release 2014.08.21.2

This moves playlist tests where they belong, i.e. to the extractors themselves.
Additionally, all our network interaction configuration for tests in test_download now applies to playlist tests as well.

Makefile

@@ -6,10 +6,10 @@ clean:
 cleanall: clean
 	rm -f youtube-dl youtube-dl.exe
 
-PREFIX=/usr/local
-BINDIR=$(PREFIX)/bin
-MANDIR=$(PREFIX)/man
-PYTHON=/usr/bin/env python
+PREFIX ?= /usr/local
+BINDIR ?= $(PREFIX)/bin
+MANDIR ?= $(PREFIX)/man
+PYTHON ?= /usr/bin/env python
 
 # set SYSCONFDIR to /etc if PREFIX=/usr or PREFIX=/usr/local
 ifeq ($(PREFIX),/usr)

README.md

@@ -17,6 +17,14 @@ If you do not have curl, you can alternatively use a recent wget:
 
 Windows users can [download a .exe file](https://yt-dl.org/latest/youtube-dl.exe) and place it in their home directory or any other location on their [PATH](http://en.wikipedia.org/wiki/PATH_%28variable%29).
 
+OS X users can install **youtube-dl** with [Homebrew](http://brew.sh/).
+
+    brew install youtube-dl
+
+You can also use pip:
+
+    sudo pip install youtube-dl
+
 Alternatively, refer to the developer instructions below for how to check out and work with the git repository. For further options, including PGP signatures, see https://rg3.github.io/youtube-dl/download.html .
 
 # DESCRIPTION
@@ -38,12 +46,6 @@ which means you can modify it, redistribute it or use it however you like.
                                      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 is restricted to one domain
-    --add-header FIELD:VALUE         specify a custom HTTP header and its value,
-                                     separated by a colon ':'. You can use this
-                                     option multiple times
     --list-extractors                List all supported extractors and the URLs
                                      they would handle
     --extractor-descriptions         Output descriptions of all supported
@@ -51,35 +53,22 @@ which means you can modify it, redistribute it or use it however you like.
     --proxy URL                      Use the specified HTTP/HTTPS proxy. Pass in
                                      an empty string (--proxy "") for direct
                                      connection
-    --no-check-certificate           Suppress HTTPS certificate validation.
-    --prefer-insecure                Use an unencrypted connection to retrieve
-                                     information about the video. (Currently
-                                     supported only for YouTube)
-    --cache-dir DIR                  Location in the filesystem where youtube-dl
-                                     can store some downloaded information
-                                     permanently. By default $XDG_CACHE_HOME
-                                     /youtube-dl or ~/.cache/youtube-dl . At the
-                                     moment, only YouTube player files (for
-                                     videos with obfuscated signatures) are
-                                     cached, but that may change.
-    --no-cache-dir                   Disable filesystem caching
     --socket-timeout None            Time to wait before giving up, in seconds
-    --bidi-workaround                Work around terminals that lack
-                                     bidirectional text support. Requires bidiv
-                                     or fribidi executable in PATH
     --default-search PREFIX          Use this prefix for unqualified URLs. For
                                      example "gvsearch2:" downloads two videos
                                      from google videos for  youtube-dl "large
                                      apple". Use the value "auto" to let
-                                     youtube-dl guess. The default value "error"
-                                     just throws an error.
+                                     youtube-dl guess ("auto_warning" to emit a
+                                     warning when guessing). "error" just throws
+                                     an error. The default value "fixup_error"
+                                     repairs broken URLs, but emits an error if
+                                     this is not possible instead of searching.
     --ignore-config                  Do not read configuration files. When given
                                      in the global configuration file /etc
                                      /youtube-dl.conf: do not read the user
                                      configuration in ~/.config/youtube-dl.conf
                                      (%APPDATA%/youtube-dl/config.txt on
                                      Windows)
-    --encoding ENCODING              Force the specified encoding (experimental)
 
 ## Video Selection:
     --playlist-start NUMBER          playlist video to start at (default is 1)
@@ -125,9 +114,9 @@ which means you can modify it, redistribute it or use it however you like.
                                      of SIZE.
 
 ## Filesystem Options:
-    -t, --title                      use title in file name (default)
+    -a, --batch-file FILE            file containing URLs to download ('-' for
+                                     stdin)
     --id                             use only video ID in file name
-    -l, --literal                    [deprecated] alias of --title
     -A, --auto-number                number downloaded files starting from 00000
     -o, --output TEMPLATE            output filename template. Use %(title)s to
                                      get the title, %(uploader)s for the
@@ -160,18 +149,15 @@ which means you can modify it, redistribute it or use it however you like.
     --restrict-filenames             Restrict filenames to only ASCII
                                      characters, and avoid "&" and spaces in
                                      filenames
-    -a, --batch-file FILE            file containing URLs to download ('-' for
-                                     stdin)
-    --load-info FILE                 json file containing the video information
-                                     (created with the "--write-json" option)
+    -t, --title                      [deprecated] use title in file name
+                                     (default)
+    -l, --literal                    [deprecated] alias of --title
     -w, --no-overwrites              do not overwrite files
     -c, --continue                   force resume of partially downloaded files.
                                      By default, youtube-dl will resume
                                      downloads if possible.
     --no-continue                    do not resume partially downloaded files
                                      (restart from beginning)
-    --cookies FILE                   file to read cookies from and dump cookie
-                                     jar in
     --no-part                        do not use .part files
     --no-mtime                       do not use the Last-modified header to set
                                      the file modification time
@@ -181,6 +167,19 @@ which means you can modify it, redistribute it or use it however you like.
     --write-annotations              write video annotations to a .annotation
                                      file
     --write-thumbnail                write thumbnail image to disk
+    --load-info FILE                 json file containing the video information
+                                     (created with the "--write-json" option)
+    --cookies FILE                   file to read cookies from and dump cookie
+                                     jar in
+    --cache-dir DIR                  Location in the filesystem where youtube-dl
+                                     can store some downloaded information
+                                     permanently. By default $XDG_CACHE_HOME
+                                     /youtube-dl or ~/.cache/youtube-dl . At the
+                                     moment, only YouTube player files (for
+                                     videos with obfuscated signatures) are
+                                     cached, but that may change.
+    --no-cache-dir                   Disable filesystem caching
+    --rm-cache-dir                   Delete all filesystem cache files
 
 ## Verbosity / Simulation Options:
     -q, --quiet                      activates quiet mode
@@ -210,6 +209,22 @@ which means you can modify it, redistribute it or use it however you like.
                                      problems
     --print-traffic                  Display sent and read HTTP traffic
 
+## Workarounds:
+    --encoding ENCODING              Force the specified encoding (experimental)
+    --no-check-certificate           Suppress HTTPS certificate validation.
+    --prefer-insecure                Use an unencrypted connection to retrieve
+                                     information about the video. (Currently
+                                     supported only for YouTube)
+    --user-agent UA                  specify a custom user agent
+    --referer REF                    specify a custom referer, use if the video
+                                     access is restricted to one domain
+    --add-header FIELD:VALUE         specify a custom HTTP header and its value,
+                                     separated by a colon ':'. You can use this
+                                     option multiple times
+    --bidi-workaround                Work around terminals that lack
+                                     bidirectional text support. Requires bidiv
+                                     or fribidi executable in PATH
+
 ## Video Format Options:
     -f, --format FORMAT              video format code, specify the order of
                                      preference using slashes: "-f 22/17/18".

test/helper.py

@@ -117,8 +117,9 @@ def expect_info_dict(self, expected_dict, got_dict):
                 u'invalid value for field %s, expected %r, got %r' % (info_field, expected, got))
 
     # Check for the presence of mandatory fields
-    for key in ('id', 'url', 'title', 'ext'):
-        self.assertTrue(got_dict.get(key), 'Missing mandatory field %s' % key)
+    if got_dict.get('_type') != 'playlist':
+        for key in ('id', 'url', 'title', 'ext'):
+            self.assertTrue(got_dict.get(key), 'Missing mandatory field %s' % key)
     # Check for mandatory fields that are automatically set by YoutubeDL
     for key in ['webpage_url', 'extractor', 'extractor_key']:
         self.assertTrue(got_dict.get(key), u'Missing field: %s' % key)
@@ -137,8 +138,8 @@ def expect_info_dict(self, expected_dict, got_dict):
 
 
 def assertRegexpMatches(self, text, regexp, msg=None):
-    if hasattr(self, 'assertRegexpMatches'):
-        return self.assertRegexpMatches(text, regexp, msg)
+    if hasattr(self, 'assertRegexp'):
+        return self.assertRegexp(text, regexp, msg)
     else:
         m = re.match(regexp, text)
         if not m:
@@ -148,3 +149,10 @@ def assertRegexpMatches(self, text, regexp, msg=None):
             else:
                 msg = note + ', ' + msg
             self.assertTrue(m, msg)
+
+
+def assertGreaterEqual(self, got, expected, msg=None):
+    if not (got >= expected):
+        if msg is None:
+            msg = '%r not greater than or equal to %r' % (got, expected)
+        self.assertTrue(got >= expected, msg)

test/test_all_urls.py

@@ -15,7 +15,6 @@ from youtube_dl.extractor import (
     FacebookIE,
     gen_extractors,
     JustinTVIE,
-    PBSIE,
     YoutubeIE,
 )
 
@@ -100,6 +99,7 @@ class TestAllURLsMatching(unittest.TestCase):
 
     def test_facebook_matching(self):
         self.assertTrue(FacebookIE.suitable('https://www.facebook.com/Shiniknoh#!/photo.php?v=10153317450565268'))
+        self.assertTrue(FacebookIE.suitable('https://www.facebook.com/cindyweather?fref=ts#!/photo.php?v=10152183998945793'))
 
     def test_no_duplicates(self):
         ies = gen_extractors()

test/test_download.py

@@ -10,7 +10,6 @@ from test.helper import (
     get_params,
     gettestcases,
     expect_info_dict,
-    md5,
     try_rm,
     report_warning,
 )
@@ -24,7 +23,6 @@ import socket
 import youtube_dl.YoutubeDL
 from youtube_dl.utils import (
     compat_http_client,
-    compat_str,
     compat_urllib_error,
     compat_HTTPError,
     DownloadError,
@@ -65,15 +63,21 @@ def generator(test_case):
     def test_template(self):
         ie = youtube_dl.extractor.get_info_extractor(test_case['name'])
         other_ies = [get_info_extractor(ie_key) for ie_key in test_case.get('add_ie', [])]
+        is_playlist = any(k.startswith('playlist') for k in test_case)
+        test_cases = test_case.get(
+            'playlist', [] if is_playlist else [test_case])
+
         def print_skipping(reason):
             print('Skipping %s: %s' % (test_case['name'], reason))
         if not ie.working():
             print_skipping('IE marked as not _WORKING')
             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')):
+
+        for tc in test_cases:
+            info_dict = tc.get('info_dict', {})
+            if not tc.get('file') and not (info_dict.get('id') and info_dict.get('ext')):
                 raise Exception('Test definition incorrect. The output file cannot be known. Are both \'id\' and \'ext\' keys present?')
+
         if 'skip' in test_case:
             print_skipping(test_case['skip'])
             return
@@ -83,6 +87,9 @@ def generator(test_case):
                 return
 
         params = get_params(test_case.get('params', {}))
+        if is_playlist and 'playlist' not in test_case:
+            params.setdefault('extract_flat', True)
+            params.setdefault('skip_download', True)
 
         ydl = YoutubeDL(params)
         ydl.add_default_info_extractors()
@@ -95,7 +102,6 @@ def generator(test_case):
         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])
         def try_rm_tcs_files():
             for tc in test_cases:
                 tc_filename = get_tc_filename(tc)
@@ -107,7 +113,10 @@ def generator(test_case):
             try_num = 1
             while True:
                 try:
-                    ydl.download([test_case['url']])
+                    # We're not using .download here sine that is just a shim
+                    # for outside error handling, and returns the exit code
+                    # instead of the result dict.
+                    res_dict = ydl.extract_info(test_case['url'])
                 except (DownloadError, ExtractorError) as err:
                     # Check if the exception is not a network related one
                     if not err.exc_info[0] in (compat_urllib_error.URLError, socket.timeout, UnavailableVideoError, compat_http_client.BadStatusLine) or (err.exc_info[0] == compat_HTTPError and err.exc_info[1].code == 503):
@@ -123,6 +132,17 @@ def generator(test_case):
                 else:
                     break
 
+            if is_playlist:
+                self.assertEqual(res_dict['_type'], 'playlist')
+                expect_info_dict(self, test_case.get('info_dict', {}), res_dict)
+            if 'playlist_mincount' in test_case:
+                self.assertGreaterEqual(
+                    len(res_dict['entries']),
+                    test_case['playlist_mincount'],
+                    'Expected at least %d in playlist %s, but got only %d' % (
+                        test_case['playlist_mincount'], test_case['url'],
+                        len(res_dict['entries'])))
+
             for tc in test_cases:
                 tc_filename = get_tc_filename(tc)
                 if not test_case.get('params', {}).get('skip_download', False):

test/test_playlists.py

@@ -1,6 +1,17 @@
 #!/usr/bin/env python
 # encoding: utf-8
 
+## DEPRECATED FILE!
+# Add new tests to the extractors themselves, like this:
+# _TEST = {
+#    'url': 'http://example.com/playlist/42',
+#    'playlist_mincount': 99,
+#    'info_dict': {
+#        'id': '42',
+#        'title': 'Playlist number forty-two',
+#    }
+# }
+
 from __future__ import unicode_literals
 
 # Allow direct execution
@@ -11,6 +22,7 @@ sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
 
 from test.helper import (
     assertRegexpMatches,
+    assertGreaterEqual,
     expect_info_dict,
     FakeYDL,
 )
@@ -71,8 +83,8 @@ class TestPlaylists(unittest.TestCase):
         ie = DailymotionUserIE(dl)
         result = ie.extract('https://www.dailymotion.com/user/nqtv')
         self.assertIsPlaylist(result)
+        assertGreaterEqual(self, len(result['entries']), 100)
         self.assertEqual(result['title'], 'Rémi Gaillard')
-        self.assertTrue(len(result['entries']) >= 100)
 
     def test_vimeo_channel(self):
         dl = FakeYDL()
@@ -111,7 +123,7 @@ class TestPlaylists(unittest.TestCase):
         ie = VineUserIE(dl)
         result = ie.extract('https://vine.co/Visa')
         self.assertIsPlaylist(result)
-        self.assertTrue(len(result['entries']) >= 47)
+        assertGreaterEqual(self, len(result['entries']), 47)
 
     def test_ustream_channel(self):
         dl = FakeYDL()
@@ -119,7 +131,7 @@ class TestPlaylists(unittest.TestCase):
         result = ie.extract('http://www.ustream.tv/channel/channeljapan')
         self.assertIsPlaylist(result)
         self.assertEqual(result['id'], '10874166')
-        self.assertTrue(len(result['entries']) >= 54)
+        assertGreaterEqual(self, len(result['entries']), 54)
 
     def test_soundcloud_set(self):
         dl = FakeYDL()
@@ -127,7 +139,7 @@ class TestPlaylists(unittest.TestCase):
         result = ie.extract('https://soundcloud.com/the-concept-band/sets/the-royal-concept-ep')
         self.assertIsPlaylist(result)
         self.assertEqual(result['title'], 'The Royal Concept EP')
-        self.assertTrue(len(result['entries']) >= 6)
+        assertGreaterEqual(self, len(result['entries']), 6)
 
     def test_soundcloud_user(self):
         dl = FakeYDL()
@@ -135,7 +147,7 @@ class TestPlaylists(unittest.TestCase):
         result = ie.extract('https://soundcloud.com/the-concept-band')
         self.assertIsPlaylist(result)
         self.assertEqual(result['id'], '9615865')
-        self.assertTrue(len(result['entries']) >= 12)
+        assertGreaterEqual(self, len(result['entries']), 12)
 
     def test_soundcloud_likes(self):
         dl = FakeYDL()
@@ -143,7 +155,7 @@ class TestPlaylists(unittest.TestCase):
         result = ie.extract('https://soundcloud.com/the-concept-band/likes')
         self.assertIsPlaylist(result)
         self.assertEqual(result['id'], '9615865')
-        self.assertTrue(len(result['entries']) >= 1)
+        assertGreaterEqual(self, len(result['entries']), 1)
 
     def test_soundcloud_playlist(self):
         dl = FakeYDL()
@@ -153,7 +165,7 @@ class TestPlaylists(unittest.TestCase):
         self.assertEqual(result['id'], '4110309')
         self.assertEqual(result['title'], 'TILT Brass - Bowery Poetry Club, August \'03 [Non-Site SCR 02]')
         assertRegexpMatches(
-            self, result['description'], r'TILT Brass - Bowery Poetry Club')
+            self, result['description'], r'.*?TILT Brass - Bowery Poetry Club')
         self.assertEqual(len(result['entries']), 6)
 
     def test_livestream_event(self):
@@ -162,7 +174,7 @@ class TestPlaylists(unittest.TestCase):
         result = ie.extract('http://new.livestream.com/tedx/cityenglish')
         self.assertIsPlaylist(result)
         self.assertEqual(result['title'], 'TEDCity2.0 (English)')
-        self.assertTrue(len(result['entries']) >= 4)
+        assertGreaterEqual(self, len(result['entries']), 4)
 
     def test_livestreamoriginal_folder(self):
         dl = FakeYDL()
@@ -170,7 +182,7 @@ class TestPlaylists(unittest.TestCase):
         result = ie.extract('https://www.livestream.com/newplay/folder?dirId=a07bf706-d0e4-4e75-a747-b021d84f2fd3')
         self.assertIsPlaylist(result)
         self.assertEqual(result['id'], 'a07bf706-d0e4-4e75-a747-b021d84f2fd3')
-        self.assertTrue(len(result['entries']) >= 28)
+        assertGreaterEqual(self, len(result['entries']), 28)
 
     def test_nhl_videocenter(self):
         dl = FakeYDL()
@@ -187,15 +199,15 @@ class TestPlaylists(unittest.TestCase):
         result = ie.extract('http://bambuser.com/channel/pixelversity')
         self.assertIsPlaylist(result)
         self.assertEqual(result['title'], 'pixelversity')
-        self.assertTrue(len(result['entries']) >= 60)
+        assertGreaterEqual(self, len(result['entries']), 60)
 
     def test_bandcamp_album(self):
         dl = FakeYDL()
         ie = BandcampAlbumIE(dl)
-        result = ie.extract('http://mpallante.bandcamp.com/album/nightmare-night-ep')
+        result = ie.extract('http://nightbringer.bandcamp.com/album/hierophany-of-the-open-grave')
         self.assertIsPlaylist(result)
-        self.assertEqual(result['title'], 'Nightmare Night EP')
-        self.assertTrue(len(result['entries']) >= 4)
+        self.assertEqual(result['title'], 'Hierophany of the Open Grave')
+        assertGreaterEqual(self, len(result['entries']), 9)
         
     def test_smotri_community(self):
         dl = FakeYDL()
@@ -204,7 +216,7 @@ class TestPlaylists(unittest.TestCase):
         self.assertIsPlaylist(result)
         self.assertEqual(result['id'], 'kommuna')
         self.assertEqual(result['title'], 'КПРФ')
-        self.assertTrue(len(result['entries']) >= 4)
+        assertGreaterEqual(self, len(result['entries']), 4)
         
     def test_smotri_user(self):
         dl = FakeYDL()
@@ -213,7 +225,7 @@ class TestPlaylists(unittest.TestCase):
         self.assertIsPlaylist(result)
         self.assertEqual(result['id'], 'inspector')
         self.assertEqual(result['title'], 'Inspector')
-        self.assertTrue(len(result['entries']) >= 9)
+        assertGreaterEqual(self, len(result['entries']), 9)
 
     def test_AcademicEarthCourse(self):
         dl = FakeYDL()
@@ -232,7 +244,7 @@ class TestPlaylists(unittest.TestCase):
         self.assertIsPlaylist(result)
         self.assertEqual(result['id'], 'dvoe_iz_lartsa')
         self.assertEqual(result['title'], 'Двое из ларца (2006 - 2008)')
-        self.assertTrue(len(result['entries']) >= 24)
+        assertGreaterEqual(self, len(result['entries']), 24)
 
     def test_ivi_compilation_season(self):
         dl = FakeYDL()
@@ -241,7 +253,7 @@ class TestPlaylists(unittest.TestCase):
         self.assertIsPlaylist(result)
         self.assertEqual(result['id'], 'dvoe_iz_lartsa/season1')
         self.assertEqual(result['title'], 'Двое из ларца (2006 - 2008) 1 сезон')
-        self.assertTrue(len(result['entries']) >= 12)
+        assertGreaterEqual(self, len(result['entries']), 12)
         
     def test_imdb_list(self):
         dl = FakeYDL()
@@ -260,7 +272,7 @@ class TestPlaylists(unittest.TestCase):
         self.assertEqual(result['id'], 'cryptography')
         self.assertEqual(result['title'], 'Journey into cryptography')
         self.assertEqual(result['description'], 'How have humans protected their secret messages through history? What has changed today?')
-        self.assertTrue(len(result['entries']) >= 3)
+        assertGreaterEqual(self, len(result['entries']), 3)
 
     def test_EveryonesMixtape(self):
         dl = FakeYDL()
@@ -277,7 +289,7 @@ class TestPlaylists(unittest.TestCase):
         result = ie.extract('http://rutube.ru/tags/video/1800/')
         self.assertIsPlaylist(result)
         self.assertEqual(result['id'], '1800')
-        self.assertTrue(len(result['entries']) >= 68)
+        assertGreaterEqual(self, len(result['entries']), 68)
 
     def test_rutube_person(self):
         dl = FakeYDL()
@@ -285,7 +297,7 @@ class TestPlaylists(unittest.TestCase):
         result = ie.extract('http://rutube.ru/video/person/313878/')
         self.assertIsPlaylist(result)
         self.assertEqual(result['id'], '313878')
-        self.assertTrue(len(result['entries']) >= 37)
+        assertGreaterEqual(self, len(result['entries']), 37)
 
     def test_multiple_brightcove_videos(self):
         # https://github.com/rg3/youtube-dl/issues/2283
@@ -322,7 +334,7 @@ class TestPlaylists(unittest.TestCase):
         self.assertIsPlaylist(result)
         self.assertEqual(result['id'], '10')
         self.assertEqual(result['title'], 'Who are the hackers?')
-        self.assertTrue(len(result['entries']) >= 6)
+        assertGreaterEqual(self, len(result['entries']), 6)
 
     def test_toypics_user(self):
         dl = FakeYDL()
@@ -330,7 +342,7 @@ class TestPlaylists(unittest.TestCase):
         result = ie.extract('http://videos.toypics.net/Mikey')
         self.assertIsPlaylist(result)
         self.assertEqual(result['id'], 'Mikey')
-        self.assertTrue(len(result['entries']) >= 17)
+        assertGreaterEqual(self, len(result['entries']), 17)
 
     def test_xtube_user(self):
         dl = FakeYDL()
@@ -338,7 +350,7 @@ class TestPlaylists(unittest.TestCase):
         result = ie.extract('http://www.xtube.com/community/profile.php?user=greenshowers')
         self.assertIsPlaylist(result)
         self.assertEqual(result['id'], 'greenshowers')
-        self.assertTrue(len(result['entries']) >= 155)
+        assertGreaterEqual(self, len(result['entries']), 155)
 
     def test_InstagramUser(self):
         dl = FakeYDL()
@@ -346,7 +358,7 @@ class TestPlaylists(unittest.TestCase):
         result = ie.extract('http://instagram.com/porsche')
         self.assertIsPlaylist(result)
         self.assertEqual(result['id'], 'porsche')
-        self.assertTrue(len(result['entries']) >= 2)
+        assertGreaterEqual(self, len(result['entries']), 2)
         test_video = next(
             e for e in result['entries']
             if e['id'] == '614605558512799803_462752227')
@@ -385,7 +397,7 @@ class TestPlaylists(unittest.TestCase):
         self.assertEqual(result['id'], '152147')
         self.assertEqual(
             result['title'], 'Brace Yourself - Today\'s Weirdest News')
-        self.assertTrue(len(result['entries']) >= 10)
+        assertGreaterEqual(self, len(result['entries']), 10)
 
     def test_TeacherTubeUser(self):
         dl = FakeYDL()
@@ -393,7 +405,7 @@ class TestPlaylists(unittest.TestCase):
         result = ie.extract('http://www.teachertube.com/user/profile/rbhagwati2')
         self.assertIsPlaylist(result)
         self.assertEqual(result['id'], 'rbhagwati2')
-        self.assertTrue(len(result['entries']) >= 179)
+        assertGreaterEqual(self, len(result['entries']), 179)
 
 if __name__ == '__main__':
     unittest.main()

test/test_swfinterp.py

@@ -7,6 +7,7 @@ import unittest
 sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
 
 
+import errno
 import io
 import json
 import re

test/test_utils.py

@@ -280,7 +280,7 @@ class TestUtil(unittest.TestCase):
         d = json.loads(stripped)
         self.assertEqual(d, [{"id": "532cb", "x": 3}])
 
-    def test_uppercase_escpae(self):
+    def test_uppercase_escape(self):
         self.assertEqual(uppercase_escape(u'aä'), u'aä')
         self.assertEqual(uppercase_escape(u'\\U0001d550'), u'𝕐')
 

test/test_youtube_signature.py

@@ -1,5 +1,7 @@
 #!/usr/bin/env python
 
+from __future__ import unicode_literals
+
 # Allow direct execution
 import os
 import sys
@@ -16,41 +18,65 @@ from youtube_dl.utils import compat_str, compat_urlretrieve
 
