
+                'url': video_url,
+                'title': info.find('titre').text,
+                'thumbnail': compat_urlparse.urljoin('http://pluzz.francetv.fr', thumbnail_path),
+                'description': info.find('synopsis').text,
+                }
+
+
+class PluzzIE(FranceTVBaseInfoExtractor):
+    IE_NAME = u'pluzz.francetv.fr'
+    _VALID_URL = r'https?://pluzz\.francetv\.fr/videos/(.*?)\.html'
+
+    # Can't use tests, videos expire in 7 days
+
+    def _real_extract(self, url):
+        title = re.match(self._VALID_URL, url).group(1)
+        webpage = self._download_webpage(url, title)
+        video_id = self._search_regex(
+            r'data-diffusion="(\d+)"', webpage, 'ID')
+        return self._extract_video(video_id)
+
+
+class FranceTvInfoIE(FranceTVBaseInfoExtractor):
+    IE_NAME = u'francetvinfo.fr'
+    _VALID_URL = r'https?://www\.francetvinfo\.fr/replay.*/(?P<title>.+).html'
+
+    _TEST = {
+        u'url': u'http://www.francetvinfo.fr/replay-jt/france-3/soir-3/jt-grand-soir-3-lundi-26-aout-2013_393427.html',
+        u'file': u'84981923.mp4',
+        u'info_dict': {
+            u'title': u'Soir 3',
+        },
+        u'params': {
+            u'skip_download': True,
+        },
+    }
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        page_title = mobj.group('title')
+        webpage = self._download_webpage(url, page_title)
+        video_id = self._search_regex(r'id-video=(\d+?)"', webpage, u'video id')
+        return self._extract_video(video_id)
+
+
+class France2IE(FranceTVBaseInfoExtractor):
+    IE_NAME = u'france2.fr'
+    _VALID_URL = r'''(?x)https?://www\.france2\.fr/
+        (?:
+            emissions/.*?/videos/(?P<id>\d+)
+        |   emission/(?P<key>[^/?]+)
+        )'''
+
+    _TEST = {
+        u'url': u'http://www.france2.fr/emissions/13h15-le-samedi-le-dimanche/videos/75540104',
+        u'file': u'75540104.mp4',
+        u'info_dict': {
+            u'title': u'13h15, le samedi...',
+            u'description': u'md5:2e5b58ba7a2d3692b35c792be081a03d',
+        },
+        u'params': {
+            u'skip_download': True,
+        },
+    }
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        if mobj.group('key'):
+            webpage = self._download_webpage(url, mobj.group('key'))
+            video_id = self._html_search_regex(
+                r'''(?x)<div\s+class="video-player">\s*
+                    <a\s+href="http://videos.francetv.fr/video/([0-9]+)"\s+
+                    class="francetv-video-player">''',
+                webpage, u'video ID')
+        else:
+            video_id = mobj.group('id')
+        return self._extract_video(video_id)
+
+
+class GenerationQuoiIE(InfoExtractor):
+    IE_NAME = u'france2.fr:generation-quoi'
+    _VALID_URL = r'https?://generation-quoi\.france2\.fr/portrait/(?P<name>.*)(\?|$)'
+
+    _TEST = {
+        u'url': u'http://generation-quoi.france2.fr/portrait/garde-a-vous',
+        u'file': u'k7FJX8VBcvvLmX4wA5Q.mp4',
+        u'info_dict': {
+            u'title': u'Génération Quoi - Garde à Vous',
+            u'uploader': u'Génération Quoi',
+        },
+        u'params': {
+            # It uses Dailymotion
+            u'skip_download': True,
+        },
+    }
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        name = mobj.group('name')
+        info_url = compat_urlparse.urljoin(url, '/medias/video/%s.json' % name)
+        info_json = self._download_webpage(info_url, name)
+        info = json.loads(info_json)
+        return self.url_result('http://www.dailymotion.com/video/%s' % info['id'],
+            ie='Dailymotion')

youtube_dl/extractor/freesound.py

@@ -0,0 +1,36 @@
+import re
+
+from .common import InfoExtractor
+from ..utils import determine_ext
+
+class FreesoundIE(InfoExtractor):
+    _VALID_URL = r'(?:https?://)?(?:www\.)?freesound\.org/people/([^/]+)/sounds/(?P<id>[^/]+)'
+    _TEST = {
+        u'url': u'http://www.freesound.org/people/miklovan/sounds/194503/',
+        u'file': u'194503.mp3',
+        u'md5': u'12280ceb42c81f19a515c745eae07650',
+        u'info_dict': {
+            u"title": u"gulls in the city.wav",
+            u"uploader" : u"miklovan",
+            u'description': u'the sounds of seagulls in the city',
+        }
+    }
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        music_id = mobj.group('id')
+        webpage = self._download_webpage(url, music_id)
+        title = self._html_search_regex(r'<div id="single_sample_header">.*?<a href="#">(.+?)</a>',
+                                webpage, 'music title', flags=re.DOTALL)
+        music_url = self._og_search_property('audio', webpage, 'music url')
+        description = self._html_search_regex(r'<div id="sound_description">(.*?)</div>',
+                                webpage, 'description', fatal=False, flags=re.DOTALL)
+
+        return [{
+            'id':       music_id,
+            'title':    title,            
+            'url':      music_url,
+            'uploader': self._og_search_property('audio:artist', webpage, 'music uploader'),
+            'ext':      determine_ext(music_url),
+            'description': description,
+        }]

youtube_dl/extractor/funnyordie.py

@@ -21,20 +21,15 @@ class FunnyOrDieIE(InfoExtractor):
         video_id = mobj.group('id')
         webpage = self._download_webpage(url, video_id)
 
-        video_url = self._html_search_regex(r'<video[^>]*>\s*<source[^>]*>\s*<source src="(?P<url>[^"]+)"',
+        video_url = self._search_regex(
+            [r'type="video/mp4" src="(.*?)"', r'src="([^>]*?)" type=\'video/mp4\''],
             webpage, u'video URL', flags=re.DOTALL)
 
-        title = self._html_search_regex((r"<h1 class='player_page_h1'.*?>(?P<title>.*?)</h1>",
-            r'<title>(?P<title>[^<]+?)</title>'), webpage, 'title', flags=re.DOTALL)
-
-        video_description = self._html_search_regex(r'<meta property="og:description" content="(?P<desc>.*?)"',
-            webpage, u'description', fatal=False, flags=re.DOTALL)
-
         info = {
             'id': video_id,
             'url': video_url,
             'ext': 'mp4',
-            'title': title,
-            'description': video_description,
+            'title': self._og_search_title(webpage),
+            'description': self._og_search_description(webpage),
         }
         return [info]

youtube_dl/extractor/gamespot.py

@@ -1,45 +1,59 @@
 import re
-import xml.etree.ElementTree
+import json
 
 from .common import InfoExtractor
 from ..utils import (
-    unified_strdate,
+    compat_urllib_parse,
+    compat_urlparse,
+    unescapeHTML,
+    get_meta_content,
 )
 
+
 class GameSpotIE(InfoExtractor):
-    _VALID_URL = r'(?:http://)?(?:www\.)?gamespot\.com/([^/]+)/videos/([^/]+)-([^/d]+)/'
+    _VALID_URL = r'(?:http://)?(?:www\.)?gamespot\.com/.*-(?P<page_id>\d+)/?'
     _TEST = {
         u"url": u"http://www.gamespot.com/arma-iii/videos/arma-iii-community-guide-sitrep-i-6410818/",
-        u"file": u"6410818.mp4",
-        u"md5": u"5569d64ca98db01f0177c934fe8c1e9b",
+        u"file": u"gs-2300-6410818.mp4",
+        u"md5": u"b2a30deaa8654fcccd43713a6b6a4825",
         u"info_dict": {
-            u"title": u"Arma III - Community Guide: SITREP I",
-            u"upload_date": u"20130627", 
+            u"title": u"Arma 3 - Community Guide: SITREP I",
+            u'description': u'Check out this video where some of the basics of Arma 3 is explained.',
         }
     }
 
-
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
-        video_id = mobj.group(3).split("-")[-1]
-        info_url = "http://www.gamespot.com/pages/video_player/xml.php?id="+str(video_id)
-        info_xml = self._download_webpage(info_url, video_id)
-        doc = xml.etree.ElementTree.fromstring(info_xml)
-        clip_el = doc.find('./playList/clip')
+        page_id = video_id = mobj.group('page_id')
+        webpage = self._download_webpage(url, page_id)
+        data_video_json = self._search_regex(r'data-video=\'(.*?)\'', webpage, u'data video')
+        data_video = json.loads(unescapeHTML(data_video_json))
 
-        video_url = clip_el.find('./URI').text
-        title = clip_el.find('./title').text
-        ext = video_url.rpartition('.')[2]
-        thumbnail_url = clip_el.find('./screenGrabURI').text
-        view_count = int(clip_el.find('./views').text)
-        upload_date = unified_strdate(clip_el.find('./postDate').text)
+        # Transform the manifest url to a link to the mp4 files
+        # they are used in mobile devices.
+        f4m_url = data_video['videoStreams']['f4m_stream']
+        f4m_path = compat_urlparse.urlparse(f4m_url).path
+        QUALITIES_RE = r'((,\d+)+,?)'
+        qualities = self._search_regex(QUALITIES_RE, f4m_path, u'qualities').strip(',').split(',')
+        http_path = f4m_path[1:].split('/', 1)[1]
+        http_template = re.sub(QUALITIES_RE, r'%s', http_path)
+        http_template = http_template.replace('.csmil/manifest.f4m', '')
+        http_template = compat_urlparse.urljoin('http://video.gamespotcdn.com/', http_template)
+        formats = []
+        for q in qualities:
+            formats.append({
+                'url': http_template % q,
+                'ext': 'mp4',
+                'format_id': q,
+            })
 
-        return [{
-            'id'          : video_id,
-            'url'         : video_url,
-            'ext'         : ext,
-            'title'       : title,
-            'thumbnail'   : thumbnail_url,
-            'upload_date' : upload_date,
-            'view_count'  : view_count,
-        }]
+        info = {
+            'id': data_video['guid'],
+            'title': compat_urllib_parse.unquote(data_video['title']),
+            'formats': formats,
+            'description': get_meta_content('description', webpage),
+            'thumbnail': self._og_search_thumbnail(webpage),
+        }
+        # TODO: Remove when #980 has been merged
+        info.update(formats[-1])
+        return info

youtube_dl/extractor/gametrailers.py

@@ -1,68 +1,36 @@
 import re
 
-from .common import InfoExtractor
-from ..utils import (
-    compat_urllib_parse,
+from .mtv import MTVIE, _media_xml_tag
 
