release 2013.11.02

The 'Content-Type' header must be set for disabling the family filter.
The 'flashversion' cookie  is only needed for AnyClip videos.
Added tests for standard metacafe videos and for age-restricted videos.
Also set the 'age_limit' field.

.gitignore

@@ -18,3 +18,11 @@ youtube-dl.tar.gz
 cover/
 updates_key.pem
 *.egg-info
+*.srt
+*.sbv
+*.vtt
+*.flv
+*.mp4
+*.part
+test/testdata
+.tox

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

@@ -19,7 +19,10 @@ which means you can modify it, redistribute it or use it however you like.
     -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
+    -i, --ignore-errors        continue on download errors, for example to to
+                               skip unavailable videos in a playlist
+    --abort-on-error           Abort downloading of further videos (in the
+                               playlist or the command line) if an error occurs
     --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
@@ -29,6 +32,11 @@ 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 DIR            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)
@@ -45,11 +53,16 @@ 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
@@ -65,7 +78,10 @@ which means you can modify it, redistribute it or use it however you like.
                                %(uploader_id)s for the uploader nickname if
                                different, %(autonumber)s to get an automatically
                                incremented number, %(ext)s for the filename
-                               extension, %(upload_date)s for the upload date
+                               extension, %(format)s for the format description
+                               (like "22 - 1280x720" or "HD"),%(format_id)s for
+                               the unique id of the format (like Youtube's
+                               itags: "137"),%(upload_date)s for the upload date
                                (YYYYMMDD), %(extractor)s for the provider
                                (youtube, metacafe, etc), %(id)s for the video id
                                , %(playlist)s for the playlist the video is in,
@@ -90,6 +106,7 @@ 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:
@@ -110,28 +127,31 @@ which means you can modify it, redistribute it or use it however you like.
     -v, --verbose              print various debugging information
     --dump-intermediate-pages  print downloaded pages to debug problems(very
                                verbose)
+    --write-pages              Write downloaded pages to files in the current
+                               directory
 
 ## 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
@@ -153,6 +173,9 @@ 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

@@ -1,14 +1,18 @@
-__youtube-dl()
+__youtube_dl()
 {
     local cur prev opts
     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
 }
 
-complete -F __youtube-dl youtube-dl
+complete -F __youtube_dl youtube-dl

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/check-porn.py

@@ -0,0 +1,39 @@
+#!/usr/bin/env python
+
+"""
+This script employs a VERY basic heuristic ('porn' in webpage.lower()) to check
+if we are not 'age_limit' tagging some porn site
+"""
+
+# Allow direct execution
+import os
+import sys
+sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
+
+from test.helper import get_testcases
+from youtube_dl.utils import compat_urllib_request
+
+for test in get_testcases():
+    try:
+        webpage = compat_urllib_request.urlopen(test['url'], timeout=10).read()
+    except:
+        print('\nFail: {0}'.format(test['name']))
+        continue
+
+    webpage = webpage.decode('utf8', 'replace')
+
+    if 'porn' in webpage.lower() and ('info_dict' not in test
+                                      or 'age_limit' not in test['info_dict']
+                                      or test['info_dict']['age_limit'] != 18):
+        print('\nPotential missing age_limit check: {0}'.format(test['name']))
+
+    elif 'porn' not in webpage.lower() and ('info_dict' in test and
+                                            'age_limit' in test['info_dict'] and
+                                            test['info_dict']['age_limit'] == 18):
+        print('\nPotential false negative: {0}'.format(test['name']))
+
+    else:
+        sys.stdout.write('.')
+    sys.stdout.flush()
+
+print()

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/$version html/downloads/"
@@ -85,12 +85,9 @@ ROOT=$(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
-    read -p "Is it good, can I push? (y/n) " -n 1
-    if [[ ! $REPLY =~ ^[Yy]$ ]]; then exit 1; fi
-    echo
     git push "$ROOT" gh-pages
     git push "$ORIGIN_URL" gh-pages
 )

devscripts/youtube_genalgo.py

@@ -1,103 +0,0 @@
-#!/usr/bin/env python
-
-# Generate youtube signature algorithm from test cases
-
-import sys
-
-tests = [
-    # 92 - vflQw-fB4 2013/07/17
-    ("qwertyuioplkjhgfdsazxcvbnm1234567890QWERTYUIOPLKJHGFDSAZXCVBNM!@#$%^&*()_-+={[]}|:;?/>.<'`~\"",
-     "mrtyuioplkjhgfdsazxcvbnq1234567890QWERTY}IOPLKJHGFDSAZXCVBNM!@#$%^&*()_-+={[]\"|:;"),
-    # 90
-    ("qwertyuioplkjhgfdsazxcvbnm1234567890QWERTYUIOPLKJHGFDSAZXCVBNM!@#$%^&*()_-+={[]}|:;?/>.<'`",
-     "mrtyuioplkjhgfdsazxcvbne1234567890QWER[YUIOPLKJHGFDSAZXCVBNM!@#$%^&*()_-+={`]}|"),
-    # 88
-    ("qwertyuioplkjhgfdsazxcvbnm1234567890QWERTYUIOPLKJHGFDSAZXCVBNM!@#$%^&*()_-+={[]}|:;?/>.<",
-     "J:|}][{=+-_)(*&;%$#@>MNBVCXZASDFGH^KLPOIUYTREWQ0987654321mnbvcxzasdfghrklpoiuytej"),
-    # 87 - vflART1Nf 2013/07/24
-    ("qwertyuioplkjhgfdsazxcvbnm1234567890QWERTYUIOPLKJHGFDSAZXCVBNM!@#$^&*()_-+={[]}|:;?/>.<",
-     "tyuioplkjhgfdsazxcv<nm1234567890QWERTYUIOPLKJHGFDSAZXCVBNM!@#$^&*()_-+={[]}|:;?/>"),
-    # 86 - vflm_D8eE 2013/07/31
-    ("qwertyuioplkjhgfdsazxcvbnm1234567890QWERTYUIOPLKJHGFDSAZXCVBNM!@#$%^&*()_-+={[|};?/>.<",
-     ">.1}|[{=+-_)(*&^%$#@!MNBVCXZASDFGHJK<POIUYTREW509876L432/mnbvcxzasdfghjklpoiuytre"),
-    # 85 - vflSAFCP9 2013/07/19
-    ("qwertyuioplkjhgfdsazxcvbnm1234567890QWERTYUIOPLKJHGFDSAZXCVBNM!@#$%^&*()_-+={[};?/>.<",
-     "ertyuiqplkjhgfdsazx$vbnm1234567890QWERTYUIOPLKJHGFDSAZXCVBNM!@#<%^&*()_-+={[};?/c"),
-    # 84
-    ("qwertyuioplkjhgfdsazxcvbnm1234567890QWERTYUIOPLKJHGFDSAZXCVBNM!@#$%^&*()_-+={[};?>.<",
-     "<.>?;}[{=+-_)(*&^%$#@!MNBVCXZASDFGHJKLPOIUYTREWe098765432rmnbvcxzasdfghjklpoiuyt1"),
-    # 83 - vflTWC9KW 2013/08/01
-    ("qwertyuioplkjhgfdsazxcvbnm1234567890QWERTYUIOPLKJHGFDSAZXCVBNM!#$%^&*()_+={[};?/>.<",
-     "qwertyuioplkjhg>dsazxcvbnm1234567890QWERTYUIOPLKJHGFDSAZXCVBNM!#$%^&*()_+={[};?/f"),
-    # 82
-    ("qwertyuioplkjhgfdsazxcvbnm1234567890QWERTYUIOPLKHGFDSAZXCVBNM!@#$%^&*(-+={[};?/>.<",
-     "Q>/?;}[{=+-(*<^%$#@!MNBVCXZASDFGHKLPOIUY8REWT0q&7654321mnbvcxzasdfghjklpoiuytrew9"),
-    # 81 - vflLC8JvQ 2013/07/25
-    ("qwertyuioplkjhgfdsazxcvbnm1234567890QWERTYUIOPLKHGFDSAZXCVBNM!@#$%^&*(-+={[};?/>.",
-     "C>/?;}[{=+-(*&^%$#@!MNBVYXZASDFGHKLPOIU.TREWQ0q87659321mnbvcxzasdfghjkl4oiuytrewp"),
-    # 79 - vflLC8JvQ 2013/07/25 (sporadic)
-    ("qwertyuioplkjhgfdsazxcvbnm1234567890QWERTYUIOPLKHGFDSAZXCVBNM!@#$%^&*(-+={[};?/",
-     "Z?;}[{=+-(*&^%$#@!MNBVCXRASDFGHKLPOIUYT/EWQ0q87659321mnbvcxzasdfghjkl4oiuytrewp"),
-]
-
-tests_age_gate = [
-    # 86 - vflqinMWD
-    ("qwertyuioplkjhgfdsazxcvbnm1234567890QWERTYUIOPLKJHGFDSAZXCVBNM!@#$%^&*()_-+={[|};?/>.<",
-     "ertyuioplkjhgfdsazxcvbnm1234567890QWERTYUIOPLKJHGFDSAZXCVBNM!/#$%^&*()_-+={[|};?@"),
-]
-
-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))
-    print(u'    Age gate:')
-    print(genall(tests_age_gate))
-
-if __name__ == '__main__':
-    main()

setup.py

@@ -8,8 +8,10 @@ import sys
 
 try:
     from setuptools import setup
+    setuptools_available = True
 except ImportError:
     from distutils.core import setup
+    setuptools_available = False
 
 try:
     # This will create an exe that needs Microsoft Visual C++ 2008
@@ -43,13 +45,16 @@ if len(sys.argv) >= 2 and sys.argv[1] == 'py2exe':
     params = py2exe_params
 else:
     params = {
-        'scripts': ['bin/youtube-dl'],
         'data_files': [  # Installing system-wide would require sudo...
             ('etc/bash_completion.d', ['youtube-dl.bash-completion']),
             ('share/doc/youtube_dl', ['README.txt']),
             ('share/man/man1/', ['youtube-dl.1'])
         ]
     }
+    if setuptools_available:
+        params['entry_points'] = {'console_scripts': ['youtube-dl = youtube_dl:main']}
+    else:
+        params['scripts'] = ['bin/youtube-dl']
 
 # Get the version from youtube_dl/version.py without importing the package
 exec(compile(open('youtube_dl/version.py').read(),
@@ -63,6 +68,7 @@ setup(
     ' YouTube.com and other video sites.',
     url='https://github.com/rg3/youtube-dl',
     author='Ricardo Garcia',
+    author_email='ytdl@yt-dl.org',
     maintainer='Philipp Hagemeister',
     maintainer_email='phihag@phihag.de',
     packages=['youtube_dl', 'youtube_dl.extractor'],

test/helper.py

@@ -1,38 +1,80 @@
+import errno
 import io
+import hashlib
 import json
 import os.path
+import re
+import types
+import sys
 
 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
+from youtube_dl.utils import preferredencoding
 
-# 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
+
+
+def report_warning(message):
+    '''
+    Print the message to stderr, it will be prefixed with 'WARNING:'
+    If stderr is a tty file the 'WARNING:' will be colored
+    '''
+    if sys.stderr.isatty() and os.name != 'nt':
+        _msg_header = u'\033[0;33mWARNING:\033[0m'
+    else:
+        _msg_header = u'WARNING:'
+    output = u'%s %s\n' % (_msg_header, message)
+    if 'b' in getattr(sys.stderr, 'mode', '') or sys.version_info[0] < 3:
+        output = output.encode(preferredencoding())
+    sys.stderr.write(output)
+
 
 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 +84,6 @@ 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 +1,133 @@
+#!/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'url': u'http://example.com/meh'},
+            {u'format_id': u'good', u'url': u'http://example.com/good'},
+            {u'format_id': u'great', u'url': u'http://example.com/great'},
+            {u'format_id': u'excellent', u'url': u'http://example.com/exc'},
+        ]
+        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')
+
+    def test_format_selection(self):
+        formats = [
+            {u'format_id': u'35', u'ext': u'mp4'},
+            {u'format_id': u'45', u'ext': u'webm'},
+            {u'format_id': u'47', u'ext': u'webm'},
+            {u'format_id': u'2', u'ext': u'flv'},
+        ]
+        info_dict = {u'formats': formats, u'extractor': u'test'}
+
+        ydl = YDL({'format': u'20/47'})
+        ydl.process_ie_result(info_dict)
+        downloaded = ydl.downloaded_info_dicts[0]
+        self.assertEqual(downloaded['format_id'], u'47')
+
+        ydl = YDL({'format': u'20/71/worst'})
+        ydl.process_ie_result(info_dict)
+        downloaded = ydl.downloaded_info_dicts[0]
+        self.assertEqual(downloaded['format_id'], u'35')
+
+        ydl = YDL()
+        ydl.process_ie_result(info_dict)
+        downloaded = ydl.downloaded_info_dicts[0]
+        self.assertEqual(downloaded['format_id'], u'2')
+
+        ydl = YDL({'format': u'webm/mp4'})
+        ydl.process_ie_result(info_dict)
+        downloaded = ydl.downloaded_info_dicts[0]
+        self.assertEqual(downloaded['format_id'], u'47')
+
+        ydl = YDL({'format': u'3gp/40/mp4'})
+        ydl.process_ie_result(info_dict)
+        downloaded = ydl.downloaded_info_dicts[0]
+        self.assertEqual(downloaded['format_id'], u'35')
+
+
+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()
@@ -62,15 +98,12 @@ class TestAllURLsMatching(unittest.TestCase):
                     self.assertFalse(ie.suitable(url), '%s should not match URL %r' % (type(ie).__name__, url))
 
     def test_keywords(self):
-        ies = gen_extractors()
-        matching_ies = lambda url: [ie.IE_NAME for ie in ies
-                                    if ie.suitable(url) and ie.IE_NAME != 'generic']
-        self.assertEqual(matching_ies(':ytsubs'), ['youtube:subscriptions'])
-        self.assertEqual(matching_ies(':ytsubscriptions'), ['youtube:subscriptions'])
-        self.assertEqual(matching_ies(':thedailyshow'), ['ComedyCentral'])
-        self.assertEqual(matching_ies(':tds'), ['ComedyCentral'])
-        self.assertEqual(matching_ies(':colbertreport'), ['ComedyCentral'])
-        self.assertEqual(matching_ies(':cr'), ['ComedyCentral'])
+        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__':