 _TESTS = [
     (
-        u'https://s.ytimg.com/yts/jsbin/html5player-vflHOr_nV.js',
-        u'js',
+        'https://s.ytimg.com/yts/jsbin/html5player-vflHOr_nV.js',
+        'js',
         86,
-        u'>=<;:/.-[+*)(\'&%$#"!ZYX0VUTSRQPONMLKJIHGFEDCBA\\yxwvutsrqponmlkjihgfedcba987654321',
+        '>=<;:/.-[+*)(\'&%$#"!ZYX0VUTSRQPONMLKJIHGFEDCBA\\yxwvutsrqponmlkjihgfedcba987654321',
     ),
     (
-        u'https://s.ytimg.com/yts/jsbin/html5player-vfldJ8xgI.js',
-        u'js',
+        'https://s.ytimg.com/yts/jsbin/html5player-vfldJ8xgI.js',
+        'js',
         85,
-        u'3456789a0cdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRS[UVWXYZ!"#$%&\'()*+,-./:;<=>?@',
+        '3456789a0cdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRS[UVWXYZ!"#$%&\'()*+,-./:;<=>?@',
     ),
     (
-        u'https://s.ytimg.com/yts/jsbin/html5player-vfle-mVwz.js',
-        u'js',
+        'https://s.ytimg.com/yts/jsbin/html5player-vfle-mVwz.js',
+        'js',
         90,
-        u']\\[@?>=<;:/.-,+*)(\'&%$#"hZYXWVUTSRQPONMLKJIHGFEDCBAzyxwvutsrqponmlkjiagfedcb39876',
+        ']\\[@?>=<;:/.-,+*)(\'&%$#"hZYXWVUTSRQPONMLKJIHGFEDCBAzyxwvutsrqponmlkjiagfedcb39876',
     ),
     (
-        u'https://s.ytimg.com/yts/jsbin/html5player-en_US-vfl0Cbn9e.js',
-        u'js',
+        'https://s.ytimg.com/yts/jsbin/html5player-en_US-vfl0Cbn9e.js',
+        'js',
         84,
-        u'O1I3456789abcde0ghijklmnopqrstuvwxyzABCDEFGHfJKLMN2PQRSTUVW@YZ!"#$%&\'()*+,-./:;<=',
+        'O1I3456789abcde0ghijklmnopqrstuvwxyzABCDEFGHfJKLMN2PQRSTUVW@YZ!"#$%&\'()*+,-./:;<=',
     ),
     (
-        u'https://s.ytimg.com/yts/jsbin/html5player-en_US-vflXGBaUN.js',
-        u'js',
-        u'2ACFC7A61CA478CD21425E5A57EBD73DDC78E22A.2094302436B2D377D14A3BBA23022D023B8BC25AA',
-        u'A52CB8B320D22032ABB3A41D773D2B6342034902.A22E87CDD37DBE75A5E52412DC874AC16A7CFCA2',
+        'https://s.ytimg.com/yts/jsbin/html5player-en_US-vflXGBaUN.js',
+        'js',
+        '2ACFC7A61CA478CD21425E5A57EBD73DDC78E22A.2094302436B2D377D14A3BBA23022D023B8BC25AA',
+        'A52CB8B320D22032ABB3A41D773D2B6342034902.A22E87CDD37DBE75A5E52412DC874AC16A7CFCA2',
     ),
     (
-        u'http://s.ytimg.com/yts/swfbin/player-vfl5vIhK2/watch_as3.swf',
-        u'swf',
+        'http://s.ytimg.com/yts/swfbin/player-vfl5vIhK2/watch_as3.swf',
+        'swf',
         86,
-        u'O1I3456789abcde0ghijklmnopqrstuvwxyzABCDEFGHfJKLMN2PQRSTUVWXY\\!"#$%&\'()*+,-./:;<=>?'
+        'O1I3456789abcde0ghijklmnopqrstuvwxyzABCDEFGHfJKLMN2PQRSTUVWXY\\!"#$%&\'()*+,-./:;<=>?'
     ),
+    (
+        'http://s.ytimg.com/yts/swfbin/player-vflmDyk47/watch_as3.swf',
+        'swf',
+        'F375F75BF2AFDAAF2666E43868D46816F83F13E81C46.3725A8218E446A0DECD33F79DC282994D6AA92C92C9',
+        '9C29AA6D499282CD97F33DCED0A644E8128A5273.64C18E31F38361864D86834E6662FAADFA2FB57F'
+    ),
+    (
+        'https://s.ytimg.com/yts/jsbin/html5player-en_US-vflBb0OQx.js',
+        'js',
+        84,
+        '123456789abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQ0STUVWXYZ!"#$%&\'()*+,@./:;<=>'
+    ),
+    (
+        'https://s.ytimg.com/yts/jsbin/html5player-en_US-vfl9FYC6l.js',
+        'js',
+        83,
+        '123456789abcdefghijklmnopqr0tuvwxyzABCDETGHIJKLMNOPQRS>UVWXYZ!"#$%&\'()*+,-./:;<=F'
+    ),
+    (
+        'https://s.ytimg.com/yts/jsbin/html5player-en_US-vflCGk6yw/html5player.js',
+        'js',
+        '4646B5181C6C3020DF1D9C7FCFEA.AD80ABF70C39BD369CCCAE780AFBB98FA6B6CB42766249D9488C288',
+        '82C8849D94266724DC6B6AF89BBFA087EACCD963.B93C07FBA084ACAEFCF7C9D1FD0203C6C1815B6B'
+    )
 ]
 
 
@@ -63,7 +89,7 @@ class TestSignature(unittest.TestCase):
 
 
 def make_tfunc(url, stype, sig_input, expected_sig):
-    m = re.match(r'.*-([a-zA-Z0-9_-]+)(?:/watch_as3)?\.[a-z]+$', url)
+    m = re.match(r'.*-([a-zA-Z0-9_-]+)(?:/watch_as3|/html5player)?\.[a-z]+$', url)
     assert m, '%r should follow URL format' % url
     test_id = m.group(1)
 

youtube_dl/FileDownloader.py

@@ -1,12 +0,0 @@
-# Legacy file for backwards compatibility, use youtube_dl.downloader instead!
-from .downloader import FileDownloader as RealFileDownloader
-from .downloader import get_suitable_downloader
-
-
-# This class reproduces the old behaviour of FileDownloader
-class FileDownloader(RealFileDownloader):
-    def _do_download(self, filename, info_dict):
-        real_fd = get_suitable_downloader(info_dict)(self.ydl, self.params)
-        for ph in self._progress_hooks:
-            real_fd.add_progress_hook(ph)
-        return real_fd.download(filename, info_dict)

youtube_dl/YoutubeDL.py

@@ -162,6 +162,7 @@ class YoutubeDL(object):
     default_search:    Prepend this string if an input url is not valid.
                        'auto' for elaborate guessing
     encoding:          Use this encoding instead of the system-specified.
+    extract_flat:      Do not resolve URLs, return the immediate result.
 
     The following parameters are not used by YoutubeDL itself, they are used by
     the FileDownloader:
@@ -275,7 +276,7 @@ class YoutubeDL(object):
             return message
 
         assert hasattr(self, '_output_process')
-        assert type(message) == type('')
+        assert isinstance(message, compat_str)
         line_count = message.count('\n') + 1
         self._output_process.stdin.write((message + '\n').encode('utf-8'))
         self._output_process.stdin.flush()
@@ -303,7 +304,7 @@ class YoutubeDL(object):
 
     def to_stderr(self, message):
         """Print message to stderr."""
-        assert type(message) == type('')
+        assert isinstance(message, compat_str)
         if self.params.get('logger'):
             self.params['logger'].error(message)
         else:
@@ -558,7 +559,12 @@ class YoutubeDL(object):
         Returns the resolved ie_result.
         """
 
-        result_type = ie_result.get('_type', 'video') # If not given we suppose it's a video, support the default old system
+        result_type = ie_result.get('_type', 'video')
+
+        if self.params.get('extract_flat', False):
+            if result_type in ('url', 'url_transparent'):
+                return ie_result
+
         if result_type == 'video':
             self.add_extra_info(ie_result, extra_info)
             return self.process_video_result(ie_result, download=download)
@@ -849,7 +855,7 @@ class YoutubeDL(object):
         # Keep for backwards compatibility
         info_dict['stitle'] = info_dict['title']
 
-        if not 'format' in info_dict:
+        if 'format' not in info_dict:
             info_dict['format'] = info_dict['ext']
 
         reason = self._match_entry(info_dict)
@@ -999,7 +1005,7 @@ class YoutubeDL(object):
                     if info_dict.get('requested_formats') is not None:
                         downloaded = []
                         success = True
-                        merger = FFmpegMergerPP(self)
+                        merger = FFmpegMergerPP(self, not self.params.get('keepvideo'))
                         if not merger._get_executable():
                             postprocessors = []
                             self.report_warning('You have requested multiple '
@@ -1197,6 +1203,10 @@ class YoutubeDL(object):
             if res:
                 res += ', '
             res += format_bytes(fdict['filesize'])
+        elif fdict.get('filesize_approx') is not None:
+            if res:
+                res += ', '
+            res += '~' + format_bytes(fdict['filesize_approx'])
         return res
 
     def list_formats(self, info_dict):
@@ -1230,14 +1240,18 @@ class YoutubeDL(object):
         if not self.params.get('verbose'):
             return
 
-        write_string(
+        if type('') is not compat_str:
+            # Python 2.6 on SLES11 SP1 (https://github.com/rg3/youtube-dl/issues/3326)
+            self.report_warning(
+                'Your Python is broken! Update to a newer and supported version')
+
+        encoding_str = (
             '[debug] Encodings: locale %s, fs %s, out %s, pref %s\n' % (
                 locale.getpreferredencoding(),
                 sys.getfilesystemencoding(),
                 sys.stdout.encoding,
-                self.get_encoding()),
-            encoding=None
-        )
+                self.get_encoding()))
+        write_string(encoding_str, encoding=None)
 
         self._write_string('[debug] youtube-dl version ' + __version__ + '\n')
         try:

youtube_dl/__init__.py

@@ -66,18 +66,21 @@ __authors__  = (
     'Naglis Jonaitis',
     'Charles Chen',
     'Hassaan Ali',
+    'Dobrosław Żybort',
+    'David Fabijan',
+    'Sebastian Haas',
+    'Alexander Kirk',
 )
 
 __license__ = 'Public Domain'
 
 import codecs
 import io
-import locale
 import optparse
 import os
 import random
-import re
 import shlex
+import shutil
 import sys
 
 
@@ -99,7 +102,7 @@ from .utils import (
     write_string,
 )
 from .update import update_self
-from .FileDownloader import (
+from .downloader import (
     FileDownloader,
 )
 from .extractor import gen_extractors
@@ -224,6 +227,7 @@ def parseOpts(overrideArguments=None):
     downloader     = optparse.OptionGroup(parser, 'Download Options')
     postproc       = optparse.OptionGroup(parser, 'Post-processing Options')
     filesystem     = optparse.OptionGroup(parser, 'Filesystem Options')
+    workarounds    = optparse.OptionGroup(parser, 'Workarounds')
     verbosity      = optparse.OptionGroup(parser, 'Verbosity / Simulation Options')
 
     general.add_option('-h', '--help',
@@ -240,14 +244,6 @@ def parseOpts(overrideArguments=None):
     general.add_option('--dump-user-agent',
             action='store_true', dest='dump_user_agent',
             help='display the current browser identification', default=False)
-    general.add_option('--user-agent',
-            dest='user_agent', help='specify a custom user agent', metavar='UA')
-    general.add_option('--referer',
-            dest='referer', help='specify a custom referer, use if the video access is restricted to one domain',
-            metavar='REF', default=None)
-    general.add_option('--add-header',
-            dest='headers', help='specify a custom HTTP header and its value, separated by a colon \':\'. You can use this option multiple times', action="append",
-            metavar='FIELD:VALUE')
     general.add_option('--list-extractors',
             action='store_true', dest='list_extractors',
             help='List all supported extractors and the URLs they would handle', default=False)
@@ -257,33 +253,17 @@ def parseOpts(overrideArguments=None):
     general.add_option(
         '--proxy', dest='proxy', default=None, metavar='URL',
         help='Use the specified HTTP/HTTPS proxy. Pass in an empty string (--proxy "") for direct connection')
-    general.add_option('--no-check-certificate', action='store_true', dest='no_check_certificate', default=False, help='Suppress HTTPS certificate validation.')
-    general.add_option(
-        '--prefer-insecure', '--prefer-unsecure', action='store_true', dest='prefer_insecure',
-        help='Use an unencrypted connection to retrieve information about the video. (Currently supported only for YouTube)')
-    general.add_option(
-        '--cache-dir', dest='cachedir', default=get_cachedir(), metavar='DIR',
-        help='Location in the filesystem where youtube-dl can store some downloaded information permanently. By default $XDG_CACHE_HOME/youtube-dl or ~/.cache/youtube-dl . At the moment, only YouTube player files (for videos with obfuscated signatures) are cached, but that may change.')
-    general.add_option(
-        '--no-cache-dir', action='store_const', const=None, dest='cachedir',
-        help='Disable filesystem caching')
     general.add_option(
         '--socket-timeout', dest='socket_timeout',
         type=float, default=None, help=u'Time to wait before giving up, in seconds')
-    general.add_option(
-        '--bidi-workaround', dest='bidi_workaround', action='store_true',
-        help=u'Work around terminals that lack bidirectional text support. Requires bidiv or fribidi executable in PATH')
     general.add_option(
         '--default-search',
         dest='default_search', metavar='PREFIX',
-        help='Use this prefix for unqualified URLs. For example "gvsearch2:" downloads two videos from google videos for  youtube-dl "large apple". Use the value "auto" to let youtube-dl guess. The default value "error" just throws an error.')
+        help='Use this prefix for unqualified URLs. For example "gvsearch2:" downloads two videos from google videos for  youtube-dl "large apple". Use the value "auto" to let youtube-dl guess ("auto_warning" to emit a warning when guessing). "error" just throws an error. The default value "fixup_error" repairs broken URLs, but emits an error if this is not possible instead of searching.')
     general.add_option(
         '--ignore-config',
         action='store_true',
         help='Do not read configuration files. When given in the global configuration file /etc/youtube-dl.conf: do not read the user configuration in ~/.config/youtube-dl.conf (%APPDATA%/youtube-dl/config.txt on Windows)')
-    general.add_option(
-        '--encoding', dest='encoding', metavar='ENCODING',
-        help='Force the specified encoding (experimental)')
 
     selection.add_option(
         '--playlist-start',
@@ -384,6 +364,33 @@ def parseOpts(overrideArguments=None):
             help='do not automatically adjust the buffer size. By default, the buffer size is automatically resized from an initial value of SIZE.', default=False)
     downloader.add_option('--test', action='store_true', dest='test', default=False, help=optparse.SUPPRESS_HELP)
 
+    workarounds.add_option(
+        '--encoding', dest='encoding', metavar='ENCODING',
+        help='Force the specified encoding (experimental)')
+    workarounds.add_option(
+        '--no-check-certificate', action='store_true',
+        dest='no_check_certificate', default=False,
+        help='Suppress HTTPS certificate validation.')
+    workarounds.add_option(
+        '--prefer-insecure', '--prefer-unsecure', action='store_true', dest='prefer_insecure',
+        help='Use an unencrypted connection to retrieve information about the video. (Currently supported only for YouTube)')
+    workarounds.add_option(
+        '--user-agent', metavar='UA',
+        dest='user_agent', help='specify a custom user agent')
+    workarounds.add_option(
+        '--referer', metavar='REF',
+        dest='referer', default=None,
+        help='specify a custom referer, use if the video access is restricted to one domain',
+    )
+    workarounds.add_option(
+        '--add-header', metavar='FIELD:VALUE',
+        dest='headers', action='append',
+        help='specify a custom HTTP header and its value, separated by a colon \':\'. You can use this option multiple times',
+    )
+    workarounds.add_option(
+        '--bidi-workaround', dest='bidi_workaround', action='store_true',
+        help=u'Work around terminals that lack bidirectional text support. Requires bidiv or fribidi executable in PATH')
+
     verbosity.add_option('-q', '--quiet',
             action='store_true', dest='quiet', help='activates quiet mode', default=False)
     verbosity.add_option(
@@ -441,12 +448,10 @@ def parseOpts(overrideArguments=None):
             help='Display sent and read HTTP traffic')
 
 
-    filesystem.add_option('-t', '--title',
-            action='store_true', dest='usetitle', help='use title in file name (default)', default=False)
+    filesystem.add_option('-a', '--batch-file',
+            dest='batchfile', metavar='FILE', help='file containing URLs to download (\'-\' for stdin)')
     filesystem.add_option('--id',
             action='store_true', dest='useid', help='use only video ID in file name', default=False)
-    filesystem.add_option('-l', '--literal',
-            action='store_true', dest='usetitle', help='[deprecated] alias of --title', default=False)
     filesystem.add_option('-A', '--auto-number',
             action='store_true', dest='autonumber',
             help='number downloaded files starting from 00000', default=False)
@@ -472,11 +477,10 @@ def parseOpts(overrideArguments=None):
     filesystem.add_option('--restrict-filenames',
             action='store_true', dest='restrictfilenames',
             help='Restrict filenames to only ASCII characters, and avoid "&" and spaces in filenames', default=False)
-    filesystem.add_option('-a', '--batch-file',
-            dest='batchfile', metavar='FILE', help='file containing URLs to download (\'-\' for stdin)')
-    filesystem.add_option('--load-info',
-            dest='load_info_filename', metavar='FILE',
-            help='json file containing the video information (created with the "--write-json" option)')
+    filesystem.add_option('-t', '--title',
+            action='store_true', dest='usetitle', help='[deprecated] use title in file name (default)', default=False)
+    filesystem.add_option('-l', '--literal',
+            action='store_true', dest='usetitle', help='[deprecated] alias of --title', default=False)
     filesystem.add_option('-w', '--no-overwrites',
             action='store_true', dest='nooverwrites', help='do not overwrite files', default=False)
     filesystem.add_option('-c', '--continue',
@@ -484,8 +488,6 @@ def parseOpts(overrideArguments=None):
     filesystem.add_option('--no-continue',
             action='store_false', dest='continue_dl',
             help='do not resume partially downloaded files (restart from beginning)')
-    filesystem.add_option('--cookies',
-            dest='cookiefile', metavar='FILE', help='file to read cookies from and dump cookie jar in')
     filesystem.add_option('--no-part',
             action='store_true', dest='nopart', help='do not use .part files', default=False)
     filesystem.add_option('--no-mtime',
@@ -503,6 +505,20 @@ def parseOpts(overrideArguments=None):
     filesystem.add_option('--write-thumbnail',
             action='store_true', dest='writethumbnail',
             help='write thumbnail image to disk', default=False)
+    filesystem.add_option('--load-info',
+            dest='load_info_filename', metavar='FILE',
+            help='json file containing the video information (created with the "--write-json" option)')
+    filesystem.add_option('--cookies',
+            dest='cookiefile', metavar='FILE', help='file to read cookies from and dump cookie jar in')
+    filesystem.add_option(
+        '--cache-dir', dest='cachedir', default=get_cachedir(), metavar='DIR',
+        help='Location in the filesystem where youtube-dl can store some downloaded information permanently. By default $XDG_CACHE_HOME/youtube-dl or ~/.cache/youtube-dl . At the moment, only YouTube player files (for videos with obfuscated signatures) are cached, but that may change.')
+    filesystem.add_option(
+        '--no-cache-dir', action='store_const', const=None, dest='cachedir',
+        help='Disable filesystem caching')
+    filesystem.add_option(
+        '--rm-cache-dir', action='store_true', dest='rm_cachedir',
+        help='Delete all filesystem cache files')
 
 
     postproc.add_option('-x', '--extract-audio', action='store_true', dest='extractaudio', default=False,
@@ -536,6 +552,7 @@ def parseOpts(overrideArguments=None):
     parser.add_option_group(downloader)
     parser.add_option_group(filesystem)
     parser.add_option_group(verbosity)
+    parser.add_option_group(workarounds)
     parser.add_option_group(video_format)
     parser.add_option_group(subtitles)
     parser.add_option_group(authentication)
@@ -635,7 +652,7 @@ def _real_main(argv=None):
             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')
+                _SEARCHES = (u'cute kittens', u'slithering pythons', u'falling cat', u'angry poodle', u'purple fish', u'running tortoise', u'sleeping bunny')
                 _COUNTS = (u'', u'5', u'10', u'all')
                 desc += u' (Example: "%s%s:%s" )' % (ie.SEARCH_KEY, random.choice(_COUNTS), random.choice(_SEARCHES))
             compat_print(desc)
@@ -696,7 +713,7 @@ def _real_main(argv=None):
         date = DateRange.day(opts.date)
     else:
         date = DateRange(opts.dateafter, opts.datebefore)
-    if opts.default_search not in ('auto', 'auto_warning', None) and ':' not in opts.default_search:
+    if opts.default_search not in ('auto', 'auto_warning', 'error', 'fixup_error', None) and ':' not in opts.default_search:
         parser.error(u'--default-search invalid; did you forget a colon (:) at the end?')
 
     # Do not download videos when there are audio-only formats
@@ -835,9 +852,26 @@ def _real_main(argv=None):
         if opts.update_self:
             update_self(ydl.to_screen, opts.verbose)
 
+        # Remove cache dir
+        if opts.rm_cachedir:
+            if opts.cachedir is None:
+                ydl.to_screen(u'No cache dir specified (Did you combine --no-cache-dir and --rm-cache-dir?)')
+            else:
+                if ('.cache' not in opts.cachedir) or ('youtube-dl' not in opts.cachedir):
+                    ydl.to_screen(u'Not removing directory %s - this does not look like a cache dir')
+                    retcode = 141
+                else:
+                    ydl.to_screen(
+                        u'Removing cache dir %s .' % opts.cachedir,
+                        skip_eol=True)
+                    if os.path.exists(opts.cachedir):
+                        ydl.to_screen(u'.', skip_eol=True)
+                        shutil.rmtree(opts.cachedir)
+                    ydl.to_screen(u'.')
+
         # Maybe do nothing
         if (len(all_urls) < 1) and (opts.load_info_filename is None):
-            if not opts.update_self:
+            if not (opts.update_self or opts.rm_cachedir):
                 parser.error(u'you must provide at least one URL')
             else:
                 sys.exit()

youtube_dl/downloader/common.py

@@ -292,7 +292,7 @@ class FileDownloader(object):
 
     def real_download(self, filename, info_dict):
         """Real download process. Redefine in subclasses."""
-        raise NotImplementedError(u'This method must be implemented by sublcasses')
+        raise NotImplementedError(u'This method must be implemented by subclasses')
 
     def _hook_progress(self, status):
         for ph in self._progress_hooks:

youtube_dl/downloader/f4m.py

@@ -220,6 +220,7 @@ class F4mFD(FileDownloader):
 
     def real_download(self, filename, info_dict):
         man_url = info_dict['url']
+        requested_bitrate = info_dict.get('tbr')
         self.to_screen('[download] Downloading f4m manifest')
         manifest = self.ydl.urlopen(man_url).read()
         self.report_destination(filename)
@@ -233,8 +234,14 @@ class F4mFD(FileDownloader):
 
         doc = etree.fromstring(manifest)
         formats = [(int(f.attrib.get('bitrate', -1)), f) for f in doc.findall(_add_ns('media'))]
-        formats = sorted(formats, key=lambda f: f[0])
-        rate, media = formats[-1]
+        if requested_bitrate is None:
+            # get the best format
+            formats = sorted(formats, key=lambda f: f[0])
+            rate, media = formats[-1]
+        else:
+            rate, media = list(filter(
+                lambda f: int(f[0]) == requested_bitrate, formats))[0]
+
         base_url = compat_urlparse.urljoin(man_url, media.attrib['url'])
         bootstrap = base64.b64decode(doc.find(_add_ns('bootstrapInfo')).text)
         metadata = base64.b64decode(media.find(_add_ns('metadata')).text)