-    ExtractorError,
-)
-
-class GametrailersIE(InfoExtractor):
+class GametrailersIE(MTVIE):
+    """
+    Gametrailers use the same videos system as MTVIE, it just changes the feed
+    url, where the uri is and the method to get the thumbnails.
+    """
     _VALID_URL = r'http://www.gametrailers.com/(?P<type>videos|reviews|full-episodes)/(?P<id>.*?)/(?P<title>.*)'
     _TEST = {
         u'url': u'http://www.gametrailers.com/videos/zbvr8i/mirror-s-edge-2-e3-2013--debut-trailer',
-        u'file': u'zbvr8i.flv',
-        u'md5': u'c3edbc995ab4081976e16779bd96a878',
+        u'file': u'70e9a5d7-cf25-4a10-9104-6f3e7342ae0d.mp4',
+        u'md5': u'4c8e67681a0ea7ec241e8c09b3ea8cf7',
         u'info_dict': {
-            u"title": u"E3 2013: Debut Trailer"
+            u'title': u'E3 2013: Debut Trailer',
+            u'description': u'Faith is back!  Check out the World Premiere trailer for Mirror\'s Edge 2 straight from the EA Press Conference at E3 2013!',
         },
-        u'skip': u'Requires rtmpdump'
     }
+    # Overwrite MTVIE properties we don't want
+    _TESTS = []
+
+    _FEED_URL = 'http://www.gametrailers.com/feeds/mrss'
+
+    def _get_thumbnail_url(self, uri, itemdoc):
+        search_path = '%s/%s' % (_media_xml_tag('group'), _media_xml_tag('thumbnail'))
+        return itemdoc.find(search_path).attrib['url']
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
-        if mobj is None:
-            raise ExtractorError(u'Invalid URL: %s' % url)
         video_id = mobj.group('id')
-        video_type = mobj.group('type')
         webpage = self._download_webpage(url, video_id)
-        if video_type == 'full-episodes':
-            mgid_re = r'data-video="(?P<mgid>mgid:.*?)"'
-        else:
-            mgid_re = r'data-contentId=\'(?P<mgid>mgid:.*?)\''
-        mgid = self._search_regex(mgid_re, webpage, u'mgid')
-        data = compat_urllib_parse.urlencode({'uri': mgid, 'acceptMethods': 'fms'})
-
-        info_page = self._download_webpage('http://www.gametrailers.com/feeds/mrss?' + data,
-                                           video_id, u'Downloading video info')
-        links_webpage = self._download_webpage('http://www.gametrailers.com/feeds/mediagen/?' + data,
-                                               video_id, u'Downloading video urls info')
-
-        self.report_extraction(video_id)
-        info_re = r'''<title><!\[CDATA\[(?P<title>.*?)\]\]></title>.*
-                      <description><!\[CDATA\[(?P<description>.*?)\]\]></description>.*
-                      <image>.*
-                        <url>(?P<thumb>.*?)</url>.*
-                      </image>'''
-
-        m_info = re.search(info_re, info_page, re.VERBOSE|re.DOTALL)
-        if m_info is None:
-            raise ExtractorError(u'Unable to extract video info')
-        video_title = m_info.group('title')
-        video_description = m_info.group('description')
-        video_thumb = m_info.group('thumb')
-
-        m_urls = list(re.finditer(r'<src>(?P<url>.*)</src>', links_webpage))
-        if m_urls is None or len(m_urls) == 0:
-            raise ExtractorError(u'Unable to extract video url')
-        # They are sorted from worst to best quality
-        video_url = m_urls[-1].group('url')
-
-        return {'url':         video_url,
-                'id':          video_id,
-                'title':       video_title,
-                # Videos are actually flv not mp4
-                'ext':         'flv',
-                'thumbnail':   video_thumb,
-                'description': video_description,
-                }
+        mgid = self._search_regex([r'data-video="(?P<mgid>mgid:.*?)"',
+                                   r'data-contentId=\'(?P<mgid>mgid:.*?)\''],
+                                  webpage, u'mgid')
+        return self._get_videos_info(mgid)

youtube_dl/extractor/generic.py

@@ -1,3 +1,5 @@
+# encoding: utf-8
+
 import os
 import re
 
@@ -6,23 +8,41 @@ from ..utils import (
     compat_urllib_error,
     compat_urllib_parse,
     compat_urllib_request,
+    compat_urlparse,
 
     ExtractorError,
+    smuggle_url,
+    unescapeHTML,
 )
+from .brightcove import BrightcoveIE
+
 
 class GenericIE(InfoExtractor):
     IE_DESC = u'Generic downloader that works on some sites'
     _VALID_URL = r'.*'
     IE_NAME = u'generic'
-    _TEST = {
-        u'url': u'http://www.hodiho.fr/2013/02/regis-plante-sa-jeep.html',
-        u'file': u'13601338388002.mp4',
-        u'md5': u'85b90ccc9d73b4acd9138d3af4c27f89',
-        u'info_dict': {
-            u"uploader": u"www.hodiho.fr", 
-            u"title": u"R\u00e9gis plante sa Jeep"
+    _TESTS = [
+        {
+            u'url': u'http://www.hodiho.fr/2013/02/regis-plante-sa-jeep.html',
+            u'file': u'13601338388002.mp4',
+            u'md5': u'85b90ccc9d73b4acd9138d3af4c27f89',
+            u'info_dict': {
+                u"uploader": u"www.hodiho.fr",
+                u"title": u"R\u00e9gis plante sa Jeep"
+            }
+        },
+        # embedded vimeo video
+        {
+            u'url': u'http://skillsmatter.com/podcast/home/move-semanticsperfect-forwarding-and-rvalue-references',
+            u'file': u'22444065.mp4',
+            u'md5': u'2903896e23df39722c33f015af0666e2',
+            u'info_dict': {
+                u'title': u'ACCU 2011: Move Semantics,Perfect Forwarding, and Rvalue references- Scott Meyers- 13/04/2011',
+                u"uploader_id": u"skillsmatter",
+                u"uploader": u"Skills Matter",
+            }
         }
-    }
+    ]
 
     def report_download_webpage(self, video_id):
         """Report webpage download."""
@@ -91,8 +111,18 @@ class GenericIE(InfoExtractor):
         return new_url
 
     def _real_extract(self, url):
-        new_url = self._test_redirect(url)
-        if new_url: return [self.url_result(new_url)]
+        parsed_url = compat_urlparse.urlparse(url)
+        if not parsed_url.scheme:
+            self._downloader.report_warning('The url doesn\'t specify the protocol, trying with http')
+            return self.url_result('http://' + url)
+
+        try:
+            new_url = self._test_redirect(url)
+            if new_url:
+                return [self.url_result(new_url)]
+        except compat_urllib_error.HTTPError:
+            # This may be a stupid server that doesn't like HEAD, our UA, or so
+            pass
 
         video_id = url.split('/')[-1]
         try:
@@ -100,9 +130,24 @@ class GenericIE(InfoExtractor):
         except ValueError:
             # since this is the last-resort InfoExtractor, if
             # this error is thrown, it'll be thrown here
-            raise ExtractorError(u'Invalid URL: %s' % url)
+            raise ExtractorError(u'Failed to download URL: %s' % url)
 
         self.report_extraction(video_id)
+        # Look for BrightCove:
+        m_brightcove = re.search(r'<object[^>]+?class=([\'"])[^>]*?BrightcoveExperience.*?\1.+?</object>', webpage, re.DOTALL)
+        if m_brightcove is not None:
+            self.to_screen(u'Brightcove video detected.')
+            bc_url = BrightcoveIE._build_brighcove_url(m_brightcove.group())
+            return self.url_result(bc_url, 'Brightcove')
+
+        # Look for embedded Vimeo player
+        mobj = re.search(
+            r'<iframe\s+src="(https?://player.vimeo.com/video/.*?)"', webpage)
+        if mobj:
+            player_url = unescapeHTML(mobj.group(1))
+            surl = smuggle_url(player_url, {'Referer': url})
+            return self.url_result(surl, 'Vimeo')
+
         # Start with something easy: JW Player in SWFObject
         mobj = re.search(r'flashvars: [\'"](?:.*&)?file=(http[^\'"&]*)', webpage)
         if mobj is None:
@@ -122,15 +167,19 @@ class GenericIE(InfoExtractor):
             if m_video_type is not None:
                 mobj = re.search(r'<meta.*?property="og:video".*?content="(.*?)"', webpage)
         if mobj is None:
-            raise ExtractorError(u'Invalid URL: %s' % url)
+            # HTML5 video
+            mobj = re.search(r'<video[^<]*(?:>.*?<source.*?)? src="([^"]+)"', webpage, flags=re.DOTALL)
+        if mobj is None:
+            raise ExtractorError(u'Unsupported URL: %s' % url)
 
         # It's possible that one of the regexes
         # matched, but returned an empty group:
         if mobj.group(1) is None:
-            raise ExtractorError(u'Invalid URL: %s' % url)
+            raise ExtractorError(u'Did not find a valid video URL at %s' % url)
 
-        video_url = compat_urllib_parse.unquote(mobj.group(1))
-        video_id = os.path.basename(video_url)
+        video_url = mobj.group(1)
+        video_url = compat_urlparse.urljoin(url, video_url)
+        video_id = compat_urllib_parse.unquote(os.path.basename(video_url))
 
         # here's a fun little line of code for you:
         video_extension = os.path.splitext(video_id)[1][1:]

youtube_dl/extractor/googleplus.py

@@ -40,7 +40,9 @@ class GooglePlusIE(InfoExtractor):
         self.report_extraction(video_id)
 
         # Extract update date
-        upload_date = self._html_search_regex('title="Timestamp">(.*?)</a>',
+        upload_date = self._html_search_regex(
+            r'''(?x)<a.+?class="o-T-s\s[^"]+"\s+style="display:\s*none"\s*>
+                    ([0-9]{4}-[0-9]{2}-[0-9]{2})</a>''',
             webpage, u'upload date', fatal=False)
         if upload_date:
             # Convert timestring to a format suitable for filename
@@ -57,8 +59,8 @@ class GooglePlusIE(InfoExtractor):
             webpage, 'title', default=u'NA')
 