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['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 +1,39 @@
 #!/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,
+    report_warning
+)
+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,
+    compat_HTTPError,
+    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 +50,12 @@ 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
@@ -77,15 +68,17 @@ def generator(test_case):
         if not ie._WORKING:
             print_skipping('IE marked as not _WORKING')
             return
-        if 'playlist' not in test_case and not test_case['file']:
-            print_skipping('No output file specified')
-            return
+        if 'playlist' not in test_case:
+            info_dict = test_case.get('info_dict', {})
+            if not test_case.get('file') and not (info_dict.get('id') and info_dict.get('ext')):
+                print_skipping('The output file cannot be know, the "file" '
+                    'key is missing or the info_dict is incomplete')
+                return
         if 'skip' in 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()
@@ -95,44 +88,55 @@ def generator(test_case):
                 finished_hook_called.add(status['filename'])
         ydl.fd.add_progress_hook(_hook)
 
+        def get_tc_filename(tc):
+            return tc.get('file') or ydl.prepare_filename(tc.get('info_dict', {}))
+
         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')
+        def try_rm_tcs_files():
+            for tc in test_cases:
+                tc_filename = get_tc_filename(tc)
+                try_rm(tc_filename)
+                try_rm(tc_filename + '.part')
+                try_rm(tc_filename + '.info.json')
+        try_rm_tcs_files()
         try:
-            for retry in range(1, RETRIES + 1):
+            try_num = 1
+            while True:
                 try:
                     ydl.download([test_case['url']])
                 except (DownloadError, ExtractorError) as err:
-                    if retry == RETRIES: raise
-
                     # Check if the exception is not a network related one
-                    if not err.exc_info[0] in (compat_urllib_error.URLError, socket.timeout, UnavailableVideoError):
+                    if not err.exc_info[0] in (compat_urllib_error.URLError, socket.timeout, UnavailableVideoError) or (err.exc_info[0] == compat_HTTPError and err.exc_info[1].code == 503):
                         raise
 
-                    print('Retrying: {0} failed tries\n\n##########\n\n'.format(retry))
+                    if try_num == RETRIES:
+                        report_warning(u'Failed due to network errors, skipping...')
+                        return
+
+                    print('Retrying: {0} failed tries\n\n##########\n\n'.format(try_num))
+
+                    try_num += 1
                 else:
                     break
 
             for tc in test_cases:
+                tc_filename = get_tc_filename(tc)
                 if not test_case.get('params', {}).get('skip_download', False):
-                    self.assertTrue(os.path.exists(tc['file']), msg='Missing file ' + tc['file'])
-                    self.assertTrue(tc['file'] in finished_hook_called)
-                self.assertTrue(os.path.exists(tc['file'] + '.info.json'))
+                    self.assertTrue(os.path.exists(tc_filename), msg='Missing file ' + tc_filename)
+                    self.assertTrue(tc_filename in finished_hook_called)
+                self.assertTrue(os.path.exists(tc_filename + '.info.json'))
                 if 'md5' in tc:
-                    md5_for_file = _file_md5(tc['file'])
+                    md5_for_file = _file_md5(tc_filename)
                     self.assertEqual(md5_for_file, tc['md5'])
-                with io.open(tc['file'] + '.info.json', encoding='utf-8') as infof:
+                with io.open(tc_filename + '.info.json', encoding='utf-8') as infof:
                     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))
@@ -145,10 +149,7 @@ def generator(test_case):
                 for key in ('id', 'url', 'title', 'ext'):
                     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_tcs_files()
 
     return test_template
 

test/test_playlists.py

@@ -1,17 +1,27 @@
 #!/usr/bin/env python
+# encoding: utf-8
 
-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 DailymotionPlaylistIE, VimeoChannelIE
-from youtube_dl.utils import *
+from test.helper import FakeYDL, global_setup
+global_setup()
+
+
+from youtube_dl.extractor import (
+    DailymotionPlaylistIE,
+    DailymotionUserIE,
+    VimeoChannelIE,
+    UstreamChannelIE,
+    SoundcloudUserIE,
+    LivestreamIE,
+    NHLVideocenterIE,
+)
 
-from helper import FakeYDL
 
 class TestPlaylists(unittest.TestCase):
     def assertIsPlaylist(self, info):
@@ -26,6 +36,14 @@ class TestPlaylists(unittest.TestCase):
         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)
@@ -34,5 +52,38 @@ class TestPlaylists(unittest.TestCase):
         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

@@ -1,23 +1,30 @@
 #!/usr/bin/env python
-
-# Various small unit tests
-
-import sys
-import unittest
-import xml.etree.ElementTree
+# 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 find_xpath_attr
+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')
@@ -127,5 +134,42 @@ class TestUtil(unittest.TestCase):
         self.assertEqual(find_xpath_attr(doc, './/node', 'x', 'a'), doc[1])
         self.assertEqual(find_xpath_attr(doc, './/node', 'y', 'c'), doc[2])
 
+    def test_meta_parser(self):
+        testhtml = 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_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)
@@ -368,6 +401,7 @@ class FileDownloader(object):
             self._hook_progress({
                 'filename': filename,
                 'status': 'finished',
+                'total_bytes': os.path.getsize(encodeFilename(filename)),
             })
             return True
 
@@ -514,13 +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,
@@ -528,6 +563,8 @@ class FileDownloader(object):
                 'tmpfilename': tmpfilename,
                 'filename': filename,
                 'status': 'downloading',
+                'eta': eta,
+                'speed': speed,
             })
 
             # Apply rate limit
@@ -538,7 +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)
@@ -570,6 +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'-'):
@@ -129,7 +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']
@@ -169,7 +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']
@@ -199,7 +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)
@@ -232,3 +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."""
@@ -184,10 +216,10 @@ class YoutubeDL(object):
         If stderr is a tty file the 'WARNING:' will be colored
         '''
         if sys.stderr.isatty() and os.name != 'nt':
-            _msg_header=u'\033[0;33mWARNING:\033[0m'
+            _msg_header = u'\033[0;33mWARNING:\033[0m'
         else:
-            _msg_header=u'WARNING:'
-        warning_message=u'%s %s' % (_msg_header,message)
+            _msg_header = u'WARNING:'
+        warning_message = u'%s %s' % (_msg_header, message)
         self.to_stderr(warning_message)
 
     def report_error(self, message, tb=None):
@@ -202,19 +234,6 @@ class YoutubeDL(object):
         error_message = u'%s %s' % (_msg_header, message)
         self.trouble(error_message, tb)
 
-    def slow_down(self, start_time, byte_counter):
-        """Sleep if the download speed is over the rate limit."""
-        rate_limit = self.params.get('ratelimit', None)
-        if rate_limit is None or byte_counter == 0:
-            return
-        now = time.time()
-        elapsed = now - start_time
-        if elapsed <= 0.0:
-            return
-        speed = float(byte_counter) / elapsed
-        if speed > rate_limit:
-            time.sleep((byte_counter - rate_limit * (now - start_time)) / rate_limit)
-
     def report_writedescription(self, descfn):
         """ Report that the description file is being written """
         self.to_screen(u'[info] Writing video description to: ' + descfn)
@@ -227,6 +246,10 @@ 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:
@@ -249,16 +272,18 @@ class YoutubeDL(object):
                 autonumber_size = 5
             autonumber_templ = u'%0' + str(autonumber_size) + u'd'
             template_dict['autonumber'] = autonumber_templ % self._num_downloads
-            if template_dict['playlist_index'] is not None:
+            if template_dict.get('playlist_index') is not None:
                 template_dict['playlist_index'] = u'%05d' % template_dict['playlist_index']
 
-            sanitize = lambda k,v: sanitize_filename(
+            sanitize = lambda k, v: sanitize_filename(
                 u'NA' if v is None else compat_str(v),
                 restricted=self.params.get('restrictfilenames'),
-                is_id=(k==u'id'))
-            template_dict = dict((k, sanitize(k, v)) for k,v in template_dict.items())
+                is_id=(k == u'id'))
+            template_dict = dict((k, sanitize(k, v))
+                                 for k, v in template_dict.items())
 
-            filename = self.params['outtmpl'] % template_dict
+            tmpl = os.path.expanduser(self.params['outtmpl'])
+            filename = tmpl % template_dict
             return filename
         except KeyError as err:
             self.report_error(u'Erroneous output template')
@@ -284,6 +309,13 @@ 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 +326,7 @@ 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
 
@@ -349,13 +379,7 @@ 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':
             ie_result.update(extra_info)
-            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
+            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
@@ -366,7 +390,7 @@ class YoutubeDL(object):
         elif result_type == 'playlist':
             # We process each entry in the playlist
             playlist = ie_result.get('title', None) or ie_result.get('id', None)
-            self.to_screen(u'[download] Downloading playlist: %s'  % playlist)
+            self.to_screen(u'[download] Downloading playlist: %s' % playlist)
 
             playlist_results = []
 
@@ -384,12 +408,12 @@ class YoutubeDL(object):
             self.to_screen(u"[%s] playlist '%s': Collected %d video ids (downloading %d of them)" %
                 (ie_result['extractor'], playlist, n_all_entries, n_entries))
 
-            for i,entry in enumerate(entries,1):
-                self.to_screen(u'[download] Downloading video #%s of %s' %(i, n_entries))
+            for i, entry in enumerate(entries, 1):
+                self.to_screen(u'[download] Downloading video #%s of %s' % (i, n_entries))
                 extra = {
-                         'playlist': playlist, 
-                         'playlist_index': i + playliststart,
-                         }
+                    'playlist': playlist,
+                    'playlist_index': i + playliststart,
+                }
                 if not 'extractor' in entry:
                     # We set the extractor, if it's an url it will be set then to
                     # the new extractor, but if it's already a video we must make
@@ -413,6 +437,107 @@ class YoutubeDL(object):
         else:
             raise Exception('Invalid result type: %s' % result_type)
 
+    def select_format(self, format_spec, available_formats):
+        if format_spec == 'best' or format_spec is None:
+            return available_formats[-1]
+        elif format_spec == 'worst':
+            return available_formats[0]
+        else:
+            extensions = [u'mp4', u'flv', u'webm', u'3gp']
+            if format_spec in extensions:
+                filter_f = lambda f: f['ext'] == format_spec
+            else:
+                filter_f = lambda f: f['format_id'] == format_spec
+            matches = list(filter(filter_f, available_formats))
+            if matches:
+                return matches[-1]
+        return None
+
+    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']:
+            if download:
+                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_id') is None:
+                format['format_id'] = compat_str(i)
+            if format.get('format') is None:
+                format['format'] = u'{id} - {res}{note}'.format(
+                    id=format['format_id'],
+                    res=self.format_resolution(format),
+                    note=u' ({0})'.format(format['format_note']) if format.get('format_note') is not None else '',
+                )
+            # Automatically determine file extension if missing
+            if 'ext' not in format:
+                format['ext'] = determine_ext(format['url'])
+
+        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')
+        if req_format is None:
+            req_format = 'best'
+        formats_to_download = []
+        # The -1 is for supporting YoutubeIE
+        if req_format in ('-1', 'all'):
+            formats_to_download = formats
+        else:
+            # We can accept formats requestd in the format: 34/5/best, we pick
+            # the first that is available, starting from left
+            req_formats = req_format.split('/')
+            for rf in req_formats:
+                selected_format = self.select_format(rf, formats)
+                if selected_format is not None:
+                    formats_to_download = [selected_format]
+                    break
+        if not formats_to_download:
+            raise ExtractorError(u'requested format not available',
+                                 expected=True)
+
+        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."""
 