youtube_dl/extractor/__init__.py

@@ -1,3 +1,4 @@
+from .abc import ABCIE
 from .academicearth import AcademicEarthCourseIE
 from .addanime import AddAnimeIE
 from .adultswim import AdultSwimIE
@@ -76,6 +77,10 @@ from .ebaumsworld import EbaumsWorldIE
 from .ehow import EHowIE
 from .eighttracks import EightTracksIE
 from .eitb import EitbIE
+from .ellentv import (
+    EllenTVIE,
+    EllenTVClipsIE,
+)
 from .elpais import ElPaisIE
 from .empflix import EmpflixIE
 from .engadget import EngadgetIE
@@ -111,9 +116,11 @@ from .funnyordie import FunnyOrDieIE
 from .gamekings import GamekingsIE
 from .gameone import GameOneIE
 from .gamespot import GameSpotIE
+from .gamestar import GameStarIE
 from .gametrailers import GametrailersIE
 from .gdcvault import GDCVaultIE
 from .generic import GenericIE
+from .godtube import GodTubeIE
 from .googleplus import GooglePlusIE
 from .googlesearch import GoogleSearchIE
 from .gorillavid import GorillaVidIE
@@ -123,6 +130,7 @@ from .helsinki import HelsinkiIE
 from .hentaistigma import HentaiStigmaIE
 from .hotnewhiphop import HotNewHipHopIE
 from .howcast import HowcastIE
+from .howstuffworks import HowStuffWorksIE
 from .huffpost import HuffPostIE
 from .hypem import HypemIE
 from .iconosquare import IconosquareIE
@@ -140,8 +148,10 @@ from .ivi import (
     IviIE,
     IviCompilationIE
 )
+from .izlesene import IzleseneIE
 from .jadorecettepub import JadoreCettePubIE
 from .jeuxvideo import JeuxVideoIE
+from .jove import JoveIE
 from .jukebox import JukeboxIE
 from .justintv import JustinTVIE
 from .jpopsukitv import JpopsukiIE
@@ -151,6 +161,7 @@ from .khanacademy import KhanAcademyIE
 from .kickstarter import KickStarterIE
 from .keek import KeekIE
 from .kontrtube import KontrTubeIE
+from .krasview import KrasViewIE
 from .ku6 import Ku6IE
 from .la7 import LA7IE
 from .lifenews import LifeNewsIE
@@ -172,10 +183,12 @@ from .mdr import MDRIE
 from .metacafe import MetacafeIE
 from .metacritic import MetacriticIE
 from .mit import TechTVMITIE, MITIE, OCWMITIE
+from .mitele import MiTeleIE
 from .mixcloud import MixcloudIE
 from .mlb import MLBIE
 from .mpora import MporaIE
 from .mofosex import MofosexIE
+from .mojvideo import MojvideoIE
 from .mooshare import MooshareIE
 from .morningstar import MorningstarIE
 from .motherless import MotherlessIE
@@ -219,9 +232,12 @@ from .nrk import (
 from .ntv import NTVIE
 from .nytimes import NYTimesIE
 from .nuvid import NuvidIE
-from .oe1 import OE1IE
 from .ooyala import OoyalaIE
-from .orf import ORFIE
+from .orf import (
+    ORFTVthekIE,
+    ORFOE1IE,
+    ORFFM4IE,
+)
 from .parliamentliveuk import ParliamentLiveUKIE
 from .pbs import PBSIE
 from .photobucket import PhotobucketIE
@@ -242,6 +258,7 @@ from .ro220 import Ro220IE
 from .rottentomatoes import RottenTomatoesIE
 from .roxwel import RoxwelIE
 from .rtbf import RTBFIE
+from .rtlnl import RtlXlIE
 from .rtlnow import RTLnowIE
 from .rts import RTSIE
 from .rtve import RTVEALaCartaIE
@@ -258,6 +275,7 @@ from .savefrom import SaveFromIE
 from .scivee import SciVeeIE
 from .screencast import ScreencastIE
 from .servingsys import ServingSysIE
+from .shared import SharedIE
 from .sina import SinaIE
 from .slideshare import SlideshareIE
 from .slutload import SlutloadIE
@@ -267,6 +285,8 @@ from .smotri import (
     SmotriUserIE,
     SmotriBroadcastIE,
 )
+from .snotr import SnotrIE
+from .sockshare import SockshareIE
 from .sohu import SohuIE
 from .soundcloud import (
     SoundcloudIE,
@@ -318,6 +338,8 @@ from .tumblr import TumblrIE
 from .tutv import TutvIE
 from .tvigle import TvigleIE
 from .tvp import TvpIE
+from .tvplay import TVPlayIE
+from .ubu import UbuIE
 from .udemy import (
     UdemyIE,
     UdemyCourseIE
@@ -339,6 +361,7 @@ from .videofyme import VideofyMeIE
 from .videopremium import VideoPremiumIE
 from .videott import VideoTtIE
 from .videoweed import VideoWeedIE
+from .vidme import VidmeIE
 from .vimeo import (
     VimeoIE,
     VimeoChannelIE,
@@ -372,6 +395,7 @@ from .wistia import WistiaIE
 from .worldstarhiphop import WorldStarHipHopIE
 from .wrzuta import WrzutaIE
 from .xbef import XBefIE
+from .xboxclips import XboxClipsIE
 from .xhamster import XHamsterIE
 from .xnxx import XNXXIE
 from .xvideos import XVideosIE

youtube_dl/extractor/abc.py

@@ -0,0 +1,48 @@
+from __future__ import unicode_literals
+
+import re
+import json
+
+from .common import InfoExtractor
+
+
+class ABCIE(InfoExtractor):
+    IE_NAME = 'abc.net.au'
+    _VALID_URL = r'http://www\.abc\.net\.au/news/[^/]+/[^/]+/(?P<id>\d+)'
+
+    _TEST = {
+        'url': 'http://www.abc.net.au/news/2014-07-25/bringing-asylum-seekers-to-australia-would-give/5624716',
+        'md5': 'dad6f8ad011a70d9ddf887ce6d5d0742',
+        'info_dict': {
+            'id': '5624716',
+            'ext': 'mp4',
+            'title': 'Bringing asylum seekers to Australia would give them right to asylum claims: professor',
+            'description': 'md5:ba36fa5e27e5c9251fd929d339aea4af',
+        },
+    }
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        video_id = mobj.group('id')
+        webpage = self._download_webpage(url, video_id)
+
+        urls_info_json = self._search_regex(
+            r'inlineVideoData\.push\((.*?)\);', webpage, 'video urls',
+            flags=re.DOTALL)
+        urls_info = json.loads(urls_info_json.replace('\'', '"'))
+        formats = [{
+            'url': url_info['url'],
+            'width': int(url_info['width']),
+            'height': int(url_info['height']),
+            'tbr': int(url_info['bitrate']),
+            'filesize': int(url_info['filesize']),
+        } for url_info in urls_info]
+        self._sort_formats(formats)
+
+        return {
+            'id': video_id,
+            'title': self._og_search_title(webpage),
+            'formats': formats,
+            'description': self._og_search_description(webpage),
+            'thumbnail': self._og_search_thumbnail(webpage),
+        }

youtube_dl/extractor/allocine.py

@@ -32,7 +32,7 @@ class AllocineIE(InfoExtractor):
             'id': '19540403',
             'ext': 'mp4',
             'title': 'Planes 2 Bande-annonce VF',
-            'description': 'md5:c4b1f7bd682a91de6491ada267ec0f4d',
+            'description': 'md5:eeaffe7c2d634525e21159b93acf3b1e',
             'thumbnail': 're:http://.*\.jpg',
         },
     }, {
@@ -42,7 +42,7 @@ class AllocineIE(InfoExtractor):
             'id': '19544709',
             'ext': 'mp4',
             'title': 'Dragons 2 - Bande annonce finale VF',
-            'description': 'md5:e74a4dc750894bac300ece46c7036490',
+            'description': 'md5:71742e3a74b0d692c7fce0dd2017a4ac',
             'thumbnail': 're:http://.*\.jpg',
         },
     }]

youtube_dl/extractor/appletrailers.py

@@ -6,6 +6,7 @@ import json
 from .common import InfoExtractor
 from ..utils import (
     compat_urlparse,
+    int_or_none,
 )
 
 
@@ -110,8 +111,8 @@ class AppleTrailersIE(InfoExtractor):
                 formats.append({
                     'url': format_url,
                     'format': format['type'],
-                    'width': format['width'],
-                    'height': int(format['height']),
+                    'width': int_or_none(format['width']),
+                    'height': int_or_none(format['height']),
                 })
 
             self._sort_formats(formats)

youtube_dl/extractor/ard.py

@@ -7,23 +7,34 @@ from .common import InfoExtractor
 from ..utils import (
     determine_ext,
     ExtractorError,
+    qualities,
+    compat_urllib_parse_urlparse,
+    compat_urllib_parse,
 )
 
 
 class ARDIE(InfoExtractor):
-    _VALID_URL = r'^https?://(?:(?:www\.)?ardmediathek\.de|mediathek\.daserste\.de)/(?:.*/)(?P<video_id>[^/\?]+)(?:\?.*)?'
+    _VALID_URL = r'^https?://(?:(?:www\.)?ardmediathek\.de|mediathek\.daserste\.de)/(?:.*/)(?P<video_id>[0-9]+|[^0-9][^/\?]+)[^/\?]*(?:\?.*)?'
 
-    _TEST = {
-        'url': 'http://www.ardmediathek.de/das-erste/guenther-jauch/edward-snowden-im-interview-held-oder-verraeter?documentId=19288786',
-        'file': '19288786.mp4',
-        'md5': '515bf47ce209fb3f5a61b7aad364634c',
+    _TESTS = [{
+        'url': 'http://mediathek.daserste.de/sendungen_a-z/328454_anne-will/22429276_vertrauen-ist-gut-spionieren-ist-besser-geht',
+        'file': '22429276.mp4',
+        'md5': '469751912f1de0816a9fc9df8336476c',
         'info_dict': {
-            'title': 'Edward Snowden im Interview - Held oder Verräter?',
-            'description': 'Edward Snowden hat alles aufs Spiel gesetzt, um die weltweite \xdcberwachung durch die Geheimdienste zu enttarnen. Nun stellt sich der ehemalige NSA-Mitarbeiter erstmals weltweit in einem TV-Interview den Fragen eines NDR-Journalisten. Die Sendung vom Sonntagabend.',
-            'thumbnail': 'http://www.ardmediathek.de/ard/servlet/contentblob/19/28/87/90/19288790/bild/2250037',
+            'title': 'Vertrauen ist gut, Spionieren ist besser - Geht so deutsch-amerikanische Freundschaft?',
+            'description': 'Das Erste Mediathek [ARD]: Vertrauen ist gut, Spionieren ist besser - Geht so deutsch-amerikanische Freundschaft?, Anne Will, Über die Spionage-Affäre diskutieren Clemens Binninger, Katrin Göring-Eckardt, Georg Mascolo, Andrew B. Denison und Constanze Kurz.. Das Video zur Sendung Anne Will am Mittwoch, 16.07.2014',
         },
         'skip': 'Blocked outside of Germany',
-    }
+    }, {
+        'url': 'http://www.ardmediathek.de/tv/Tatort/Das-Wunder-von-Wolbeck-Video-tgl-ab-20/Das-Erste/Video?documentId=22490580&bcastId=602916',
+        'info_dict': {
+            'id': '22490580',
+            'ext': 'mp4',
+            'title': 'Das Wunder von Wolbeck (Video tgl. ab 20 Uhr)',
+            'description': 'Auf einem restaurierten Hof bei Wolbeck wird der Heilpraktiker Raffael Lembeck eines morgens von seiner Frau Stella tot aufgefunden. Das Opfer war offensichtlich in seiner Praxis zu Fall gekommen und ist dann verblutet, erklärt Prof. Boerne am Tatort.',
+        },
+        'skip': 'Blocked outside of Germany',
+    }]
 
     def _real_extract(self, url):
         # determine video id from url
@@ -35,48 +46,78 @@ class ARDIE(InfoExtractor):
         else:
             video_id = m.group('video_id')
 
+        urlp = compat_urllib_parse_urlparse(url)
+        url = urlp._replace(path=compat_urllib_parse.quote(urlp.path.encode('utf-8'))).geturl()
+
         webpage = self._download_webpage(url, video_id)
 
+        if '>Der gewünschte Beitrag ist nicht mehr verfügbar.<' in webpage:
+            raise ExtractorError('Video %s is no longer available' % video_id, expected=True)
+
         title = self._html_search_regex(
             [r'<h1(?:\s+class="boxTopHeadline")?>(.*?)</h1>',
              r'<meta name="dcterms.title" content="(.*?)"/>',
              r'<h4 class="headline">(.*?)</h4>'],
             webpage, 'title')
         description = self._html_search_meta(
-            'dcterms.abstract', webpage, 'description')
-        thumbnail = self._og_search_thumbnail(webpage)
+            'dcterms.abstract', webpage, 'description', default=None)
+        if description is None:
+            description = self._html_search_meta(
+                'description', webpage, 'meta description')
 
+        # Thumbnail is sometimes not present.
+        # It is in the mobile version, but that seems to use a different URL
+        # structure altogether.
+        thumbnail = self._og_search_thumbnail(webpage, default=None)
 
-        media_info = self._download_json(
-            'http://www.ardmediathek.de/play/media/%s' % video_id, video_id)
-        # The second element of the _mediaArray contains the standard http urls
-        streams = media_info['_mediaArray'][1]['_mediaStreamArray']
-        if not streams:
-            if '"fsk"' in webpage:
-                raise ExtractorError('This video is only available after 20:00')
+        media_streams = re.findall(r'''(?x)
+            mediaCollection\.addMediaStream\([0-9]+,\s*[0-9]+,\s*"[^"]*",\s*
+            "([^"]+)"''', webpage)
 
-        formats = []
+        if media_streams:
+            QUALITIES = qualities(['lo', 'hi', 'hq'])
+            formats = []
+            for furl in set(media_streams):
+                if furl.endswith('.f4m'):
+                    fid = 'f4m'
+                else:
+                    fid_m = re.match(r'.*\.([^.]+)\.[^.]+$', furl)
+                    fid = fid_m.group(1) if fid_m else None
+                formats.append({
+                    'quality': QUALITIES(fid),
+                    'format_id': fid,
+                    'url': furl,
+                })
+        else:  # request JSON file
+            media_info = self._download_json(
+                'http://www.ardmediathek.de/play/media/%s' % video_id, video_id)
+            # The second element of the _mediaArray contains the standard http urls
+            streams = media_info['_mediaArray'][1]['_mediaStreamArray']
+            if not streams:
+                if '"fsk"' in webpage:
+                    raise ExtractorError('This video is only available after 20:00')
 
-        for s in streams:
-            if type(s['_stream']) == list:
-                for index, url in enumerate(s['_stream'][::-1]):
-                    quality = s['_quality'] + index
-                    formats.append({
-                        'quality': quality,
-                        'url': url,
-                        'format_id': '%s-%s' % (determine_ext(url), quality)
+            formats = []
+            for s in streams:
+                if type(s['_stream']) == list:
+                    for index, url in enumerate(s['_stream'][::-1]):
+                        quality = s['_quality'] + index
+                        formats.append({
+                            'quality': quality,
+                            'url': url,
+                            'format_id': '%s-%s' % (determine_ext(url), quality)
                         })
-                continue
+                    continue
 
-            format = {
-                'quality': s['_quality'],
-                'url': s['_stream'],
-            }
+                format = {
+                    'quality': s['_quality'],
+                    'url': s['_stream'],
+                }
 
-            format['format_id'] = '%s-%s' % (
-                determine_ext(format['url']), format['quality'])
+                format['format_id'] = '%s-%s' % (
+                    determine_ext(format['url']), format['quality'])
 
-            formats.append(format)
+                formats.append(format)
 
         self._sort_formats(formats)
 

youtube_dl/extractor/arte.py

@@ -109,15 +109,19 @@ class ArteTVPlus7IE(InfoExtractor):
             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
+        # TODO: Might want not to drop videos that does not match requested language
+        # but to process those formats with lower precedence
         formats = filter(_match_lang, all_formats)
-        formats = list(formats) # in python3 filter returns an iterator
+        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' or f['versionCode'] == 'VA' for f in all_formats):
-                formats = all_formats
-            else:
-                raise ExtractorError(u'The formats list is empty')
+            # Sometimes there are neither videos of requested lang code
+            # nor original version videos available
+            # For such cases we just take all_formats as is
+            formats = all_formats
+            if not formats:
+                raise ExtractorError('The formats list is empty')
 
         if re.match(r'[A-Z]Q', formats[0]['quality']) is not None:
             def sort_key(f):

youtube_dl/extractor/blinkx.py

@@ -52,7 +52,7 @@ class BlinkxIE(InfoExtractor):
                     'height': int(m['h']),
                 })
             elif m['type'] == 'original':
-                duration = m['d']
+                duration = float(m['d'])
             elif m['type'] == 'youtube':
                 yt_id = m['link']
                 self.to_screen('Youtube video detected: %s' % yt_id)

youtube_dl/extractor/bloomberg.py

@@ -10,7 +10,7 @@ class BloombergIE(InfoExtractor):
 
     _TEST = {
         'url': 'http://www.bloomberg.com/video/shah-s-presentation-on-foreign-exchange-strategies-qurhIVlJSB6hzkVi229d8g.html',
-        'md5': '7bf08858ff7c203c870e8a6190e221e5',
+        # The md5 checksum changes
         'info_dict': {
             'id': 'qurhIVlJSB6hzkVi229d8g',
             'ext': 'flv',
@@ -31,8 +31,7 @@ class BloombergIE(InfoExtractor):
         return {
             'id': name.split('-')[-1],
             'title': title,
-            'url': f4m_url,
-            'ext': 'flv',
+            'formats': self._extract_f4m_formats(f4m_url, name),
             'description': self._og_search_description(webpage),
             'thumbnail': self._og_search_thumbnail(webpage),
         }

youtube_dl/extractor/br.py

@@ -7,12 +7,13 @@ from .common import InfoExtractor
 from ..utils import (
     ExtractorError,
     int_or_none,
+    parse_duration,
 )
 
 
 class BRIE(InfoExtractor):
     IE_DESC = 'Bayerischer Rundfunk Mediathek'
-    _VALID_URL = r'https?://(?:www\.)?br\.de/(?:[a-z0-9\-]+/)+(?P<id>[a-z0-9\-]+)\.html'
+    _VALID_URL = r'https?://(?:www\.)?br\.de/(?:[a-z0-9\-_]+/)+(?P<id>[a-z0-9\-_]+)\.html'
     _BASE_URL = 'http://www.br.de'
 
     _TESTS = [
@@ -22,8 +23,9 @@ class BRIE(InfoExtractor):
             'info_dict': {
                 'id': '25e279aa-1ffd-40fd-9955-5325bd48a53a',
                 'ext': 'mp4',
-                'title': 'Am 1. und 2. August in Oberammergau',
-                'description': 'md5:dfd224e5aa6819bc1fcbb7826a932021',
+                'title': 'Wenn das Traditions-Theater wackelt',
+                'description': 'Heimatsound-Festival 2014: Wenn das Traditions-Theater wackelt',
+                'duration': 34,
             }
         },
         {
@@ -34,6 +36,7 @@ class BRIE(InfoExtractor):
                 'ext': 'mp4',
                 'title': 'Über den Pass',
                 'description': 'Die Eroberung der Alpen: Über den Pass',
+                'duration': 2588,
             }
         },
         {
@@ -44,6 +47,7 @@ class BRIE(InfoExtractor):
                 'ext': 'aac',
                 'title': '"Keine neuen Schulden im nächsten Jahr"',
                 'description': 'Haushaltsentwurf: "Keine neuen Schulden im nächsten Jahr"',
+                'duration': 64,
             }
         },
         {
@@ -54,6 +58,7 @@ class BRIE(InfoExtractor):
                 'ext': 'mp4',
                 'title': 'Umweltbewusster Häuslebauer',
                 'description': 'Uwe Erdelt: Umweltbewusster Häuslebauer',
+                'duration': 116,
             }
         },
         {
@@ -64,6 +69,7 @@ class BRIE(InfoExtractor):
                 'ext': 'mp4',
                 'title': 'Folge 1 - Metaphysik',
                 'description': 'Kant für Anfänger: Folge 1 - Metaphysik',
+                'duration': 893,
                 'uploader': 'Eva Maria Steimle',
                 'upload_date': '20140117',
             }
@@ -84,6 +90,7 @@ class BRIE(InfoExtractor):
             media = {
                 'id': xml_media.get('externalId'),
                 'title': xml_media.find('title').text,
+                'duration': parse_duration(xml_media.find('duration').text),
                 'formats': self._extract_formats(xml_media.find('assets')),
                 'thumbnails': self._extract_thumbnails(xml_media.find('teaserImage/variants')),
                 'description': ' '.join(xml_media.find('shareTitle').text.splitlines()),

youtube_dl/extractor/cbs.py

@@ -1,24 +1,42 @@
+from __future__ import unicode_literals
+
 import re
 
 from .common import InfoExtractor
 
 
 class CBSIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:www\.)?cbs\.com/shows/[^/]+/video/(?P<id>[^/]+)/.*'
+    _VALID_URL = r'https?://(?:www\.)?cbs\.com/shows/[^/]+/(?:video|artist)/(?P<id>[^/]+)/.*'
 
-    _TEST = {
-        u'url': u'http://www.cbs.com/shows/garth-brooks/video/_u7W953k6la293J7EPTd9oHkSPs6Xn6_/connect-chat-feat-garth-brooks/',
-        u'file': u'4JUVEwq3wUT7.flv',
-        u'info_dict': {
-            u'title': u'Connect Chat feat. Garth Brooks',
-            u'description': u'Connect with country music singer Garth Brooks, as he chats with fans on Wednesday November 27, 2013. Be sure to tune in to Garth Brooks: Live from Las Vegas, Friday November 29, at 9/8c on CBS!',
-            u'duration': 1495,
+    _TESTS = [{
+        'url': 'http://www.cbs.com/shows/garth-brooks/video/_u7W953k6la293J7EPTd9oHkSPs6Xn6_/connect-chat-feat-garth-brooks/',
+        'info_dict': {
+            'id': '4JUVEwq3wUT7',
+            'ext': 'flv',
+            'title': 'Connect Chat feat. Garth Brooks',
+            'description': 'Connect with country music singer Garth Brooks, as he chats with fans on Wednesday November 27, 2013. Be sure to tune in to Garth Brooks: Live from Las Vegas, Friday November 29, at 9/8c on CBS!',
+            'duration': 1495,
         },
-        u'params': {
+        'params': {
             # rtmp download
-            u'skip_download': True,
+            'skip_download': True,
         },
-    }
+        '_skip': 'Blocked outside the US',
+    }, {
+        'url': 'http://www.cbs.com/shows/liveonletterman/artist/221752/st-vincent/',
+        'info_dict': {
+            'id': 'P9gjWjelt6iP',
+            'ext': 'flv',
+            'title': 'Live on Letterman - St. Vincent',
+            'description': 'Live On Letterman: St. Vincent in concert from New York\'s Ed Sullivan Theater on Tuesday, July 16, 2014.',
+            'duration': 3221,
+        },
+        'params': {
+            # rtmp download
+            'skip_download': True,
+        },
+        '_skip': 'Blocked outside the US',
+    }]
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
@@ -26,5 +44,5 @@ class CBSIE(InfoExtractor):
         webpage = self._download_webpage(url, video_id)
         real_id = self._search_regex(
             r"video\.settings\.pid\s*=\s*'([^']+)';",
-            webpage, u'real video ID')
+            webpage, 'real video ID')
         return self.url_result(u'theplatform:%s' % real_id)

youtube_dl/extractor/chilloutzone.py

@@ -42,7 +42,7 @@ class ChilloutzoneIE(InfoExtractor):
             'id': '85523671',
             'ext': 'mp4',
             'title': 'The Sunday Times - Icons',
-            'description': 'md5:3e1c0dc6047498d6728dcdaad0891762',
+            'description': 'md5:a5f7ff82e2f7a9ed77473fe666954e84',
             'uploader': 'Us',
             'uploader_id': 'usfilms',
             'upload_date': '20140131'

youtube_dl/extractor/cnet.py

@@ -43,7 +43,11 @@ class CNETIE(InfoExtractor):
             raise ExtractorError('Cannot find video data')
 
         video_id = vdata['id']
-        title = vdata['headline']
+        title = vdata.get('headline')
+        if title is None:
+            title = vdata.get('title')
+        if title is None:
+            raise ExtractorError('Cannot find title!')
         description = vdata.get('dek')
         thumbnail = vdata.get('image', {}).get('path')
         author = vdata.get('author')

youtube_dl/extractor/common.py

@@ -18,6 +18,7 @@ from ..utils import (
     clean_html,
     compiled_regex_type,
     ExtractorError,
+    int_or_none,
     RegexNotFoundError,
     sanitize_filename,
     unescapeHTML,
@@ -69,6 +70,7 @@ class InfoExtractor(object):
                     * vcodec     Name of the video codec in use
                     * container  Name of the container format
                     * filesize   The number of bytes, if known in advance
+                    * filesize_approx  An estimate for the number of bytes
                     * player_url SWF Player URL (used for rtmpdump).
                     * protocol   The protocol that will be used for the actual
                                  download, lower-case.
@@ -300,8 +302,12 @@ class InfoExtractor(object):
     def _download_json(self, url_or_request, video_id,
                        note=u'Downloading JSON metadata',
                        errnote=u'Unable to download JSON metadata',
-                       transform_source=None):
-        json_string = self._download_webpage(url_or_request, video_id, note, errnote)
+                       transform_source=None,
+                       fatal=True):
+        json_string = self._download_webpage(
+            url_or_request, video_id, note, errnote, fatal=fatal)
+        if (not fatal) and json_string is False:
+            return None
         if transform_source:
             json_string = transform_source(json_string)
         try:
@@ -368,7 +374,8 @@ class InfoExtractor(object):
         else:
             for p in pattern:
                 mobj = re.search(p, string, flags)
-                if mobj: break
+                if mobj:
+                    break
 
         if os.name != 'nt' and sys.stderr.isatty():
             _name = u'\033[0;34m%s\033[0m' % name
@@ -456,8 +463,9 @@ class InfoExtractor(object):
         return self._og_search_property('title', html, **kargs)
 
     def _og_search_video_url(self, html, name='video url', secure=True, **kargs):
-        regexes = self._og_regexes('video')
-        if secure: regexes = self._og_regexes('video:secure_url') + regexes
+        regexes = self._og_regexes('video') + self._og_regexes('video:url')
+        if secure:
+            regexes = self._og_regexes('video:secure_url') + regexes
         return self._html_search_regex(regexes, html, name, **kargs)
 
     def _og_search_url(self, html, **kargs):
@@ -468,7 +476,7 @@ class InfoExtractor(object):
             display_name = name
         return self._html_search_regex(
             r'''(?ix)<meta
-                    (?=[^>]+(?:itemprop|name|property)=["\']%s["\'])
+                    (?=[^>]+(?:itemprop|name|property)=["\']?%s["\']?)
                     [^>]+content=["\']([^"\']+)["\']''' % re.escape(name),
             html, display_name, fatal=fatal, **kwargs)
 
@@ -555,6 +563,7 @@ class InfoExtractor(object):
                 f.get('abr') if f.get('abr') is not None else -1,
                 audio_ext_preference,
                 f.get('filesize') if f.get('filesize') is not None else -1,
+                f.get('filesize_approx') if f.get('filesize_approx') is not None else -1,
                 f.get('format_id'),
             )
         formats.sort(key=_formats_key)
@@ -583,6 +592,24 @@ class InfoExtractor(object):
         self.to_screen(msg)
         time.sleep(timeout)
 
+    def _extract_f4m_formats(self, manifest_url, video_id):
+        manifest = self._download_xml(
+            manifest_url, video_id, 'Downloading f4m manifest',
+            'Unable to download f4m manifest')
+
+        formats = []
+        for media_el in manifest.findall('{http://ns.adobe.com/f4m/1.0}media'):
+            formats.append({
+                'url': manifest_url,
+                'ext': 'flv',
+                'tbr': int_or_none(media_el.attrib.get('bitrate')),
+                'width': int_or_none(media_el.attrib.get('width')),
+                'height': int_or_none(media_el.attrib.get('height')),
+            })
+        self._sort_formats(formats)
+
+        return formats
+
 
 class SearchInfoExtractor(InfoExtractor):
     """

youtube_dl/extractor/dfb.py

@@ -30,7 +30,7 @@ class DFBIE(InfoExtractor):
             video_id)
         video_info = player_info.find('video')
 