         # Step 2, Simulate clicking the image box to launch video
-        DOMAIN = 'https://plus.google.com'
-        video_page = self._search_regex(r'<a href="((?:%s)?/photos/.*?)"' % re.escape(DOMAIN),
+        DOMAIN = 'https://plus.google.com/'
+        video_page = self._search_regex(r'<a href="((?:%s)?photos/.*?)"' % re.escape(DOMAIN),
             webpage, u'video page URL')
         if not video_page.startswith(DOMAIN):
             video_page = DOMAIN + video_page

youtube_dl/extractor/hark.py

@@ -0,0 +1,37 @@
+# -*- coding: utf-8 -*-
+
+import re
+import json
+
+from .common import InfoExtractor
+from ..utils import determine_ext
+
+class HarkIE(InfoExtractor):
+    _VALID_URL = r'https?://www\.hark\.com/clips/(.+?)-.+'
+    _TEST = {
+        u'url': u'http://www.hark.com/clips/mmbzyhkgny-obama-beyond-the-afghan-theater-we-only-target-al-qaeda-on-may-23-2013',
+        u'file': u'mmbzyhkgny.mp3',
+        u'md5': u'6783a58491b47b92c7c1af5a77d4cbee',
+        u'info_dict': {
+            u'title': u"Obama: 'Beyond The Afghan Theater, We Only Target Al Qaeda' on May 23, 2013",
+            u'description': u'President Barack Obama addressed the nation live on May 23, 2013 in a speech aimed at addressing counter-terrorism policies including the use of drone strikes, detainees at Guantanamo Bay prison facility, and American citizens who are terrorists.',
+            u'duration': 11,
+        }
+    }
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        video_id = mobj.group(1)
+        json_url = "http://www.hark.com/clips/%s.json" %(video_id)
+        info_json = self._download_webpage(json_url, video_id)
+        info = json.loads(info_json)
+        final_url = info['url']
+
+        return {'id': video_id,
+                'url' : final_url,
+                'title': info['name'],
+                'ext': determine_ext(final_url),
+                'description': info['description'],
+                'thumbnail': info['image_original'],
+                'duration': info['duration'],
+                }

youtube_dl/extractor/hotnewhiphop.py

@@ -7,11 +7,11 @@ from .common import InfoExtractor
 class HotNewHipHopIE(InfoExtractor):
     _VALID_URL = r'http://www\.hotnewhiphop.com/.*\.(?P<id>.*)\.html'
     _TEST = {
-        u'url': u"http://www.hotnewhiphop.com/freddie-gibbs-lay-it-down-song.1435540.html'",
+        u'url': u"http://www.hotnewhiphop.com/freddie-gibbs-lay-it-down-song.1435540.html",
         u'file': u'1435540.mp3',
         u'md5': u'2c2cd2f76ef11a9b3b581e8b232f3d96',
         u'info_dict': {
-            u"title": u"Freddie Gibbs Songs - Lay It Down"
+            u"title": u"Freddie Gibbs - Lay It Down"
         }
     }
 
@@ -33,16 +33,12 @@ class HotNewHipHopIE(InfoExtractor):
 
         video_title = self._html_search_regex(r"<title>(.*)</title>",
             webpage_src, u'title')
-        
-        # Getting thumbnail and if not thumbnail sets correct title for WSHH candy video.
-        thumbnail = self._html_search_regex(r'"og:image" content="(.*)"',
-            webpage_src, u'thumbnail', fatal=False)
 
         results = [{
                     'id': video_id,
                     'url' : video_url,
                     'title' : video_title,
-                    'thumbnail' : thumbnail,
+                    'thumbnail' : self._og_search_thumbnail(webpage_src),
                     'ext' : 'mp3',
                     }]
-        return results
+        return results

youtube_dl/extractor/howcast.py

@@ -19,8 +19,7 @@ class HowcastIE(InfoExtractor):
         mobj = re.match(self._VALID_URL, url)
 
         video_id = mobj.group('id')
-        webpage_url = 'http://www.howcast.com/videos/' + video_id
-        webpage = self._download_webpage(webpage_url, video_id)
+        webpage = self._download_webpage(url, video_id)
 
         self.report_extraction(video_id)
 

youtube_dl/extractor/ign.py

@@ -0,0 +1,129 @@
+import re
+import json
+
+from .common import InfoExtractor
+from ..utils import (
+    determine_ext,
+)
+
+
+class IGNIE(InfoExtractor):
+    """
+    Extractor for some of the IGN sites, like www.ign.com, es.ign.com de.ign.com.
+    Some videos of it.ign.com are also supported
+    """
+
+    _VALID_URL = r'https?://.+?\.ign\.com/(?P<type>videos|show_videos|articles|(?:[^/]*/feature))(/.+)?/(?P<name_or_id>.+)'
+    IE_NAME = u'ign.com'
+
+    _CONFIG_URL_TEMPLATE = 'http://www.ign.com/videos/configs/id/%s.config'
+    _DESCRIPTION_RE = [r'<span class="page-object-description">(.+?)</span>',
+                       r'id="my_show_video">.*?<p>(.*?)</p>',
+                       ]
+
+    _TESTS = [
+        {
+            u'url': u'http://www.ign.com/videos/2013/06/05/the-last-of-us-review',
+            u'file': u'8f862beef863986b2785559b9e1aa599.mp4',
+            u'md5': u'eac8bdc1890980122c3b66f14bdd02e9',
+            u'info_dict': {
+                u'title': u'The Last of Us Review',
+                u'description': u'md5:c8946d4260a4d43a00d5ae8ed998870c',
+            }
+        },
+        {
+            u'url': u'http://me.ign.com/en/feature/15775/100-little-things-in-gta-5-that-will-blow-your-mind',
+            u'playlist': [
+                {
+                    u'file': u'5ebbd138523268b93c9141af17bec937.mp4',
+                    u'info_dict': {
+                        u'title': u'GTA 5 Video Review',
+                        u'description': u'Rockstar drops the mic on this generation of games. Watch our review of the masterly Grand Theft Auto V.',
+                    },
+                },
+                {
+                    u'file': u'638672ee848ae4ff108df2a296418ee2.mp4',
+                    u'info_dict': {
+                        u'title': u'GTA 5\'s Twisted Beauty in Super Slow Motion',
+                        u'description': u'The twisted beauty of GTA 5 in stunning slow motion.',
+                    },
+                },
+            ],
+            u'params': {
+                u'skip_download': True,
+            },
+        },
+    ]
+
+    def _find_video_id(self, webpage):
+        res_id = [r'data-video-id="(.+?)"',
+                  r'<object id="vid_(.+?)"',
+                  r'<meta name="og:image" content=".*/(.+?)-(.+?)/.+.jpg"',
+                  ]
+        return self._search_regex(res_id, webpage, 'video id')
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        name_or_id = mobj.group('name_or_id')
+        page_type = mobj.group('type')
+        webpage = self._download_webpage(url, name_or_id)
+        if page_type == 'articles':
+            video_url = self._search_regex(r'var videoUrl = "(.+?)"', webpage, u'video url')
+            return self.url_result(video_url, ie='IGN')
+        elif page_type != 'video':
+            multiple_urls = re.findall(
+                '<param name="flashvars" value="[^"]*?url=(https?://www\.ign\.com/videos/.*?)["&]',
+                webpage)
+            if multiple_urls:
+                return [self.url_result(u, ie='IGN') for u in multiple_urls]
+
+        video_id = self._find_video_id(webpage)
+        result = self._get_video_info(video_id)
+        description = self._html_search_regex(self._DESCRIPTION_RE,
+                                              webpage, 'video description',
+                                              flags=re.DOTALL)
+        result['description'] = description
+        return result
+
+    def _get_video_info(self, video_id):
+        config_url = self._CONFIG_URL_TEMPLATE % video_id
+        config = json.loads(self._download_webpage(config_url, video_id,
+                            u'Downloading video info'))
+        media = config['playlist']['media']
+        video_url = media['url']
+
+        return {'id': media['metadata']['videoId'],
+                'url': video_url,
+                'ext': determine_ext(video_url),
+                'title': media['metadata']['title'],
+                'thumbnail': media['poster'][0]['url'].replace('{size}', 'grande'),
+                }
+
+
+class OneUPIE(IGNIE):
+    """Extractor for 1up.com, it uses the ign videos system."""
+
+    _VALID_URL = r'https?://gamevideos.1up.com/(?P<type>video)/id/(?P<name_or_id>.+)'
+    IE_NAME = '1up.com'
+
+    _DESCRIPTION_RE = r'<div id="vid_summary">(.+?)</div>'
+
+    _TEST = {
+        u'url': u'http://gamevideos.1up.com/video/id/34976',
+        u'file': u'34976.mp4',
+        u'md5': u'68a54ce4ebc772e4b71e3123d413163d',
+        u'info_dict': {
+            u'title': u'Sniper Elite V2 - Trailer',
+            u'description': u'md5:5d289b722f5a6d940ca3136e9dae89cf',
+        }
+    }
+
+    # Override IGN tests
+    _TESTS = []
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        id = mobj.group('name_or_id')
+        result = super(OneUPIE, self)._real_extract(url)
+        result['id'] = id
+        return result

youtube_dl/extractor/ina.py

@@ -5,7 +5,7 @@ from .common import InfoExtractor
 
 class InaIE(InfoExtractor):
     """Information Extractor for Ina.fr"""
-    _VALID_URL = r'(?:http://)?(?:www\.)?ina\.fr/video/(?P<id>I[0-9]+)/.*'
+    _VALID_URL = r'(?:http://)?(?:www\.)?ina\.fr/video/(?P<id>I?[A-F0-9]+)/.*'
     _TEST = {
         u'url': u'www.ina.fr/video/I12055569/francois-hollande-je-crois-que-c-est-clair-video.html',
         u'file': u'I12055569.mp4',

youtube_dl/extractor/instagram.py

@@ -5,12 +5,13 @@ from .common import InfoExtractor
 class InstagramIE(InfoExtractor):
     _VALID_URL = r'(?:http://)?instagram.com/p/(.*?)/'
     _TEST = {
-        u'url': u'http://instagram.com/p/aye83DjauH/#',
+        u'url': u'http://instagram.com/p/aye83DjauH/?foo=bar#abc',
         u'file': u'aye83DjauH.mp4',
         u'md5': u'0d2da106a9d2631273e192b372806516',
         u'info_dict': {
             u"uploader_id": u"naomipq", 
-            u"title": u"Video by naomipq"
+            u"title": u"Video by naomipq",
+            u'description': u'md5:1f17f0ab29bd6fe2bfad705f58de3cb8',
         }
     }
 