@@ -448,10 +573,11 @@ 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'])
-        if self.params.get('forcethumbnail', False) and 'thumbnail' in info_dict:
+            # 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 info_dict.get('thumbnail') is not None:
             compat_print(info_dict['thumbnail'])
-        if self.params.get('forcedescription', False) and 'description' in info_dict:
+        if self.params.get('forcedescription', False) and info_dict.get('description') is not None:
             compat_print(info_dict['description'])
         if self.params.get('forcefilename', False) and filename is not None:
             compat_print(filename)
@@ -479,50 +605,50 @@ 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)
             try:
-                json_info_dict = dict((k, v) for k,v in info_dict.items() if not k in ['urlhandle'])
+                json_info_dict = dict((k, v) for k, v in info_dict.items() if not k in ['urlhandle'])
                 write_json_file(json_info_dict, encodeFilename(infofn))
             except (OSError, IOError):
                 self.report_error(u'Cannot write metadata to JSON file ' + infofn)
@@ -534,11 +660,15 @@ class YoutubeDL(object):
                 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)):
@@ -546,11 +676,11 @@ class YoutubeDL(object):
             else:
                 try:
                     success = self.fd._do_download(filename, info_dict)
-                except (OSError, IOError) as err:
-                    raise UnavailableVideoError(err)
                 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
@@ -562,6 +692,8 @@ 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():
@@ -586,7 +718,7 @@ class YoutubeDL(object):
         keep_video = None
         for pp in self._pps:
             try:
-                keep_video_wish,new_info = pp.run(info)
+                keep_video_wish, new_info = pp.run(info)
                 if keep_video_wish is not None:
                     if keep_video_wish:
                         keep_video = keep_video_wish
@@ -601,3 +733,61 @@ class YoutubeDL(object):
                 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')
+
+    @staticmethod
+    def format_resolution(format, default='unknown'):
+        if format.get('_resolution') is not None:
+            return format['_resolution']
+        if format.get('height') is not None:
+            if format.get('width') is not None:
+                res = u'%sx%s' % (format['width'], format['height'])
+            else:
+                res = u'%sp' % format['height']
+        else:
+            res = default
+        return res
+
+    def list_formats(self, info_dict):
+        def line(format):
+            return (u'%-15s%-10s%-12s%s' % (
+                format['format_id'],
+                format['ext'],
+                self.format_resolution(format),
+                format.get('format_note', ''),
+                )
+            )
+
+        formats = info_dict.get('formats', [info_dict])
+        formats_s = list(map(line, formats))
+        if len(formats) > 1:
+            formats_s[0] += (' ' if formats[0].get('format_note') else '') + '(worst)'
+            formats_s[-1] += (' ' if formats[-1].get('format_note') else '') + '(best)'
+
+        header_line = line({
+            'format_id': u'format code', 'ext': u'extension',
+            '_resolution': u'resolution', 'format_note': u'note'})
+        self.to_screen(u'[info] Available formats for %s:\n%s\n%s' %
+                       (info_dict['id'], header_line, u"\n".join(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', '--video-password']:
+            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')
@@ -131,7 +178,10 @@ def parseOpts(overrideArguments=None):
     general.add_option('-U', '--update',
             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('--abort-on-error',
+            action='store_false', dest='ignoreerrors',
+            help='Abort downloading of further videos (in the playlist or the command line) if an error occurs')
     general.add_option('--dump-user-agent',
             action='store_true', dest='dump_user_agent',
             help='display the current browser identification', default=False)
@@ -148,6 +198,12 @@ 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(), metavar='DIR',
+        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 +218,13 @@ 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 +238,8 @@ 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 +248,33 @@ 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 +316,13 @@ 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('--write-pages',
+            action='store_true', dest='write_pages', default=False,
+            help='Write downloaded pages to files in the current directory')
+    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)
@@ -269,7 +338,10 @@ def parseOpts(overrideArguments=None):
             help=('output filename template. Use %(title)s to get the title, '
                   '%(uploader)s for the uploader name, %(uploader_id)s for the uploader nickname if different, '
                   '%(autonumber)s to get an automatically incremented number, '
-                  '%(ext)s for the filename extension, %(upload_date)s for the upload date (YYYYMMDD), '
+                  '%(ext)s for the filename extension, '
+                  '%(format)s for the format description (like "22 - 1280x720" or "HD"),'
+                  '%(format_id)s for the unique id of the format (like Youtube\'s itags: "137"),'
+                  '%(upload_date)s for the upload date (YYYYMMDD), '
                   '%(extractor)s for the provider (youtube, metacafe, etc), '
                   '%(id)s for the video id , %(playlist)s for the playlist the video is in, '
                   '%(playlist_index)s for the position in the playlist and %% for a literal percent. '
@@ -303,6 +375,9 @@ 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 +395,10 @@ 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 +407,33 @@ 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:]
         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
 
@@ -372,7 +456,7 @@ 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:
@@ -399,29 +483,13 @@ def _real_main(argv=None):
             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:
-                sys.stderr.write(u'[debug] Batch file urls: ' + repr(batchurls) + u'\n')
+                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()
 
@@ -438,6 +506,8 @@ 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')
@@ -509,6 +579,11 @@ 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:
@@ -521,6 +596,10 @@ 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({
@@ -555,11 +634,13 @@ 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,
@@ -567,22 +648,27 @@ 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,
         'prefer_free_formats': opts.prefer_free_formats,
         'verbose': opts.verbose,
         'dump_intermediate_pages': opts.dump_intermediate_pages,
+        'write_pages': opts.write_pages,
         'test': opts.test,
         'keepvideo': opts.keepvideo,
         '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:
-        sys.stderr.write(u'[debug] youtube-dl version ' + __version__ + u'\n')
+        write_string(u'[debug] youtube-dl version ' + __version__ + u'\n')
         try:
             sp = subprocess.Popen(
                 ['git', 'rev-parse', '--short', 'HEAD'],
@@ -591,26 +677,36 @@ def _real_main(argv=None):
             out, err = sp.communicate()
             out = out.decode().strip()
             if re.match('[0-9a-f]+', out):
-                sys.stderr.write(u'[debug] Git HEAD: ' + out + u'\n')
+                write_string(u'[debug] Git HEAD: ' + out + u'\n')
         except:
             try:
                 sys.exc_clear()
             except:
                 pass
-        sys.stderr.write(u'[debug] Python version %s - %s' %(platform.python_version(), platform.platform()) + u'\n')
-        sys.stderr.write(u'[debug] Proxy map: ' + str(proxy_handler.proxies) + u'\n')
+        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:
@@ -629,11 +725,42 @@ 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

@@ -1,27 +1,57 @@
+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, DailymotionPlaylistIE
+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
@@ -29,6 +59,7 @@ 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
@@ -36,49 +67,84 @@ 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 .keezmovies import KeezMoviesIE
+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 .myspace import MySpaceIE
 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 .pornhub import PornHubIE
 from .pornotube import PornotubeIE
 from .rbmaradio import RBMARadioIE
 from .redtube import RedTubeIE
 from .ringtv import RingTVIE
+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 .soundcloud import SoundcloudIE, SoundcloudSetIE
+from .slashdot import SlashdotIE
+from .slideshare import SlideshareIE
+from .sohu import SohuIE
+from .soundcloud import SoundcloudIE, SoundcloudSetIE, SoundcloudUserIE
+from .southparkstudios import SouthParkStudiosIE
+from .spankwire import SpankwireIE
 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 .tube8 import Tube8IE
 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 .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 .c56 import C56IE
+from .vk import VKIE
 from .wat import WatIE
+from .websurg import WeBSurgIE
 from .weibo import WeiboIE
 from .wimp import WimpIE
 from .worldstarhiphop import WorldStarHipHopIE
@@ -98,6 +164,7 @@ from .youtube import (
     YoutubeShowIE,
     YoutubeSubscriptionsIE,
     YoutubeRecommendedIE,
+    YoutubeTruncatedURLIE,
     YoutubeWatchLaterIE,
     YoutubeFavouritesIE,
 )
@@ -111,12 +178,14 @@ _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 +1,86 @@
+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.mp4',
+        u'md5': u'72954ea10bc979ab5e2eb288b21425a0',
+        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) or \
+               ee.cause.code != 503:
+                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)
+
+        formats = []
+        for format_id in ('normal', 'hq'):
+            rex = r"var %s_video_file = '(.*?)';" % re.escape(format_id)
+            video_url = self._search_regex(rex, webpage, u'video file URLx',
+                                           fatal=False)
+            if not video_url:
+                continue
+            formats.append({
+                'format_id': format_id,
+                'url': video_url,
+            })
+        if not formats:
+            raise ExtractorError(u'Cannot find any video format!')
+        video_title = self._og_search_title(webpage)
+        video_description = self._og_search_description(webpage)
+
+        return {
+            '_type': 'video',
+            'id':  video_id,
+            'formats': formats,
+            '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

@@ -46,6 +46,8 @@ class ArchiveOrgIE(InfoExtractor):
             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',
@@ -61,7 +63,6 @@ class ArchiveOrgIE(InfoExtractor):
             info['thumbnail'] = thumbnail
 
         # TODO: Remove when #980 has been merged
-        info['url'] = formats[-1]['url']
-        info['ext'] = determine_ext(formats[-1]['url'])
+        info.update(formats[-1])
 
         return info

youtube_dl/extractor/arte.py

@@ -1,3 +1,4 @@
+# encoding: utf-8
 import re
 import json
 import xml.etree.ElementTree
@@ -7,15 +8,15 @@ from ..utils import (
     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/(?P<lang>fr|de)/(?:(?:sendungen|emissions)/)?(?P<id>.*?)/(?P<name>.*?)(\?.*)?'
     _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$'
@@ -24,7 +25,7 @@ class ArteTvIE(InfoExtractor):
 
     @classmethod
     def suitable(cls, url):
-        return any(re.match(regex, url) for regex in (cls._EMISSION_URL, cls._VIDEOS_URL, cls._LIVEWEB_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
@@ -55,14 +56,6 @@ 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:
-            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 self._extract_emission(url, video_id, lang)
-
         mobj = re.match(self._VIDEOS_URL, url)
         if mobj is not None:
             id = mobj.group('id')
@@ -80,49 +73,6 @@ class ArteTvIE(InfoExtractor):
             # self.extractLiveStream(url)
             # return
 
-    def _extract_emission(self, url, video_id, lang):
-        """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.get('VDE'),
-                     'upload_date': unified_strdate(player_info['VDA'].split(' ')[0]),
-                     'thumbnail': player_info['programImage'],
-                     'ext': 'flv',
-                     }
-
-        formats = player_info['VSR'].values()
-        def _match_lang(f):
-            # 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)
-        # We order the formats by quality
-        formats = sorted(formats, key=lambda f: int(f['height']))
-        # Prefer videos without subtitles in the same language
-        formats = sorted(formats, key=lambda f: re.match(r'VO(F|A)-STM\1', f['versionCode']) is None)
-        # 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, lang):
         """Extract from videos.arte.tv"""
         ref_xml_url = url.replace('/videos/', '/do_delegate/videos/')
@@ -172,3 +122,130 @@ class ArteTvIE(InfoExtractor):
                 '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'),
+        }
+
+        all_formats = player_info['VSR'].values()
+        # Some formats use the m3u8 protocol
+        all_formats = list(filter(lambda f: f.get('videoFormat') != 'M3U8', all_formats))
+        def _match_lang(f):
+            if f.get('versionCode') is None:
+                return True
+            # Return true if that format is in the language of the url
+            if lang == 'fr':
+                l = 'F'
+            elif lang == 'de':
+                l = 'A'
+            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, all_formats)
+        formats = list(formats) # in python3 filter returns an iterator
+        if not formats:
+            # Some videos are only available in the 'Originalversion'
+            # they aren't tagged as being in French or German
+            if all(f['versionCode'] == 'VO' for f in all_formats):
+                formats = all_formats
+            else:
+                raise ExtractorError(u'The formats list is empty')
+        # We order the formats by quality
+        if re.match(r'[A-Z]Q', formats[0]['quality']) is not None:
+            sort_key = lambda f: ['HQ', 'MQ', 'EQ', 'SQ'].index(f['quality'])
+        else:
+            sort_key = lambda f: int(f.get('height',-1))
+        formats = sorted(formats, key=sort_key)
+        # 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):
+            quality = format_info['quality']
+            m_quality = re.match(r'\w*? - (\d*)p', quality)
+            if m_quality is not None:
+                quality = m_quality.group(1)
+            if format_info.get('versionCode') is not None:
+                format_id = u'%s-%s' % (quality, format_info['versionCode'])
+            else:
+                format_id = quality
+            info = {
+                'format_id': format_id,
+                'format_note': format_info.get('versionLibelle'),
+                '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]
+
+        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 / Agence 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/bliptv.py

@@ -115,7 +115,7 @@ class BlipTVIE(InfoExtractor):
                 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,

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/brightcove.py

@@ -1,3 +1,5 @@
+# encoding: utf-8
+
 import re
 import json
 import xml.etree.ElementTree
@@ -7,6 +9,8 @@ from ..utils import (
     compat_urllib_parse,
     find_xpath_attr,
     compat_urlparse,
+
+    ExtractorError,
 )
 
 class BrightcoveIE(InfoExtractor):
@@ -14,8 +18,30 @@ class BrightcoveIE(InfoExtractor):
     _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'
 
-    # There is a test for Brigtcove in GenericIE, that way we test both the download
-    # and the detection of videos, and we don't have to find an URL that is always valid
+    _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):
@@ -23,6 +49,13 @@ class BrightcoveIE(InfoExtractor):
         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'],
@@ -65,22 +98,37 @@ class BrightcoveIE(InfoExtractor):
         playlist_info = self._download_webpage(self._PLAYLIST_URL_TEMPLATE % player_key,
                                                player_key, u'Downloading playlist information')
 