-        f4m_info = self._download_xml(video_info.find('url').text, video_id)
+        f4m_info = self._download_xml(self._proto_relative_url(video_info.find('url').text.strip()), video_id)
         token_el = f4m_info.find('token')
         manifest_url = token_el.attrib['url'] + '?' + 'hdnea=' + token_el.attrib['auth'] + '&hdcore=3.2.0'
 

youtube_dl/extractor/dropbox.py

@@ -5,24 +5,26 @@ import os.path
 import re
 
 from .common import InfoExtractor
+from ..utils import compat_urllib_parse_unquote
 
 
 class DropboxIE(InfoExtractor):
     _VALID_URL = r'https?://(?:www\.)?dropbox[.]com/s/(?P<id>[a-zA-Z0-9]{15})/(?P<title>[^?#]*)'
     _TEST = {
-        'url': 'https://www.dropbox.com/s/0qr9sai2veej4f8/THE_DOCTOR_GAMES.mp4',
-        'md5': '8ae17c51172fb7f93bdd6a214cc8c896',
+        'url': 'https://www.dropbox.com/s/nelirfsxnmcfbfh/youtube-dl%20test%20video%20%27%C3%A4%22BaW_jenozKc.mp4',
+        'md5': '8a3d905427a6951ccb9eb292f154530b',
         'info_dict': {
-            'id': '0qr9sai2veej4f8',
+            'id': 'nelirfsxnmcfbfh',
             'ext': 'mp4',
-            'title': 'THE_DOCTOR_GAMES'
+            'title': 'youtube-dl test video \'ä"BaW_jenozKc'
         }
     }
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
         video_id = mobj.group('id')
-        title = os.path.splitext(mobj.group('title'))[0]
+        fn = compat_urllib_parse_unquote(mobj.group('title'))
+        title = os.path.splitext(fn)[0]
         video_url = url + '?dl=1'
 
         return {

youtube_dl/extractor/ellentv.py

@@ -0,0 +1,79 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+import re
+import json
+
+from .common import InfoExtractor
+from ..utils import (
+    ExtractorError,
+    parse_iso8601,
+)
+
+
+class EllenTVIE(InfoExtractor):
+    _VALID_URL = r'https?://(?:www\.)?ellentv\.com/videos/(?P<id>[a-z0-9_-]+)'
+    _TEST = {
+        'url': 'http://www.ellentv.com/videos/0-7jqrsr18/',
+        'md5': 'e4af06f3bf0d5f471921a18db5764642',
+        'info_dict': {
+            'id': '0-7jqrsr18',
+            'ext': 'mp4',
+            'title': 'What\'s Wrong with These Photos? A Whole Lot',
+            'timestamp': 1406876400,
+            'upload_date': '20140801',
+        }
+    }
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        video_id = mobj.group('id')
+
+        webpage = self._download_webpage(url, video_id)
+        timestamp = parse_iso8601(self._search_regex(
+            r'<span class="publish-date"><time datetime="([^"]+)">',
+            webpage, 'timestamp'))
+
+        return {
+            'id': video_id,
+            'title': self._og_search_title(webpage),
+            'url': self._html_search_meta('VideoURL', webpage, 'url'),
+            'timestamp': timestamp,
+        }
+
+
+class EllenTVClipsIE(InfoExtractor):
+    IE_NAME = 'EllenTV:clips'
+    _VALID_URL = r'https?://(?:www\.)?ellentv\.com/episodes/(?P<id>[a-z0-9_-]+)'
+    _TEST = {
+        'url': 'http://www.ellentv.com/episodes/meryl-streep-vanessa-hudgens/',
+        'info_dict': {
+            'id': 'meryl-streep-vanessa-hudgens',
+            'title': 'Meryl Streep, Vanessa Hudgens',
+        },
+        'playlist_mincount': 9,
+    }
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        playlist_id = mobj.group('id')
+
+        webpage = self._download_webpage(url, playlist_id)
+        playlist = self._extract_playlist(webpage)
+
+        return {
+            '_type': 'playlist',
+            'id': playlist_id,
+            'title': self._og_search_title(webpage),
+            'entries': self._extract_entries(playlist)
+        }
+
+    def _extract_playlist(self, webpage):
+        json_string = self._search_regex(r'playerView.addClips\(\[\{(.*?)\}\]\);', webpage, 'json')
+        try:
+            return json.loads("[{" + json_string + "}]")
+        except ValueError as ve:
+            raise ExtractorError('Failed to download JSON', cause=ve)
+
+    def _extract_entries(self, playlist):
+        return [self.url_result(item['url'], 'EllenTV') for item in playlist]

youtube_dl/extractor/escapist.py

@@ -36,7 +36,7 @@ class EscapistIE(InfoExtractor):
             r'<meta name="description" content="([^"]*)"',
             webpage, 'description', fatal=False)
 
-        playerUrl = self._og_search_video_url(webpage, name=u'player URL')
+        playerUrl = self._og_search_video_url(webpage, name='player URL')
 
         title = self._html_search_regex(
             r'<meta name="title" content="([^"]*)"',

youtube_dl/extractor/facebook.py

@@ -20,7 +20,7 @@ from ..utils import (
 class FacebookIE(InfoExtractor):
     _VALID_URL = r'''(?x)
         https?://(?:\w+\.)?facebook\.com/
-        (?:[^#?]*\#!/)?
+        (?:[^#]*?\#!/)?
         (?:video/video\.php|photo\.php|video/embed)\?(?:.*?)
         (?:v|video_id)=(?P<id>[0-9]+)
         (?:.*)'''

youtube_dl/extractor/firedrive.py

@@ -8,7 +8,6 @@ from ..utils import (
     ExtractorError,
     compat_urllib_parse,
     compat_urllib_request,
-    determine_ext,
 )
 
 
@@ -43,7 +42,6 @@ class FiredriveIE(InfoExtractor):
         fields = dict(re.findall(r'''(?x)<input\s+
             type="hidden"\s+
             name="([^"]+)"\s+
-            (?:id="[^"]+"\s+)?
             value="([^"]*)"
             ''', webpage))
 
@@ -67,7 +65,7 @@ class FiredriveIE(InfoExtractor):
         ext = self._search_regex(r'type:\s?\'([^\']+)\',',
                                  webpage, 'extension', fatal=False)
         video_url = self._search_regex(
-            r'file:\s?\'(http[^\']+)\',', webpage, 'file url')
+            r'file:\s?loadURL\(\'(http[^\']+)\'\),', webpage, 'file url')
 
         formats = [{
             'format_id': 'sd',

youtube_dl/extractor/francetv.py

@@ -19,17 +19,35 @@ class FranceTVBaseInfoExtractor(InfoExtractor):
             + video_id, video_id, 'Downloading XML config')
 
         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/')
+        manifest_url = manifest_url.replace('/z/', '/i/')
+        
+        if manifest_url.startswith('rtmp'):
+            formats = [{'url': manifest_url, 'ext': 'flv'}]
+        else:
+            formats = []
+            available_formats = self._search_regex(r'/[^,]*,(.*?),k\.mp4', manifest_url, 'available formats')
+            for index, format_descr in enumerate(available_formats.split(',')):
+                format_info = {
+                    'url': manifest_url.replace('manifest.f4m', 'index_%d_av.m3u8' % index),
+                    'ext': 'mp4',
+                }
+                m_resolution = re.search(r'(?P<width>\d+)x(?P<height>\d+)', format_descr)
+                if m_resolution is not None:
+                    format_info.update({
+                        'width': int(m_resolution.group('width')),
+                        'height': int(m_resolution.group('height')),
+                    })
+                formats.append(format_info)
+
         thumbnail_path = info.find('image').text
 
-        return {'id': video_id,
-                'ext': 'flv' if video_url.startswith('rtmp') else 'mp4',
-                'url': video_url,
-                'title': info.find('titre').text,
-                'thumbnail': compat_urlparse.urljoin('http://pluzz.francetv.fr', thumbnail_path),
-                'description': info.find('synopsis').text,
-                }
+        return {
+            'id': video_id,
+            'title': info.find('titre').text,
+            'formats': formats,
+            'thumbnail': compat_urlparse.urljoin('http://pluzz.francetv.fr', thumbnail_path),
+            'description': info.find('synopsis').text,
+        }
 
 
 class PluzzIE(FranceTVBaseInfoExtractor):

youtube_dl/extractor/funnyordie.py

@@ -26,7 +26,7 @@ class FunnyOrDieIE(InfoExtractor):
             'id': 'e402820827',
             'ext': 'mp4',
             'title': 'Please Use This Song (Jon Lajoie)',
-            'description': 'md5:2ed27d364f5a805a6dba199faaf6681d',
+            'description': 'Please use this to sell something.  www.jonlajoie.com',
             'thumbnail': 're:^http:.*\.jpg$',
         },
     }]

youtube_dl/extractor/gamestar.py

@@ -0,0 +1,74 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+import re
+
+from .common import InfoExtractor
+from ..utils import (
+    int_or_none,
+    parse_duration,
+    str_to_int,
+    unified_strdate,
+)
+
+
+class GameStarIE(InfoExtractor):
+    _VALID_URL = r'http://www\.gamestar\.de/videos/.*,(?P<id>[0-9]+)\.html'
+    _TEST = {
+        'url': 'http://www.gamestar.de/videos/trailer,3/hobbit-3-die-schlacht-der-fuenf-heere,76110.html',
+        'md5': '96974ecbb7fd8d0d20fca5a00810cea7',
+        'info_dict': {
+            'id': '76110',
+            'ext': 'mp4',
+            'title': 'Hobbit 3: Die Schlacht der Fünf Heere - Teaser-Trailer zum dritten Teil',
+            'description': 'Der Teaser-Trailer zu Hobbit 3: Die Schlacht der Fünf Heere zeigt einige Szenen aus dem dritten Teil der Saga und kündigt den vollständigen Trailer an.',
+            'thumbnail': 'http://images.gamestar.de/images/idgwpgsgp/bdb/2494525/600x.jpg',
+            'upload_date': '20140728',
+            'duration': 17
+        }
+    }
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        video_id = mobj.group('id')
+
+        webpage = self._download_webpage(url, video_id)
+
+        og_title = self._og_search_title(webpage)
+        title = og_title.replace(' - Video bei GameStar.de', '').strip()
+
+        url = 'http://gamestar.de/_misc/videos/portal/getVideoUrl.cfm?premium=0&videoId=' + video_id
+
+        description = self._og_search_description(webpage).strip()
+
+        thumbnail = self._proto_relative_url(
+            self._og_search_thumbnail(webpage), scheme='http:')
+
+        upload_date = unified_strdate(self._html_search_regex(
+            r'<span style="float:left;font-size:11px;">Datum: ([0-9]+\.[0-9]+\.[0-9]+)&nbsp;&nbsp;',
+            webpage, 'upload_date', fatal=False))
+
+        duration = parse_duration(self._html_search_regex(
+            r'&nbsp;&nbsp;Länge: ([0-9]+:[0-9]+)</span>', webpage, 'duration',
+            fatal=False))
+
+        view_count = str_to_int(self._html_search_regex(
+            r'&nbsp;&nbsp;Zuschauer: ([0-9\.]+)&nbsp;&nbsp;', webpage,
+            'view_count', fatal=False))
+
+        comment_count = int_or_none(self._html_search_regex(
+            r'>Kommentieren \(([0-9]+)\)</a>', webpage, 'comment_count',
+            fatal=False))
+
+        return {
+            'id': video_id,
+            'title': title,
+            'url': url,
+            'ext': 'mp4',
+            'thumbnail': thumbnail,
+            'description': description,
+            'upload_date': upload_date,
+            'duration': duration,
+            'view_count': view_count,
+            'comment_count': comment_count
+        }

youtube_dl/extractor/gdcvault.py

@@ -8,6 +8,7 @@ from ..utils import (
     compat_urllib_request,
 )
 
+
 class GDCVaultIE(InfoExtractor):
     _VALID_URL = r'https?://(?:www\.)?gdcvault\.com/play/(?P<id>\d+)/(?P<name>(\w|-)+)'
     _TESTS = [
@@ -31,6 +32,15 @@ class GDCVaultIE(InfoExtractor):
                 'skip_download': True,  # Requires rtmpdump
             }
         },
+        {
+            'url': 'http://www.gdcvault.com/play/1015301/Thexder-Meets-Windows-95-or',
+            'md5': 'a5eb77996ef82118afbbe8e48731b98e',
+            'info_dict': {
+                'id': '1015301',
+                'ext': 'flv',
+                'title': 'Thexder Meets Windows 95, or Writing Great Games in the Windows 95 Environment',
+            }
+        }
     ]
 
     def _parse_mp4(self, xml_description):
@@ -103,18 +113,40 @@ class GDCVaultIE(InfoExtractor):
         webpage_url = 'http://www.gdcvault.com/play/' + video_id
         start_page = self._download_webpage(webpage_url, video_id)
 
-        xml_root = self._html_search_regex(r'<iframe src="(?P<xml_root>.*?)player.html.*?".*?</iframe>', start_page, 'xml root', None, False)
+        direct_url = self._search_regex(
+            r's1\.addVariable\("file",\s*encodeURIComponent\("(/[^"]+)"\)\);',
+            start_page, 'url', default=None)
+        if direct_url:
+            video_url = 'http://www.gdcvault.com/' + direct_url
+            title = self._html_search_regex(
+                r'<td><strong>Session Name</strong></td>\s*<td>(.*?)</td>',
+                start_page, 'title')
 
+            return {
+                'id': video_id,
+                'url': video_url,
+                'ext': 'flv',
+                'title': title,
+            }
+
+        xml_root = self._html_search_regex(
+            r'<iframe src="(?P<xml_root>.*?)player.html.*?".*?</iframe>',
+            start_page, 'xml root', default=None)
         if xml_root is None:
             # Probably need to authenticate
-            start_page = self._login(webpage_url, video_id)
-            if start_page is None:
+            login_res = self._login(webpage_url, video_id)
+            if login_res is None:
                 self.report_warning('Could not login.')
             else:
+                start_page = login_res
                 # Grab the url from the authenticated page
-                xml_root = self._html_search_regex(r'<iframe src="(?P<xml_root>.*?)player.html.*?".*?</iframe>', start_page, 'xml root')
+                xml_root = self._html_search_regex(
+                    r'<iframe src="(.*?)player.html.*?".*?</iframe>',
+                    start_page, 'xml root')
 
-        xml_name = self._html_search_regex(r'<iframe src=".*?\?xml=(?P<xml_file>.+?\.xml).*?".*?</iframe>', start_page, 'xml filename', None, False)
+        xml_name = self._html_search_regex(
+            r'<iframe src=".*?\?xml=(.+?\.xml).*?".*?</iframe>',
+            start_page, 'xml filename', default=None)
         if xml_name is None:
             # Fallback to the older format
             xml_name = self._html_search_regex(r'<iframe src=".*?\?xmlURL=xml/(?P<xml_file>.+?\.xml).*?".*?</iframe>', start_page, 'xml filename')

youtube_dl/extractor/generic.py

@@ -383,13 +383,13 @@ class GenericIE(InfoExtractor):
         if not parsed_url.scheme:
             default_search = self._downloader.params.get('default_search')
             if default_search is None:
-                default_search = 'error'
+                default_search = 'fixup_error'
 
-            if default_search in ('auto', 'auto_warning'):
+            if default_search in ('auto', 'auto_warning', 'fixup_error'):
                 if '/' in url:
                     self._downloader.report_warning('The url doesn\'t specify the protocol, trying with http')
                     return self.url_result('http://' + url)
-                else:
+                elif default_search != 'fixup_error':
                     if default_search == 'auto_warning':
                         if re.match(r'^(?:url|URL)$', url):
                             raise ExtractorError(
@@ -399,10 +399,11 @@ class GenericIE(InfoExtractor):
                             self._downloader.report_warning(
                                 'Falling back to youtube search for  %s . Set --default-search "auto" to suppress this warning.' % url)
                     return self.url_result('ytsearch:' + url)
-            elif default_search == 'error':
+
+            if default_search in ('error', 'fixup_error'):
                 raise ExtractorError(
                     ('%r is not a valid URL. '
-                     'Set --default-search "ytseach" (or run  youtube-dl "ytsearch:%s" ) to search YouTube'
+                     'Set --default-search "ytsearch" (or run  youtube-dl "ytsearch:%s" ) to search YouTube'
                     ) % (url, url), expected=True)
             else:
                 assert ':' in default_search
@@ -705,6 +706,13 @@ class GenericIE(InfoExtractor):
             url = unescapeHTML(mobj.group('url'))
             return self.url_result(url, ie='MTVServicesEmbedded')
 
+        # Look for embedded yahoo player
+        mobj = re.search(
+            r'<iframe[^>]+?src=(["\'])(?P<url>https?://(?:screen|movies)\.yahoo\.com/.+?\.html\?format=embed)\1',
+            webpage)
+        if mobj is not None:
+            return self.url_result(mobj.group('url'), 'Yahoo')
+
         # Start with something easy: JW Player in SWFObject
         found = re.findall(r'flashvars: [\'"](?:.*&)?file=(http[^\'"&]*)', webpage)
         if not found:

youtube_dl/extractor/godtube.py

@@ -0,0 +1,58 @@
+from __future__ import unicode_literals
+
+import re
+
+from .common import InfoExtractor
+from ..utils import (
+    parse_duration,
+    parse_iso8601,
+)
+
+
+class GodTubeIE(InfoExtractor):
+    _VALID_URL = r'https?://(?:www\.)?godtube\.com/watch/\?v=(?P<id>[\da-zA-Z]+)'
+    _TESTS = [
+        {
+            'url': 'https://www.godtube.com/watch/?v=0C0CNNNU',
+            'md5': '77108c1e4ab58f48031101a1a2119789',
+            'info_dict': {
+                'id': '0C0CNNNU',
+                'ext': 'mp4',
+                'title': 'Woman at the well.',
+                'duration': 159,
+                'timestamp': 1205712000,
+                'uploader': 'beverlybmusic',
+                'upload_date': '20080317',
+                'thumbnail': 're:^https?://.*\.jpg$',
+            },
+        },
+    ]
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        video_id = mobj.group('id')
+
+        config = self._download_xml(
+            'http://www.godtube.com/resource/mediaplayer/%s.xml' % video_id.lower(),
+            video_id, 'Downloading player config XML')
+
+        video_url = config.find('.//file').text
+        uploader = config.find('.//author').text
+        timestamp = parse_iso8601(config.find('.//date').text)
+        duration = parse_duration(config.find('.//duration').text)
+        thumbnail = config.find('.//image').text
+
+        media = self._download_xml(
+            'http://www.godtube.com/media/xml/?v=%s' % video_id, video_id, 'Downloading media XML')
+
+        title = media.find('.//title').text
+
+        return {
+            'id': video_id,
+            'url': video_url,
+            'title': title,
+            'thumbnail': thumbnail,
+            'timestamp': timestamp,
+            'uploader': uploader,
+            'duration': duration,
+        }

youtube_dl/extractor/howstuffworks.py

@@ -0,0 +1,134 @@
+from __future__ import unicode_literals
+
+import re
+import json
+import random
+import string
+
+from .common import InfoExtractor
+from ..utils import find_xpath_attr
+
+
+class HowStuffWorksIE(InfoExtractor):
+    _VALID_URL = r'https?://[\da-z-]+\.howstuffworks\.com/(?:[^/]+/)*\d+-(?P<id>.+?)-video\.htm'
+    _TESTS = [
+        {
+            'url': 'http://adventure.howstuffworks.com/5266-cool-jobs-iditarod-musher-video.htm',
+            'info_dict': {
+                'id': '450221',
+                'display_id': 'cool-jobs-iditarod-musher',
+                'ext': 'flv',
+                'title': 'Cool Jobs - Iditarod Musher',
+                'description': 'md5:82bb58438a88027b8186a1fccb365f90',
+                'thumbnail': 're:^https?://.*\.jpg$',
+            },
+            'params': {
+                # md5 is not consistent
+                'skip_download': True
+            }
+        },
+        {
+            'url': 'http://adventure.howstuffworks.com/39516-deadliest-catch-jakes-farewell-pots-video.htm',
+            'info_dict': {
+                'id': '553470',
+                'display_id': 'deadliest-catch-jakes-farewell-pots',
+                'ext': 'mp4',
+                'title': 'Deadliest Catch: Jake\'s Farewell Pots',
+                'description': 'md5:9632c346d5e43ee238028c9cefd8dbbc',
+                'thumbnail': 're:^https?://.*\.jpg$',
+            },
+            'params': {
+                # md5 is not consistent
+                'skip_download': True
+            }
+        },
+        {
+            'url': 'http://entertainment.howstuffworks.com/arts/2706-sword-swallowing-1-by-dan-meyer-video.htm',
+            'info_dict': {
+                'id': '440011',
+                'display_id': 'sword-swallowing-1-by-dan-meyer',
+                'ext': 'flv',
+                'title': 'Sword Swallowing #1 by Dan Meyer',
+                'description': 'md5:b2409e88172913e2e7d3d1159b0ef735',
+                'thumbnail': 're:^https?://.*\.jpg$',
+            },
+            'params': {
+                # md5 is not consistent
+                'skip_download': True
+            }
+        },
+    ]
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        display_id = mobj.group('id')
+        webpage = self._download_webpage(url, display_id)
+
+        content_id = self._search_regex(r'var siteSectionId="(\d+)";', webpage, 'content id')
+
+        mp4 = self._search_regex(
+            r'''(?xs)var\s+clip\s*=\s*{\s*
+                .+?\s*
+                content_id\s*:\s*%s\s*,\s*
+                .+?\s*
+                mp4\s*:\s*\[(.*?),?\]\s*
+                };\s*
+                videoData\.push\(clip\);''' % content_id,
+            webpage, 'mp4', fatal=False, default=None)
+
+        smil = self._download_xml(
+            'http://services.media.howstuffworks.com/videos/%s/smil-service.smil' % content_id,
+            content_id, 'Downloading video SMIL')
+
+        http_base = find_xpath_attr(
+            smil,
+            './{0}head/{0}meta'.format('{http://www.w3.org/2001/SMIL20/Language}'),
+            'name',
+            'httpBase').get('content')
+
+        def random_string(str_len=0):
+            return ''.join([random.choice(string.ascii_uppercase) for _ in range(str_len)])
+
+        URL_SUFFIX = '?v=2.11.3&fp=LNX 11,2,202,356&r=%s&g=%s' % (random_string(5), random_string(12))
+
+        formats = []
+
+        if mp4:
+            for video in json.loads('[%s]' % mp4):
+                bitrate = video['bitrate']
+                fmt = {
+                    'url': video['src'].replace('http://pmd.video.howstuffworks.com', http_base) + URL_SUFFIX,
+                    'format_id': bitrate,
+                }
+                m = re.search(r'(?P<vbr>\d+)[Kk]', bitrate)
+                if m:
+                    fmt['vbr'] = int(m.group('vbr'))
+                formats.append(fmt)
+        else:
+            for video in smil.findall(
+                    './/{0}body/{0}switch/{0}video'.format('{http://www.w3.org/2001/SMIL20/Language}')):
+                vbr = int(video.attrib['system-bitrate']) / 1000
+                formats.append({
+                    'url': '%s/%s%s' % (http_base, video.attrib['src'], URL_SUFFIX),
+                    'format_id': '%dk' % vbr,
+                    'vbr': vbr,
+                })
+
+        self._sort_formats(formats)
+
+        title = self._og_search_title(webpage)
+        TITLE_SUFFIX = ' : HowStuffWorks'
+        if title.endswith(TITLE_SUFFIX):
+            title = title[:-len(TITLE_SUFFIX)]
+
+        description = self._og_search_description(webpage)
+        thumbnail = self._og_search_thumbnail(webpage)
+
+        return {
+            'id': content_id,
+            'display_id': display_id,
+            'title': title,
+            'description': description,
+            'thumbnail': thumbnail,
+            'formats': formats,
+        }