@@ -18,25 +19,17 @@ class InstagramIE(InfoExtractor):
         mobj = re.match(self._VALID_URL, url)
         video_id = mobj.group(1)
         webpage = self._download_webpage(url, video_id)
-        video_url = self._html_search_regex(
-            r'<meta property="og:video" content="(.+?)"',
-            webpage, u'video URL')
-        thumbnail_url = self._html_search_regex(
-            r'<meta property="og:image" content="(.+?)" />',
-            webpage, u'thumbnail URL', fatal=False)
-        html_title = self._html_search_regex(
-            r'<title>(.+?)</title>',
-            webpage, u'title', flags=re.DOTALL)
-        title = re.sub(u'(?: *\(Videos?\))? \u2022 Instagram$', '', html_title).strip()
-        uploader_id = self._html_search_regex(r'content="(.*?)\'s video on Instagram',
-            webpage, u'uploader name', fatal=False)
-        ext = 'mp4'
+        uploader_id = self._search_regex(r'"owner":{"username":"(.+?)"',
+            webpage, u'uploader id', fatal=False)
+        desc = self._search_regex(r'"caption":"(.*?)"', webpage, u'description',
+            fatal=False)
 
         return [{
             'id':        video_id,
-            'url':       video_url,
-            'ext':       ext,
-            'title':     title,
-            'thumbnail': thumbnail_url,
-            'uploader_id' : uploader_id
+            'url':       self._og_search_video_url(webpage),
+            'ext':       'mp4',
+            'title':     u'Video by %s' % uploader_id,
+            'thumbnail': self._og_search_thumbnail(webpage),
+            'uploader_id' : uploader_id,
+            'description': desc,
         }]

youtube_dl/extractor/internetvideoarchive.py

@@ -0,0 +1,87 @@
+import re
+import xml.etree.ElementTree
+
+from .common import InfoExtractor
+from ..utils import (
+    compat_urlparse,
+    compat_urllib_parse,
+    xpath_with_ns,
+    determine_ext,
+)
+
+
+class InternetVideoArchiveIE(InfoExtractor):
+    _VALID_URL = r'https?://video\.internetvideoarchive\.net/flash/players/.*?\?.*?publishedid.*?'
+
+    _TEST = {
+        u'url': u'http://video.internetvideoarchive.net/flash/players/flashconfiguration.aspx?customerid=69249&publishedid=452693&playerid=247',
+        u'file': u'452693.mp4',
+        u'info_dict': {
+            u'title': u'SKYFALL',
+            u'description': u'In SKYFALL, Bond\'s loyalty to M is tested as her past comes back to haunt her. As MI6 comes under attack, 007 must track down and destroy the threat, no matter how personal the cost.',
+            u'duration': 156,
+        },
+    }
+
+    @staticmethod
+    def _build_url(query):
+        return 'http://video.internetvideoarchive.net/flash/players/flashconfiguration.aspx?' + query
+
+    @staticmethod
+    def _clean_query(query):
+        NEEDED_ARGS = ['publishedid', 'customerid']
+        query_dic = compat_urlparse.parse_qs(query)
+        cleaned_dic = dict((k,v[0]) for (k,v) in query_dic.items() if k in NEEDED_ARGS)
+        # Other player ids return m3u8 urls
+        cleaned_dic['playerid'] = '247'
+        cleaned_dic['videokbrate'] = '100000'
+        return compat_urllib_parse.urlencode(cleaned_dic)
+
+    def _real_extract(self, url):
+        query = compat_urlparse.urlparse(url).query
+        query_dic = compat_urlparse.parse_qs(query)
+        video_id = query_dic['publishedid'][0]
+        url = self._build_url(query)
+
+        flashconfiguration_xml = self._download_webpage(url, video_id,
+            u'Downloading flash configuration')
+        flashconfiguration = xml.etree.ElementTree.fromstring(flashconfiguration_xml.encode('utf-8'))
+        file_url = flashconfiguration.find('file').text
+        file_url = file_url.replace('/playlist.aspx', '/mrssplaylist.aspx')
+        # Replace some of the parameters in the query to get the best quality
+        # and http links (no m3u8 manifests)
+        file_url = re.sub(r'(?<=\?)(.+)$',
+            lambda m: self._clean_query(m.group()),
+            file_url)
+        info_xml = self._download_webpage(file_url, video_id,
+            u'Downloading video info')
+        info = xml.etree.ElementTree.fromstring(info_xml.encode('utf-8'))
+        item = info.find('channel/item')
+
+        def _bp(p):
+            return xpath_with_ns(p,
+                {'media': 'http://search.yahoo.com/mrss/',
+                'jwplayer': 'http://developer.longtailvideo.com/trac/wiki/FlashFormats'})
+        formats = []
+        for content in item.findall(_bp('media:group/media:content')):
+            attr = content.attrib
+            f_url = attr['url']
+            formats.append({
+                'url': f_url,
+                'ext': determine_ext(f_url),
+                'width': int(attr['width']),
+                'bitrate': int(attr['bitrate']),
+            })
+        formats = sorted(formats, key=lambda f: f['bitrate'])
+
+        info = {
+            'id': video_id,
+            'title': item.find('title').text,
+            'formats': formats,
+            'thumbnail': item.find(_bp('media:thumbnail')).attrib['url'],
+            'description': item.find('description').text,
+            'duration': int(attr['duration']),
+        }
+        # TODO: Remove when #980 has been merged
+        info.update(formats[-1])
+        return info

youtube_dl/extractor/jeuxvideo.py

@@ -0,0 +1,52 @@
+# coding: utf-8
+
+import json
+import re
+import xml.etree.ElementTree
+
+from .common import InfoExtractor
+
+
+class JeuxVideoIE(InfoExtractor):
+    _VALID_URL = r'http://.*?\.jeuxvideo\.com/.*/(.*?)-\d+\.htm'
+
+    _TEST = {
+        u'url': u'http://www.jeuxvideo.com/reportages-videos-jeux/0004/00046170/tearaway-playstation-vita-gc-2013-tearaway-nous-presente-ses-papiers-d-identite-00115182.htm',
+        u'file': u'5182.mp4',
+        u'md5': u'046e491afb32a8aaac1f44dd4ddd54ee',
+        u'info_dict': {
+            u'title': u'GC 2013 : Tearaway nous présente ses papiers d\'identité',
+            u'description': u'Lorsque les développeurs de LittleBigPlanet proposent un nouveau titre, on ne peut que s\'attendre à un résultat original et fort attrayant.\n',
+        },
+    }
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        title = re.match(self._VALID_URL, url).group(1)
+        webpage = self._download_webpage(url, title)
+        xml_link = self._html_search_regex(
+            r'<param name="flashvars" value="config=(.*?)" />',
+            webpage, u'config URL')
+        
+        video_id = self._search_regex(
+            r'http://www\.jeuxvideo\.com/config/\w+/\d+/(.*?)/\d+_player\.xml',
+            xml_link, u'video ID')
+
+        xml_config = self._download_webpage(
+            xml_link, title, u'Downloading XML config')
+        config = xml.etree.ElementTree.fromstring(xml_config.encode('utf-8'))
+        info_json = self._search_regex(
+            r'(?sm)<format\.json>(.*?)</format\.json>',
+            xml_config, u'JSON information')
+        info = json.loads(info_json)['versions'][0]
+        
+        video_url = 'http://video720.jeuxvideo.com/' + info['file']
+
+        return {
+            'id': video_id,
+            'title': config.find('titre_video').text,
+            'ext': 'mp4',
+            'url': video_url,
+            'description': self._og_search_description(webpage),
+            'thumbnail': config.find('image').text,
+        }

youtube_dl/extractor/kankan.py

@@ -0,0 +1,39 @@
+import re
+
+from .common import InfoExtractor
+from ..utils import determine_ext
+
+
+class KankanIE(InfoExtractor):
+    _VALID_URL = r'https?://(?:.*?\.)?kankan\.com/.+?/(?P<id>\d+)\.shtml'
+    
+    _TEST = {
+        u'url': u'http://yinyue.kankan.com/vod/48/48863.shtml',
+        u'file': u'48863.flv',
+        u'md5': u'29aca1e47ae68fc28804aca89f29507e',
+        u'info_dict': {
+            u'title': u'Ready To Go',
+        },
+    }
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        video_id = mobj.group('id')
+        webpage = self._download_webpage(url, video_id)
+
+        title = self._search_regex(r'(?:G_TITLE=|G_MOVIE_TITLE = )[\'"](.+?)[\'"]', webpage, u'video title')
+        surls = re.search(r'surls:\[\'.+?\'\]|lurl:\'.+?\.flv\'', webpage).group(0)
+        gcids = re.findall(r"http://.+?/.+?/(.+?)/", surls)
+        gcid = gcids[-1]
+
+        video_info_page = self._download_webpage('http://p2s.cl.kankan.com/getCdnresource_flv?gcid=%s' % gcid,
+                                                 video_id, u'Downloading video url info')
+        ip = self._search_regex(r'ip:"(.+?)"', video_info_page, u'video url ip')
+        path = self._search_regex(r'path:"(.+?)"', video_info_page, u'video url path')
+        video_url = 'http://%s%s' % (ip, path)
+
+        return {'id': video_id,
+                'title': title,
+                'url': video_url,
+                'ext': determine_ext(video_url),
+                }

youtube_dl/extractor/keek.py

@@ -4,10 +4,10 @@ from .common import InfoExtractor
 
 
 class KeekIE(InfoExtractor):
-    _VALID_URL = r'http://(?:www\.)?keek\.com/(?:!|\w+/keeks/)(?P<videoID>\w+)'
+    _VALID_URL = r'https?://(?:www\.)?keek\.com/(?:!|\w+/keeks/)(?P<videoID>\w+)'
     IE_NAME = u'keek'
     _TEST = {
-        u'url': u'http://www.keek.com/ytdl/keeks/NODfbab',
+        u'url': u'https://www.keek.com/ytdl/keeks/NODfbab',
         u'file': u'NODfbab.mp4',
         u'md5': u'9b0636f8c0f7614afa4ea5e4c6e57e83',
         u'info_dict': {
@@ -24,8 +24,7 @@ class KeekIE(InfoExtractor):
         thumbnail = u'http://cdn.keek.com/keek/thumbnail/%s/w100/h75' % video_id
         webpage = self._download_webpage(url, video_id)
 
-        video_title = self._html_search_regex(r'<meta property="og:title" content="(?P<title>.*?)"',
-            webpage, u'title')
+        video_title = self._og_search_title(webpage)
 
         uploader = self._html_search_regex(r'<div class="user-name-and-bio">[\S\s]+?<h2>(?P<uploader>.+?)</h2>',
             webpage, u'uploader', fatal=False)

youtube_dl/extractor/kickstarter.py

@@ -0,0 +1,37 @@
+import re
+
+from .common import InfoExtractor
+
+
+class KickStarterIE(InfoExtractor):
+    _VALID_URL = r'https?://www\.kickstarter\.com/projects/(?P<id>\d*)/.*'
+    _TEST = {
+        u"url": u"https://www.kickstarter.com/projects/1404461844/intersection-the-story-of-josh-grant?ref=home_location",
+        u"file": u"1404461844.mp4",
+        u"md5": u"c81addca81327ffa66c642b5d8b08cab",
+        u"info_dict": {
+            u"title": u"Intersection: The Story of Josh Grant by Kyle Cowling",
+        },
+    }
+
+    def _real_extract(self, url):
+        m = re.match(self._VALID_URL, url)
+        video_id = m.group('id')
+        webpage_src = self._download_webpage(url, video_id)
+
+        video_url = self._search_regex(r'data-video="(.*?)">',
+            webpage_src, u'video URL')
+        if 'mp4' in video_url:
+            ext = 'mp4'
+        else:
+            ext = 'flv'
+        video_title = self._html_search_regex(r"<title>(.*?)</title>",
+            webpage_src, u'title').rpartition(u'\u2014 Kickstarter')[0].strip()
+
+        results = [{
+                    'id': video_id,
+                    'url': video_url,
+                    'title': video_title,
+                    'ext': ext,
+                    }]
+        return results

youtube_dl/extractor/liveleak.py

@@ -33,11 +33,9 @@ class LiveLeakIE(InfoExtractor):
         video_url = self._search_regex(r'file: "(.*?)",',
             webpage, u'video URL')
 