-        playlist_info = json.loads(playlist_info)['videoList']
+        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):
-        renditions = video_info['renditions']
-        renditions = sorted(renditions, key=lambda r: r['size'])
-        best_format = renditions[-1]
+        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'),
+        }
 
-        return {'id': video_info['id'],
-                'title': video_info['displayName'],
+        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',
-                'description': video_info.get('shortDescription'),
-                'thumbnail': video_info.get('videoStillURL') or video_info.get('thumbnailURL'),
-                'uploader': video_info.get('publisherName'),
-                }
+            })
+        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

@@ -12,8 +12,8 @@ class C56IE(InfoExtractor):
 
     _TEST ={
         u'url': u'http://www.56.com/u39/v_OTM0NDA3MTY.html',
-        u'file': u'93440716.mp4',
-        u'md5': u'9dc07b5c8e978112a6441f9e75d2b59e',
+        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',
         },

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

@@ -1,3 +1,4 @@
+# encoding: utf-8
 import re
 import xml.etree.ElementTree
 
@@ -5,24 +6,29 @@ from .common import InfoExtractor
 from ..utils import unified_strdate
 
 class CanalplusIE(InfoExtractor):
-    _VALID_URL = r'https?://www\.canalplus\.fr/.*?\?vid=(?P<id>\d+)'
+    _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-divertissement/pid3351-c-le-petit-journal.html?vid=889861',
-        u'file': u'889861.flv',
-        u'md5': u'590a888158b5f0d6832f84001fbf3e99',
+        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'Le Petit Journal 20/06/13 - La guerre des drone',
-            u'upload_date': u'20130620',
+            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,
         },
-        u'skip': u'Requires rtmpdump'
     }
 
     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')
@@ -43,4 +49,6 @@ class CanalplusIE(InfoExtractor):
                 '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 +1,88 @@
+# 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)
+        url = self._html_search_regex(r'\'streamer\': \'(?P<url>[^\']+)\'', playerdata, u'url')
+
+        sd_file = self._html_search_regex(r'\'file\': \'(?P<sd_file>[^\']+)\'', playerdata, u'sd_file')
+        hd_file = self._html_search_regex(r'\'?file\'?: "(?P<hd_file>[^"]+)"', playerdata, u'hd_file')
+        video_thumbnail = self._html_search_regex(r'\'image\': \'(?P<thumbnail>[^\']+)\'', playerdata, u'thumbnail', fatal=False)
+
+        formats = [
+            {
+                'url': url,
+                'play_path': 'mp4:' + sd_file,
+                'ext': 'flv',
+                'format': 'sd',
+                'format_id': 'sd',
+            },
+            {
+                'url': url,
+                'play_path': 'mp4:' + hd_file,
+                'ext': 'flv',
+                'format': 'hd',
+                'format_id': 'hd',
+            },
+        ]
+
+        return {
+            'id': video_id,
+            'title': video_title,
+            'formats': formats,
+            'description': video_description,
+            'upload_date': video_date,
+            'thumbnail': video_thumbnail,
+        }

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

@@ -4,6 +4,7 @@ import xml.etree.ElementTree
 from .common import InfoExtractor
 from ..utils import (
     compat_urllib_parse_urlparse,
+    determine_ext,
 
     ExtractorError,
 )
@@ -12,7 +13,7 @@ from ..utils import (
 class CollegeHumorIE(InfoExtractor):
     _VALID_URL = r'^(?:https?://)?(?:www\.)?collegehumor\.com/(video|embed|e)/(?P<videoid>[0-9]+)/?(?P<shorttitle>.*)$'
 
-    _TEST = {
+    _TESTS = [{
         u'url': u'http://www.collegehumor.com/video/6902724/comic-con-cosplay-catastrophe',
         u'file': u'6902724.mp4',
         u'md5': u'1264c12ad95dca142a9f0bf7968105a0',
@@ -20,7 +21,16 @@ class CollegeHumorIE(InfoExtractor):
             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)
@@ -49,25 +59,29 @@ class CollegeHumorIE(InfoExtractor):
             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'
-        manifestXml = self._download_webpage(manifest_url, video_id,
-                                             u'Downloading XML manifest',
-                                             u'Unable to download video info XML')
+        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(info['thumbnail'])
-
-        info['url'] = url_pr.scheme + '://' + url_pr.netloc + video_id[:-2].replace('.csmil','').replace(',','')
-        info['ext'] = 'mp4'
-        return [info]
+        return info

youtube_dl/extractor/comedycentral.py

@@ -51,12 +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
@@ -64,11 +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)
@@ -155,40 +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

@@ -14,6 +14,8 @@ from ..utils import (
     clean_html,
     compiled_regex_type,
     ExtractorError,
+    RegexNotFoundError,
+    sanitize_filename,
     unescapeHTML,
 )
 
@@ -35,6 +37,8 @@ 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)
@@ -47,12 +51,28 @@ 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 the format_id, width, height.
+                                and format_note fields if missing.
+                    * format_id A short description of the format
+                                ("mp4_h264_opus" or "19")
+                    * format_note Additional info about the format
+                                ("3D" or "DASH video")
+                    * 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.
@@ -77,7 +97,13 @@ 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):
@@ -107,6 +133,11 @@ 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]
@@ -122,7 +153,7 @@ 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) """
@@ -133,12 +164,17 @@ class InfoExtractor(object):
 
         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()
@@ -147,6 +183,17 @@ class InfoExtractor(object):
             self.to_screen(u'Dumping request to ' + url)
             dump = base64.b64encode(webpage_bytes).decode('ascii')
             self._downloader.to_screen(dump)
+        if self._downloader.params.get('write_pages', False):
+            try:
+                url = url_or_request.get_full_url()
+            except AttributeError:
+                url = url_or_request
+            raw_filename = ('%s_%s.dump' % (video_id, url))
+            filename = sanitize_filename(raw_filename, restricted=True)
+            self.to_screen(u'Saving request to ' + filename)
+            with open(filename, 'wb') as outf:
+                outf.write(webpage_bytes)
+
         content = webpage_bytes.decode(encoding, 'replace')
         return (content, urlh)
 
@@ -197,7 +244,7 @@ class InfoExtractor(object):
         Perform a regex search on the given string, using a single or a list of
         patterns returning the first matching group.
         In case of failure return a default value or raise a WARNING or a
-        ExtractorError, depending on fatal, specifying the field name.
+        RegexNotFoundError, depending on fatal, specifying the field name.
         """
         if isinstance(pattern, (str, compat_str, compiled_regex_type)):
             mobj = re.search(pattern, string, flags)
@@ -217,7 +264,7 @@ class InfoExtractor(object):
         elif default is not None:
             return default
         elif fatal:
-            raise ExtractorError(u'Unable to extract %s' % _name)
+            raise RegexNotFoundError(u'Unable to extract %s' % _name)
         else:
             self._downloader.report_warning(u'unable to extract %s; '
                 u'please report this issue on http://yt-dl.org/bug' % _name)
@@ -283,10 +330,19 @@ class InfoExtractor(object):
     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 _og_search_video_url(self, html, name='video url', secure=True, **kargs):
+        regexes = [self._og_regex('video')]
+        if secure: regexes.insert(0, self._og_regex('video:secure_url'))
+        return self._html_search_regex(regexes, 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):
     """
@@ -325,7 +381,7 @@ 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/dailymotion.py

@@ -3,28 +3,78 @@ import json
 import itertools
 
 from .common import InfoExtractor
+from .subtitles import SubtitlesInfoExtractor
+
 from ..utils import (
     compat_urllib_request,
+    compat_str,
     get_element_by_attribute,
     get_element_by_id,
+    orderedSet,
 
     ExtractorError,
 )
 
-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')
+        request.add_header('Cookie', 'ff=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\""
+
+    _FORMATS = [
+        (u'stream_h264_ld_url', u'ld'),
+        (u'stream_h264_url', u'standard'),
+        (u'stream_h264_hq_url', u'hq'),
+        (u'stream_h264_hd_url', u'hd'),
+        (u'stream_h264_hd1080_url', u'hd180'),
+    ]
+
+    _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',
+        },
+        # age-restricted video
+        {
+            u'url': u'http://www.dailymotion.com/video/xyh2zz_leanna-decker-cyber-girl-of-the-year-desires-nude-playboy-plus_redband',
+            u'file': u'xyh2zz.mp4',
+            u'md5': u'0d667a7b9cebecc3c89ee93099c4159d',
+            u'info_dict': {
+                u'title': 'Leanna Decker - Cyber Girl Of The Year Desires Nude [Playboy Plus]',
+                u'uploader': 'HotWaves1012',
+                u'age_limit': 18,
+            }
+
         }
-    }
+    ]
 
     def _real_extract(self, url):
         # Extract id and simplified title from URL
@@ -32,20 +82,29 @@ 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)
 
+        # 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')
+
         video_uploader = self._search_regex([r'(?im)<span class="owner[^\"]+?">[^<]+?<a [^>]+?>([^<]+?)</a>',
                                              # Looking for official user
                                              r'<(?:span|a) .*?rel="author".*?>([^<]+?)</'],
-                                            webpage, 'video uploader')
+                                            webpage, 'video uploader', fatal=False)
+        age_limit = self._rta_search(webpage)
 
         video_upload_date = None
         mobj = re.search(r'<div class="[^"]*uploaded_cont[^"]*" title="[^"]*">([0-9]{2})-([0-9]{2})-([0-9]{4})</div>', webpage)
@@ -55,56 +114,115 @@ class DailymotionIE(InfoExtractor):
         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')
+        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:
+        formats = []
+        for (key, format_id) in self._FORMATS:
+            video_url = info.get(key)
+            if video_url is not None:
+                m_size = re.search(r'H264-(\d+)x(\d+)', video_url)
+                if m_size is not None:
+                    width, height = m_size.group(1), m_size.group(2)
+                else:
+                    width, height = None, None
+                formats.append({
+                    'url': video_url,
+                    'ext': 'mp4',
+                    'format_id': format_id,
+                    'width': width,
+                    'height': height,
+                })
+        if not formats:
             raise ExtractorError(u'Unable to extract video URL')
-        video_url = info[max_quality]
 
-        return [{
+        # 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,
+            'formats': formats,
             'uploader': video_uploader,
             'upload_date':  video_upload_date,
             'title':    self._og_search_title(webpage),
-            'ext':      video_extension,
-            'thumbnail': info['thumbnail_url']
-        }]
+            'subtitles':    video_subtitles,
+            'thumbnail': info['thumbnail_url'],
+            'age_limit': age_limit,
+        }
+
+    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(InfoExtractor):
+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 _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
-        playlist_id =  mobj.group('id')
+    def _extract_entries(self, id):
         video_ids = []
-
         for pagenum in itertools.count(1):
-            webpage = self._download_webpage('https://www.dailymotion.com/playlist/%s/%s' % (playlist_id, pagenum),
-                                             playlist_id, u'Downloading page %s' % pagenum)
+            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="(.+?)" data-ext-id', playlist_el))
+            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)
 
-        entries = [self.url_result('http://www.dailymotion.com/video/%s' % video_id, 'Dailymotion')
-                   for video_id in video_ids]
         return {'_type': 'playlist',
                 'id': playlist_id,
                 'title': get_element_by_id(u'playlist_name', webpage),
-                'entries': entries,
+                '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/dreisat.py

@@ -54,6 +54,7 @@ class DreiSatIE(InfoExtractor):
             '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,
@@ -79,7 +80,6 @@ class DreiSatIE(InfoExtractor):
         }
 
         # TODO: Remove when #980 has been merged
-        info['url'] = formats[-1]['url']
-        info['ext'] = determine_ext(formats[-1]['url'])
+        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/eighttracks.py

@@ -101,7 +101,7 @@ class EightTracksIE(InfoExtractor):
         first_url = 'http://8tracks.com/sets/%s/play?player=sm&mix_id=%s&format=jsonh' % (session, mix_id)
         next_url = first_url
         res = []
-        for i in itertools.count():
+        for i in range(track_count):
             api_json = self._download_webpage(next_url, playlist_id,
                 note=u'Downloading song information %s/%s' % (str(i+1), track_count),
                 errnote=u'Failed to download song information')
@@ -116,7 +116,5 @@ class EightTracksIE(InfoExtractor):
                 'ext': 'm4a',
             }
             res.append(info)
-            if api_data['set']['at_last_track']:
-                break
             next_url = 'http://8tracks.com/sets/%s/next?player=sm&mix_id=%s&format=jsonh&track_id=%s' % (session, mix_id, track_data['id'])
         return res

youtube_dl/extractor/exfm.py

@@ -11,14 +11,14 @@ class ExfmIE(InfoExtractor):
     _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'url': u'http://ex.fm/song/eh359',
+            u'file': u'44216187.mp3',
+            u'md5': u'e45513df5631e6d760970b14cc0c11e7',
             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"title": u"Test House \"Love Is Not Enough\" (Extended Mix) DeadJournalist Exclusive",
+                u"uploader": u"deadjournalist",
+                u'upload_date': u'20120424',
+                u'description': u'Test House \"Love Is Not Enough\" (Extended Mix) DeadJournalist Exclusive',
             },
             u'note': u'Soundcloud song',
         },

youtube_dl/extractor/facebook.py

@@ -19,7 +19,8 @@ class FacebookIE(InfoExtractor):
     """Information Extractor for Facebook"""
 