youtube_dl/extractor/izlesene.py

@@ -0,0 +1,97 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+import re
+
+from .common import InfoExtractor
+from ..utils import (
+    get_element_by_id,
+    parse_iso8601,
+    determine_ext,
+    int_or_none,
+    str_to_int,
+)
+
+
+class IzleseneIE(InfoExtractor):
+    _VALID_URL = r'https?://(?:(?:www|m)\.)?izlesene\.com/(?:video|embedplayer)/(?:[^/]+/)?(?P<id>[0-9]+)'
+    _STREAM_URL = 'http://panel.izlesene.com/api/streamurl/{id:}/{format:}'
+    _TEST = {
+        'url': 'http://www.izlesene.com/video/sevincten-cildirtan-dogum-gunu-hediyesi/7599694',
+        'md5': '4384f9f0ea65086734b881085ee05ac2',
+        'info_dict': {
+            'id': '7599694',
+            'ext': 'mp4',
+            'title': 'Sevinçten Çıldırtan Doğum Günü Hediyesi',
+            'description': 'Annesi oğluna doğum günü hediyesi olarak minecraft cd si alıyor, ve çocuk hunharca seviniyor',
+            'thumbnail': 're:^http://.*\.jpg',
+            'uploader_id': 'pelikzzle',
+            'timestamp': 1404298698,
+            'upload_date': '20140702',
+            'duration': 95.395,
+            'age_limit': 0,
+        }
+    }
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        video_id = mobj.group('id')
+        url = 'http://www.izlesene.com/video/%s' % video_id
+
+        webpage = self._download_webpage(url, video_id)
+
+        title = self._og_search_title(webpage)
+        description = self._og_search_description(webpage)
+        thumbnail = self._og_search_thumbnail(webpage)
+
+        uploader = self._html_search_regex(
+            r"adduserUsername\s*=\s*'([^']+)';", webpage, 'uploader', fatal=False, default='')
+        timestamp = parse_iso8601(self._html_search_meta(
+            'uploadDate', webpage, 'upload date', fatal=False))
+
+        duration = int_or_none(self._html_search_regex(
+            r'"videoduration"\s*:\s*"([^"]+)"', webpage, 'duration', fatal=False))
+        if duration:
+            duration /= 1000.0
+
+        view_count = str_to_int(get_element_by_id('videoViewCount', webpage))
+        comment_count = self._html_search_regex(
+            r'comment_count\s*=\s*\'([^\']+)\';', webpage, 'uploader', fatal=False)
+
+        family_friendly = self._html_search_meta(
+            'isFamilyFriendly', webpage, 'age limit', fatal=False)
+
+        content_url = self._html_search_meta(
+            'contentURL', webpage, 'content URL', fatal=False)
+        ext = determine_ext(content_url, 'mp4')
+
+        # Might be empty for some videos.
+        qualities = self._html_search_regex(
+            r'"quality"\s*:\s*"([^"]+)"', webpage, 'qualities', fatal=False, default='')
+
+        formats = []
+        for quality in qualities.split('|'):
+            json = self._download_json(
+                self._STREAM_URL.format(id=video_id, format=quality), video_id,
+                note='Getting video URL for "%s" quality' % quality,
+                errnote='Failed to get video URL for "%s" quality' % quality
+            )
+            formats.append({
+                'url': json.get('streamurl'),
+                'ext': ext,
+                'format_id': '%sp' % quality if quality else 'sd',
+            })
+
+        return {
+            'id': video_id,
+            'title': title,
+            'description': description,
+            'thumbnail': thumbnail,
+            'uploader_id': uploader,
+            'timestamp': timestamp,
+            'duration': duration,
+            'view_count': int_or_none(view_count),
+            'comment_count': int_or_none(comment_count),
+            'age_limit': 18 if family_friendly == 'False' else 0,
+            'formats': formats,
+        }

youtube_dl/extractor/jove.py

@@ -0,0 +1,80 @@
+from __future__ import unicode_literals
+
+import re
+
+from .common import InfoExtractor
+from ..utils import (
+    ExtractorError,
+    unified_strdate
+)
+
+
+class JoveIE(InfoExtractor):
+    _VALID_URL = r'https?://(?:www\.)?jove\.com/video/(?P<id>[0-9]+)'
+    _CHAPTERS_URL = 'http://www.jove.com/video-chapters?videoid={video_id:}'
+    _TESTS = [
+        {
+            'url': 'http://www.jove.com/video/2744/electrode-positioning-montage-transcranial-direct-current',
+            'md5': '93723888d82dbd6ba8b3d7d0cd65dd2b',
+            'info_dict': {
+                'id': '2744',
+                'ext': 'mp4',
+                'title': 'Electrode Positioning and Montage in Transcranial Direct Current Stimulation',
+                'description': 'md5:015dd4509649c0908bc27f049e0262c6',
+                'thumbnail': 're:^https?://.*\.png$',
+                'upload_date': '20110523',
+            }
+        },
+        {
+            'url': 'http://www.jove.com/video/51796/culturing-caenorhabditis-elegans-axenic-liquid-media-creation',
+            'md5': '914aeb356f416811d911996434811beb',
+            'info_dict': {
+                'id': '51796',
+                'ext': 'mp4',
+                'title': 'Culturing Caenorhabditis elegans in Axenic Liquid Media and Creation of Transgenic Worms by Microparticle Bombardment',
+                'description': 'md5:35ff029261900583970c4023b70f1dc9',
+                'thumbnail': 're:^https?://.*\.png$',
+                'upload_date': '20140802',
+            }
+        },
+
+    ]
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        video_id = mobj.group('id')
+
+        webpage = self._download_webpage(url, video_id)
+
+        chapters_id = self._html_search_regex(
+            r'/video-chapters\?videoid=([0-9]+)', webpage, 'chapters id')
+
+        chapters_xml = self._download_xml(
+            self._CHAPTERS_URL.format(video_id=chapters_id),
+            video_id, note='Downloading chapters XML',
+            errnote='Failed to download chapters XML')
+
+        video_url = chapters_xml.attrib.get('video')
+        if not video_url:
+            raise ExtractorError('Failed to get the video URL')
+
+        title = self._html_search_meta('citation_title', webpage, 'title')
+        thumbnail = self._og_search_thumbnail(webpage)
+        description = self._html_search_regex(
+            r'<div id="section_body_summary"><p class="jove_content">(.+?)</p>',
+            webpage, 'description', fatal=False)
+        publish_date = unified_strdate(self._html_search_meta(
+            'citation_publication_date', webpage, 'publish date', fatal=False))
+        comment_count = self._html_search_regex(
+            r'<meta name="num_comments" content="(\d+) Comments?"',
+            webpage, 'comment count', fatal=False)
+
+        return {
+            'id': video_id,
+            'title': title,
+            'url': video_url,
+            'thumbnail': thumbnail,
+            'description': description,
+            'upload_date': publish_date,
+            'comment_count': comment_count,
+        }

youtube_dl/extractor/justintv.py

@@ -1,5 +1,6 @@
 from __future__ import unicode_literals
 
+import itertools
 import json
 import os
 import re
@@ -43,10 +44,11 @@ class JustinTVIE(InfoExtractor):
     }
 
     # Return count of items, list of *valid* items
-    def _parse_page(self, url, video_id):
-        info_json = self._download_webpage(url, video_id,
-                                           'Downloading video info JSON',
-                                           'unable to download video info JSON')
+    def _parse_page(self, url, video_id, counter):
+        info_json = self._download_webpage(
+            url, video_id,
+            'Downloading video info JSON on page %d' % counter,
+            'Unable to download video info JSON %d' % counter)
 
         response = json.loads(info_json)
         if type(response) != list:
@@ -138,11 +140,10 @@ class JustinTVIE(InfoExtractor):
         entries = []
         offset = 0
         limit = self._JUSTIN_PAGE_LIMIT
-        while True:
-            if paged:
-                self.report_download_page(video_id, offset)
+        for counter in itertools.count(1):
             page_url = api + ('?offset=%d&limit=%d' % (offset, limit))
-            page_count, page_info = self._parse_page(page_url, video_id)
+            page_count, page_info = self._parse_page(
+                page_url, video_id, counter)
             entries.extend(page_info)
             if not paged or page_count != limit:
                 break

youtube_dl/extractor/kickstarter.py

@@ -8,7 +8,7 @@ from .common import InfoExtractor
 
 class KickStarterIE(InfoExtractor):
     _VALID_URL = r'https?://www\.kickstarter\.com/projects/(?P<id>[^/]*)/.*'
-    _TEST = {
+    _TESTS = [{
         'url': 'https://www.kickstarter.com/projects/1404461844/intersection-the-story-of-josh-grant?ref=home_location',
         'md5': 'c81addca81327ffa66c642b5d8b08cab',
         'info_dict': {
@@ -18,22 +18,45 @@ class KickStarterIE(InfoExtractor):
             'description': 'A unique motocross documentary that examines the '
                 'life and mind of one of sports most elite athletes: Josh Grant.',
         },
-    }
+    }, {
+        'note': 'Embedded video (not using the native kickstarter video service)',
+        'url': 'https://www.kickstarter.com/projects/597507018/pebble-e-paper-watch-for-iphone-and-android/posts/659178',
+        'playlist': [
+            {
+                'info_dict': {
+                    'id': '78704821',
+                    'ext': 'mp4',
+                    'uploader_id': 'pebble',
+                    'uploader': 'Pebble Technology',
+                    'title': 'Pebble iOS Notifications',
+                }
+            }
+        ],
+    }]
 
     def _real_extract(self, url):
         m = re.match(self._VALID_URL, url)
         video_id = m.group('id')
         webpage = self._download_webpage(url, video_id)
 
-        video_url = self._search_regex(r'data-video-url="(.*?)"',
-            webpage, 'video URL')
-        video_title = self._html_search_regex(r'<title>(.*?)</title>',
-            webpage, 'title').rpartition('— Kickstarter')[0].strip()
+        title = self._html_search_regex(
+            r'<title>\s*(.*?)(?:\s*&mdash; Kickstarter)?\s*</title>',
+            webpage, 'title')
+        video_url = self._search_regex(
+            r'data-video-url="(.*?)"',
+            webpage, 'video URL', default=None)
+        if video_url is None:  # No native kickstarter, look for embedded videos
+            return {
+                '_type': 'url_transparent',
+                'ie_key': 'Generic',
+                'url': url,
+                'title': title,
+            }
 
         return {
             'id': video_id,
             'url': video_url,
-            'title': video_title,
+            'title': title,
             'description': self._og_search_description(webpage),
             'thumbnail': self._og_search_thumbnail(webpage),
         }

youtube_dl/extractor/krasview.py

@@ -0,0 +1,59 @@
+# encoding: utf-8
+from __future__ import unicode_literals
+
+import json
+import re
+
+from .common import InfoExtractor
+from ..utils import (
+    int_or_none,
+    unescapeHTML,
+)
+
+
+class KrasViewIE(InfoExtractor):
+    IE_DESC = 'Красвью'
+    _VALID_URL = r'https?://krasview\.ru/video/(?P<id>\d+)'
+
+    _TEST = {
+        'url': 'http://krasview.ru/video/512228',
+        'md5': '3b91003cf85fc5db277870c8ebd98eae',
+        'info_dict': {
+            'id': '512228',
+            'ext': 'mp4',
+            'title': 'Снег, лёд, заносы',
+            'description': 'Снято в городе Нягань, в Ханты-Мансийском автономном округе.',
+            'duration': 27,
+            'thumbnail': 're:^https?://.*\.jpg',
+        },
+    }
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        video_id = mobj.group('id')
+
+        webpage = self._download_webpage(url, video_id)
+
+        flashvars = json.loads(self._search_regex(
+            r'flashvars\s*:\s*({.+?})\s*}\);', webpage, 'flashvars'))
+
+        video_url = flashvars['url']
+        title = unescapeHTML(flashvars['title'])
+        description = unescapeHTML(flashvars.get('subtitle') or self._og_search_description(webpage, default=None))
+        thumbnail = flashvars['image']
+        duration = int(flashvars['duration'])
+        filesize = int(flashvars['size'])
+        width = int_or_none(self._og_search_property('video:width', webpage, 'video width'))
+        height = int_or_none(self._og_search_property('video:height', webpage, 'video height'))
+
+        return {
+            'id': video_id,
+            'url': video_url,
+            'title': title,
+            'description': description,
+            'thumbnail': thumbnail,
+            'duration': duration,
+            'filesize': filesize,
+            'width': width,
+            'height': height,
+        }

youtube_dl/extractor/livestream.py

@@ -5,11 +5,14 @@ import json
 
 from .common import InfoExtractor
 from ..utils import (
+    compat_str,
     compat_urllib_parse_urlparse,
     compat_urlparse,
-    xpath_with_ns,
-    compat_str,
+    ExtractorError,
+    find_xpath_attr,
+    int_or_none,
     orderedSet,
+    xpath_with_ns,
 )
 
 
@@ -24,18 +27,82 @@ class LivestreamIE(InfoExtractor):
             'ext': 'mp4',
             'title': 'Live from Webster Hall NYC',
             'upload_date': '20121012',
+            'like_count': int,
+            'view_count': int,
+            'thumbnail': 're:^http://.*\.jpg$'
         }
     }
 
+    def _parse_smil(self, video_id, smil_url):
+        formats = []
+        _SWITCH_XPATH = (
+            './/{http://www.w3.org/2001/SMIL20/Language}body/'
+            '{http://www.w3.org/2001/SMIL20/Language}switch')
+        smil_doc = self._download_xml(
+            smil_url, video_id,
+            note='Downloading SMIL information',
+            errnote='Unable to download SMIL information',
+            fatal=False)
+        if smil_doc is False:  # Download failed
+            return formats
+        title_node = find_xpath_attr(
+            smil_doc, './/{http://www.w3.org/2001/SMIL20/Language}meta',
+            'name', 'title')
+        if title_node is None:
+            self.report_warning('Cannot find SMIL id')
+            switch_node = smil_doc.find(_SWITCH_XPATH)
+        else:
+            title_id = title_node.attrib['content']
+            switch_node = find_xpath_attr(
+                smil_doc, _SWITCH_XPATH, 'id', title_id)
+        if switch_node is None:
+            raise ExtractorError('Cannot find switch node')
+        video_nodes = switch_node.findall(
+            '{http://www.w3.org/2001/SMIL20/Language}video')
+
+        for vn in video_nodes:
+            tbr = int_or_none(vn.attrib.get('system-bitrate'))
+            furl = (
+                'http://livestream-f.akamaihd.net/%s?v=3.0.3&fp=WIN%%2014,0,0,145' %
+                (vn.attrib['src']))
+            if 'clipBegin' in vn.attrib:
+                furl += '&ssek=' + vn.attrib['clipBegin']
+            formats.append({
+                'url': furl,
+                'format_id': 'smil_%d' % tbr,
+                'ext': 'flv',
+                'tbr': tbr,
+                'preference': -1000,
+            })
+        return formats
+
     def _extract_video_info(self, video_data):
-        video_url = video_data.get('progressive_url_hd') or video_data.get('progressive_url')
+        video_id = compat_str(video_data['id'])
+
+        FORMAT_KEYS = (
+            ('sd', 'progressive_url'),
+            ('hd', 'progressive_url_hd'),
+        )
+        formats = [{
+            'format_id': format_id,
+            'url': video_data[key],
+            'quality': i + 1,
+        } for i, (format_id, key) in enumerate(FORMAT_KEYS)
+            if video_data.get(key)]
+
+        smil_url = video_data.get('smil_url')
+        if smil_url:
+            formats.extend(self._parse_smil(video_id, smil_url))
+        self._sort_formats(formats)
+
         return {
-            'id': compat_str(video_data['id']),
-            'url': video_url,
-            'ext': 'mp4',
+            'id': video_id,
+            'formats': formats,
             'title': video_data['caption'],
-            'thumbnail': video_data['thumbnail_url'],
+            'thumbnail': video_data.get('thumbnail_url'),
             'upload_date': video_data['updated_at'].replace('-', '')[:8],
+            'like_count': video_data.get('likes', {}).get('total'),
+            'view_count': video_data.get('views'),
         }
 
     def _real_extract(self, url):
@@ -44,16 +111,28 @@ class LivestreamIE(InfoExtractor):
         event_name = mobj.group('event_name')
         webpage = self._download_webpage(url, video_id or event_name)
 
-        if video_id is None:
-            # This is an event page:
+        og_video = self._og_search_video_url(webpage, 'player url', fatal=False, default=None)
+        if og_video is None:
             config_json = self._search_regex(
                 r'window.config = ({.*?});', webpage, 'window config')
             info = json.loads(config_json)['event']
+
+            def is_relevant(vdata, vid):
+                result = vdata['type'] == 'video'
+                if video_id is not None:
+                    result = result and compat_str(vdata['data']['id']) == vid
+                return result
+
             videos = [self._extract_video_info(video_data['data'])
-                for video_data in info['feed']['data'] if video_data['type'] == 'video']
-            return self.playlist_result(videos, info['id'], info['full_name'])
+                      for video_data in info['feed']['data']
+                      if is_relevant(video_data, video_id)]
+            if video_id is None:
+                # This is an event page:
+                return self.playlist_result(videos, info['id'], info['full_name'])
+            else:
+                if videos:
+                    return videos[0]
         else:
-            og_video = self._og_search_video_url(webpage, 'player url')
             query_str = compat_urllib_parse_urlparse(og_video).query
             query = compat_urlparse.parse_qs(query_str)
             api_url = query['play_url'][0].replace('.smil', '')

youtube_dl/extractor/mitele.py

@@ -0,0 +1,60 @@
+from __future__ import unicode_literals
+
+import re
+import json
+
+from .common import InfoExtractor
+from ..utils import (
+    compat_urllib_parse,
+    get_element_by_attribute,
+    parse_duration,
+    strip_jsonp,
+)
+
+
+class MiTeleIE(InfoExtractor):
+    IE_NAME = 'mitele.es'
+    _VALID_URL = r'http://www\.mitele\.es/[^/]+/[^/]+/[^/]+/(?P<episode>[^/]+)/'
+
+    _TEST = {
+        'url': 'http://www.mitele.es/programas-tv/diario-de/la-redaccion/programa-144/',
+        'md5': '6a75fe9d0d3275bead0cb683c616fddb',
+        'info_dict': {
+            'id': '0fce117d',
+            'ext': 'mp4',
+            'title': 'Programa 144 - Tor, la web invisible',
+            'description': 'md5:3b6fce7eaa41b2d97358726378d9369f',
+            'display_id': 'programa-144',
+            'duration': 2913,
+        },
+    }
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        episode = mobj.group('episode')
+        webpage = self._download_webpage(url, episode)
+        embed_data_json = self._search_regex(
+            r'MSV\.embedData\[.*?\]\s*=\s*({.*?});', webpage, 'embed data',
+            flags=re.DOTALL
+        ).replace('\'', '"')
+        embed_data = json.loads(embed_data_json)
+
+        info_url = embed_data['flashvars']['host']
+        info_el = self._download_xml(info_url, episode).find('./video/info')
+
+        video_link = info_el.find('videoUrl/link').text
+        token_query = compat_urllib_parse.urlencode({'id': video_link})
+        token_info = self._download_json(
+            'http://token.mitele.es/?' + token_query, episode,
+            transform_source=strip_jsonp
+        )
+
+        return {
+            'id': embed_data['videoId'],
+            'display_id': episode,
+            'title': info_el.find('title').text,
+            'url': token_info['tokenizedUrl'],
+            'description': get_element_by_attribute('class', 'text', webpage),
+            'thumbnail': info_el.find('thumb').text,
+            'duration': parse_duration(info_el.find('duration').text),
+        }

youtube_dl/extractor/mlb.py

@@ -11,8 +11,22 @@ from ..utils import (
 
 
 class MLBIE(InfoExtractor):
-    _VALID_URL = r'https?://m\.mlb\.com/video/(?:topic/[\da-z_-]+/)?v(?P<id>n?\d+)'
+    _VALID_URL = r'https?://m\.mlb\.com/(?:.*?/)?video/(?:topic/[\da-z_-]+/)?v(?P<id>n?\d+)'
     _TESTS = [
+        {
+            'url': 'http://m.mlb.com/sea/video/topic/51231442/v34698933/nymsea-ackley-robs-a-home-run-with-an-amazing-catch/?c_id=sea',
+            'md5': 'ff56a598c2cf411a9a38a69709e97079',
+            'info_dict': {
+                'id': '34698933',
+                'ext': 'mp4',
+                'title': "Ackley's spectacular catch",
+                'description': 'md5:7f5a981eb4f3cbc8daf2aeffa2215bf0',
+                'duration': 66,
+                'timestamp': 1405980600,
+                'upload_date': '20140721',
+                'thumbnail': 're:^https?://.*\.jpg$',
+            },
+        },
         {
             'url': 'http://m.mlb.com/video/topic/81536970/v34496663/mianym-stanton-practices-for-the-home-run-derby',
             'md5': 'd9c022c10d21f849f49c05ae12a8a7e9',

youtube_dl/extractor/mojvideo.py

@@ -0,0 +1,58 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+import re
+
+from .common import InfoExtractor
+from ..utils import (
+    ExtractorError,
+    parse_duration,
+)
+
+
+class MojvideoIE(InfoExtractor):
+    _VALID_URL = r'https?://(?:www\.)?mojvideo\.com/video-(?P<display_id>[^/]+)/(?P<id>[a-f0-9]+)'
+    _TEST = {
+        'url': 'http://www.mojvideo.com/video-v-avtu-pred-mano-rdecelaska-alfi-nipic/3d1ed4497707730b2906',
+        'md5': 'f7fd662cc8ce2be107b0d4f2c0483ae7',
+        'info_dict': {
+            'id': '3d1ed4497707730b2906',
+            'display_id': 'v-avtu-pred-mano-rdecelaska-alfi-nipic',
+            'ext': 'mp4',
+            'title': 'V avtu pred mano rdečelaska - Alfi Nipič',
+            'thumbnail': 're:^http://.*\.jpg$',
+            'duration': 242,
+        }
+    }
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        video_id = mobj.group('id')
+        display_id = mobj.group('display_id')
+
+        # XML is malformed
+        playerapi = self._download_webpage(
+            'http://www.mojvideo.com/playerapi.php?v=%s&t=1' % video_id, display_id)
+
+        if '<error>true</error>' in playerapi:
+            error_desc = self._html_search_regex(
+                r'<errordesc>([^<]*)</errordesc>', playerapi, 'error description', fatal=False)
+            raise ExtractorError('%s said: %s' % (self.IE_NAME, error_desc), expected=True)
+
+        title = self._html_search_regex(
+            r'<title>([^<]+)</title>', playerapi, 'title')
+        video_url = self._html_search_regex(
+            r'<file>([^<]+)</file>', playerapi, 'video URL')
+        thumbnail = self._html_search_regex(
+            r'<preview>([^<]+)</preview>', playerapi, 'thumbnail', fatal=False)
+        duration = parse_duration(self._html_search_regex(
+            r'<duration>([^<]+)</duration>', playerapi, 'duration', fatal=False))
+
+        return {
+            'id': video_id,
+            'display_id': display_id,
+            'url': video_url,
+            'title': title,
+            'thumbnail': thumbnail,
+            'duration': duration,
+        }

youtube_dl/extractor/nbc.py

@@ -4,7 +4,11 @@ import re
 import json
 
 from .common import InfoExtractor
-from ..utils import find_xpath_attr, compat_str
+from ..utils import (
+    compat_str,
+    ExtractorError,
+    find_xpath_attr,
+)
 
 
 class NBCIE(InfoExtractor):
@@ -85,11 +89,25 @@ class NBCNewsIE(InfoExtractor):
                 flags=re.MULTILINE)
             bootstrap = json.loads(bootstrap_json)
             info = bootstrap['results'][0]['video']
-            playlist_url = info['fallbackPlaylistUrl'] + '?form=MPXNBCNewsAPI'
             mpxid = info['mpxId']
-            all_videos = self._download_json(playlist_url, title)['videos']
-            # The response contains additional videos
-            info = next(v for v in all_videos if v['mpxId'] == mpxid)
+
+            base_urls = [
+                info['fallbackPlaylistUrl'],
+                info['associatedPlaylistUrl'],
+            ]
+
+            for base_url in base_urls:
+                playlist_url = base_url + '?form=MPXNBCNewsAPI'
+                all_videos = self._download_json(playlist_url, title)['videos']
+
+                try:
+                    info = next(v for v in all_videos if v['mpxId'] == mpxid)
+                    break
+                except StopIteration:
+                    continue
+
+            if info is None:
+                raise ExtractorError('Could not find video in playlists')
 
             return {
                 '_type': 'url',

youtube_dl/extractor/nowness.py

@@ -1,3 +1,4 @@
+# encoding: utf-8
 from __future__ import unicode_literals
 
 import re
@@ -8,19 +9,34 @@ from ..utils import ExtractorError
 
 
 class NownessIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:www\.)?nowness\.com/[^?#]*?/(?P<id>[0-9]+)/(?P<slug>[^/]+?)(?:$|[?#])'
+    _VALID_URL = r'https?://(?:(?:www|cn)\.)?nowness\.com/[^?#]*?/(?P<id>[0-9]+)/(?P<slug>[^/]+?)(?:$|[?#])'
 
-    _TEST = {
-        'url': 'http://www.nowness.com/day/2013/6/27/3131/candor--the-art-of-gesticulation',
-        'md5': '068bc0202558c2e391924cb8cc470676',
-        'info_dict': {
-            'id': '2520295746001',
-            'ext': 'mp4',
-            'description': 'Candor: The Art of Gesticulation',
-            'uploader': 'Nowness',
-            'title': 'Candor: The Art of Gesticulation',
-        }
-    }
+    _TESTS = [
+        {
+            'url': 'http://www.nowness.com/day/2013/6/27/3131/candor--the-art-of-gesticulation',
+            'md5': '068bc0202558c2e391924cb8cc470676',
+            'info_dict': {
+                'id': '2520295746001',
+                'ext': 'mp4',
+                'title': 'Candor: The Art of Gesticulation',
+                'description': 'Candor: The Art of Gesticulation',
+                'thumbnail': 're:^https?://.*\.jpg',
+                'uploader': 'Nowness',
+            }
+        },
+        {
+            'url': 'http://cn.nowness.com/day/2014/8/7/4069/kasper-bj-rke-ft-jaakko-eino-kalevi--tnr',
+            'md5': 'e79cf125e387216f86b2e0a5b5c63aa3',
+            'info_dict': {
+                'id': '3716354522001',
+                'ext': 'mp4',
+                'title': 'Kasper Bjørke ft. Jaakko Eino Kalevi: TNR',
+                'description': 'Kasper Bjørke ft. Jaakko Eino Kalevi: TNR',
+                'thumbnail': 're:^https?://.*\.jpg',
+                'uploader': 'Nowness',
+            }
+        },
+    ]
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)

youtube_dl/extractor/oe1.py

@@ -1,40 +0,0 @@
-# coding: utf-8
-from __future__ import unicode_literals
-
-import calendar
-import datetime
-import re
-
-from .common import InfoExtractor
-
-# audios on oe1.orf.at are only available for 7 days, so we can't
-# add tests.
-
-
-class OE1IE(InfoExtractor):
-    IE_DESC = 'oe1.orf.at'
-    _VALID_URL = r'http://oe1\.orf\.at/programm/(?P<id>[0-9]+)'
-
-    def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
-        show_id = mobj.group('id')
-
-        data = self._download_json(
-            'http://oe1.orf.at/programm/%s/konsole' % show_id,
-            show_id
-        )
-
-        timestamp = datetime.datetime.strptime('%s %s' % (
-            data['item']['day_label'],
-            data['item']['time']
-        ), '%d.%m.%Y %H:%M')
-        unix_timestamp = calendar.timegm(timestamp.utctimetuple())
-
-        return {
-            'id': show_id,
-            'title': data['item']['title'],
-            'url': data['item']['url_stream'],
-            'ext': 'mp3',
-            'description': data['item'].get('info'),
-            'timestamp': unix_timestamp
-        }

youtube_dl/extractor/ooyala.py

@@ -3,23 +3,38 @@ import re
 import json
 
 from .common import InfoExtractor
-from ..utils import unescapeHTML
+from ..utils import (
+    unescapeHTML,
+    ExtractorError,
+)
 
 
 class OoyalaIE(InfoExtractor):
     _VALID_URL = r'(?:ooyala:|https?://.+?\.ooyala\.com/.*?(?:embedCode|ec)=)(?P<id>.+?)(&|$)'
 