-        video_title = self._html_search_regex(r'<meta property="og:title" content="(?P<title>.*?)"',
-            webpage, u'title').replace('LiveLeak.com -', '').strip()
+        video_title = self._og_search_title(webpage).replace('LiveLeak.com -', '').strip()
 
-        video_description = self._html_search_regex(r'<meta property="og:description" content="(?P<desc>.*?)"',
-            webpage, u'description', fatal=False)
+        video_description = self._og_search_description(webpage)
 
         video_uploader = self._html_search_regex(r'By:.*?(\w+)</a>',
             webpage, u'uploader', fatal=False)

youtube_dl/extractor/livestream.py

@@ -0,0 +1,60 @@
+import re
+import json
+
+from .common import InfoExtractor
+from ..utils import (
+    compat_urllib_parse_urlparse,
+    compat_urlparse,
+    get_meta_content,
+    ExtractorError,
+)
+
+
+class LivestreamIE(InfoExtractor):
+    _VALID_URL = r'http://new.livestream.com/.*?/(?P<event_name>.*?)(/videos/(?P<id>\d+))?/?$'
+    _TEST = {
+        u'url': u'http://new.livestream.com/CoheedandCambria/WebsterHall/videos/4719370',
+        u'file': u'4719370.mp4',
+        u'md5': u'0d2186e3187d185a04b3cdd02b828836',
+        u'info_dict': {
+            u'title': u'Live from Webster Hall NYC',
+            u'upload_date': u'20121012',
+        }
+    }
+
+    def _extract_video_info(self, video_data):
+        video_url = video_data.get('progressive_url_hd') or video_data.get('progressive_url')
+        return {'id': video_data['id'],
+                'url': video_url,
+                'ext': 'mp4',
+                'title': video_data['caption'],
+                'thumbnail': video_data['thumbnail_url'],
+                'upload_date': video_data['updated_at'].replace('-','')[:8],
+                }
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        video_id = mobj.group('id')
+        event_name = mobj.group('event_name')
+        webpage = self._download_webpage(url, video_id or event_name)
+
+        if video_id is None:
+            # This is an event page:
+            player = get_meta_content('twitter:player', webpage)
+            if player is None:
+                raise ExtractorError('Couldn\'t extract event api url')
+            api_url = player.replace('/player', '')
+            api_url = re.sub(r'^(https?://)(new\.)', r'\1api.\2', api_url)
+            info = json.loads(self._download_webpage(api_url, event_name,
+                                                     u'Downloading event info'))
+            videos = [self._extract_video_info(video_data['data'])
+                for video_data in info['feed']['data'] if video_data['type'] == u'video']
+            return self.playlist_result(videos, info['id'], info['full_name'])
+        else:
+            og_video = self._og_search_video_url(webpage, name=u'player url')
+            query_str = compat_urllib_parse_urlparse(og_video).query
+            query = compat_urlparse.parse_qs(query_str)
+            api_url = query['play_url'][0].replace('.smil', '')
+            info = json.loads(self._download_webpage(api_url, video_id,
+                                                     u'Downloading video info'))
+            return self._extract_video_info(info)

youtube_dl/extractor/metacafe.py

@@ -9,7 +9,7 @@ from ..utils import (
     compat_urllib_parse,
     compat_urllib_request,
     compat_str,
-
+    determine_ext,
     ExtractorError,
 )
 
@@ -20,7 +20,7 @@ class MetacafeIE(InfoExtractor):
     _DISCLAIMER = 'http://www.metacafe.com/family_filter/'
     _FILTER_POST = 'http://www.metacafe.com/f/index.php?inputType=filter&controllerGroup=user'
     IE_NAME = u'metacafe'
-    _TEST = {
+    _TESTS = [{
         u"add_ie": ["Youtube"],
         u"url":  u"http://metacafe.com/watch/yt-_aUehQsCQtM/the_electric_company_short_i_pbs_kids_go/",
         u"file":  u"_aUehQsCQtM.flv",
@@ -31,7 +31,16 @@ class MetacafeIE(InfoExtractor):
             u"uploader": u"PBS",
             u"uploader_id": u"PBS"
         }
-    }
+    },
+    {
+        u"url": u"http://www.metacafe.com/watch/an-dVVXnuY7Jh77J/the_andromeda_strain_1971_stop_the_bomb_part_3/",
+        u"file": u"an-dVVXnuY7Jh77J.mp4",
+        u"info_dict": {
+            u"title": u"The Andromeda Strain (1971): Stop the Bomb Part 3",
+            u"uploader": u"anyclip",
+            u"description": u"md5:38c711dd98f5bb87acf973d573442e67"
+        }
+    }]
 
 
     def report_disclaimer(self):
@@ -73,14 +82,16 @@ class MetacafeIE(InfoExtractor):
             return [self.url_result('http://www.youtube.com/watch?v=%s' % mobj2.group(1), 'Youtube')]
 
         # Retrieve video webpage to extract further information
-        webpage = self._download_webpage('http://www.metacafe.com/watch/%s/' % video_id, video_id)
+        req = compat_urllib_request.Request('http://www.metacafe.com/watch/%s/' % video_id)
+        req.headers['Cookie'] = 'flashVersion=0;'
+        webpage = self._download_webpage(req, video_id)
 
         # Extract URL, uploader and title from webpage
         self.report_extraction(video_id)
         mobj = re.search(r'(?m)&mediaURL=([^&]+)', webpage)
         if mobj is not None:
             mediaURL = compat_urllib_parse.unquote(mobj.group(1))
-            video_extension = mediaURL[-3:]
+            video_ext = mediaURL[-3:]
 
             # Extract gdaKey if available
             mobj = re.search(r'(?m)&gdaKey=(.*?)&', webpage)
@@ -90,34 +101,37 @@ class MetacafeIE(InfoExtractor):
                 gdaKey = mobj.group(1)
                 video_url = '%s?__gda__=%s' % (mediaURL, gdaKey)
         else:
-            mobj = re.search(r' name="flashvars" value="(.*?)"', webpage)
-            if mobj is None:
-                raise ExtractorError(u'Unable to extract media URL')
-            vardict = compat_parse_qs(mobj.group(1))
-            if 'mediaData' not in vardict:
-                raise ExtractorError(u'Unable to extract media URL')
-            mobj = re.search(r'"mediaURL":"(?P<mediaURL>http.*?)",(.*?)"key":"(?P<key>.*?)"', vardict['mediaData'][0])
-            if mobj is None:
-                raise ExtractorError(u'Unable to extract media URL')
-            mediaURL = mobj.group('mediaURL').replace('\\/', '/')
-            video_extension = mediaURL[-3:]
-            video_url = '%s?__gda__=%s' % (mediaURL, mobj.group('key'))
+            mobj = re.search(r'<video src="([^"]+)"', webpage)
+            if mobj:
+                video_url = mobj.group(1)
+                video_ext = 'mp4'
+            else:
+                mobj = re.search(r' name="flashvars" value="(.*?)"', webpage)
+                if mobj is None:
+                    raise ExtractorError(u'Unable to extract media URL')
+                vardict = compat_parse_qs(mobj.group(1))
+                if 'mediaData' not in vardict:
+                    raise ExtractorError(u'Unable to extract media URL')
+                mobj = re.search(r'"mediaURL":"(?P<mediaURL>http.*?)",(.*?)"key":"(?P<key>.*?)"', vardict['mediaData'][0])
+                if mobj is None:
+                    raise ExtractorError(u'Unable to extract media URL')
+                mediaURL = mobj.group('mediaURL').replace('\\/', '/')
+                video_url = '%s?__gda__=%s' % (mediaURL, mobj.group('key'))
+                video_ext = determine_ext(video_url)
 
-        mobj = re.search(r'(?im)<title>(.*) - Video</title>', webpage)
-        if mobj is None:
-            raise ExtractorError(u'Unable to extract title')
-        video_title = mobj.group(1).decode('utf-8')
+        video_title = self._html_search_regex(r'(?im)<title>(.*) - Video</title>', webpage, u'title')
+        description = self._og_search_description(webpage)
+        video_uploader = self._html_search_regex(
+                r'submitter=(.*?);|googletag\.pubads\(\)\.setTargeting\("(?:channel|submiter)","([^"]+)"\);',
+                webpage, u'uploader nickname', fatal=False)
 
-        mobj = re.search(r'submitter=(.*?);', webpage)
-        if mobj is None:
-            raise ExtractorError(u'Unable to extract uploader nickname')
-        video_uploader = mobj.group(1)
-
-        return [{
-            'id':       video_id.decode('utf-8'),
-            'url':      video_url.decode('utf-8'),
-            'uploader': video_uploader.decode('utf-8'),
+        return {
+            '_type':    'video',
+            'id':       video_id,
+            'url':      video_url,
+            'description': description,
+            'uploader': video_uploader,
             'upload_date':  None,
             'title':    video_title,
-            'ext':      video_extension.decode('utf-8'),
-        }]
+            'ext':      video_ext,
+        }

youtube_dl/extractor/metacritic.py

@@ -0,0 +1,55 @@
+import re
+import xml.etree.ElementTree
+import operator
+
+from .common import InfoExtractor
+
+
+class MetacriticIE(InfoExtractor):
+    _VALID_URL = r'https?://www\.metacritic\.com/.+?/trailers/(?P<id>\d+)'
+
+    _TEST = {
+        u'url': u'http://www.metacritic.com/game/playstation-4/infamous-second-son/trailers/3698222',
+        u'file': u'3698222.mp4',
+        u'info_dict': {
+            u'title': u'inFamous: Second Son - inSide Sucker Punch: Smoke & Mirrors',
+            u'description': u'Take a peak behind-the-scenes to see how Sucker Punch brings smoke into the universe of inFAMOUS Second Son on the PS4.',
+            u'duration': 221,
+        },
+    }
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        video_id = mobj.group('id')
+        webpage = self._download_webpage(url, video_id)
+        # The xml is not well formatted, there are raw '&'
+        info_xml = self._download_webpage('http://www.metacritic.com/video_data?video=' + video_id,
+            video_id, u'Downloading info xml').replace('&', '&amp;')
+        info = xml.etree.ElementTree.fromstring(info_xml.encode('utf-8'))
+
+        clip = next(c for c in info.findall('playList/clip') if c.find('id').text == video_id)
+        formats = []
+        for videoFile in clip.findall('httpURI/videoFile'):
+            rate_str = videoFile.find('rate').text
+            video_url = videoFile.find('filePath').text
+            formats.append({
+                'url': video_url,
+                'ext': 'mp4',
+                'format_id': rate_str,
+                'rate': int(rate_str),
+            })
+        formats.sort(key=operator.itemgetter('rate'))
+
+        description = self._html_search_regex(r'<b>Description:</b>(.*?)</p>',
+            webpage, u'description', flags=re.DOTALL)
+
+        info = {
+            'id': video_id,
+            'title': clip.find('title').text,
+            'formats': formats,
+            'description': description,
+            'duration': int(clip.find('duration').text),
+        }
+        # TODO: Remove when #980 has been merged
+        info.update(formats[-1])
+        return info