     _VALID_URL = r'^(?:https?://)?(?:\w+\.)?facebook\.com/(?:video/video|photo)\.php\?(?:.*?)v=(?P<ID>\d+)(?:.*)'
-    _LOGIN_URL = 'https://login.facebook.com/login.php?m&next=http%3A%2F%2Fm.facebook.com%2Fhome.php&'
+    _LOGIN_URL = 'https://www.facebook.com/login.php?next=http%3A%2F%2Ffacebook.com%2Fhome.php&login_attempt=1'
+    _CHECKPOINT_URL = 'https://www.facebook.com/checkpoint/?next=http%3A%2F%2Ffacebook.com%2Fhome.php&_fb_noscript=1'
     _NETRC_MACHINE = 'facebook'
     IE_NAME = u'facebook'
     _TEST = {
@@ -36,50 +37,56 @@ class FacebookIE(InfoExtractor):
         """Report attempt to log in."""
         self.to_screen(u'Logging in')
 
-    def _real_initialize(self):
-        if self._downloader is None:
-            return
-
-        useremail = 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:
-            useremail = 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:
-                    useremail = 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
-
+    def _login(self):
+        (useremail, password) = self._get_login_info()
         if useremail is None:
             return
 
-        # Log in
+        login_page_req = compat_urllib_request.Request(self._LOGIN_URL)
+        login_page_req.add_header('Cookie', 'locale=en_US')
+        self.report_login()
+        login_page = self._download_webpage(login_page_req, None, note=False,
+            errnote=u'Unable to download login page')
+        lsd = self._search_regex(r'"lsd":"(\w*?)"', login_page, u'lsd')
+        lgnrnd = self._search_regex(r'name="lgnrnd" value="([^"]*?)"', login_page, u'lgnrnd')
+
         login_form = {
             'email': useremail,
             'pass': password,
-            'login': 'Log+In'
+            'lsd': lsd,
+            'lgnrnd': lgnrnd,
+            'next': 'http://facebook.com/home.php',
+            'default_persistent': '0',
+            'legacy_return': '1',
+            'timezone': '-60',
+            'trynum': '1',
             }
         request = compat_urllib_request.Request(self._LOGIN_URL, compat_urllib_parse.urlencode(login_form))
+        request.add_header('Content-Type', 'application/x-www-form-urlencoded')
         try:
-            self.report_login()
             login_results = compat_urllib_request.urlopen(request).read()
             if re.search(r'<form(.*)name="login"(.*)</form>', login_results) is not None:
                 self._downloader.report_warning(u'unable to log in: bad username/password, or exceded login rate limit (~3/min). Check credentials or wait.')
                 return
+
+            check_form = {
+                'fb_dtsg': self._search_regex(r'"fb_dtsg":"(.*?)"', login_results, u'fb_dtsg'),
+                'nh': self._search_regex(r'name="nh" value="(\w*?)"', login_results, u'nh'),
+                'name_action_selected': 'dont_save',
+                'submit[Continue]': self._search_regex(r'<input value="(.*?)" name="submit\[Continue\]"', login_results, u'continue'),
+            }
+            check_req = compat_urllib_request.Request(self._CHECKPOINT_URL, compat_urllib_parse.urlencode(check_form))
+            check_req.add_header('Content-Type', 'application/x-www-form-urlencoded')
+            check_response = compat_urllib_request.urlopen(check_req).read()
+            if re.search(r'id="checkpointSubmitButton"', check_response) is not None:
+                self._downloader.report_warning(u'Unable to confirm login, you have to login in your brower and authorize the login.')
         except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
             self._downloader.report_warning(u'unable to log in: %s' % compat_str(err))
             return
 
+    def _real_initialize(self):
+        self._login()
+
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
         if mobj is None:
@@ -93,7 +100,13 @@ class FacebookIE(InfoExtractor):
         AFTER = '.forEach(function(variable) {swf.addVariable(variable[0], variable[1]);});'
         m = re.search(re.escape(BEFORE) + '(.*?)' + re.escape(AFTER), webpage)
         if not m:
-            raise ExtractorError(u'Cannot parse data')
+            m_msg = re.search(r'class="[^"]*uiInterstitialContent[^"]*"><div>(.*?)</div>', webpage)
+            if m_msg is not None:
+                raise ExtractorError(
+                    u'The video is not available, Facebook said: "%s"' % m_msg.group(1),
+                    expected=True)
+            else:
+                raise ExtractorError(u'Cannot parse data')
         data = dict(json.loads(m.group(1)))
         params_raw = compat_urllib_parse.unquote(data['params'])
         params = json.loads(params_raw)
@@ -106,8 +119,8 @@ 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 +1,58 @@
+# encoding: utf-8
+import re
+import xml.etree.ElementTree
+
+from .common import InfoExtractor
+from ..utils import (
+    determine_ext,
+)
+
+
+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 = self._html_search_regex(r'<p class="Content Copy">(.*?)</p>', webpage, u'description')
+        info = {
+            'id': video_id,
+            'title': self._og_search_title(webpage),
+            'formats': formats,
+            'description': descr,
+            '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',

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/funnyordie.py

@@ -21,17 +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)
-
         info = {
             'id': video_id,
             'url': video_url,
             'ext': 'mp4',
-            'title': title,
+            'title': self._og_search_title(webpage),
             'description': self._og_search_description(webpage),
         }
         return [info]

youtube_dl/extractor/gamespot.py

@@ -1,55 +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/.*-(?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"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)
-        page_id = mobj.group('page_id')
+        page_id = video_id = mobj.group('page_id')
         webpage = self._download_webpage(url, page_id)
-        video_id = self._html_search_regex([r'"og:video" content=".*?\?id=(\d+)"',
-                                            r'http://www\.gamespot\.com/videoembed/(\d+)'],
-                                           webpage, 'video id')
-        data = compat_urllib_parse.urlencode({'id': video_id, 'newplayer': '1'})
-        info_url = 'http://www.gamespot.com/pages/video_player/xml.php?' + data
-        info_xml = self._download_webpage(info_url, video_id)
-        doc = xml.etree.ElementTree.fromstring(info_xml)
-        clip_el = doc.find('./playList/clip')
+        data_video_json = self._search_regex(r'data-video=\'(.*?)\'', webpage, u'data video')
+        data_video = json.loads(unescapeHTML(data_video_json))
 
-        http_urls = [{'url': node.find('filePath').text,
-                      'rate': int(node.find('rate').text)}
-            for node in clip_el.find('./httpURI')]
-        best_quality = sorted(http_urls, key=lambda f: f['rate'])[-1]
-        video_url = best_quality['url']
-        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/generic.py

@@ -8,11 +8,15 @@ 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'.*'
@@ -21,23 +25,33 @@ class GenericIE(InfoExtractor):
         {
             u'url': u'http://www.hodiho.fr/2013/02/regis-plante-sa-jeep.html',
             u'file': u'13601338388002.mp4',
-            u'md5': u'85b90ccc9d73b4acd9138d3af4c27f89',
+            u'md5': u'6e15c93721d7ec9e9ca3fdbf07982cfd',
             u'info_dict': {
                 u"uploader": u"www.hodiho.fr",
                 u"title": u"R\u00e9gis plante sa Jeep"
             }
         },
+        # embedded vimeo video
         {
-            u'url': u'http://www.8tv.cat/8aldia/videos/xavier-sala-i-martin-aquesta-tarda-a-8-al-dia/',
-            u'file': u'2371591881001.mp4',
-            u'md5': u'9e80619e0a94663f0bdc849b4566af19',
-            u'note': u'Test Brightcove downloads and detection in GenericIE',
+            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'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',
+                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",
             }
         },
+        # bandcamp page with custom domain
+        {
+            u'url': u'http://bronyrock.com/track/the-pony-mash',
+            u'file': u'3235767654.mp3',
+            u'info_dict': {
+                u'title': u'The Pony Mash',
+                u'uploader': u'M_Pallante',
+            },
+            u'skip': u'There is a limit of 200 free downloads / month for the test song',
+        },
     ]
 
     def report_download_webpage(self, video_id):
@@ -107,8 +121,18 @@ 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:
@@ -116,16 +140,37 @@ 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 BrigthCove:
-        m_brightcove = re.search(r'<object.+?class=([\'"]).*?BrightcoveExperience.*?\1.+?</object>', webpage, re.DOTALL)
+        # 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[^>]+?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')
+
+        # Look for embedded YouTube player
+        mobj = re.search(
+            r'<iframe[^>]+?src="(https?://(?:www\.)?youtube.com/embed/.+?)"', webpage)
+        if mobj:
+            surl = unescapeHTML(mobj.group(1))
+            return self.url_result(surl, 'Youtube')
+
+        # Look for Bandcamp pages with custom domain
+        mobj = re.search(r'<meta property="og:url"[^>]*?content="(.*?bandcamp\.com.*?)"', webpage)
+        if mobj is not None:
+            burl = unescapeHTML(mobj.group(1))
+            return self.url_result(burl, 'Bandcamp')
+
         # Start with something easy: JW Player in SWFObject
         mobj = re.search(r'flashvars: [\'"](?:.*&)?file=(http[^\'"&]*)', webpage)
         if mobj is None:
@@ -145,15 +190,19 @@ 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

@@ -40,8 +40,10 @@ class GooglePlusIE(InfoExtractor):
         self.report_extraction(video_id)
 
         # Extract update date
-        upload_date = self._html_search_regex('title="Timestamp">(.*?)</a>',
-            webpage, u'upload date', fatal=False)
+        upload_date = self._html_search_regex(
+            r'''(?x)<a.+?class="o-U-s\s[^"]+"\s+style="display:\s*none"\s*>
+                    ([0-9]{4}-[0-9]{2}-[0-9]{2})</a>''',
+            webpage, u'upload date', fatal=False, flags=re.VERBOSE)
         if upload_date:
             # Convert timestring to a format suitable for filename
             upload_date = datetime.datetime.strptime(upload_date, "%Y-%m-%d")
@@ -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"
         }
     }
 

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

@@ -13,7 +13,7 @@ class IGNIE(InfoExtractor):
     Some videos of it.ign.com are also supported
     """
 
-    _VALID_URL = r'https?://.+?\.ign\.com/(?:videos|show_videos)(/.+)?/(?P<name_or_id>.+)'
+    _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'
@@ -21,15 +21,39 @@ class IGNIE(InfoExtractor):
                        r'id="my_show_video">.*?<p>(.*?)</p>',
                        ]
 
-    _TEST = {
-        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',
-        }
-    }
+    _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="(.+?)"',
@@ -41,7 +65,18 @@ class IGNIE(InfoExtractor):
     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,
@@ -68,7 +103,7 @@ class IGNIE(InfoExtractor):
 class OneUPIE(IGNIE):
     """Extractor for 1up.com, it uses the ign videos system."""
 
-    _VALID_URL = r'https?://gamevideos.1up.com/video/id/(?P<name_or_id>.+)'
+    _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>'
@@ -83,6 +118,9 @@ class OneUPIE(IGNIE):
         }
     }
 
+    # Override IGN tests
+    _TESTS = []
+
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
         id = mobj.group('name_or_id')

youtube_dl/extractor/instagram.py

@@ -26,7 +26,7 @@ class InstagramIE(InfoExtractor):
 
         return [{
             'id':        video_id,
-            'url':       self._og_search_video_url(webpage),
+            'url':       self._og_search_video_url(webpage, secure=False),
             'ext':       'mp4',
             'title':     u'Video by %s' % uploader_id,
             'thumbnail': self._og_search_thumbnail(webpage),

youtube_dl/extractor/internetvideoarchive.py

@@ -0,0 +1,84 @@
+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': 153,
+        },
+    }
+
+    @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'])
+
+        return {
+            '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']),
+        }

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

@@ -21,8 +21,10 @@ class KankanIE(InfoExtractor):
         video_id = mobj.group('id')
         webpage = self._download_webpage(url, video_id)
 
-        title = self._search_regex(r'G_TITLE=[\'"](.+?)[\'"]', webpage, u'video title')
-        gcid = self._search_regex(r'lurl:[\'"]http://.+?/.+?/(.+?)/', webpage, u'gcid')
+        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')

youtube_dl/extractor/keezmovies.py

@@ -0,0 +1,61 @@
+import os
+import re
+
+from .common import InfoExtractor
+from ..utils import (
+    compat_urllib_parse_urlparse,
+    compat_urllib_request,
+    compat_urllib_parse,
+)
+from ..aes import (
+    aes_decrypt_text
+)
+
+class KeezMoviesIE(InfoExtractor):
+    _VALID_URL = r'^(?:https?://)?(?:www\.)?(?P<url>keezmovies\.com/video/.+?(?P<videoid>[0-9]+))'
+    _TEST = {
+        u'url': u'http://www.keezmovies.com/video/petite-asian-lady-mai-playing-in-bathtub-1214711',
+        u'file': u'1214711.mp4',
+        u'md5': u'6e297b7e789329923fcf83abb67c9289',
+        u'info_dict': {
+            u"title": u"Petite Asian Lady Mai Playing In Bathtub",
+            u"age_limit": 18,
+        }
+    }
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        video_id = mobj.group('videoid')
+        url = 'http://www.' + mobj.group('url')
+
+        req = compat_urllib_request.Request(url)
+        req.add_header('Cookie', 'age_verified=1')
+        webpage = self._download_webpage(req, video_id)
+
+        # embedded video
+        mobj = re.search(r'href="([^"]+)"></iframe>', webpage)
+        if mobj:
+            embedded_url = mobj.group(1)
+            return self.url_result(embedded_url)
+
+        video_title = self._html_search_regex(r'<h1 [^>]*>([^<]+)', webpage, u'title')
+        video_url = compat_urllib_parse.unquote(self._html_search_regex(r'video_url=(.+?)&amp;', webpage, u'video_url'))
+        if webpage.find('encrypted=true')!=-1:
+            password = self._html_search_regex(r'video_title=(.+?)&amp;', webpage, u'password')
+            video_url = aes_decrypt_text(video_url, password, 32).decode('utf-8')
+        path = compat_urllib_parse_urlparse( video_url ).path
+        extension = os.path.splitext( path )[1][1:]
+        format = path.split('/')[4].split('_')[:2]
+        format = "-".join( format )
+
+        age_limit = self._rta_search(webpage)
+
+        return {
+            'id': video_id,
+            'title': video_title,
+            'url': video_url,
+            'ext': extension,
+            'format': format,
+            'format_id': format,
+            'age_limit': age_limit,
+        }