-    _TEST = {
-        # From http://it.slashdot.org/story/13/04/25/178216/recovering-data-from-broken-hard-drives-and-ssds-video
-        'url': 'http://player.ooyala.com/player.js?embedCode=pxczE2YjpfHfn1f3M-ykG_AmJRRn0PD8',
-        'md5': '3f5cceb3a7bf461d6c29dc466cf8033c',
-        'info_dict': {
-            'id': 'pxczE2YjpfHfn1f3M-ykG_AmJRRn0PD8',
-            'ext': 'mp4',
-            'title': 'Explaining Data Recovery from Hard Drives and SSDs',
-            'description': 'How badly damaged does a drive have to be to defeat Russell and his crew? Apparently, smashed to bits.',
+    _TESTS = [
+        {
+            # From http://it.slashdot.org/story/13/04/25/178216/recovering-data-from-broken-hard-drives-and-ssds-video
+            'url': 'http://player.ooyala.com/player.js?embedCode=pxczE2YjpfHfn1f3M-ykG_AmJRRn0PD8',
+            'md5': '3f5cceb3a7bf461d6c29dc466cf8033c',
+            'info_dict': {
+                'id': 'pxczE2YjpfHfn1f3M-ykG_AmJRRn0PD8',
+                'ext': 'mp4',
+                'title': 'Explaining Data Recovery from Hard Drives and SSDs',
+                'description': 'How badly damaged does a drive have to be to defeat Russell and his crew? Apparently, smashed to bits.',
+            },
+        }, {
+            # Only available for ipad
+            'url': 'http://player.ooyala.com/player.js?embedCode=x1b3lqZDq9y_7kMyC2Op5qo-p077tXD0',
+            'md5': '4b9754921fddb68106e48c142e2a01e6',
+            'info_dict': {
+                'id': 'x1b3lqZDq9y_7kMyC2Op5qo-p077tXD0',
+                'ext': 'mp4',
+                'title': 'Simulation Overview - Levels of Simulation',
+                'description': '',
+            },
         },
-    }
+    ]
 
     @staticmethod
     def _url_for_embed_code(embed_code):
@@ -47,13 +62,30 @@ class OoyalaIE(InfoExtractor):
         player = self._download_webpage(player_url, embedCode)
         mobile_url = self._search_regex(r'mobile_player_url="(.+?)&device="',
                                         player, 'mobile player url')
-        mobile_player = self._download_webpage(mobile_url, embedCode)
-        videos_info = self._search_regex(
-            r'var streams=window.oo_testEnv\?\[\]:eval\("\((\[{.*?}\])\)"\);',
-            mobile_player, 'info').replace('\\"','"')
-        videos_more_info = self._search_regex(r'eval\("\(({.*?\\"promo\\".*?})\)"', mobile_player, 'more info').replace('\\"','"')
+        # Looks like some videos are only available for particular devices
+        # (e.g. http://player.ooyala.com/player.js?embedCode=x1b3lqZDq9y_7kMyC2Op5qo-p077tXD0
+        # is only available for ipad)
+        # Working around with fetching URLs for all the devices found starting with 'unknown'
+        # until we succeed or eventually fail for each device.
+        devices = re.findall(r'device\s*=\s*"([^"]+)";', player)
+        devices.remove('unknown')
+        devices.insert(0, 'unknown')
+        for device in devices:
+            mobile_player = self._download_webpage(
+                '%s&device=%s' % (mobile_url, device), embedCode,
+                'Downloading mobile player JS for %s device' % device)
+            videos_info = self._search_regex(
+                r'var streams=window.oo_testEnv\?\[\]:eval\("\((\[{.*?}\])\)"\);',
+                mobile_player, 'info', fatal=False, default=None)
+            if videos_info:
+                break
+        if not videos_info:
+            raise ExtractorError('Unable to extract info')
+        videos_info = videos_info.replace('\\"', '"')
+        videos_more_info = self._search_regex(
+            r'eval\("\(({.*?\\"promo\\".*?})\)"', mobile_player, 'more info').replace('\\"', '"')
         videos_info = json.loads(videos_info)
-        videos_more_info =json.loads(videos_more_info)
+        videos_more_info = json.loads(videos_more_info)
 
         if videos_more_info.get('lineup'):
             videos = [self._extract_result(info, more_info) for (info, more_info) in zip(videos_info, videos_more_info['lineup'])]

youtube_dl/extractor/orf.py

@@ -3,6 +3,8 @@ from __future__ import unicode_literals
 
 import json
 import re
+import calendar
+import datetime
 
 from .common import InfoExtractor
 from ..utils import (
@@ -12,7 +14,9 @@ from ..utils import (
 )
 
 
-class ORFIE(InfoExtractor):
+class ORFTVthekIE(InfoExtractor):
+    IE_NAME = 'orf:tvthek'
+    IE_DESC = 'ORF TVthek'
     _VALID_URL = r'https?://tvthek\.orf\.at/(?:programs/.+?/episodes|topics/.+?|program/[^/]+)/(?P<id>\d+)'
 
     _TEST = {
@@ -105,3 +109,73 @@ class ORFIE(InfoExtractor):
             'entries': entries,
             'id': playlist_id,
         }
+
+
+# Audios on ORF radio are only available for 7 days, so we can't add tests.
+
+
+class ORFOE1IE(InfoExtractor):
+    IE_NAME = 'orf:oe1'
+    IE_DESC = 'Radio Österreich 1'
+    _VALID_URL = r'http://oe1\.orf\.at/programm/(?P<id>[0-9]+)'
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        show_id = mobj.group('id')
+
+        data = self._download_json(
+            'http://oe1.orf.at/programm/%s/konsole' % show_id,
+            show_id
+        )
+
+        timestamp = datetime.datetime.strptime('%s %s' % (
+            data['item']['day_label'],
+            data['item']['time']
+        ), '%d.%m.%Y %H:%M')
+        unix_timestamp = calendar.timegm(timestamp.utctimetuple())
+
+        return {
+            'id': show_id,
+            'title': data['item']['title'],
+            'url': data['item']['url_stream'],
+            'ext': 'mp3',
+            'description': data['item'].get('info'),
+            'timestamp': unix_timestamp
+        }
+
+
+class ORFFM4IE(InfoExtractor):
+    IE_DESC = 'orf:fm4'
+    IE_DESC = 'radio FM4'
+    _VALID_URL = r'http://fm4\.orf\.at/7tage/?#(?P<date>[0-9]+)/(?P<show>\w+)'
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        show_date = mobj.group('date')
+        show_id = mobj.group('show')
+
+        data = self._download_json(
+            'http://audioapi.orf.at/fm4/json/2.0/broadcasts/%s/4%s' % (show_date, show_id),
+            show_id
+        )
+
+        def extract_entry_dict(info, title, subtitle):
+            return {
+                'id': info['loopStreamId'].replace('.mp3', ''),
+                'url': 'http://loopstream01.apa.at/?channel=fm4&id=%s' % info['loopStreamId'],
+                'title': title,
+                'description': subtitle,
+                'duration': (info['end'] - info['start']) / 1000,
+                'timestamp': info['start'] / 1000,
+                'ext': 'mp3'
+            }
+
+        entries = [extract_entry_dict(t, data['title'], data['subtitle']) for t in data['streams']]
+
+        return {
+            '_type': 'playlist',
+            'id': show_id,
+            'title': data['title'],
+            'description': data['subtitle'],
+            'entries': entries
+        }

youtube_dl/extractor/pbs.py

@@ -20,25 +20,60 @@ class PBSIE(InfoExtractor):
         )
     '''
 
-    _TEST = {
-        'url': 'http://www.pbs.org/tpt/constitution-usa-peter-sagal/watch/a-more-perfect-union/',
-        'md5': 'ce1888486f0908d555a8093cac9a7362',
-        'info_dict': {
-            'id': '2365006249',
-            'ext': 'mp4',
-            'title': 'A More Perfect Union',
-            'description': 'md5:ba0c207295339c8d6eced00b7c363c6a',
-            'duration': 3190,
+    _TESTS = [
+        {
+            'url': 'http://www.pbs.org/tpt/constitution-usa-peter-sagal/watch/a-more-perfect-union/',
+            'md5': 'ce1888486f0908d555a8093cac9a7362',
+            'info_dict': {
+                'id': '2365006249',
+                'ext': 'mp4',
+                'title': 'A More Perfect Union',
+                'description': 'md5:ba0c207295339c8d6eced00b7c363c6a',
+                'duration': 3190,
+            },
         },
-    }
+        {
+            'url': 'http://www.pbs.org/wgbh/pages/frontline/losing-iraq/',
+            'md5': '143c98aa54a346738a3d78f54c925321',
+            'info_dict': {
+                'id': '2365297690',
+                'ext': 'mp4',
+                'title': 'Losing Iraq',
+                'description': 'md5:f5bfbefadf421e8bb8647602011caf8e',
+                'duration': 5050,
+            },
+        },
+        {
+            'url': 'http://www.pbs.org/newshour/bb/education-jan-june12-cyberschools_02-23/',
+            'md5': 'b19856d7f5351b17a5ab1dc6a64be633',
+            'info_dict': {
+                'id': '2201174722',
+                'ext': 'mp4',
+                'title': 'Cyber Schools Gain Popularity, but Quality Questions Persist',
+                'description': 'md5:5871c15cba347c1b3d28ac47a73c7c28',
+                'duration': 801,
+            },
+        },
+    ]
 
-    def _real_extract(self, url):
+    def _extract_ids(self, url):
         mobj = re.match(self._VALID_URL, url)
 
         presumptive_id = mobj.group('presumptive_id')
         display_id = presumptive_id
         if presumptive_id:
             webpage = self._download_webpage(url, display_id)
+
+            MEDIA_ID_REGEXES = [
+                r"div\s*:\s*'videoembed'\s*,\s*mediaid\s*:\s*'(\d+)'",  # frontline video embed
+                r'class="coveplayerid">([^<]+)<',                       # coveplayer
+            ]
+
+            media_id = self._search_regex(
+                MEDIA_ID_REGEXES, webpage, 'media ID', fatal=False, default=None)
+            if media_id:
+                return media_id, presumptive_id
+
             url = self._search_regex(
                 r'<iframe\s+id=["\']partnerPlayer["\'].*?\s+src=["\'](.*?)["\']>',
                 webpage, 'player URL')
@@ -57,6 +92,11 @@ class PBSIE(InfoExtractor):
             video_id = mobj.group('id')
             display_id = video_id
 
+        return video_id, display_id
+
+    def _real_extract(self, url):
+        video_id, display_id = self._extract_ids(url)
+
         info_url = 'http://video.pbs.org/videoInfo/%s?format=json' % video_id
         info = self._download_json(info_url, display_id)
 

youtube_dl/extractor/reverbnation.py

@@ -1,23 +1,23 @@
 from __future__ import unicode_literals
 
 import re
-import time
 
 from .common import InfoExtractor
-from ..utils import strip_jsonp
+from ..utils import str_or_none
 
 
 class ReverbNationIE(InfoExtractor):
     _VALID_URL = r'^https?://(?:www\.)?reverbnation\.com/.*?/song/(?P<id>\d+).*?$'
     _TESTS = [{
         'url': 'http://www.reverbnation.com/alkilados/song/16965047-mona-lisa',
-        'file': '16965047.mp3',
         'md5': '3da12ebca28c67c111a7f8b262d3f7a7',
         'info_dict': {
+            "id": "16965047",
+            "ext": "mp3",
             "title": "MONA LISA",
             "uploader": "ALKILADOS",
-            "uploader_id": 216429,
-            "thumbnail": "//gp1.wac.edgecastcdn.net/802892/production_public/Photo/13761700/image/1366002176_AVATAR_MONA_LISA.jpg"
+            "uploader_id": "216429",
+            "thumbnail": "re:^https://gp1\.wac\.edgecastcdn\.net/.*?\.jpg$"
         },
     }]
 
@@ -26,10 +26,8 @@ class ReverbNationIE(InfoExtractor):
         song_id = mobj.group('id')
 
         api_res = self._download_json(
-            'https://api.reverbnation.com/song/%s?callback=api_response_5&_=%d'
-                % (song_id, int(time.time() * 1000)),
+            'https://api.reverbnation.com/song/%s' % song_id,
             song_id,
-            transform_source=strip_jsonp,
             note='Downloading information of song %s' % song_id
         )
 
@@ -38,8 +36,9 @@ class ReverbNationIE(InfoExtractor):
             'title': api_res.get('name'),
             'url': api_res.get('url'),
             'uploader': api_res.get('artist', {}).get('name'),
-            'uploader_id': api_res.get('artist', {}).get('id'),
-            'thumbnail': api_res.get('image', api_res.get('thumbnail')),
+            'uploader_id': str_or_none(api_res.get('artist', {}).get('id')),
+            'thumbnail': self._proto_relative_url(
+                api_res.get('image', api_res.get('thumbnail'))),
             'ext': 'mp3',
             'vcodec': 'none',
         }

youtube_dl/extractor/rtlnl.py

@@ -0,0 +1,52 @@
+from __future__ import unicode_literals
+
+import re
+
+from .common import InfoExtractor
+
+
+class RtlXlIE(InfoExtractor):
+    IE_NAME = 'rtlxl.nl'
+    _VALID_URL = r'https?://www\.rtlxl\.nl/#!/[^/]+/(?P<uuid>[^/?]+)'
+
+    _TEST = {
+        'url': 'http://www.rtlxl.nl/#!/rtl-nieuws-132237/6e4203a6-0a5e-3596-8424-c599a59e0677',
+        'info_dict': {
+            'id': '6e4203a6-0a5e-3596-8424-c599a59e0677',
+            'ext': 'flv',
+            'title': 'RTL Nieuws - Laat',
+            'description': 'Dagelijks het laatste nieuws uit binnen- en '
+                'buitenland. Voor nog meer nieuws kunt u ook gebruikmaken van '
+                'onze mobiele apps.',
+            'timestamp': 1408051800,
+            'upload_date': '20140814',
+        },
+        'params': {
+            # We download the first bytes of the first fragment, it can't be
+            # processed by the f4m downloader beacuse it isn't complete
+            'skip_download': True,
+        },
+    }
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        uuid = mobj.group('uuid')
+
+        info = self._download_json(
+            'http://www.rtl.nl/system/s4m/vfd/version=2/uuid=%s/fmt=flash/' % uuid,
+            uuid)
+        meta = info['meta']
+        material = info['material'][0]
+        episode_info = info['episodes'][0]
+
+        f4m_url = 'http://manifest.us.rtl.nl' + material['videopath']
+        progname = info['abstracts'][0]['name']
+        subtitle = material['title'] or info['episodes'][0]['name']
+
+        return {
+            'id': uuid,
+            'title': '%s - %s' % (progname, subtitle), 
+            'formats': self._extract_f4m_formats(f4m_url, uuid),
+            'timestamp': material['original_date'],
+            'description': episode_info['synopsis'],
+        }

youtube_dl/extractor/rtlnow.py

@@ -92,16 +92,7 @@ class RTLnowIE(InfoExtractor):
         },
         {
             'url': 'http://www.n-tvnow.de/deluxe-alles-was-spass-macht/thema-ua-luxushotel-fuer-vierbeiner.php?container_id=153819&player=1&season=0',
-            'info_dict': {
-                'id': '153819',
-                'ext': 'flv',
-                'title': 'Deluxe - Alles was Spaß macht - Thema u.a.: Luxushotel für Vierbeiner',
-                'description': 'md5:c3705e1bb32e1a5b2bcd634fc065c631',
-                'thumbnail': 'http://autoimg.static-fra.de/ntvnow/383157/1500x1500/image2.jpg',
-                'upload_date': '20140221',
-                'duration': 2429,
-            },
-            'skip': 'Only works from Germany',
+            'only_matching': True,
         },
     ]
 

youtube_dl/extractor/rtve.py

@@ -17,7 +17,7 @@ class RTVEALaCartaIE(InfoExtractor):
 
     _TEST = {
         'url': 'http://www.rtve.es/alacarta/videos/balonmano/o-swiss-cup-masculina-final-espana-suecia/2491869/',
-        'md5': '18fcd45965bdd076efdb12cd7f6d7b9e',
+        'md5': '1d49b7e1ca7a7502c56a4bf1b60f1b43',
         'info_dict': {
             'id': '2491869',
             'ext': 'mp4',

youtube_dl/extractor/savefrom.py

@@ -20,7 +20,7 @@ class SaveFromIE(InfoExtractor):
             'upload_date': '20120816',
             'uploader': 'Howcast',
             'uploader_id': 'Howcast',
-            'description': 'md5:4f0aac94361a12e1ce57d74f85265175',
+            'description': 're:(?s).* Hi, my name is Rene Dreifuss\. And I\'m here to show you some MMA.*',
         },
         'params': {
             'skip_download': True

youtube_dl/extractor/shared.py

@@ -0,0 +1,57 @@
+from __future__ import unicode_literals
+
+import re
+import base64
+
+from .common import InfoExtractor
+from ..utils import (
+    ExtractorError,
+    compat_urllib_request,
+    compat_urllib_parse,
+    int_or_none,
+)
+
+
+class SharedIE(InfoExtractor):
+    _VALID_URL = r'http://shared\.sx/(?P<id>[\da-z]{10})'
+
+    _TEST = {
+        'url': 'http://shared.sx/0060718775',
+        'md5': '106fefed92a8a2adb8c98e6a0652f49b',
+        'info_dict': {
+            'id': '0060718775',
+            'ext': 'mp4',
+            'title': 'Bmp4',
+        },
+    }
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        video_id = mobj.group('id')
+
+        page = self._download_webpage(url, video_id)
+
+        if re.search(r'>File does not exist<', page) is not None:
+            raise ExtractorError('Video %s does not exist' % video_id, expected=True)
+
+        download_form = dict(re.findall(r'<input type="hidden" name="([^"]+)" value="([^"]*)"', page))
+
+        request = compat_urllib_request.Request(url, compat_urllib_parse.urlencode(download_form))
+        request.add_header('Content-Type', 'application/x-www-form-urlencoded')
+
+        video_page = self._download_webpage(request, video_id, 'Downloading video page')
+
+        video_url = self._html_search_regex(r'data-url="([^"]+)"', video_page, 'video URL')
+        title = base64.b64decode(self._html_search_meta('full:title', page, 'title')).decode('utf-8')
+        filesize = int_or_none(self._html_search_meta('full:size', page, 'file size', fatal=False))
+        thumbnail = self._html_search_regex(
+            r'data-poster="([^"]+)"', video_page, 'thumbnail', fatal=False, default=None)
+
+        return {
+            'id': video_id,
+            'url': video_url,
+            'ext': 'mp4',
+            'filesize': filesize,
+            'title': title,
+            'thumbnail': thumbnail,
+        }

youtube_dl/extractor/snotr.py

@@ -0,0 +1,68 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+import re
+
+from .common import InfoExtractor
+from ..utils import (
+    float_or_none,
+    str_to_int,
+    parse_duration,
+)
+
+
+class SnotrIE(InfoExtractor):
+    _VALID_URL = r'http?://(?:www\.)?snotr\.com/video/(?P<id>\d+)/([\w]+)'
+    _TESTS = [{
+        'url': 'http://www.snotr.com/video/13708/Drone_flying_through_fireworks',
+        'info_dict': {
+            'id': '13708',
+            'ext': 'flv',
+            'title': 'Drone flying through fireworks!',
+            'duration': 247,
+            'filesize_approx': 98566144,
+            'description': 'A drone flying through Fourth of July Fireworks',
+        }
+    }, {
+        'url': 'http://www.snotr.com/video/530/David_Letteman_-_George_W_Bush_Top_10',
+        'info_dict': {
+            'id': '530',
+            'ext': 'flv',
+            'title': 'David Letteman - George W. Bush Top 10',
+            'duration': 126,
+            'filesize_approx': 8912896,
+            'description': 'The top 10 George W. Bush moments, brought to you by David Letterman!',
+        }
+    }]
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        video_id = mobj.group('id')
+
+        webpage = self._download_webpage(url, video_id)
+        title = self._og_search_title(webpage)
+
+        description = self._og_search_description(webpage)
+        video_url = "http://cdn.videos.snotr.com/%s.flv" % video_id
+
+        view_count = str_to_int(self._html_search_regex(
+            r'<p>\n<strong>Views:</strong>\n([\d,\.]+)</p>',
+            webpage, 'view count', fatal=False))
+
+        duration = parse_duration(self._html_search_regex(
+            r'<p>\n<strong>Length:</strong>\n\s*([0-9:]+).*?</p>',
+            webpage, 'duration', fatal=False))
+
+        filesize_approx = float_or_none(self._html_search_regex(
+            r'<p>\n<strong>Filesize:</strong>\n\s*([0-9.]+)\s*megabyte</p>',
+            webpage, 'filesize', fatal=False), invscale=1024 * 1024)
+
+        return {
+            'id': video_id,
+            'description': description,
+            'title': title,
+            'url': video_url,
+            'view_count': view_count,
+            'duration': duration,
+            'filesize_approx': filesize_approx,
+        }

youtube_dl/extractor/sockshare.py

@@ -0,0 +1,80 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+from ..utils import (
+    ExtractorError,
+    compat_urllib_parse,
+    compat_urllib_request,
+    determine_ext,
+)
+import re
+
+from .common import InfoExtractor
+
+
+class SockshareIE(InfoExtractor):
+    _VALID_URL = r'https?://(?:www\.)?sockshare\.com/file/(?P<id>[0-9A-Za-z]+)'
+    _FILE_DELETED_REGEX = r'This file doesn\'t exist, or has been removed\.</div>'
+    _TEST = {
+        'url': 'http://www.sockshare.com/file/437BE28B89D799D7',
+        'md5': '9d0bf1cfb6dbeaa8d562f6c97506c5bd',
+        'info_dict': {
+            'id': '437BE28B89D799D7',
+            'title': 'big_buck_bunny_720p_surround.avi',
+            'ext': 'avi',
+            'thumbnail': 're:^http://.*\.jpg$',
+        }
+    }
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        video_id = mobj.group('id')
+
+        url = 'http://sockshare.com/file/%s' % video_id
+        webpage = self._download_webpage(url, video_id)
+
+        if re.search(self._FILE_DELETED_REGEX, webpage) is not None:
+            raise ExtractorError('Video %s does not exist' % video_id,
+                                 expected=True)
+
+        confirm_hash = self._html_search_regex(r'''(?x)<input\s+
+            type="hidden"\s+
+            value="([^"]*)"\s+
+            name="hash"
+            ''', webpage, 'hash')
+
+        fields = {
+            "hash": confirm_hash,
+            "confirm": "Continue as Free User"
+        }
+
+        post = compat_urllib_parse.urlencode(fields)
+        req = compat_urllib_request.Request(url, post)
+        # Apparently, this header is required for confirmation to work.
+        req.add_header('Host', 'www.sockshare.com')
+        req.add_header('Content-type', 'application/x-www-form-urlencoded')
+
+        webpage = self._download_webpage(
+            req, video_id, 'Downloading video page')
+
+        video_url = self._html_search_regex(
+            r'<a href="([^"]*)".+class="download_file_link"',
+            webpage, 'file url')
+        video_url = "http://www.sockshare.com" + video_url
+        title = self._html_search_regex(r'<h1>(.+)<strong>', webpage, 'title')
+        thumbnail = self._html_search_regex(
+            r'<img\s+src="([^"]*)".+?name="bg"',
+            webpage, 'thumbnail')
+
+        formats = [{
+            'format_id': 'sd',
+            'url': video_url,
+            'ext': determine_ext(title),
+        }]
+
+        return {
+            'id': video_id,
+            'title': title,
+            'thumbnail': thumbnail,
+            'formats': formats,
+        }

youtube_dl/extractor/soundcloud.py

@@ -82,10 +82,10 @@ class SoundcloudIE(InfoExtractor):
         # downloadable song
         {
             'url': 'https://soundcloud.com/oddsamples/bus-brakes',
-            'md5': 'fee7b8747b09bb755cefd4b853e7249a',
+            'md5': '7624f2351f8a3b2e7cd51522496e7631',
             'info_dict': {
                 'id': '128590877',
-                'ext': 'wav',
+                'ext': 'mp3',
                 'title': 'Bus Brakes',
                 'description': 'md5:0170be75dd395c96025d210d261c784e',
                 'uploader': 'oddsamples',

youtube_dl/extractor/steam.py

@@ -53,7 +53,7 @@ class SteamIE(InfoExtractor):
             'ext': 'mp4',
             'upload_date': '20140329',
             'title': 'FRONTIERS - Final Greenlight Trailer',
-            'description': 'md5:6df4fe8dd494ae811869672b0767e025',
+            'description': 'md5:dc96a773669d0ca1b36c13c1f30250d9',
             'uploader': 'AAD Productions',
             'uploader_id': 'AtomicAgeDogGames',
         }

youtube_dl/extractor/streamcloud.py

@@ -1,4 +1,6 @@
 # coding: utf-8
+from __future__ import unicode_literals
+
 import re
 import time
 
@@ -10,18 +12,18 @@ from ..utils import (
 
 
 class StreamcloudIE(InfoExtractor):
-    IE_NAME = u'streamcloud.eu'
+    IE_NAME = 'streamcloud.eu'
     _VALID_URL = r'https?://streamcloud\.eu/(?P<id>[a-zA-Z0-9_-]+)/(?P<fname>[^#?]*)\.html'
 