youtube_dl/extractor/mit.py

@@ -0,0 +1,74 @@
+import re
+import json
+
+from .common import InfoExtractor
+from ..utils import (
+    clean_html,
+    get_element_by_id,
+)
+
+
+class TechTVMITIE(InfoExtractor):
+    IE_NAME = u'techtv.mit.edu'
+    _VALID_URL = r'https?://techtv\.mit\.edu/(videos|embeds)/(?P<id>\d+)'
+
+    _TEST = {
+        u'url': u'http://techtv.mit.edu/videos/25418-mit-dna-learning-center-set',
+        u'file': u'25418.mp4',
+        u'md5': u'1f8cb3e170d41fd74add04d3c9330e5f',
+        u'info_dict': {
+            u'title': u'MIT DNA Learning Center Set',
+            u'description': u'md5:82313335e8a8a3f243351ba55bc1b474',
+        },
+    }
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        video_id = mobj.group('id')
+        raw_page = self._download_webpage(
+            'http://techtv.mit.edu/videos/%s' % video_id, video_id)
+        clean_page = re.compile(u'<!--.*?-->', re.S).sub(u'', raw_page)
+
+        base_url = self._search_regex(r'ipadUrl: \'(.+?cloudfront.net/)',
+            raw_page, u'base url')
+        formats_json = self._search_regex(r'bitrates: (\[.+?\])', raw_page,
+            u'video formats')
+        formats = json.loads(formats_json)
+        formats = sorted(formats, key=lambda f: f['bitrate'])
+
+        title = get_element_by_id('edit-title', clean_page)
+        description = clean_html(get_element_by_id('edit-description', clean_page))
+        thumbnail = self._search_regex(r'playlist:.*?url: \'(.+?)\'',
+            raw_page, u'thumbnail', flags=re.DOTALL)
+
+        return {'id': video_id,
+                'title': title,
+                'url': base_url + formats[-1]['url'].replace('mp4:', ''),
+                'ext': 'mp4',
+                'description': description,
+                'thumbnail': thumbnail,
+                }
+
+
+class MITIE(TechTVMITIE):
+    IE_NAME = u'video.mit.edu'
+    _VALID_URL = r'https?://video\.mit\.edu/watch/(?P<title>[^/]+)'
+
+    _TEST = {
+        u'url': u'http://video.mit.edu/watch/the-government-is-profiling-you-13222/',
+        u'file': u'21783.mp4',
+        u'md5': u'7db01d5ccc1895fc5010e9c9e13648da',
+        u'info_dict': {
+            u'title': u'The Government is Profiling You',
+            u'description': u'md5:ad5795fe1e1623b73620dbfd47df9afd',
+        },
+    }
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        page_title = mobj.group('title')
+        webpage = self._download_webpage(url, page_title)
+        self.to_screen('%s: Extracting %s url' % (page_title, TechTVMITIE.IE_NAME))
+        embed_url = self._search_regex(r'<iframe .*?src="(.+?)"', webpage,
+            u'embed url')
+        return self.url_result(embed_url, ie='TechTVMIT')

youtube_dl/extractor/mixcloud.py

@@ -5,34 +5,27 @@ import socket
 from .common import InfoExtractor
 from ..utils import (
     compat_http_client,
-    compat_str,
     compat_urllib_error,
     compat_urllib_request,
-
-    ExtractorError,
+    unified_strdate,
 )
 
 
 class MixcloudIE(InfoExtractor):
-    _WORKING = False # New API, but it seems good http://www.mixcloud.com/developers/documentation/
     _VALID_URL = r'^(?:https?://)?(?:www\.)?mixcloud\.com/([\w\d-]+)/([\w\d-]+)'
     IE_NAME = u'mixcloud'
 
-    def report_download_json(self, file_id):
-        """Report JSON download."""
-        self.to_screen(u'Downloading json')
-
-    def get_urls(self, jsonData, fmt, bitrate='best'):
-        """Get urls from 'audio_formats' section in json"""
-        try:
-            bitrate_list = jsonData[fmt]
-            if bitrate is None or bitrate == 'best' or bitrate not in bitrate_list:
-                bitrate = max(bitrate_list) # select highest
-
-            url_list = jsonData[fmt][bitrate]
-        except TypeError: # we have no bitrate info.
-            url_list = jsonData[fmt]
-        return url_list
+    _TEST = {
+        u'url': u'http://www.mixcloud.com/dholbach/cryptkeeper/',
+        u'file': u'dholbach-cryptkeeper.mp3',
+        u'info_dict': {
+            u'title': u'Cryptkeeper',
+            u'description': u'After quite a long silence from myself, finally another Drum\'n\'Bass mix with my favourite current dance floor bangers.',
+            u'uploader': u'Daniel Holbach',
+            u'uploader_id': u'dholbach',
+            u'upload_date': u'20111115',
+        },
+    }
 
     def check_urls(self, url_list):
         """Returns 1st active url from list"""
@@ -45,71 +38,32 @@ class MixcloudIE(InfoExtractor):
 
         return None
 
-    def _print_formats(self, formats):
-        print('Available formats:')
-        for fmt in formats.keys():
-            for b in formats[fmt]:
-                try:
-                    ext = formats[fmt][b][0]
-                    print('%s\t%s\t[%s]' % (fmt, b, ext.split('.')[-1]))
-                except TypeError: # we have no bitrate info
-                    ext = formats[fmt][0]
-                    print('%s\t%s\t[%s]' % (fmt, '??', ext.split('.')[-1]))
-                    break
-
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
-        if mobj is None:
-            raise ExtractorError(u'Invalid URL: %s' % url)
-        # extract uploader & filename from url
-        uploader = mobj.group(1).decode('utf-8')
-        file_id = uploader + "-" + mobj.group(2).decode('utf-8')
 
-        # construct API request
-        file_url = 'http://www.mixcloud.com/api/1/cloudcast/' + '/'.join(url.split('/')[-3:-1]) + '.json'
-        # retrieve .json file with links to files
-        request = compat_urllib_request.Request(file_url)
-        try:
-            self.report_download_json(file_url)
-            jsonData = compat_urllib_request.urlopen(request).read()
-        except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
-            raise ExtractorError(u'Unable to retrieve file: %s' % compat_str(err))
+        uploader = mobj.group(1)
+        cloudcast_name = mobj.group(2)
+        track_id = '-'.join((uploader, cloudcast_name))
+        api_url = 'http://api.mixcloud.com/%s/%s/' % (uploader, cloudcast_name)
+        webpage = self._download_webpage(url, track_id)
+        json_data = self._download_webpage(api_url, track_id,
+            u'Downloading cloudcast info')
+        info = json.loads(json_data)
 
-        # parse JSON
-        json_data = json.loads(jsonData)
-        player_url = json_data['player_swf_url']
-        formats = dict(json_data['audio_formats'])
+        preview_url = self._search_regex(r'data-preview-url="(.+?)"', webpage, u'preview url')
+        song_url = preview_url.replace('/previews/', '/cloudcasts/originals/')
+        template_url = re.sub(r'(stream\d*)', 'stream%d', song_url)
+        final_song_url = self.check_urls(template_url % i for i in range(30))
 
-        req_format = self._downloader.params.get('format', None)
-
-        if self._downloader.params.get('listformats', None):
-            self._print_formats(formats)
-            return
-
-        if req_format is None or req_format == 'best':
-            for format_param in formats.keys():
-                url_list = self.get_urls(formats, format_param)
-                # check urls
-                file_url = self.check_urls(url_list)
-                if file_url is not None:
-                    break # got it!
-        else:
-            if req_format not in formats:
-                raise ExtractorError(u'Format is not available')
-
-            url_list = self.get_urls(formats, req_format)
-            file_url = self.check_urls(url_list)
-            format_param = req_format
-
-        return [{
-            'id': file_id.decode('utf-8'),
-            'url': file_url.decode('utf-8'),
-            'uploader': uploader.decode('utf-8'),
-            'upload_date': None,
-            'title': json_data['name'],
-            'ext': file_url.split('.')[-1].decode('utf-8'),
-            'format': (format_param is None and u'NA' or format_param.decode('utf-8')),
-            'thumbnail': json_data['thumbnail_url'],
-            'description': json_data['description'],
-            'player_url': player_url.decode('utf-8'),
-        }]
+        return {
+            'id': track_id,
+            'title': info['name'],
+            'url': final_song_url,
+            'ext': 'mp3',
+            'description': info['description'],
+            'thumbnail': info['pictures'].get('extra_large'),
+            'uploader': info['user']['name'],
+            'uploader_id': info['user']['username'],
+            'upload_date': unified_strdate(info['created_time']),
+            'view_count': info['play_count'],
+        }

youtube_dl/extractor/mtv.py

@@ -1,28 +1,119 @@
 import re
-import socket
 import xml.etree.ElementTree
 
 from .common import InfoExtractor
 from ..utils import (
-    compat_http_client,
-    compat_str,
-    compat_urllib_error,
-    compat_urllib_request,
-
+    compat_urllib_parse,
     ExtractorError,
 )
 