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/livestream.py

@@ -2,7 +2,12 @@ import re
 import json
 
 from .common import InfoExtractor
-from ..utils import compat_urllib_parse_urlparse, compat_urlparse
+from ..utils import (
+    compat_urllib_parse_urlparse,
+    compat_urlparse,
+    get_meta_content,
+    ExtractorError,
+)
 
 
 class LivestreamIE(InfoExtractor):
@@ -35,10 +40,9 @@ class LivestreamIE(InfoExtractor):
 
         if video_id is None:
             # This is an event page:
-            api_url = self._search_regex(r'event_design_eventId: \'(.+?)\'',
-                                         webpage, 'api url')
-            info = json.loads(self._download_webpage(api_url, event_name,
-                                                     u'Downloading event info'))
+            config_json = self._search_regex(r'window.config = ({.*?});',
+                webpage, u'window config')
+            info = json.loads(config_json)['event']
             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'])

youtube_dl/extractor/metacafe.py

@@ -20,10 +20,12 @@ 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'
-    _TESTS = [{
+    _TESTS = [
+    # Youtube video
+    {
         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",
+        u"file":  u"_aUehQsCQtM.mp4",
         u"info_dict": {
             u"upload_date": u"20090102",
             u"title": u"The Electric Company | \"Short I\" | PBS KIDS GO!",
@@ -32,15 +34,42 @@ class MetacafeIE(InfoExtractor):
             u"uploader_id": u"PBS"
         }
     },
+    # Normal metacafe video
+    {
+        u'url': u'http://www.metacafe.com/watch/11121940/news_stuff_you_wont_do_with_your_playstation_4/',
+        u'md5': u'6e0bca200eaad2552e6915ed6fd4d9ad',
+        u'info_dict': {
+            u'id': u'11121940',
+            u'ext': u'mp4',
+            u'title': u'News: Stuff You Won\'t Do with Your PlayStation 4',
+            u'uploader': u'ign',
+            u'description': u'Sony released a massive FAQ on the PlayStation Blog detailing the PS4\'s capabilities and limitations.',
+        },
+    },
+    # AnyClip video
     {
         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"
-        }
-    }]
+            u"description": u"md5:38c711dd98f5bb87acf973d573442e67",
+        },
+    },
+    # age-restricted video
+    {
+        u'url': u'http://www.metacafe.com/watch/5186653/bbc_internal_christmas_tape_79_uncensored_outtakes_etc/',
+        u'md5': u'98dde7c1a35d02178e8ab7560fe8bd09',
+        u'info_dict': {
+            u'id': u'5186653',
+            u'ext': u'mp4',
+            u'title': u'BBC INTERNAL Christmas Tape \'79 - UNCENSORED Outtakes, Etc.',
+            u'uploader': u'Dwayne Pipe',
+            u'description': u'md5:950bf4c581e2c059911fa3ffbe377e4b',
+            u'age_limit': 18,
+        },
+    },
+    ]
 
 
     def report_disclaimer(self):
@@ -62,6 +91,7 @@ class MetacafeIE(InfoExtractor):
             'submit': "Continue - I'm over 18",
             }
         request = compat_urllib_request.Request(self._FILTER_POST, compat_urllib_parse.urlencode(disclaimer_form))
+        request.add_header('Content-Type', 'application/x-www-form-urlencoded')
         try:
             self.report_age_confirmation()
             compat_urllib_request.urlopen(request).read()
@@ -83,7 +113,12 @@ class MetacafeIE(InfoExtractor):
 
         # Retrieve video webpage to extract further information
         req = compat_urllib_request.Request('http://www.metacafe.com/watch/%s/' % video_id)
-        req.headers['Cookie'] = 'flashVersion=0;'
+
+        # AnyClip videos require the flashversion cookie so that we get the link
+        # to the mp4 file
+        mobj_an = re.match(r'^an-(.*?)$', video_id)
+        if mobj_an:
+            req.headers['Cookie'] = 'flashVersion=0;'
         webpage = self._download_webpage(req, video_id)
 
         # Extract URL, uploader and title from webpage
@@ -122,9 +157,14 @@ class MetacafeIE(InfoExtractor):
         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\("channel","([^"]+)"\);',
+                r'submitter=(.*?);|googletag\.pubads\(\)\.setTargeting\("(?:channel|submiter)","([^"]+)"\);',
                 webpage, u'uploader nickname', fatal=False)
 
+        if re.search(r'"contentRating":"restricted"', webpage) is not None:
+            age_limit = 18
+        else:
+            age_limit = 0
+
         return {
             '_type':    'video',
             'id':       video_id,
@@ -134,4 +174,5 @@ class MetacafeIE(InfoExtractor):
             'upload_date':  None,
             'title':    video_title,
             'ext':      video_ext,
+            'age_limit': age_limit,
         }

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

@@ -54,46 +54,57 @@ class MTVIE(InfoExtractor):
     def _get_thumbnail_url(self, uri, itemdoc):
         return 'http://mtv.mtvnimages.com/uri/' + uri
 
-    def _extract_video_url(self, metadataXml):
+    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')
 
-        # 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']
-            rtmp_video_url = rendition.find('./src').text
-        except KeyError:
-            raise ExtractorError('Invalid rendition field.')
-        video_url = self._transform_rtmp_url(rtmp_video_url)
-        return {'ext': ext, 'url': video_url, 'format': format}
+        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']
+        # Remove the templates, like &device={device}
+        mediagen_url = re.sub(r'&[^=]*?={.*?}(?=(&|$))', u'', mediagen_url)
         if 'acceptMethods' not in mediagen_url:
             mediagen_url += '&acceptMethods=fms'
         mediagen_page = self._download_webpage(mediagen_url, video_id,
                                                u'Downloading video urls')
-        video_info = self._extract_video_url(mediagen_page)
 
         description_node = itemdoc.find('description')
         if description_node is not None:
-            description = description_node.text
+            description = description_node.text.strip()
         else:
             description = None
-        video_info.update({'title': itemdoc.find('title').text,
-                           'id': video_id,
-                           'thumbnail': self._get_thumbnail_url(uri, itemdoc),
-                           'description': description,
-                           })
-        return video_info
+
+        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)

youtube_dl/extractor/myspace.py

@@ -0,0 +1,48 @@
+import re
+import json
+
+from .common import InfoExtractor
+from ..utils import (
+    compat_str,
+)
+
+
+class MySpaceIE(InfoExtractor):
+    _VALID_URL = r'https?://myspace\.com/([^/]+)/video/[^/]+/(?P<id>\d+)'
+
+    _TEST = {
+        u'url': u'https://myspace.com/coldplay/video/viva-la-vida/100008689',
+        u'info_dict': {
+            u'id': u'100008689',
+            u'ext': u'flv',
+            u'title': u'Viva La Vida',
+            u'description': u'The official Viva La Vida video, directed by Hype Williams',
+            u'uploader': u'Coldplay',
+            u'uploader_id': u'coldplay',
+        },
+        u'params': {
+            # rtmp download
+            u'skip_download': True,
+        },
+    }
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        video_id = mobj.group('id')
+        webpage = self._download_webpage(url, video_id)
+        context = json.loads(self._search_regex(r'context = ({.*?});', webpage,
+            u'context'))
+        video = context['video']
+        rtmp_url, play_path = video['streamUrl'].split(';', 1)
+
+        return {
+            'id': compat_str(video['mediaId']),
+            'title': video['title'],
+            'url': rtmp_url,
+            'play_path': play_path,
+            'ext': 'flv',
+            'description': video['description'],
+            'thumbnail': video['imageUrl'],
+            'uploader': video['artistName'],
+            'uploader_id': video['artistUsername'],
+        }

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/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'tab0"[^>]*?>(.*?)</td>',
+            webpage, u'playlist title', flags=re.DOTALL).lower().capitalize()
+
+        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],
+        }

youtube_dl/extractor/nowvideo.py

@@ -0,0 +1,46 @@
+import re
+
+from .common import InfoExtractor
+from ..utils import compat_urlparse
+
+
+class NowVideoIE(InfoExtractor):
+    _VALID_URL = r'(?:https?://)?(?:www\.)?nowvideo\.ch/video/(?P<id>\w+)'
+    _TEST = {
+        u'url': u'http://www.nowvideo.ch/video/0mw0yow7b6dxa',
+        u'file': u'0mw0yow7b6dxa.flv',
+        u'md5': u'f8fbbc8add72bd95b7850c6a02fc8817',
+        u'info_dict': {
+            u"title": u"youtubedl test video _BaW_jenozKc.mp4"
+        }
+    }
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+
+        video_id = mobj.group('id')
+        webpage_url = 'http://www.nowvideo.ch/video/' + video_id
+        embed_url = 'http://embed.nowvideo.ch/embed.php?v=' + video_id
+        webpage = self._download_webpage(webpage_url, video_id)
+        embed_page = self._download_webpage(embed_url, video_id,
+            u'Downloading embed page')
+
+        self.report_extraction(video_id)
+
+        video_title = self._html_search_regex(r'<h4>(.*)</h4>',
+            webpage, u'video title')
+
+        video_key = self._search_regex(r'var fkzd="(.*)";',
+            embed_page, u'video key')
+
+        api_call = "http://www.nowvideo.ch/api/player.api.php?file={0}&numOfErrors=0&cid=1&key={1}".format(video_id, video_key)
+        api_response = self._download_webpage(api_call, video_id,
+            u'Downloading API page')
+        video_url = compat_urlparse.parse_qs(api_response)[u'url'][0]
+
+        return [{
+            'id':        video_id,
+            'url':       video_url,
+            'ext':       'flv',
+            'title':     video_title,
+        }]

youtube_dl/extractor/ooyala.py

@@ -18,11 +18,15 @@ class OoyalaIE(InfoExtractor):
         },
     }
 
+    @staticmethod
+    def _url_for_embed_code(embed_code):
+        return 'http://player.ooyala.com/player.js?embedCode=%s' % embed_code
+
     def _extract_result(self, info, more_info):
         return {'id': info['embedCode'],
                 'ext': 'mp4',
                 'title': unescapeHTML(info['title']),
-                'url': info['url'],
+                'url': info.get('ipad_url') or info['url'],
                 'description': unescapeHTML(more_info['description']),
                 'thumbnail': more_info['promo'],
                 }
@@ -35,7 +39,9 @@ class OoyalaIE(InfoExtractor):
         mobile_url = self._search_regex(r'mobile_player_url="(.+?)&device="',
                                         player, u'mobile player url')
         mobile_player = self._download_webpage(mobile_url, embedCode)
-        videos_info = self._search_regex(r'eval\("\((\[{.*?stream_redirect.*?}\])\)"\);', mobile_player, u'info').replace('\\"','"')
+        videos_info = self._search_regex(
+            r'var streams=window.oo_testEnv\?\[\]:eval\("\((\[{.*?}\])\)"\);',
+            mobile_player, u'info').replace('\\"','"')
         videos_more_info = self._search_regex(r'eval\("\(({.*?\\"promo\\".*?})\)"', mobile_player, u'more info').replace('\\"','"')
         videos_info = json.loads(videos_info)
         videos_more_info =json.loads(videos_more_info)