     _TEST = {
-        u'url': u'http://streamcloud.eu/skp9j99s4bpz/youtube-dl_test_video_____________-BaW_jenozKc.mp4.html',
-        u'file': u'skp9j99s4bpz.mp4',
-        u'md5': u'6bea4c7fa5daaacc2a946b7146286686',
-        u'info_dict': {
-            u'title': u'youtube-dl test video  \'/\\ ä ↭',
-            u'duration': 9,
+        'url': 'http://streamcloud.eu/skp9j99s4bpz/youtube-dl_test_video_____________-BaW_jenozKc.mp4.html',
+        'md5': '6bea4c7fa5daaacc2a946b7146286686',
+        'info_dict': {
+            'id': 'skp9j99s4bpz',
+            'ext': 'mp4',
+            'title': 'youtube-dl test video  \'/\\ ä ↭',
         },
-        u'skip': u'Only available from the EU'
+        'skip': 'Only available from the EU'
     }
 
     def _real_extract(self, url):
@@ -46,21 +48,17 @@ class StreamcloudIE(InfoExtractor):
         req = compat_urllib_request.Request(url, post, headers)
 
         webpage = self._download_webpage(
-            req, video_id, note=u'Downloading video page ...')
+            req, video_id, note='Downloading video page ...')
         title = self._html_search_regex(
-            r'<h1[^>]*>([^<]+)<', webpage, u'title')
+            r'<h1[^>]*>([^<]+)<', webpage, 'title')
         video_url = self._search_regex(
-            r'file:\s*"([^"]+)"', webpage, u'video URL')
-        duration_str = self._search_regex(
-            r'duration:\s*"?([0-9]+)"?', webpage, u'duration', fatal=False)
-        duration = None if duration_str is None else int(duration_str)
+            r'file:\s*"([^"]+)"', webpage, 'video URL')
         thumbnail = self._search_regex(
-            r'image:\s*"([^"]+)"', webpage, u'thumbnail URL', fatal=False)
+            r'image:\s*"([^"]+)"', webpage, 'thumbnail URL', fatal=False)
 
         return {
             'id': video_id,
             'title': title,
             'url': video_url,
-            'duration': duration,
             'thumbnail': thumbnail,
         }

youtube_dl/extractor/swrmediathek.py

@@ -8,7 +8,7 @@ from ..utils import parse_duration
 
 
 class SWRMediathekIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:www\.)?swrmediathek\.de/player\.htm\?show=(?P<id>[\da-f]{8}-[\da-f]{4}-[\da-f]{4}-[\da-f]{4}-[\da-f]{12})'
+    _VALID_URL = r'https?://(?:www\.)?swrmediathek\.de/(?:content/)?player\.htm\?show=(?P<id>[\da-f]{8}-[\da-f]{4}-[\da-f]{4}-[\da-f]{4}-[\da-f]{12})'
 
     _TESTS = [{
         'url': 'http://swrmediathek.de/player.htm?show=849790d0-dab8-11e3-a953-0026b975f2e6',
@@ -52,6 +52,20 @@ class SWRMediathekIE(InfoExtractor):
             'uploader': 'SWR 2',
             'uploader_id': '284670',
         }
+    }, {
+        'url': 'http://swrmediathek.de/content/player.htm?show=52dc7e00-15c5-11e4-84bc-0026b975f2e6',
+        'md5': '881531487d0633080a8cc88d31ef896f',
+        'info_dict': {
+            'id': '52dc7e00-15c5-11e4-84bc-0026b975f2e6',
+            'ext': 'mp4',
+            'title': 'Familienspaß am Bodensee',
+            'description': 'md5:0b591225a32cfde7be1629ed49fe4315',
+            'thumbnail': 're:http://.*\.jpg',
+            'duration': 1784,
+            'upload_date': '20140727',
+            'uploader': 'SWR Fernsehen BW',
+            'uploader_id': '281130',
+        }
     }]
 
     def _real_extract(self, url):

youtube_dl/extractor/tagesschau.py

@@ -19,16 +19,6 @@ class TagesschauIE(InfoExtractor):
             'description': 'md5:69da3c61275b426426d711bde96463ab',
             'thumbnail': 're:^http:.*\.jpg$',
         },
-    }, {
-        'url': 'http://www.tagesschau.de/multimedia/video/video-5964.html',
-        'md5': '66652566900963a3f962333579eeffcf',
-        'info_dict': {
-            'id': '5964',
-            'ext': 'mp4',
-            'title': 'Nahost-Konflikt: Israel bombadiert Ziele im Gazastreifen und Westjordanland',
-            'description': 'md5:07bfc78c48eec3145ed4805299a1900a',
-            'thumbnail': 're:http://.*\.jpg',
-        },
     }]
 
     _FORMATS = {

youtube_dl/extractor/teachertube.py

@@ -62,7 +62,7 @@ class TeacherTubeIE(InfoExtractor):
 
         webpage = self._download_webpage(url, video_id)
 
-        title = self._html_search_meta('title', webpage, 'title')
+        title = self._html_search_meta('title', webpage, 'title', fatal=True)
         TITLE_SUFFIX = ' - TeacherTube'
         if title.endswith(TITLE_SUFFIX):
             title = title[:-len(TITLE_SUFFIX)].strip()
@@ -101,7 +101,11 @@ class TeacherTubeUserIE(InfoExtractor):
 
     _VALID_URL = r'https?://(?:www\.)?teachertube\.com/(user/profile|collection)/(?P<user>[0-9a-zA-Z]+)/?'
 
-    _MEDIA_RE = r'(?s)"sidebar_thumb_time">[0-9:]+</div>.+?<a href="(https?://(?:www\.)?teachertube\.com/(?:video|audio)/[^"]+)">'
+    _MEDIA_RE = r'''(?sx)
+        class="?sidebar_thumb_time"?>[0-9:]+</div>
+        \s*
+        <a\s+href="(https?://(?:www\.)?teachertube\.com/(?:video|audio)/[^"]+)"
+    '''
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
@@ -111,14 +115,12 @@ class TeacherTubeUserIE(InfoExtractor):
         webpage = self._download_webpage(url, user_id)
         urls.extend(re.findall(self._MEDIA_RE, webpage))
         
-        pages = re.findall(r'/ajax-user/user-videos/%s\?page=([0-9]+)' % user_id, webpage)[1:-1]
+        pages = re.findall(r'/ajax-user/user-videos/%s\?page=([0-9]+)' % user_id, webpage)[:-1]
         for p in pages:
             more = 'http://www.teachertube.com/ajax-user/user-videos/%s?page=%s' % (user_id, p)
-            webpage = self._download_webpage(more, user_id, 'Downloading page %s/%s' % (p, len(pages) + 1))
-            urls.extend(re.findall(self._MEDIA_RE, webpage))
-
-        entries = []
-        for url in urls:
-            entries.append(self.url_result(url, 'TeacherTube'))
+            webpage = self._download_webpage(more, user_id, 'Downloading page %s/%s' % (p, len(pages)))
+            video_urls = re.findall(self._MEDIA_RE, webpage)
+            urls.extend(video_urls)
 
+        entries = [self.url_result(vurl, 'TeacherTube') for vurl in urls]
         return self.playlist_result(entries, user_id)

youtube_dl/extractor/teamcoco.py

@@ -37,7 +37,7 @@ class TeamcocoIE(InfoExtractor):
         video_id = mobj.group("video_id")
         if not video_id:
             video_id = self._html_search_regex(
-                r'<article class="video" data-id="(\d+?)"',
+                r'data-node-id="(\d+?)"',
                 webpage, 'video id')
 
         data_url = 'http://teamcoco.com/cvp/2.0/%s.xml' % video_id

youtube_dl/extractor/tenplay.py

@@ -1,8 +1,6 @@
 # coding: utf-8
 from __future__ import unicode_literals
 
-import re
-
 from .common import InfoExtractor
 
 

youtube_dl/extractor/tvplay.py

@@ -0,0 +1,85 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+import re
+
+from .common import InfoExtractor
+from ..utils import (
+    ExtractorError,
+    parse_iso8601,
+    qualities,
+)
+
+
+class TVPlayIE(InfoExtractor):
+    _VALID_URL = r'http://(?:www\.)?tvplay\.lv/parraides/[^/]+/(?P<id>\d+)'
+    _TESTS = [
+        {
+            'url': 'http://www.tvplay.lv/parraides/vinas-melo-labak/418113?autostart=true',
+            'info_dict': {
+                'id': '418113',
+                'ext': 'flv',
+                'title': 'Kādi ir īri? - Viņas melo labāk',
+                'description': 'Baiba apsmej īrus, kādi tie ir un ko viņi dara.',
+                'duration': 25,
+                'timestamp': 1406097056,
+                'upload_date': '20140723',
+            },
+            'params': {
+                # rtmp download
+                'skip_download': True,
+            },
+        },
+    ]
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        video_id = mobj.group('id')
+
+        video = self._download_json(
+            'http://playapi.mtgx.tv/v1/videos/%s' % video_id, video_id, 'Downloading video JSON')
+
+        if video['is_geo_blocked']:
+            raise ExtractorError(
+                'This content is not available in your country due to copyright reasons', expected=True)
+
+        streams = self._download_json(
+            'http://playapi.mtgx.tv/v1/videos/stream/%s' % video_id, video_id, 'Downloading streams JSON')
+
+        quality = qualities(['hls', 'medium', 'high'])
+        formats = []
+        for format_id, video_url in streams['streams'].items():
+            if not video_url:
+                continue
+            fmt = {
+                'format_id': format_id,
+                'preference': quality(format_id),
+            }
+            if video_url.startswith('rtmp'):
+                m = re.search(r'^(?P<url>rtmp://[^/]+/(?P<app>[^/]+))/(?P<playpath>.+)$', video_url)
+                if not m:
+                    continue
+                fmt.update({
+                    'ext': 'flv',
+                    'url': m.group('url'),
+                    'app': m.group('app'),
+                    'play_path': m.group('playpath'),
+                })
+            else:
+                fmt.update({
+                    'url': video_url,
+                })
+            formats.append(fmt)
+
+        self._sort_formats(formats)
+
+        return {
+            'id': video_id,
+            'title': video['title'],
+            'description': video['description'],
+            'duration': video['duration'],
+            'timestamp': parse_iso8601(video['created_at']),
+            'view_count': video['views']['total'],
+            'age_limit': video.get('age_limit', 0),
+            'formats': formats,
+        }

youtube_dl/extractor/ubu.py

@@ -0,0 +1,56 @@
+from __future__ import unicode_literals
+
+import re
+
+from .common import InfoExtractor
+from ..utils import int_or_none
+
+
+class UbuIE(InfoExtractor):
+    _VALID_URL = r'http://(?:www\.)?ubu\.com/film/(?P<id>[\da-z_-]+)\.html'
+    _TEST = {
+        'url': 'http://ubu.com/film/her_noise.html',
+        'md5': '8edd46ee8aa6b265fb5ed6cf05c36bc9',
+        'info_dict': {
+            'id': 'her_noise',
+            'ext': 'mp4',
+            'title': 'Her Noise - The Making Of (2007)',
+            'duration': 3600,
+        },
+    }
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        video_id = mobj.group('id')
+
+        webpage = self._download_webpage(url, video_id)
+
+        title = self._html_search_regex(
+            r'<title>.+?Film &amp; Video: ([^<]+)</title>', webpage, 'title')
+
+        duration = int_or_none(self._html_search_regex(
+            r'Duration: (\d+) minutes', webpage, 'duration', fatal=False, default=None))
+        if duration:
+            duration *= 60
+
+        formats = []
+
+        FORMAT_REGEXES = [
+            ['sq', r"'flashvars'\s*,\s*'file=([^']+)'"],
+            ['hq', r'href="(http://ubumexico\.centro\.org\.mx/video/[^"]+)"']
+        ]
+
+        for format_id, format_regex in FORMAT_REGEXES:
+            m = re.search(format_regex, webpage)
+            if m:
+                formats.append({
+                    'url': m.group(1),
+                    'format_id': format_id,
+                })
+
+        return {
+            'id': video_id,
+            'title': title,
+            'duration': duration,
+            'formats': formats,
+        }

youtube_dl/extractor/vevo.py

@@ -177,6 +177,7 @@ class VevoIE(InfoExtractor):
             self._downloader.report_warning(
                 'Cannot download SMIL information, falling back to JSON ..')
 
+        self._sort_formats(formats)
         timestamp_ms = int(self._search_regex(
             r'/Date\((\d+)\)/', video_info['launchDate'], 'launch date'))
 

youtube_dl/extractor/vidme.py

@@ -0,0 +1,68 @@
+from __future__ import unicode_literals
+
+import re
+
+from .common import InfoExtractor
+from ..utils import (
+    int_or_none,
+    float_or_none,
+    str_to_int,
+)
+
+
+class VidmeIE(InfoExtractor):
+    _VALID_URL = r'https?://vid\.me/(?:e/)?(?P<id>[\da-zA-Z]+)'
+    _TEST = {
+        'url': 'https://vid.me/QNB',
+        'md5': 'f42d05e7149aeaec5c037b17e5d3dc82',
+        'info_dict': {
+            'id': 'QNB',
+            'ext': 'mp4',
+            'title': 'Fishing for piranha - the easy way',
+            'description': 'source: https://www.facebook.com/photo.php?v=312276045600871',
+            'duration': 119.92,
+            'timestamp': 1406313244,
+            'upload_date': '20140725',
+            'thumbnail': 're:^https?://.*\.jpg',
+        },
+    }
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        video_id = mobj.group('id')
+
+        webpage = self._download_webpage(url, video_id)
+
+        video_url = self._html_search_regex(r'<source src="([^"]+)"', webpage, 'video URL')
+
+        title = self._og_search_title(webpage)
+        description = self._og_search_description(webpage, default='')
+        thumbnail = self._og_search_thumbnail(webpage)
+        timestamp = int_or_none(self._og_search_property('updated_time', webpage, fatal=False))
+        width = int_or_none(self._og_search_property('video:width', webpage, fatal=False))
+        height = int_or_none(self._og_search_property('video:height', webpage, fatal=False))
+        duration = float_or_none(self._html_search_regex(
+            r'data-duration="([^"]+)"', webpage, 'duration', fatal=False))
+        view_count = str_to_int(self._html_search_regex(
+            r'<span class="video_views">\s*([\d,\.]+)\s*plays?', webpage, 'view count', fatal=False))
+        like_count = str_to_int(self._html_search_regex(
+            r'class="score js-video-vote-score"[^>]+data-score="([\d,\.\s]+)">',
+            webpage, 'like count', fatal=False))
+        comment_count = str_to_int(self._html_search_regex(
+            r'class="js-comment-count"[^>]+data-count="([\d,\.\s]+)">',
+            webpage, 'comment count', fatal=False))
+
+        return {
+            'id': video_id,
+            'url': video_url,
+            'title': title,
+            'description': description,
+            'thumbnail': thumbnail,
+            'timestamp': timestamp,
+            'width': width,
+            'height': height,
+            'duration': duration,
+            'view_count': view_count,
+            'like_count': like_count,
+            'comment_count': comment_count,
+        }

youtube_dl/extractor/vimeo.py

@@ -98,7 +98,7 @@ class VimeoIE(VimeoBaseInfoExtractor, SubtitlesInfoExtractor):
             'info_dict': {
                 'id': '54469442',
                 'ext': 'mp4',
-                'title': 'Kathy Sierra: Building the minimum Badass User, Business of Software',
+                'title': 'Kathy Sierra: Building the minimum Badass User, Business of Software 2012',
                 'uploader': 'The BLN & Business of Software',
                 'uploader_id': 'theblnbusinessofsoftware',
                 'duration': 3610,
@@ -121,6 +121,21 @@ class VimeoIE(VimeoBaseInfoExtractor, SubtitlesInfoExtractor):
                 'videopassword': 'youtube-dl',
             },
         },
+        {
+            'url': 'http://vimeo.com/channels/keypeele/75629013',
+            'md5': '2f86a05afe9d7abc0b9126d229bbe15d',
+            'note': 'Video is freely available via original URL '
+                    'and protected with password when accessed via http://vimeo.com/75629013',
+            'info_dict': {
+                'id': '75629013',
+                'ext': 'mp4',
+                'title': 'Key & Peele: Terrorist Interrogation',
+                'description': 'md5:8678b246399b070816b12313e8b4eb5c',
+                'uploader_id': 'atencio',
+                'uploader': 'Peter Atencio',
+                'duration': 187,
+            },
+        },
         {
             'url': 'http://vimeo.com/76979871',
             'md5': '3363dd6ffebe3784d56f4132317fd446',
@@ -196,8 +211,6 @@ class VimeoIE(VimeoBaseInfoExtractor, SubtitlesInfoExtractor):
         video_id = mobj.group('id')
         if mobj.group('pro') or mobj.group('player'):
             url = 'http://player.vimeo.com/video/' + video_id
-        else:
-            url = 'https://vimeo.com/' + video_id
 
         # Retrieve video webpage to extract further information
         request = compat_urllib_request.Request(url, None, headers)
@@ -263,7 +276,7 @@ class VimeoIE(VimeoBaseInfoExtractor, SubtitlesInfoExtractor):
         if video_thumbnail is None:
             video_thumbs = config["video"].get("thumbs")
             if video_thumbs and isinstance(video_thumbs, dict):
-                _, video_thumbnail = sorted((int(width), t_url) for (width, t_url) in video_thumbs.items())[-1]
+                _, video_thumbnail = sorted((int(width if width.isdigit() else 0), t_url) for (width, t_url) in video_thumbs.items())[-1]
 
         # Extract video description
         video_description = None

youtube_dl/extractor/vodlocker.py

@@ -10,7 +10,7 @@ from ..utils import (
 
 
 class VodlockerIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:www\.)?vodlocker.com/(?P<id>[0-9a-zA-Z]+)(?:\..*?)?'
+    _VALID_URL = r'https?://(?:www\.)?vodlocker\.com/(?P<id>[0-9a-zA-Z]+)(?:\..*?)?'
 
     _TESTS = [{
         'url': 'http://vodlocker.com/e8wvyzz4sl42',

youtube_dl/extractor/vube.py

@@ -3,7 +3,10 @@ from __future__ import unicode_literals
 import re
 
 from .common import InfoExtractor
-from ..utils import int_or_none
+from ..utils import (
+    int_or_none,
+    compat_str,
+)
 
 
 class VubeIE(InfoExtractor):
@@ -20,12 +23,15 @@ class VubeIE(InfoExtractor):
                 'ext': 'mp4',
                 'title': 'Chiara Grispo - Price Tag by Jessie J',
                 'description': 'md5:8ea652a1f36818352428cb5134933313',
-                'thumbnail': 'http://frame.thestaticvube.com/snap/228x128/102e7e63057-5ebc-4f5c-4065-6ce4ebde131f.jpg',
+                'thumbnail': 're:^http://frame\.thestaticvube\.com/snap/[0-9x]+/102e7e63057-5ebc-4f5c-4065-6ce4ebde131f\.jpg$',
                 'uploader': 'Chiara.Grispo',
-                'uploader_id': '1u3hX0znhP',
                 'timestamp': 1388743358,
                 'upload_date': '20140103',
-                'duration': 170.56
+                'duration': 170.56,
+                'like_count': int,
+                'dislike_count': int,
+                'comment_count': int,
+                'categories': ['pop', 'music', 'cover', 'singing', 'jessie j', 'price tag', 'chiara grispo'],
             }
         },
         {
@@ -36,12 +42,33 @@ class VubeIE(InfoExtractor):
                 'ext': 'mp4',
                 'title': 'My 7 year old Sister and I singing "Alive" by Krewella',
                 'description': 'md5:40bcacb97796339f1690642c21d56f4a',
-                'thumbnail': 'http://frame.thestaticvube.com/snap/228x128/102265d5a9f-0f17-4f6b-5753-adf08484ee1e.jpg',
+                'thumbnail': 're:^http://frame\.thestaticvube\.com/snap/[0-9x]+/102265d5a9f-0f17-4f6b-5753-adf08484ee1e\.jpg$',
                 'uploader': 'Seraina',
-                'uploader_id': 'XU9VE2BQ2q',
                 'timestamp': 1396492438,
                 'upload_date': '20140403',
-                'duration': 240.107
+                'duration': 240.107,
+                'like_count': int,
+                'dislike_count': int,
+                'comment_count': int,
+                'categories': ['seraina', 'jessica', 'krewella', 'alive'],
+            }
+        }, {
+            'url': 'http://vube.com/vote/Siren+Gene/0nmsMY5vEq?n=2&t=s',
+            'md5': '0584fc13b50f887127d9d1007589d27f',
+            'info_dict': {
+                'id': '0nmsMY5vEq',
+                'ext': 'mp4',
+                'title': 'Frozen - Let It Go Cover by Siren Gene',
+                'description': 'My rendition of "Let It Go" originally sung by Idina Menzel.',
+                'thumbnail': 're:^http://frame\.thestaticvube\.com/snap/[0-9x]+/10283ab622a-86c9-4681-51f2-30d1f65774af\.jpg$',
+                'uploader': 'Siren',
+                'timestamp': 1395448018,
+                'upload_date': '20140322',
+                'duration': 221.788,
+                'like_count': int,
+                'dislike_count': int,
+                'comment_count': int,
+                'categories': ['let it go', 'cover', 'idina menzel', 'frozen', 'singing', 'disney', 'siren gene'],
             }
         }
     ]
@@ -51,39 +78,52 @@ class VubeIE(InfoExtractor):
         video_id = mobj.group('id')
 
         video = self._download_json(
-            'http://vube.com/api/v2/video/%s' % video_id, video_id, 'Downloading video JSON')
+            'http://vube.com/t-api/v1/video/%s' % video_id, video_id, 'Downloading video JSON')
 
         public_id = video['public_id']
 
-        formats = [
-            {
-                'url': 'http://video.thestaticvube.com/video/%s/%s.mp4' % (fmt['media_resolution_id'], public_id),
-                'height': int(fmt['height']),
-                'abr': int(fmt['audio_bitrate']),
-                'vbr': int(fmt['video_bitrate']),
-                'format_id': fmt['media_resolution_id']
-            } for fmt in video['mtm'] if fmt['transcoding_status'] == 'processed'
-        ]
+        formats = []
+
+        for media in video['media'].get('video', []) + video['media'].get('audio', []):
+            if media['transcoding_status'] != 'processed':
+                continue
+            fmt = {
+                'url': 'http://video.thestaticvube.com/video/%s/%s.mp4' % (media['media_resolution_id'], public_id),
+                'abr': int(media['audio_bitrate']),
+                'format_id': compat_str(media['media_resolution_id']),
+            }
+            vbr = int(media['video_bitrate'])
+            if vbr:
+                fmt.update({
+                    'vbr': vbr,
+                    'height': int(media['height']),
+                })
+            formats.append(fmt)
 
         self._sort_formats(formats)
 
         title = video['title']
         description = video.get('description')
-        thumbnail = video['thumbnail_src']
-        if thumbnail.startswith('//'):
-            thumbnail = 'http:' + thumbnail
-        uploader = video['user_alias']
-        uploader_id = video['user_url_id']
-        timestamp = int(video['upload_time'])
+        thumbnail = self._proto_relative_url(video.get('thumbnail_src'), scheme='http:')
+        uploader = video.get('user_alias') or video.get('channel')
+        timestamp = int_or_none(video.get('upload_time'))
         duration = video['duration']
         view_count = video.get('raw_view_count')
         like_count = video.get('total_likes')
-        dislike_count= video.get('total_hates')
+        dislike_count = video.get('total_hates')
 
-        comment = self._download_json(
-            'http://vube.com/api/video/%s/comment' % video_id, video_id, 'Downloading video comment JSON')
+        comments = video.get('comments')
+        comment_count = None
+        if comments is None:
+            comment_data = self._download_json(
+                'http://vube.com/api/video/%s/comment' % video_id,
+                video_id, 'Downloading video comment JSON', fatal=False)
+            if comment_data is not None:
+                comment_count = int_or_none(comment_data.get('total'))
+        else:
+            comment_count = len(comments)
 
-        comment_count = int_or_none(comment.get('total'))
+        categories = [tag['text'] for tag in video['tags']]
 
         return {
             'id': video_id,
@@ -92,11 +132,11 @@ class VubeIE(InfoExtractor):
             'description': description,
             'thumbnail': thumbnail,
             'uploader': uploader,
-            'uploader_id': uploader_id,
             'timestamp': timestamp,
             'duration': duration,
             'view_count': view_count,
             'like_count': like_count,
             'dislike_count': dislike_count,
             'comment_count': comment_count,
+            'categories': categories,
         }

youtube_dl/extractor/wdr.py

@@ -55,14 +55,14 @@ class WDRIE(InfoExtractor):
             },
         },
         {
-            'url': 'http://www.funkhauseuropa.de/av/audiosuepersongsoulbossanova100-audioplayer.html',
-            'md5': '24e83813e832badb0a8d7d1ef9ef0691',
+            'url': 'http://www.funkhauseuropa.de/av/audioflaviacoelhoamaramar100-audioplayer.html',
+            'md5': '99a1443ff29af19f6c52cf6f4dc1f4aa',
             'info_dict': {
-                'id': 'mdb-463528',
+                'id': 'mdb-478135',
                 'ext': 'mp3',
-                'title': 'Süpersong: Soul Bossa Nova',
+                'title': 'Flavia Coelho: Amar é Amar',
                 'description': 'md5:7b29e97e10dfb6e265238b32fa35b23a',
-                'upload_date': '20140630',
+                'upload_date': '20140717',
             },
         },
     ]
@@ -81,7 +81,7 @@ class WDRIE(InfoExtractor):
             ]
             return self.playlist_result(entries, page_id)
 