+def _media_xml_tag(tag):
+    return '{http://search.yahoo.com/mrss/}%s' % tag
 
 class MTVIE(InfoExtractor):
-    _VALID_URL = r'^(?P<proto>https?://)?(?:www\.)?mtv\.com/videos/[^/]+/(?P<videoid>[0-9]+)/[^/]+$'
-    _WORKING = False
+    _VALID_URL = r'^https?://(?:www\.)?mtv\.com/videos/.+?/(?P<videoid>[0-9]+)/[^/]+$'
+
+    _FEED_URL = 'http://www.mtv.com/player/embed/AS3/rss/'
+
+    _TESTS = [
+        {
+            u'url': u'http://www.mtv.com/videos/misc/853555/ours-vh1-storytellers.jhtml',
+            u'file': u'853555.mp4',
+            u'md5': u'850f3f143316b1e71fa56a4edfd6e0f8',
+            u'info_dict': {
+                u'title': u'Taylor Swift - "Ours (VH1 Storytellers)"',
+                u'description': u'Album: Taylor Swift performs "Ours" for VH1 Storytellers at Harvey Mudd College.',
+            },
+        },
+        {
+            u'url': u'http://www.mtv.com/videos/taylor-swift/916187/everything-has-changed-ft-ed-sheeran.jhtml',
+            u'file': u'USCJY1331283.mp4',
+            u'md5': u'73b4e7fcadd88929292fe52c3ced8caf',
+            u'info_dict': {
+                u'title': u'Everything Has Changed',
+                u'upload_date': u'20130606',
+                u'uploader': u'Taylor Swift',
+            },
+            u'skip': u'VEVO is only available in some countries',
+        },
+    ]
+
+    @staticmethod
+    def _id_from_uri(uri):
+        return uri.split(':')[-1]
+
+    # This was originally implemented for ComedyCentral, but it also works here
+    @staticmethod
+    def _transform_rtmp_url(rtmp_video_url):
+        m = re.match(r'^rtmpe?://.*?/(?P<finalid>gsp\..+?/.*)$', rtmp_video_url)
+        if not m:
+            raise ExtractorError(u'Cannot transform RTMP url')
+        base = 'http://mtvnmobile.vo.llnwd.net/kip0/_pxn=1+_pxI0=Ripod-h264+_pxL0=undefined+_pxM0=+_pxK=18639+_pxE=mp4/44620/mtvnorigin/'
+        return base + m.group('finalid')
+
+    def _get_thumbnail_url(self, uri, itemdoc):
+        return 'http://mtv.mtvnimages.com/uri/' + uri
+
+    def _extract_video_formats(self, metadataXml):
+        if '/error_country_block.swf' in metadataXml:
+            raise ExtractorError(u'This video is not available from your country.', expected=True)
+        mdoc = xml.etree.ElementTree.fromstring(metadataXml.encode('utf-8'))
+        renditions = mdoc.findall('.//rendition')
+
+        formats = []
+        for rendition in mdoc.findall('.//rendition'):
+            try:
+                _, _, ext = rendition.attrib['type'].partition('/')
+                rtmp_video_url = rendition.find('./src').text
+                formats.append({'ext': ext,
+                                'url': self._transform_rtmp_url(rtmp_video_url),
+                                'format_id': rendition.get('bitrate'),
+                                'width': int(rendition.get('width')),
+                                'height': int(rendition.get('height')),
+                                })
+            except (KeyError, TypeError):
+                raise ExtractorError('Invalid rendition field.')
+        return formats
+
+    def _get_video_info(self, itemdoc):
+        uri = itemdoc.find('guid').text
+        video_id = self._id_from_uri(uri)
+        self.report_extraction(video_id)
+        mediagen_url = itemdoc.find('%s/%s' % (_media_xml_tag('group'), _media_xml_tag('content'))).attrib['url']
+        if 'acceptMethods' not in mediagen_url:
+            mediagen_url += '&acceptMethods=fms'
+        mediagen_page = self._download_webpage(mediagen_url, video_id,
+                                               u'Downloading video urls')
+
+        description_node = itemdoc.find('description')
+        if description_node is not None:
+            description = description_node.text.strip()
+        else:
+            description = None
+
+        info = {
+            'title': itemdoc.find('title').text,
+            'formats': self._extract_video_formats(mediagen_page),
+            'id': video_id,
+            'thumbnail': self._get_thumbnail_url(uri, itemdoc),
+            'description': description,
+        }
+
+        # TODO: Remove when #980 has been merged
+        info.update(info['formats'][-1])
+
+        return info
+
+    def _get_videos_info(self, uri):
+        video_id = self._id_from_uri(uri)
+        data = compat_urllib_parse.urlencode({'uri': uri})
+        infoXml = self._download_webpage(self._FEED_URL +'?' + data, video_id,
+                                         u'Downloading info')
+        idoc = xml.etree.ElementTree.fromstring(infoXml.encode('utf-8'))
+        return [self._get_video_info(item) for item in idoc.findall('.//item')]
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
-        if mobj is None:
-            raise ExtractorError(u'Invalid URL: %s' % url)
-        if not mobj.group('proto'):
-            url = 'http://' + url
         video_id = mobj.group('videoid')
 
         webpage = self._download_webpage(url, video_id)
@@ -35,46 +126,5 @@ class MTVIE(InfoExtractor):
             self.to_screen(u'Vevo video detected: %s' % vevo_id)
             return self.url_result('vevo:%s' % vevo_id, ie='Vevo')
 
-        #song_name = self._html_search_regex(r'<meta name="mtv_vt" content="([^"]+)"/>',
-        #    webpage, u'song name', fatal=False)
-
-        video_title = self._html_search_regex(r'<meta name="mtv_an" content="([^"]+)"/>',
-            webpage, u'title')
-
-        mtvn_uri = self._html_search_regex(r'<meta name="mtvn_uri" content="([^"]+)"/>',
-            webpage, u'mtvn_uri', fatal=False)
-
-        content_id = self._search_regex(r'MTVN.Player.defaultPlaylistId = ([0-9]+);',
-            webpage, u'content id', fatal=False)
-
-        videogen_url = 'http://www.mtv.com/player/includes/mediaGen.jhtml?uri=' + mtvn_uri + '&id=' + content_id + '&vid=' + video_id + '&ref=www.mtvn.com&viewUri=' + mtvn_uri
-        self.report_extraction(video_id)
-        request = compat_urllib_request.Request(videogen_url)
-        try:
-            metadataXml = compat_urllib_request.urlopen(request).read()
-        except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
-            raise ExtractorError(u'Unable to download video metadata: %s' % compat_str(err))
-
-        mdoc = xml.etree.ElementTree.fromstring(metadataXml)
-        renditions = mdoc.findall('.//rendition')
-
-        # For now, always pick the highest quality.
-        rendition = renditions[-1]
-
-        try:
-            _,_,ext = rendition.attrib['type'].partition('/')
-            format = ext + '-' + rendition.attrib['width'] + 'x' + rendition.attrib['height'] + '_' + rendition.attrib['bitrate']
-            video_url = rendition.find('./src').text
-        except KeyError:
-            raise ExtractorError('Invalid rendition field.')
-
-        info = {
-            'id': video_id,
-            'url': video_url,
-            'upload_date': None,
-            'title': video_title,
-            'ext': ext,
-            'format': format,
-        }
-
-        return [info]
+        uri = self._html_search_regex(r'/uri/(.*?)\?', webpage, u'uri')
+        return self._get_videos_info(uri)

youtube_dl/extractor/muzu.py

@@ -0,0 +1,64 @@
+import re
+import json
+
+from .common import InfoExtractor
+from ..utils import (
+    compat_urllib_parse,
+    determine_ext,
+)
+
+
+class MuzuTVIE(InfoExtractor):
+    _VALID_URL = r'https?://www.muzu.tv/(.+?)/(.+?)/(?P<id>\d+)'
+    IE_NAME = u'muzu.tv'
+
+    _TEST = {
+        u'url': u'http://www.muzu.tv/defected/marcashken-featuring-sos-cat-walk-original-mix-music-video/1981454/',
+        u'file': u'1981454.mp4',
+        u'md5': u'98f8b2c7bc50578d6a0364fff2bfb000',
+        u'info_dict': {
+            u'title': u'Cat Walk (Original Mix)',
+            u'description': u'md5:90e868994de201b2570e4e5854e19420',
+            u'uploader': u'MarcAshken featuring SOS',
+        },
+    }
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        video_id = mobj.group('id')
+
+        info_data = compat_urllib_parse.urlencode({'format': 'json',
+                                                   'url': url,
+                                                   })
+        video_info_page = self._download_webpage('http://www.muzu.tv/api/oembed/?%s' % info_data,
+                                                 video_id, u'Downloading video info')
+        info = json.loads(video_info_page)
+
+        player_info_page = self._download_webpage('http://player.muzu.tv/player/playerInit?ai=%s' % video_id,
+                                                  video_id, u'Downloading player info')
+        video_info = json.loads(player_info_page)['videos'][0]
+        for quality in ['1080' , '720', '480', '360']:
+            if video_info.get('v%s' % quality):
+                break
+
+        data = compat_urllib_parse.urlencode({'ai': video_id,
+                                              # Even if each time you watch a video the hash changes,
+                                              # it seems to work for different videos, and it will work
+                                              # even if you use any non empty string as a hash
+                                              'viewhash': 'VBNff6djeV4HV5TRPW5kOHub2k',
+                                              'device': 'web',
+                                              'qv': quality,
+                                              })
+        video_url_page = self._download_webpage('http://player.muzu.tv/player/requestVideo?%s' % data,
+                                                video_id, u'Downloading video url')
+        video_url_info = json.loads(video_url_page)
+        video_url = video_url_info['url']
+
+        return {'id': video_id,
+                'title': info['title'],
+                'url': video_url,
+                'ext': determine_ext(video_url),
+                'thumbnail': info['thumbnail_url'],
+                'description': info['description'],
+                'uploader': info['author_name'],
+                }

youtube_dl/extractor/myvideo.py

@@ -2,11 +2,13 @@ import binascii
 import base64
 import hashlib
 import re
+import json
 
 from .common import InfoExtractor
 from ..utils import (
     compat_ord,
     compat_urllib_parse,
+    compat_urllib_request,
 
     ExtractorError,
 )
@@ -16,7 +18,7 @@ from ..utils import (
 class MyVideoIE(InfoExtractor):
     """Information Extractor for myvideo.de."""
 