youtube_dl/extractor/orf.py

@@ -0,0 +1,54 @@
+# coding: utf-8
+
+import re
+import xml.etree.ElementTree
+import json
+
+from .common import InfoExtractor
+from ..utils import (
+    compat_urlparse,
+    ExtractorError,
+    find_xpath_attr,
+)
+
+class ORFIE(InfoExtractor):
+    _VALID_URL = r'https?://tvthek.orf.at/(programs/.+?/episodes|topics/.+?)/(?P<id>\d+)'
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        playlist_id = mobj.group('id')
+        webpage = self._download_webpage(url, playlist_id)
+
+        flash_xml = self._search_regex('ORF.flashXML = \'(.+?)\'', webpage, u'flash xml')
+        flash_xml = compat_urlparse.parse_qs('xml='+flash_xml)['xml'][0]
+        flash_config = xml.etree.ElementTree.fromstring(flash_xml.encode('utf-8'))
+        playlist_json = self._search_regex(r'playlist\': \'(\[.*?\])\'', webpage, u'playlist').replace(r'\"','"')
+        playlist = json.loads(playlist_json)
+
+        videos = []
+        ns = '{http://tempuri.org/XMLSchema.xsd}'
+        xpath = '%(ns)sPlaylist/%(ns)sItems/%(ns)sItem' % {'ns': ns}
+        webpage_description = self._og_search_description(webpage)
+        for (i, (item, info)) in enumerate(zip(flash_config.findall(xpath), playlist), 1):
+            # Get best quality url
+            rtmp_url = None
+            for q in ['Q6A', 'Q4A', 'Q1A']:
+                video_url = find_xpath_attr(item, '%sVideoUrl' % ns, 'quality', q)
+                if video_url is not None:
+                    rtmp_url = video_url.text
+                    break
+            if rtmp_url is None:
+                raise ExtractorError(u'Couldn\'t get video url: %s' % info['id'])
+            description = self._html_search_regex(
+                r'id="playlist_entry_%s".*?<p>(.*?)</p>' % i, webpage,
+                u'description', default=webpage_description, flags=re.DOTALL)
+            videos.append({
+                '_type': 'video',
+                'id': info['id'],
+                'title': info['title'],
+                'url': rtmp_url,
+                'ext': 'flv',
+                'description': description,
+                })
+
+        return videos

youtube_dl/extractor/pbs.py

@@ -0,0 +1,34 @@
+import re
+import json
+
+from .common import InfoExtractor
+
+
+class PBSIE(InfoExtractor):
+    _VALID_URL = r'https?://video.pbs.org/video/(?P<id>\d+)/?'
+
+    _TEST = {
+        u'url': u'http://video.pbs.org/video/2365006249/',
+        u'file': u'2365006249.mp4',
+        u'md5': 'ce1888486f0908d555a8093cac9a7362',
+        u'info_dict': {
+            u'title': u'A More Perfect Union',
+            u'description': u'md5:ba0c207295339c8d6eced00b7c363c6a',
+            u'duration': 3190,
+        },
+    }
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        video_id = mobj.group('id')
+        info_url = 'http://video.pbs.org/videoInfo/%s?format=json' % video_id
+        info_page = self._download_webpage(info_url, video_id)
+        info =json.loads(info_page)
+        return {'id': video_id,
+                'title': info['title'],
+                'url': info['alternate_encoding']['url'],
+                'ext': 'mp4',
+                'description': info['program'].get('description'),
+                'thumbnail': info.get('image_url'),
+                'duration': info.get('duration'),
+                }

youtube_dl/extractor/pornhub.py

@@ -0,0 +1,69 @@
+import os
+import re
+
+from .common import InfoExtractor
+from ..utils import (
+    compat_urllib_parse_urlparse,
+    compat_urllib_request,
+    compat_urllib_parse,
+    unescapeHTML,
+)
+from ..aes import (
+    aes_decrypt_text
+)
+
+class PornHubIE(InfoExtractor):
+    _VALID_URL = r'^(?:https?://)?(?:www\.)?(?P<url>pornhub\.com/view_video\.php\?viewkey=(?P<videoid>[0-9]+))'
+    _TEST = {
+        u'url': u'http://www.pornhub.com/view_video.php?viewkey=648719015',
+        u'file': u'648719015.mp4',
+        u'md5': u'882f488fa1f0026f023f33576004a2ed',
+        u'info_dict': {
+            u"uploader": u"BABES-COM", 
+            u"title": u"Seductive Indian beauty strips down and fingers her pink pussy",
+            u"age_limit": 18
+        }
+    }
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        video_id = mobj.group('videoid')
+        url = 'http://www.' + mobj.group('url')
+
+        req = compat_urllib_request.Request(url)
+        req.add_header('Cookie', 'age_verified=1')
+        webpage = self._download_webpage(req, video_id)
+
+        video_title = self._html_search_regex(r'<h1 [^>]+>([^<]+)', webpage, u'title')
+        video_uploader = self._html_search_regex(r'<b>From: </b>(?:\s|<[^>]*>)*(.+?)<', webpage, u'uploader', fatal=False)
+        thumbnail = self._html_search_regex(r'"image_url":"([^"]+)', webpage, u'thumbnail', fatal=False)
+        if thumbnail:
+            thumbnail = compat_urllib_parse.unquote(thumbnail)
+
+        video_urls = list(map(compat_urllib_parse.unquote , re.findall(r'"quality_[0-9]{3}p":"([^"]+)', webpage)))
+        if webpage.find('"encrypted":true') != -1:
+            password = self._html_search_regex(r'"video_title":"([^"]+)', webpage, u'password').replace('+', ' ')
+            video_urls = list(map(lambda s: aes_decrypt_text(s, password, 32).decode('utf-8'), video_urls))
+
+        formats = []
+        for video_url in video_urls:
+            path = compat_urllib_parse_urlparse( video_url ).path
+            extension = os.path.splitext( path )[1][1:]
+            format = path.split('/')[5].split('_')[:2]
+            format = "-".join( format )
+            formats.append({
+                'url': video_url,
+                'ext': extension,
+                'format': format,
+                'format_id': format,
+            })
+        formats.sort(key=lambda format: list(map(lambda s: s.zfill(6), format['format'].split('-'))))
+
+        return {
+            'id': video_id,
+            'uploader': video_uploader,
+            'title': video_title,
+            'thumbnail': thumbnail,
+            'formats': formats,
+            'age_limit': 18,
+        }

youtube_dl/extractor/pornotube.py

@@ -16,7 +16,8 @@ class PornotubeIE(InfoExtractor):
         u'md5': u'374dd6dcedd24234453b295209aa69b6',
         u'info_dict': {
             u"upload_date": u"20090708", 
-            u"title": u"Marilyn-Monroe-Bathing"
+            u"title": u"Marilyn-Monroe-Bathing",
+            u"age_limit": 18
         }
     }
 
@@ -38,6 +39,7 @@ class PornotubeIE(InfoExtractor):
         VIDEO_UPLOADED_RE = r'<div class="video_added_by">Added (?P<date>[0-9\/]+) by'
         upload_date = self._html_search_regex(VIDEO_UPLOADED_RE, webpage, u'upload date', fatal=False)
         if upload_date: upload_date = unified_strdate(upload_date)
+        age_limit = self._rta_search(webpage)
 
         info = {'id': video_id,
                 'url': video_url,
@@ -45,6 +47,7 @@ class PornotubeIE(InfoExtractor):
                 'upload_date': upload_date,
                 'title': video_title,
                 'ext': 'flv',
-                'format': 'flv'}
+                'format': 'flv',
+                'age_limit': age_limit}
 
         return [info]

youtube_dl/extractor/redtube.py

@@ -10,11 +10,12 @@ class RedTubeIE(InfoExtractor):
         u'file': u'66418.mp4',
         u'md5': u'7b8c22b5e7098a3e1c09709df1126d2d',
         u'info_dict': {
-            u"title": u"Sucked on a toilet"
+            u"title": u"Sucked on a toilet",
+            u"age_limit": 18,
         }
     }
 
-    def _real_extract(self,url):
+    def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
 
         video_id = mobj.group('id')
@@ -23,15 +24,21 @@ class RedTubeIE(InfoExtractor):
 
         self.report_extraction(video_id)
 
-        video_url = self._html_search_regex(r'<source src="(.+?)" type="video/mp4">',
-            webpage, u'video URL')
+        video_url = self._html_search_regex(
+            r'<source src="(.+?)" type="video/mp4">', webpage, u'video URL')
 
-        video_title = self._html_search_regex('<h1 class="videoTitle slidePanelMovable">(.+?)</h1>',
+        video_title = self._html_search_regex(
+            r'<h1 class="videoTitle slidePanelMovable">(.+?)</h1>',
             webpage, u'title')
 
-        return [{
-            'id':       video_id,
-            'url':      video_url,
-            'ext':      video_extension,
-            'title':    video_title,
-        }]
+        # No self-labeling, but they describe themselves as
+        # "Home of Videos Porno"
+        age_limit = 18
+
+        return {
+            'id':        video_id,
+            'url':       video_url,
+            'ext':       video_extension,
+            'title':     video_title,
+            'age_limit': age_limit,
+        }

youtube_dl/extractor/ro220.py

@@ -0,0 +1,42 @@
+import re
+
+from .common import InfoExtractor
+from ..utils import (
+    clean_html,
+    compat_parse_qs,
+)
+
+
+class Ro220IE(InfoExtractor):
+    IE_NAME = '220.ro'
+    _VALID_URL = r'(?x)(?:https?://)?(?:www\.)?220\.ro/(?P<category>[^/]+)/(?P<shorttitle>[^/]+)/(?P<video_id>[^/]+)'
+    _TEST = {
+        u"url": u"http://www.220.ro/sport/Luati-Le-Banii-Sez-4-Ep-1/LYV6doKo7f/",
+        u'file': u'LYV6doKo7f.mp4',
+        u'md5': u'03af18b73a07b4088753930db7a34add',
+        u'info_dict': {
+            u"title": u"Luati-le Banii sez 4 ep 1",
+            u"description": u"Iata-ne reveniti dupa o binemeritata vacanta. Va astept si pe Facebook cu pareri si comentarii.",
+        }
+    }
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        video_id = mobj.group('video_id')
+
+        webpage = self._download_webpage(url, video_id)
+        flashVars_str = self._search_regex(
+            r'<param name="flashVars" value="([^"]+)"',
+            webpage, u'flashVars')
+        flashVars = compat_parse_qs(flashVars_str)
+
+        info = {
+            '_type': 'video',
+            'id': video_id,
+            'ext': 'mp4',
+            'url': flashVars['videoURL'][0],
+            'title': flashVars['title'][0],
+            'description': clean_html(flashVars['desc'][0]),
+            'thumbnail': flashVars['preview'][0],
+        }
+        return info

youtube_dl/extractor/rottentomatoes.py

@@ -0,0 +1,16 @@
+from .videodetective import VideoDetectiveIE
+
+
+# It just uses the same method as videodetective.com,
+# the internetvideoarchive.com is extracted from the og:video property
+class RottenTomatoesIE(VideoDetectiveIE):
+    _VALID_URL = r'https?://www\.rottentomatoes\.com/m/[^/]+/trailers/(?P<id>\d+)'
+
+    _TEST = {
+        u'url': u'http://www.rottentomatoes.com/m/toy_story_3/trailers/11028566/',
+        u'file': '613340.mp4',
+        u'info_dict': {
+            u'title': u'TOY STORY 3',
+            u'description': u'From the creators of the beloved TOY STORY films, comes a story that will reunite the gang in a whole new way.',
+        },
+    }