-    _VALID_URL = r'(?:http://)?(?:www\.)?myvideo\.de/watch/([0-9]+)/([^?/]+).*'
+    _VALID_URL = r'(?:http://)?(?:www\.)?myvideo\.de/(?:[^/]+/)?watch/([0-9]+)/([^?/]+).*'
     IE_NAME = u'myvideo'
     _TEST = {
         u'url': u'http://www.myvideo.de/watch/8229274/bowling_fail_or_win',
@@ -85,6 +87,20 @@ class MyVideoIE(InfoExtractor):
                 'ext':      video_ext,
             }]
 
+        mobj = re.search(r'data-video-service="/service/data/video/%s/config' % video_id, webpage)
+        if mobj is not None:
+            request = compat_urllib_request.Request('http://www.myvideo.de/service/data/video/%s/config' % video_id, '')
+            response = self._download_webpage(request, video_id,
+                                              u'Downloading video info')
+            info = json.loads(base64.b64decode(response).decode('utf-8'))
+            return {'id': video_id,
+                    'title': info['title'],
+                    'url': info['streaming_url'].replace('rtmpe', 'rtmpt'),
+                    'play_path': info['filename'],
+                    'ext': 'flv',
+                    'thumbnail': info['thumbnail'][0]['url'],
+                    }
+
         # try encxml
         mobj = re.search('var flashvars={(.+?)}', webpage)
         if mobj is None:

youtube_dl/extractor/naver.py

@@ -0,0 +1,73 @@
+# encoding: utf-8
+import re
+import xml.etree.ElementTree
+
+from .common import InfoExtractor
+from ..utils import (
+    compat_urllib_parse,
+    ExtractorError,
+)
+
+
+class NaverIE(InfoExtractor):
+    _VALID_URL = r'https?://tvcast\.naver\.com/v/(?P<id>\d+)'
+
+    _TEST = {
+        u'url': u'http://tvcast.naver.com/v/81652',
+        u'file': u'81652.mp4',
+        u'info_dict': {
+            u'title': u'[9월 모의고사 해설강의][수학_김상희] 수학 A형 16~20번',
+            u'description': u'합격불변의 법칙 메가스터디 | 메가스터디 수학 김상희 선생님이 9월 모의고사 수학A형 16번에서 20번까지 해설강의를 공개합니다.',
+            u'upload_date': u'20130903',
+        },
+    }
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        video_id = mobj.group(1)
+        webpage = self._download_webpage(url, video_id)
+        m_id = re.search(r'var rmcPlayer = new nhn.rmcnmv.RMCVideoPlayer\("(.+?)", "(.+?)"',
+            webpage)
+        if m_id is None:
+            raise ExtractorError(u'couldn\'t extract vid and key')
+        vid = m_id.group(1)
+        key = m_id.group(2)
+        query = compat_urllib_parse.urlencode({'vid': vid, 'inKey': key,})
+        query_urls = compat_urllib_parse.urlencode({
+            'masterVid': vid,
+            'protocol': 'p2p',
+            'inKey': key,
+        })
+        info_xml = self._download_webpage(
+            'http://serviceapi.rmcnmv.naver.com/flash/videoInfo.nhn?' + query,
+            video_id, u'Downloading video info')
+        urls_xml = self._download_webpage(
+            'http://serviceapi.rmcnmv.naver.com/flash/playableEncodingOption.nhn?' + query_urls,
+            video_id, u'Downloading video formats info')
+        info = xml.etree.ElementTree.fromstring(info_xml.encode('utf-8'))
+        urls = xml.etree.ElementTree.fromstring(urls_xml.encode('utf-8'))
+
+        formats = []
+        for format_el in urls.findall('EncodingOptions/EncodingOption'):
+            domain = format_el.find('Domain').text
+            if domain.startswith('rtmp'):
+                continue
+            formats.append({
+                'url': domain + format_el.find('uri').text,
+                'ext': 'mp4',
+                'width': int(format_el.find('width').text),
+                'height': int(format_el.find('height').text),
+            })
+
+        info = {
+            'id': video_id,
+            'title': info.find('Subject').text,
+            'formats': formats,
+            'description': self._og_search_description(webpage),
+            'thumbnail': self._og_search_thumbnail(webpage),
+            'upload_date': info.find('WriteDate').text.replace('.', ''),
+            'view_count': int(info.find('PlayCount').text),
+        }
+        # TODO: Remove when #980 has been merged
+        info.update(formats[-1])
+        return info

youtube_dl/extractor/nba.py

@@ -30,8 +30,7 @@ class NBAIE(InfoExtractor):
         video_url = u'http://ht-mobile.cdn.turner.com/nba/big' + video_id + '_nba_1280x720.mp4'
 
         shortened_video_id = video_id.rpartition('/')[2]
-        title = self._html_search_regex(r'<meta property="og:title" content="(.*?)"',
-            webpage, 'title', default=shortened_video_id).replace('NBA.com: ', '')
+        title = self._og_search_title(webpage, default=shortened_video_id).replace('NBA.com: ', '')
 
         # It isn't there in the HTML it returns to us
         # uploader_date = self._html_search_regex(r'<b>Date:</b> (.*?)</div>', webpage, 'upload_date', fatal=False)

youtube_dl/extractor/nbc.py

@@ -0,0 +1,33 @@
+import re
+import xml.etree.ElementTree
+
+from .common import InfoExtractor
+from ..utils import find_xpath_attr, compat_str
+
+
+class NBCNewsIE(InfoExtractor):
+    _VALID_URL = r'https?://www\.nbcnews\.com/video/.+?/(?P<id>\d+)'
+
+    _TEST = {
+        u'url': u'http://www.nbcnews.com/video/nbc-news/52753292',
+        u'file': u'52753292.flv',
+        u'md5': u'47abaac93c6eaf9ad37ee6c4463a5179',
+        u'info_dict': {
+            u'title': u'Crew emerges after four-month Mars food study',
+            u'description': u'md5:24e632ffac72b35f8b67a12d1b6ddfc1',
+        },
+    }
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        video_id = mobj.group('id')
+        info_xml = self._download_webpage('http://www.nbcnews.com/id/%s/displaymode/1219' % video_id, video_id)
+        info = xml.etree.ElementTree.fromstring(info_xml.encode('utf-8')).find('video')
+
+        return {'id': video_id,
+                'title': info.find('headline').text,
+                'ext': 'flv',
+                'url': find_xpath_attr(info, 'media', 'type', 'flashVideo').text,
+                'description': compat_str(info.find('caption').text),
+                'thumbnail': find_xpath_attr(info, 'media', 'type', 'thumbnail').text,
+                }

youtube_dl/extractor/newgrounds.py

@@ -0,0 +1,38 @@
+import json
+import re
+
+from .common import InfoExtractor
+from ..utils import determine_ext
+
+
+class NewgroundsIE(InfoExtractor):
+    _VALID_URL = r'(?:https?://)?(?:www\.)?newgrounds\.com/audio/listen/(?P<id>\d+)'
+    _TEST = {
+        u'url': u'http://www.newgrounds.com/audio/listen/549479',
+        u'file': u'549479.mp3',
+        u'md5': u'fe6033d297591288fa1c1f780386f07a',
+        u'info_dict': {
+            u"title": u"B7 - BusMode",
+            u"uploader": u"Burn7",
+        }
+    }
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        music_id = mobj.group('id')
+        webpage = self._download_webpage(url, music_id)
+        
+        title = self._html_search_regex(r',"name":"([^"]+)",', webpage, u'music title')
+        uploader = self._html_search_regex(r',"artist":"([^"]+)",', webpage, u'music uploader')
+        
+        music_url_json_string = self._html_search_regex(r'({"url":"[^"]+"),', webpage, u'music url') + '}'
+        music_url_json = json.loads(music_url_json_string)
+        music_url = music_url_json['url']
+
+        return {
+            'id':       music_id,
+            'title':    title,
+            'url':      music_url,
+            'uploader': uploader,
+            'ext':      determine_ext(music_url),
+        }

youtube_dl/extractor/nhl.py

@@ -0,0 +1,120 @@
+import re
+import json
+import xml.etree.ElementTree
+
+from .common import InfoExtractor
+from ..utils import (
+    compat_urlparse,
+    compat_urllib_parse,
+    determine_ext,
+    unified_strdate,
+)
+
+
+class NHLBaseInfoExtractor(InfoExtractor):
+    @staticmethod
+    def _fix_json(json_string):
+        return json_string.replace('\\\'', '\'')
+
+    def _extract_video(self, info):
+        video_id = info['id']
+        self.report_extraction(video_id)
+
+        initial_video_url = info['publishPoint']
+        data = compat_urllib_parse.urlencode({
+            'type': 'fvod',
+            'path': initial_video_url.replace('.mp4', '_sd.mp4'),
+        })
+        path_url = 'http://video.nhl.com/videocenter/servlets/encryptvideopath?' + data
+        path_response = self._download_webpage(path_url, video_id,
+            u'Downloading final video url')
+        path_doc = xml.etree.ElementTree.fromstring(path_response)
+        video_url = path_doc.find('path').text
+
+        join = compat_urlparse.urljoin
+        return {
+            'id': video_id,
+            'title': info['name'],
+            'url': video_url,
+            'ext': determine_ext(video_url),
+            'description': info['description'],
+            'duration': int(info['duration']),
+            'thumbnail': join(join(video_url, '/u/'), info['bigImage']),
+            'upload_date': unified_strdate(info['releaseDate'].split('.')[0]),
+        }
+
+
+class NHLIE(NHLBaseInfoExtractor):
+    IE_NAME = u'nhl.com'
+    _VALID_URL = r'https?://video(?P<team>\.[^.]*)?\.nhl\.com/videocenter/console\?.*?(?<=[?&])id=(?P<id>\d+)'
+
+    _TEST = {
+        u'url': u'http://video.canucks.nhl.com/videocenter/console?catid=6?id=453614',
+        u'file': u'453614.mp4',
+        u'info_dict': {
+            u'title': u'Quick clip: Weise 4-3 goal vs Flames',
+            u'description': u'Dale Weise scores his first of the season to put the Canucks up 4-3.',
+            u'duration': 18,
+            u'upload_date': u'20131006',
+        },
+    }
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        video_id = mobj.group('id')
+        json_url = 'http://video.nhl.com/videocenter/servlets/playlist?ids=%s&format=json' % video_id
+        info_json = self._download_webpage(json_url, video_id,
+            u'Downloading info json')
+        info_json = self._fix_json(info_json)
+        info = json.loads(info_json)[0]
+        return self._extract_video(info)
+
+
+class NHLVideocenterIE(NHLBaseInfoExtractor):
+    IE_NAME = u'nhl.com:videocenter'
+    IE_DESC = u'Download the first 12 videos from a videocenter category'
+    _VALID_URL = r'https?://video\.(?P<team>[^.]*)\.nhl\.com/videocenter/(console\?.*?catid=(?P<catid>[^&]+))?'
+
+    @classmethod
+    def suitable(cls, url):
+        if NHLIE.suitable(url):
+            return False
+        return super(NHLVideocenterIE, cls).suitable(url)
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        team = mobj.group('team')
+        webpage = self._download_webpage(url, team)
+        cat_id = self._search_regex(
+            [r'var defaultCatId = "(.+?)";',
+             r'{statusIndex:0,index:0,.*?id:(.*?),'],
+            webpage, u'category id')
+        playlist_title = self._html_search_regex(
+            r'\?catid=%s">(.*?)</a>' % cat_id,
+            webpage, u'playlist title', flags=re.DOTALL)
+
+        data = compat_urllib_parse.urlencode({
+            'cid': cat_id,
+            # This is the default value
+            'count': 12,
+            'ptrs': 3,
+            'format': 'json',
+        })
+        path = '/videocenter/servlets/browse?' + data
+        request_url = compat_urlparse.urljoin(url, path)
+        response = self._download_webpage(request_url, playlist_title)
+        response = self._fix_json(response)
+        if not response.strip():
+            self._downloader.report_warning(u'Got an empty reponse, trying '
+                                            u'adding the "newvideos" parameter')
+            response = self._download_webpage(request_url + '&newvideos=true',
+                playlist_title)
+            response = self._fix_json(response)
+        videos = json.loads(response)
+
+        return {
+            '_type': 'playlist',
+            'title': playlist_title,
+            'id': cat_id,
+            'entries': [self._extract_video(i) for i in videos],
+        }