youtube_dl/extractor/rtlnow.py

@@ -0,0 +1,155 @@
+# encoding: utf-8
+import re
+
+from .common import InfoExtractor
+from ..utils import (
+    clean_html,
+    ExtractorError,
+)
+
+class RTLnowIE(InfoExtractor):
+    """Information Extractor for RTL NOW, RTL2 NOW, RTL NITRO, SUPER RTL NOW, VOX NOW and n-tv NOW"""
+    _VALID_URL = r'(?:http://)?(?P<url>(?P<base_url>rtl-now\.rtl\.de/|rtl2now\.rtl2\.de/|(?:www\.)?voxnow\.de/|(?:www\.)?rtlnitronow\.de/|(?:www\.)?superrtlnow\.de/|(?:www\.)?n-tvnow\.de/)[a-zA-Z0-9-]+/[a-zA-Z0-9-]+\.php\?(?:container_id|film_id)=(?P<video_id>[0-9]+)&player=1(?:&season=[0-9]+)?(?:&.*)?)'
+    _TESTS = [{
+        u'url': u'http://rtl-now.rtl.de/ahornallee/folge-1.php?film_id=90419&player=1&season=1',
+        u'file': u'90419.flv',
+        u'info_dict': {
+            u'upload_date': u'20070416', 
+            u'title': u'Ahornallee - Folge 1 - Der Einzug',
+            u'description': u'Folge 1 - Der Einzug',
+        },
+        u'params': {
+            u'skip_download': True,
+        },
+        u'skip': u'Only works from Germany',
+    },
+    {
+        u'url': u'http://rtl2now.rtl2.de/aerger-im-revier/episode-15-teil-1.php?film_id=69756&player=1&season=2&index=5',
+        u'file': u'69756.flv',
+        u'info_dict': {
+            u'upload_date': u'20120519', 
+            u'title': u'Ärger im Revier - Ein junger Ladendieb, ein handfester Streit...',
+            u'description': u'Ärger im Revier - Ein junger Ladendieb, ein handfester Streit u.a.',
+            u'thumbnail': u'http://autoimg.static-fra.de/rtl2now/219850/1500x1500/image2.jpg',
+        },
+        u'params': {
+            u'skip_download': True,
+        },
+        u'skip': u'Only works from Germany',
+    },
+    {
+        u'url': u'www.voxnow.de/voxtours/suedafrika-reporter-ii.php?film_id=13883&player=1&season=17',
+        u'file': u'13883.flv',
+        u'info_dict': {
+            u'upload_date': u'20090627', 
+            u'title': u'Voxtours - Südafrika-Reporter II',
+            u'description': u'Südafrika-Reporter II',
+        },
+        u'params': {
+            u'skip_download': True,
+        },
+    },
+    {
+        u'url': u'http://superrtlnow.de/medicopter-117/angst.php?film_id=99205&player=1',
+        u'file': u'99205.flv',
+        u'info_dict': {
+            u'upload_date': u'20080928', 
+            u'title': u'Medicopter 117 - Angst!',
+            u'description': u'Angst!',
+            u'thumbnail': u'http://autoimg.static-fra.de/superrtlnow/287529/1500x1500/image2.jpg'
+        },
+        u'params': {
+            u'skip_download': True,
+        },
+    },
+    {
+        u'url': u'http://www.rtlnitronow.de/recht-ordnung/stadtpolizei-frankfurt-gerichtsvollzieher-leipzig.php?film_id=129679&player=1&season=1',
+        u'file': u'129679.flv',
+        u'info_dict': {
+            u'upload_date': u'20131016', 
+            u'title': u'Recht & Ordnung - Stadtpolizei Frankfurt/ Gerichtsvollzieher...',
+            u'description': u'Stadtpolizei Frankfurt/ Gerichtsvollzieher Leipzig',
+        },
+        u'params': {
+            u'skip_download': True,
+        },
+    },
+    {
+        u'url': u'http://www.n-tvnow.de/top-gear/episode-1-2013-01-01-00-00-00.php?film_id=124903&player=1&season=10',
+        u'file': u'124903.flv',
+        u'info_dict': {
+            u'upload_date': u'20130101', 
+            u'title': u'Top Gear vom 01.01.2013',
+            u'description': u'Episode 1',
+        },
+        u'params': {
+            u'skip_download': True,
+        },
+        u'skip': u'Only works from Germany',
+    }]
+
+
+    def _real_extract(self,url):
+        mobj = re.match(self._VALID_URL, url)
+
+        webpage_url = u'http://' + mobj.group('url')
+        video_page_url = u'http://' + mobj.group('base_url')
+        video_id = mobj.group(u'video_id')
+
+        webpage = self._download_webpage(webpage_url, video_id)
+
+        note_m = re.search(r'''(?sx)
+            <div[ ]style="margin-left:[ ]20px;[ ]font-size:[ ]13px;">(.*?)
+            <div[ ]id="playerteaser">''', webpage)
+        if note_m:
+            msg = clean_html(note_m.group(1))
+            raise ExtractorError(msg)
+
+        video_title = self._html_search_regex(r'<title>(?P<title>[^<]+?)( \| [^<]*)?</title>',
+            webpage, u'title')
+        playerdata_url = self._html_search_regex(r'\'playerdata\': \'(?P<playerdata_url>[^\']+)\'',
+            webpage, u'playerdata_url')
+
+        playerdata = self._download_webpage(playerdata_url, video_id)
+        mobj = re.search(r'<title><!\[CDATA\[(?P<description>.+?)(?:\s+- (?:Sendung )?vom (?P<upload_date_d>[0-9]{2})\.(?P<upload_date_m>[0-9]{2})\.(?:(?P<upload_date_Y>[0-9]{4})|(?P<upload_date_y>[0-9]{2})) [0-9]{2}:[0-9]{2} Uhr)?\]\]></title>', playerdata)
+        if mobj:
+            video_description = mobj.group(u'description')
+            if mobj.group('upload_date_Y'):
+                video_upload_date = mobj.group('upload_date_Y')
+            elif mobj.group('upload_date_y'):
+                video_upload_date = u'20' + mobj.group('upload_date_y')
+            else:
+                video_upload_date = None
+            if video_upload_date:
+                video_upload_date += mobj.group('upload_date_m')+mobj.group('upload_date_d')
+        else:
+            video_description = None
+            video_upload_date = None
+            self._downloader.report_warning(u'Unable to extract description and upload date')
+
+        # Thumbnail: not every video has an thumbnail
+        mobj = re.search(r'<meta property="og:image" content="(?P<thumbnail>[^"]+)">', webpage)
+        if mobj:
+            video_thumbnail = mobj.group(u'thumbnail')
+        else:
+            video_thumbnail = None
+
+        mobj = re.search(r'<filename [^>]+><!\[CDATA\[(?P<url>rtmpe://(?:[^/]+/){2})(?P<play_path>[^\]]+)\]\]></filename>', playerdata)
+        if mobj is None:
+            raise ExtractorError(u'Unable to extract media URL')
+        video_url = mobj.group(u'url')
+        video_play_path = u'mp4:' + mobj.group(u'play_path')
+        video_player_url = video_page_url + u'includes/vodplayer.swf'
+
+        return [{
+            'id':          video_id,
+            'url':         video_url,
+            'play_path':   video_play_path,
+            'page_url':    video_page_url,
+            'player_url':  video_player_url,
+            'ext':         'flv',
+            'title':       video_title,
+            'description': video_description,
+            'upload_date': video_upload_date,
+            'thumbnail':   video_thumbnail,
+        }]

youtube_dl/extractor/rutube.py

@@ -0,0 +1,58 @@
+# encoding: utf-8
+import re
+import json
+
+from .common import InfoExtractor
+from ..utils import (
+    compat_urlparse,
+    compat_str,
+    ExtractorError,
+)
+
+
+class RutubeIE(InfoExtractor):
+    _VALID_URL = r'https?://rutube.ru/video/(?P<long_id>\w+)'
+
+    _TEST = {
+        u'url': u'http://rutube.ru/video/3eac3b4561676c17df9132a9a1e62e3e/',
+        u'file': u'3eac3b4561676c17df9132a9a1e62e3e.mp4',
+        u'info_dict': {
+            u'title': u'Раненный кенгуру забежал в аптеку',
+            u'uploader': u'NTDRussian',
+            u'uploader_id': u'29790',
+        },
+        u'params': {
+            # It requires ffmpeg (m3u8 download)
+            u'skip_download': True,
+        },
+    }
+
+    def _get_api_response(self, short_id, subpath):
+        api_url = 'http://rutube.ru/api/play/%s/%s/?format=json' % (subpath, short_id)
+        response_json = self._download_webpage(api_url, short_id,
+            u'Downloading %s json' % subpath)
+        return json.loads(response_json)
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        long_id = mobj.group('long_id')
+        webpage = self._download_webpage(url, long_id)
+        og_video = self._og_search_video_url(webpage)
+        short_id = compat_urlparse.urlparse(og_video).path[1:]
+        options = self._get_api_response(short_id, 'options')
+        trackinfo = self._get_api_response(short_id, 'trackinfo')
+        # Some videos don't have the author field
+        author = trackinfo.get('author') or {}
+        m3u8_url = trackinfo['video_balancer'].get('m3u8')
+        if m3u8_url is None:
+            raise ExtractorError(u'Couldn\'t find m3u8 manifest url')
+
+        return {
+            'id': trackinfo['id'],
+            'title': trackinfo['title'],
+            'url': m3u8_url,
+            'ext': 'mp4',
+            'thumbnail': options['thumbnail_url'],
+            'uploader': author.get('name'),
+            'uploader_id': compat_str(author['id']) if author else None,
+        }

youtube_dl/extractor/slashdot.py

@@ -0,0 +1,23 @@
+import re
+
+from .common import InfoExtractor
+
+
+class SlashdotIE(InfoExtractor):
+    _VALID_URL = r'https?://tv.slashdot.org/video/\?embed=(?P<id>.*?)(&|$)'
+
+    _TEST = {
+        u'url': u'http://tv.slashdot.org/video/?embed=JscHMzZDplD0p-yNLOzTfzC3Q3xzJaUz',
+        u'file': u'JscHMzZDplD0p-yNLOzTfzC3Q3xzJaUz.mp4',
+        u'md5': u'd2222e7a4a4c1541b3e0cf732fb26735',
+        u'info_dict': {
+            u'title': u' Meet the Stampede Supercomputing Cluster\'s Administrator',
+        },
+    }
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        video_id = mobj.group('id')
+        webpage = self._download_webpage(url, video_id)
+        ooyala_url = self._search_regex(r'<script src="(.*?)"', webpage, 'ooyala url')
+        return self.url_result(ooyala_url, 'Ooyala')

youtube_dl/extractor/slideshare.py

@@ -0,0 +1,47 @@
+import re
+import json
+
+from .common import InfoExtractor
+from ..utils import (
+    compat_urlparse,
+    ExtractorError,
+)
+
+
+class SlideshareIE(InfoExtractor):
+    _VALID_URL = r'https?://www\.slideshare\.net/[^/]+?/(?P<title>.+?)($|\?)'
+
+    _TEST = {
+        u'url': u'http://www.slideshare.net/Dataversity/keynote-presentation-managing-scale-and-complexity',
+        u'file': u'25665706.mp4',
+        u'info_dict': {
+            u'title': u'Managing Scale and Complexity',
+            u'description': u'This was a keynote presentation at the NoSQL Now! 2013 Conference & Expo (http://www.nosqlnow.com). This presentation was given by Adrian Cockcroft from Netflix',
+        },
+    }
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        page_title = mobj.group('title')
+        webpage = self._download_webpage(url, page_title)
+        slideshare_obj = self._search_regex(
+            r'var slideshare_object =  ({.*?}); var user_info =',
+            webpage, u'slideshare object')
+        info = json.loads(slideshare_obj)
+        if info['slideshow']['type'] != u'video':
+            raise ExtractorError(u'Webpage type is "%s": only video extraction is supported for Slideshare' % info['slideshow']['type'], expected=True)
+
+        doc = info['doc']
+        bucket = info['jsplayer']['video_bucket']
+        ext = info['jsplayer']['video_extension']
+        video_url = compat_urlparse.urljoin(bucket, doc + '-SD.' + ext)
+
+        return {
+            '_type': 'video',
+            'id': info['slideshow']['id'],
+            'title': info['slideshow']['title'],
+            'ext': ext,
+            'url': video_url,
+            'thumbnail': info['slideshow']['pin_image_url'],
+            'description': self._og_search_description(webpage),
+        }

youtube_dl/extractor/sohu.py

@@ -0,0 +1,94 @@
+# encoding: utf-8
+
+import json
+import re
+
+from .common import InfoExtractor
+from ..utils import ExtractorError
+
+
+class SohuIE(InfoExtractor):
+    _VALID_URL = r'https?://(?P<mytv>my\.)?tv\.sohu\.com/.+?/(?(mytv)|n)(?P<id>\d+)\.shtml.*?'
+
+    _TEST = {
+        u'url': u'http://tv.sohu.com/20130724/n382479172.shtml#super',
+        u'file': u'382479172.mp4',
+        u'md5': u'bde8d9a6ffd82c63a1eefaef4eeefec7',
+        u'info_dict': {
+            u'title': u'MV：Far East Movement《The Illest》',
+        },
+    }
+
+    def _real_extract(self, url):
+
+        def _fetch_data(vid_id, mytv=False):
+            if mytv:
+                base_data_url = 'http://my.tv.sohu.com/play/videonew.do?vid='
+            else:
+                base_data_url = u'http://hot.vrs.sohu.com/vrs_flash.action?vid='
+            data_url = base_data_url + str(vid_id)
+            data_json = self._download_webpage(
+                data_url, video_id,
+                note=u'Downloading JSON data for ' + str(vid_id))
+            return json.loads(data_json)
+
+        mobj = re.match(self._VALID_URL, url)
+        video_id = mobj.group('id')
+        mytv = mobj.group('mytv') is not None
+
+        webpage = self._download_webpage(url, video_id)
+        raw_title = self._html_search_regex(r'(?s)<title>(.+?)</title>',
+                                            webpage, u'video title')
+        title = raw_title.partition('-')[0].strip()
+
+        vid = self._html_search_regex(r'var vid ?= ?["\'](\d+)["\']', webpage,
+                                      u'video path')
+        data = _fetch_data(vid, mytv)
+
+        QUALITIES = ('ori', 'super', 'high', 'nor')
+        vid_ids = [data['data'][q + 'Vid']
+                   for q in QUALITIES
+                   if data['data'][q + 'Vid'] != 0]
+        if not vid_ids:
+            raise ExtractorError(u'No formats available for this video')
+
+        # For now, we just pick the highest available quality
+        vid_id = vid_ids[-1]
+
+        format_data = data if vid == vid_id else _fetch_data(vid_id, mytv)
+        part_count = format_data['data']['totalBlocks']
+        allot = format_data['allot']
+        prot = format_data['prot']
+        clipsURL = format_data['data']['clipsURL']
+        su = format_data['data']['su']
+
+        playlist = []
+        for i in range(part_count):
+            part_url = ('http://%s/?prot=%s&file=%s&new=%s' %
+                        (allot, prot, clipsURL[i], su[i]))
+            part_str = self._download_webpage(
+                part_url, video_id,
+                note=u'Downloading part %d of %d' % (i+1, part_count))
+
+            part_info = part_str.split('|')
+            video_url = '%s%s?key=%s' % (part_info[0], su[i], part_info[3])
+
+            video_info = {
+                'id': '%s_part%02d' % (video_id, i + 1),
+                'title': title,
+                'url': video_url,
+                'ext': 'mp4',
+            }
+            playlist.append(video_info)
+
+        if len(playlist) == 1:
+            info = playlist[0]
+            info['id'] = video_id
+        else:
+            info = {
+                '_type': 'playlist',
+                'entries': playlist,
+                'id': video_id,
+            }
+
+        return info