-        flashvars = compat_urlparse.parse_qs(
+        flashvars = compat_parse_qs(
             self._html_search_regex(r'<param name="flashvars" value="([^"]+)"', webpage, 'flashvars'))
 
         page_id = flashvars['trackerClipId'][0]

youtube_dl/extractor/xboxclips.py

@@ -0,0 +1,57 @@
+# encoding: utf-8
+from __future__ import unicode_literals
+
+import re
+
+from .common import InfoExtractor
+from ..utils import (
+    parse_iso8601,
+    float_or_none,
+    int_or_none,
+)
+
+
+class XboxClipsIE(InfoExtractor):
+    _VALID_URL = r'https?://(?:www\.)?xboxclips\.com/video\.php\?.*vid=(?P<id>[\w-]{36})'
+    _TEST = {
+        'url': 'https://xboxclips.com/video.php?uid=2533274823424419&gamertag=Iabdulelah&vid=074a69a9-5faf-46aa-b93b-9909c1720325',
+        'md5': 'fbe1ec805e920aeb8eced3c3e657df5d',
+        'info_dict': {
+            'id': '074a69a9-5faf-46aa-b93b-9909c1720325',
+            'ext': 'mp4',
+            'title': 'Iabdulelah playing Upload Studio',
+            'filesize_approx': 28101836.8,
+            'timestamp': 1407388500,
+            'upload_date': '20140807',
+            'duration': 56,
+        }
+    }
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        video_id = mobj.group('id')
+
+        webpage = self._download_webpage(url, video_id)
+
+        video_url = self._html_search_regex(
+            r'>Link: <a href="([^"]+)">', webpage, 'video URL')
+        title = self._html_search_regex(
+            r'<title>XboxClips \| ([^<]+)</title>', webpage, 'title')
+        timestamp = parse_iso8601(self._html_search_regex(
+            r'>Recorded: ([^<]+)<', webpage, 'upload date', fatal=False))
+        filesize = float_or_none(self._html_search_regex(
+            r'>Size: ([\d\.]+)MB<', webpage, 'file size', fatal=False), invscale=1024 * 1024)
+        duration = int_or_none(self._html_search_regex(
+            r'>Duration: (\d+) Seconds<', webpage, 'duration', fatal=False))
+        view_count = int_or_none(self._html_search_regex(
+            r'>Views: (\d+)<', webpage, 'view count', fatal=False))
+
+        return {
+            'id': video_id,
+            'url': video_url,
+            'title': title,
+            'timestamp': timestamp,
+            'filesize_approx': filesize,
+            'duration': duration,
+            'view_count': view_count,
+        }

youtube_dl/extractor/yahoo.py

@@ -15,7 +15,7 @@ from ..utils import (
 
 class YahooIE(InfoExtractor):
     IE_DESC = 'Yahoo screen and movies'
-    _VALID_URL = r'https?://(?:screen|movies)\.yahoo\.com/.*?-(?P<id>[0-9]+)(?:-[a-z]+)?\.html'
+    _VALID_URL = r'(?P<url>https?://(?:screen|movies)\.yahoo\.com/.*?-(?P<id>[0-9]+)(?:-[a-z]+)?\.html)'
     _TESTS = [
         {
             'url': 'http://screen.yahoo.com/julian-smith-travis-legg-watch-214727115.html',
@@ -46,12 +46,23 @@ class YahooIE(InfoExtractor):
                 'title': 'The World Loves Spider-Man',
                 'description': '''People all over the world are celebrating the release of \"The Amazing Spider-Man 2.\" We're taking a look at the enthusiastic response Spider-Man has received from viewers all over the world.''',
             }
-        }
+        },
+        {
+            'url': 'https://screen.yahoo.com/community/community-sizzle-reel-203225340.html?format=embed',
+            'md5': '60e8ac193d8fb71997caa8fce54c6460',
+            'info_dict': {
+                'id': '4fe78544-8d48-39d8-97cd-13f205d9fcdb',
+                'ext': 'mp4',
+                'title': "Yahoo Saves 'Community'",
+                'description': 'md5:4d4145af2fd3de00cbb6c1d664105053',
+            }
+        },
     ]
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
         video_id = mobj.group('id')
+        url = mobj.group('url')
         webpage = self._download_webpage(url, video_id)
 
         items_json = self._search_regex(

youtube_dl/extractor/youtube.py

@@ -1,15 +1,12 @@
 # coding: utf-8
 
-import collections
 import errno
 import io
 import itertools
 import json
 import os.path
 import re
-import struct
 import traceback
-import zlib
 
 from .common import InfoExtractor, SearchInfoExtractor
 from .subtitles import SubtitlesInfoExtractor
@@ -347,15 +344,22 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor):
         """Indicate the download will use the RTMP protocol."""
         self.to_screen(u'RTMP download detected')
 
-    def _extract_signature_function(self, video_id, player_url, slen):
+    def _signature_cache_id(self, example_sig):
+        """ Return a string representation of a signature """
+        return u'.'.join(compat_str(len(part)) for part in example_sig.split('.'))
+
+    def _extract_signature_function(self, video_id, player_url, example_sig):
         id_m = re.match(
-            r'.*-(?P<id>[a-zA-Z0-9_-]+)(?:/watch_as3)?\.(?P<ext>[a-z]+)$',
+            r'.*-(?P<id>[a-zA-Z0-9_-]+)(?:/watch_as3|/html5player)?\.(?P<ext>[a-z]+)$',
             player_url)
+        if not id_m:
+            raise ExtractorError('Cannot identify player %r' % player_url)
         player_type = id_m.group('ext')
         player_id = id_m.group('id')
 
         # Read from filesystem cache
-        func_id = '%s_%s_%d' % (player_type, player_id, slen)
+        func_id = '%s_%s_%s' % (
+            player_type, player_id, self._signature_cache_id(example_sig))
         assert os.path.basename(func_id) == func_id
         cache_dir = get_cachedir(self._downloader.params)
 
@@ -370,6 +374,13 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor):
                 return lambda s: u''.join(s[i] for i in cache_spec)
             except IOError:
                 pass  # No cache available
+            except ValueError:
+                try:
+                    file_size = os.path.getsize(cache_fn)
+                except (OSError, IOError) as oe:
+                    file_size = str(oe)
+                self._downloader.report_warning(
+                    u'Cache %s failed (%s)' % (cache_fn, file_size))
 
         if player_type == 'js':
             code = self._download_webpage(
@@ -389,7 +400,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor):
 
         if cache_enabled:
             try:
-                test_string = u''.join(map(compat_chr, range(slen)))
+                test_string = u''.join(map(compat_chr, range(len(example_sig))))
                 cache_res = res(test_string)
                 cache_spec = [ord(c) for c in cache_res]
                 try:
@@ -405,7 +416,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor):
 
         return res
 
-    def _print_sig_code(self, func, slen):
+    def _print_sig_code(self, func, example_sig):
         def gen_sig_code(idxs):
             def _genslice(start, end, step):
                 starts = u'' if start == 0 else str(start)
@@ -434,11 +445,14 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor):
             else:
                 yield _genslice(start, i, step)
 
-        test_string = u''.join(map(compat_chr, range(slen)))
+        test_string = u''.join(map(compat_chr, range(len(example_sig))))
         cache_res = func(test_string)
         cache_spec = [ord(c) for c in cache_res]
         expr_code = u' + '.join(gen_sig_code(cache_spec))
-        code = u'if len(s) == %d:\n    return %s\n' % (slen, expr_code)
+        signature_id_tuple = '(%s)' % (
+            ', '.join(compat_str(len(p)) for p in example_sig.split('.')))
+        code = (u'if tuple(len(p) for p in s.split(\'.\')) == %s:\n'
+                u'    return %s\n') % (signature_id_tuple, expr_code)
         self.to_screen(u'Extracted signature function:\n' + code)
 
     def _parse_sig_js(self, jscode):
@@ -466,20 +480,20 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor):
         if player_url.startswith(u'//'):
             player_url = u'https:' + player_url
         try:
-            player_id = (player_url, len(s))
+            player_id = (player_url, self._signature_cache_id(s))
             if player_id not in self._player_cache:
                 func = self._extract_signature_function(
-                    video_id, player_url, len(s)
+                    video_id, player_url, s
                 )
                 self._player_cache[player_id] = func
             func = self._player_cache[player_id]
             if self._downloader.params.get('youtube_print_sig_code'):
-                self._print_sig_code(func, len(s))
+                self._print_sig_code(func, s)
             return func(s)
         except Exception as e:
             tb = traceback.format_exc()
             raise ExtractorError(
-                u'Automatic signature extraction failed: ' + tb, cause=e)
+                u'Signature extraction failed: ' + tb, cause=e)
 
     def _get_available_subtitles(self, video_id, webpage):
         try:
@@ -609,14 +623,12 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor):
             age_gate = True
             # We simulate the access to the video from www.youtube.com/v/{video_id}
             # this can be viewed without login into Youtube
-            data = compat_urllib_parse.urlencode({'video_id': video_id,
-                                                  'el': 'player_embedded',
-                                                  'gl': 'US',
-                                                  'hl': 'en',
-                                                  'eurl': 'https://youtube.googleapis.com/v/' + video_id,
-                                                  'asv': 3,
-                                                  'sts':'1588',
-                                                  })
+            data = compat_urllib_parse.urlencode({
+                'video_id': video_id,
+                'eurl': 'https://youtube.googleapis.com/v/' + video_id,
+                'sts': self._search_regex(
+                    r'"sts"\s*:\s*(\d+)', video_webpage, 'sts'),
+            })
             video_info_url = proto + '://www.youtube.com/get_video_info?' + data
             video_info_webpage = self._download_webpage(video_info_url, video_id,
                                     note=False,
@@ -809,50 +821,54 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor):
             url_map = {}
             for url_data_str in encoded_url_map.split(','):
                 url_data = compat_parse_qs(url_data_str)
-                if 'itag' in url_data and 'url' in url_data:
-                    url = url_data['url'][0]
-                    if 'sig' in url_data:
-                        url += '&signature=' + url_data['sig'][0]
-                    elif 's' in url_data:
-                        encrypted_sig = url_data['s'][0]
+                if 'itag' not in url_data or 'url' not in url_data:
+                    continue
+                format_id = url_data['itag'][0]
+                url = url_data['url'][0]
 
-                        if not age_gate:
-                            jsplayer_url_json = self._search_regex(
-                                r'"assets":.+?"js":\s*("[^"]+")',
-                                video_webpage, u'JS player URL')
-                            player_url = json.loads(jsplayer_url_json)
+                if 'sig' in url_data:
+                    url += '&signature=' + url_data['sig'][0]
+                elif 's' in url_data:
+                    encrypted_sig = url_data['s'][0]
+
+                    if not age_gate:
+                        jsplayer_url_json = self._search_regex(
+                            r'"assets":.+?"js":\s*("[^"]+")',
+                            video_webpage, u'JS player URL')
+                        player_url = json.loads(jsplayer_url_json)
+                    if player_url is None:
+                        player_url_json = self._search_regex(
+                            r'ytplayer\.config.*?"url"\s*:\s*("[^"]+")',
+                            video_webpage, u'age gate player URL')
+                        player_url = json.loads(player_url_json)
+
+                    if self._downloader.params.get('verbose'):
                         if player_url is None:
-                            player_url_json = self._search_regex(
-                                r'ytplayer\.config.*?"url"\s*:\s*("[^"]+")',
-                                video_webpage, u'age gate player URL')
-                            player_url = json.loads(player_url_json)
-
-                        if self._downloader.params.get('verbose'):
-                            if player_url is None:
-                                player_version = 'unknown'
-                                player_desc = 'unknown'
+                            player_version = 'unknown'
+                            player_desc = 'unknown'
+                        else:
+                            if player_url.endswith('swf'):
+                                player_version = self._search_regex(
+                                    r'-(.+?)(?:/watch_as3)?\.swf$', player_url,
+                                    u'flash player', fatal=False)
+                                player_desc = 'flash player %s' % player_version
                             else:
-                                if player_url.endswith('swf'):
-                                    player_version = self._search_regex(
-                                        r'-(.+)\.swf$', player_url,
-                                        u'flash player', fatal=False)
-                                    player_desc = 'flash player %s' % player_version
-                                else:
-                                    player_version = self._search_regex(
-                                        r'html5player-(.+?)\.js', video_webpage,
-                                        'html5 player', fatal=False)
-                                    player_desc = u'html5 player %s' % player_version
+                                player_version = self._search_regex(
+                                    r'html5player-([^/]+?)(?:/html5player)?\.js',
+                                    player_url,
+                                    'html5 player', fatal=False)
+                                player_desc = u'html5 player %s' % player_version
 
-                            parts_sizes = u'.'.join(compat_str(len(part)) for part in encrypted_sig.split('.'))
-                            self.to_screen(u'encrypted signature length %d (%s), itag %s, %s' %
-                                (len(encrypted_sig), parts_sizes, url_data['itag'][0], player_desc))
+                        parts_sizes = self._signature_cache_id(encrypted_sig)
+                        self.to_screen(u'{%s} signature length %s, %s' %
+                            (format_id, parts_sizes, player_desc))
 
-                        signature = self._decrypt_signature(
-                            encrypted_sig, video_id, player_url, age_gate)
-                        url += '&signature=' + signature
-                    if 'ratebypass' not in url:
-                        url += '&ratebypass=yes'
-                    url_map[url_data['itag'][0]] = url
+                    signature = self._decrypt_signature(
+                        encrypted_sig, video_id, player_url, age_gate)
+                    url += '&signature=' + signature
+                if 'ratebypass' not in url:
+                    url += '&ratebypass=yes'
+                url_map[format_id] = url
             formats = _map_to_format_list(url_map)
         elif video_info.get('hlsvp'):
             manifest_url = video_info['hlsvp'][0]

youtube_dl/jsinterp.py

@@ -1,5 +1,6 @@
 from __future__ import unicode_literals
 
+import json
 import re
 
 from .utils import (
@@ -40,8 +41,9 @@ class JSInterpreter(object):
             assign = lambda v: v
             expr = stmt[len('return '):]
         else:
-            raise ExtractorError(
-                'Cannot determine left side of statement in %r' % stmt)
+            # Try interpreting it as an expression
+            expr = stmt
+            assign = lambda v: v
 
         v = self.interpret_expression(expr, local_vars, allow_recursion)
         return assign(v)
@@ -53,35 +55,63 @@ class JSInterpreter(object):
         if expr.isalpha():
             return local_vars[expr]
 
-        m = re.match(r'^(?P<in>[a-z]+)\.(?P<member>.*)$', expr)
-        if m:
-            member = m.group('member')
-            variable = m.group('in')
+        try:
+            return json.loads(expr)
+        except ValueError:
+            pass
 
-            if variable not in local_vars:
+        m = re.match(
+            r'^(?P<var>[a-zA-Z0-9_]+)\.(?P<member>[^(]+)(?:\(+(?P<args>[^()]*)\))?$',
+            expr)
+        if m:
+            variable = m.group('var')
+            member = m.group('member')
+            arg_str = m.group('args')
+
+            if variable in local_vars:
+                obj = local_vars[variable]
+            else:
                 if variable not in self._objects:
                     self._objects[variable] = self.extract_object(variable)
                 obj = self._objects[variable]
-                key, args = member.split('(', 1)
-                args = args.strip(')')
-                argvals = [int(v) if v.isdigit() else local_vars[v]
-                           for v in args.split(',')]
-                return obj[key](argvals)
 
-            val = local_vars[variable]
-            if member == 'split("")':
-                return list(val)
-            if member == 'join("")':
-                return ''.join(val)
-            if member == 'length':
-                return len(val)
-            if member == 'reverse()':
-                return val[::-1]
-            slice_m = re.match(r'slice\((?P<idx>.*)\)', member)
-            if slice_m:
-                idx = self.interpret_expression(
-                    slice_m.group('idx'), local_vars, allow_recursion - 1)
-                return val[idx:]
+            if arg_str is None:
+                # Member access
+                if member == 'length':
+                    return len(obj)
+                return obj[member]
+
+            assert expr.endswith(')')
+            # Function call
+            if arg_str == '':
+                argvals = tuple()
+            else:
+                argvals = tuple([
+                    self.interpret_expression(v, local_vars, allow_recursion)
+                    for v in arg_str.split(',')])
+
+            if member == 'split':
+                assert argvals == ('',)
+                return list(obj)
+            if member == 'join':
+                assert len(argvals) == 1
+                return argvals[0].join(obj)
+            if member == 'reverse':
+                assert len(argvals) == 0
+                obj.reverse()
+                return obj
+            if member == 'slice':
+                assert len(argvals) == 1
+                return obj[argvals[0]:]
+            if member == 'splice':
+                assert isinstance(obj, list)
+                index, howMany = argvals
+                res = []
+                for i in range(index, min(index + howMany, len(obj))):
+                    res.append(obj.pop(index))
+                return res
+
+            return obj[member](argvals)
 
         m = re.match(
             r'^(?P<in>[a-z]+)\[(?P<idx>.+)\]$', expr)
@@ -103,10 +133,11 @@ class JSInterpreter(object):
             r'^(?P<func>[a-zA-Z$]+)\((?P<args>[a-z0-9,]+)\)$', expr)
         if m:
             fname = m.group('func')
+            argvals = tuple([
+                int(v) if v.isdigit() else local_vars[v]
+                for v in m.group('args').split(',')])
             if fname not in self._functions:
                 self._functions[fname] = self.extract_function(fname)
-            argvals = [int(v) if v.isdigit() else local_vars[v]
-                       for v in m.group('args').split(',')]
             return self._functions[fname](argvals)
         raise ExtractorError('Unsupported JS expression %r' % expr)
 
@@ -114,13 +145,13 @@ class JSInterpreter(object):
         obj = {}
         obj_m = re.search(
             (r'(?:var\s+)?%s\s*=\s*\{' % re.escape(objname)) +
-            r'\s*(?P<fields>([a-zA-Z$]+\s*:\s*function\(.*?\)\s*\{.*?\})*)' +
+            r'\s*(?P<fields>([a-zA-Z$0-9]+\s*:\s*function\(.*?\)\s*\{.*?\})*)' +
             r'\}\s*;',
             self.code)
         fields = obj_m.group('fields')
         # Currently, it only supports function definitions
         fields_m = re.finditer(
-            r'(?P<key>[a-zA-Z$]+)\s*:\s*function'
+            r'(?P<key>[a-zA-Z$0-9]+)\s*:\s*function'
             r'\((?P<args>[a-z,]+)\){(?P<code>[^}]+)}',
             fields)
         for f in fields_m:

youtube_dl/postprocessor/ffmpeg.py

@@ -18,14 +18,15 @@ from ..utils import (
 )
 
 
-
 class FFmpegPostProcessorError(PostProcessingError):
     pass
 
+
 class FFmpegPostProcessor(PostProcessor):
-    def __init__(self,downloader=None):
+    def __init__(self, downloader=None, deletetempfiles=False):
         PostProcessor.__init__(self, downloader)
         self._exes = self.detect_executables()
+        self._deletetempfiles = deletetempfiles
 
     @staticmethod
     def detect_executables():
@@ -60,6 +61,9 @@ class FFmpegPostProcessor(PostProcessor):
             stderr = stderr.decode('utf-8', 'replace')
             msg = stderr.strip().split('\n')[-1]
             raise FFmpegPostProcessorError(msg)
+        if self._deletetempfiles:
+            for ipath in input_paths:
+                os.remove(ipath)
 
     def run_ffmpeg(self, path, out_path, opts):
         self.run_ffmpeg_multiple_files([path], out_path, opts)

youtube_dl/swfinterp.py

@@ -2,12 +2,12 @@ from __future__ import unicode_literals
 
 import collections
 import io
-import struct
 import zlib
 
 from .utils import (
     compat_str,
     ExtractorError,
+    struct_unpack,
 )
 
 
@@ -23,17 +23,17 @@ def _extract_tags(file_contents):
             file_contents[:1])
 
     # Determine number of bits in framesize rectangle
-    framesize_nbits = struct.unpack('!B', content[:1])[0] >> 3
+    framesize_nbits = struct_unpack('!B', content[:1])[0] >> 3
     framesize_len = (5 + 4 * framesize_nbits + 7) // 8
 
     pos = framesize_len + 2 + 2
     while pos < len(content):
-        header16 = struct.unpack('<H', content[pos:pos + 2])[0]
+        header16 = struct_unpack('<H', content[pos:pos + 2])[0]
         pos += 2
         tag_code = header16 >> 6
         tag_len = header16 & 0x3f
         if tag_len == 0x3f:
-            tag_len = struct.unpack('<I', content[pos:pos + 4])[0]
+            tag_len = struct_unpack('<I', content[pos:pos + 4])[0]
             pos += 4
         assert pos + tag_len <= len(content), \
             ('Tag %d ends at %d+%d - that\'s longer than the file (%d)'
@@ -99,7 +99,7 @@ def _read_int(reader):
     for _ in range(5):
         buf = reader.read(1)
         assert len(buf) == 1
-        b = struct.unpack('<B', buf)[0]
+        b = struct_unpack('<B', buf)[0]
         res = res | ((b & 0x7f) << shift)
         if b & 0x80 == 0:
             break
@@ -111,7 +111,7 @@ def _u30(reader):
     res = _read_int(reader)
     assert res & 0xf0000000 == 0
     return res
-u32 = _read_int
+_u32 = _read_int
 
 
 def _s32(reader):
@@ -125,7 +125,7 @@ def _s24(reader):
     bs = reader.read(3)
     assert len(bs) == 3
     last_byte = b'\xff' if (ord(bs[2:3]) >= 0x80) else b'\x00'
-    return struct.unpack('<i', bs + last_byte)[0]
+    return struct_unpack('<i', bs + last_byte)[0]
 
 
 def _read_string(reader):
@@ -144,7 +144,7 @@ def _read_bytes(count, reader):
 
 def _read_byte(reader):
     resb = _read_bytes(1, reader=reader)
-    res = struct.unpack('<B', resb)[0]
+    res = struct_unpack('<B', resb)[0]
     return res
 
 
@@ -470,8 +470,7 @@ class SWFInterpreter(object):
 
                     mname = self.multinames[index]
                     assert isinstance(obj, _AVMClass)
-                    construct_method = self.extract_function(
-                        obj, mname)
+
                     # We do not actually call the constructor for now;
                     # we just pretend it does nothing
                     stack.append(obj.make_object())

youtube_dl/utils.py

@@ -24,6 +24,7 @@ import socket
 import struct
 import subprocess
 import sys
+import tempfile
 import traceback
 import xml.etree.ElementTree
 import zlib
@@ -91,11 +92,9 @@ except ImportError:
     compat_subprocess_get_DEVNULL = lambda: open(os.path.devnull, 'w')
 
 try:
-    from urllib.parse import parse_qs as compat_parse_qs
-except ImportError: # Python 2
-    # HACK: The following is the correct parse_qs implementation from cpython 3's stdlib.
-    # Python 2's version is apparently totally broken
-    def _unquote(string, encoding='utf-8', errors='replace'):
+    from urllib.parse import unquote as compat_urllib_parse_unquote
+except ImportError:
+    def compat_urllib_parse_unquote(string, encoding='utf-8', errors='replace'):
         if string == '':
             return string
         res = string.split('%')
@@ -130,6 +129,13 @@ except ImportError: # Python 2
             string += pct_sequence.decode(encoding, errors)
         return string
 
+
+try:
+    from urllib.parse import parse_qs as compat_parse_qs
+except ImportError: # Python 2
+    # HACK: The following is the correct parse_qs implementation from cpython 3's stdlib.
+    # Python 2's version is apparently totally broken
+
     def _parse_qsl(qs, keep_blank_values=False, strict_parsing=False,
                 encoding='utf-8', errors='replace'):
         qs, _coerce_result = qs, unicode
@@ -149,10 +155,12 @@ except ImportError: # Python 2
                     continue
             if len(nv[1]) or keep_blank_values:
                 name = nv[0].replace('+', ' ')
-                name = _unquote(name, encoding=encoding, errors=errors)
+                name = compat_urllib_parse_unquote(
+                    name, encoding=encoding, errors=errors)
                 name = _coerce_result(name)
                 value = nv[1].replace('+', ' ')
-                value = _unquote(value, encoding=encoding, errors=errors)
+                value = compat_urllib_parse_unquote(
+                    value, encoding=encoding, errors=errors)
                 value = _coerce_result(value)
                 r.append((name, value))
         return r
@@ -221,22 +229,40 @@ else:
         assert type(s) == type(u'')
         print(s)
 
-# In Python 2.x, json.dump expects a bytestream.
-# In Python 3.x, it writes to a character stream
-if sys.version_info < (3,0):
-    def write_json_file(obj, fn):
-        with open(fn, 'wb') as f:
-            json.dump(obj, f)
-else:
-    def write_json_file(obj, fn):
-        with open(fn, 'w', encoding='utf-8') as f:
-            json.dump(obj, f)
 
-if sys.version_info >= (2,7):
+def write_json_file(obj, fn):
+    """ Encode obj as JSON and write it to fn, atomically """
+
+    # In Python 2.x, json.dump expects a bytestream.
+    # In Python 3.x, it writes to a character stream
+    if sys.version_info < (3, 0):
+        mode = 'wb'
+        encoding = None
+    else:
+        mode = 'w'
+        encoding = 'utf-8'
+    tf = tempfile.NamedTemporaryFile(
+        suffix='.tmp', prefix=os.path.basename(fn) + '.',
+        dir=os.path.dirname(fn),
+        delete=False)
+
+    try:
+        with tf:
+            json.dump(obj, tf)
+        os.rename(tf.name, fn)
+    except:
+        try:
+            os.remove(tf.name)
+        except OSError:
+            pass
+        raise
+
+
+if sys.version_info >= (2, 7):
     def find_xpath_attr(node, xpath, key, val):
         """ Find the xpath xpath[@key=val] """
-        assert re.match(r'^[a-zA-Z]+$', key)
-        assert re.match(r'^[a-zA-Z0-9@\s:._]*$', val)
+        assert re.match(r'^[a-zA-Z-]+$', key)
+        assert re.match(r'^[a-zA-Z0-9@\s:._-]*$', val)
         expr = xpath + u"[@%s='%s']" % (key, val)
         return node.find(expr)
 else:
@@ -820,6 +846,7 @@ def unified_strdate(date_str):
         '%b %dnd %Y %I:%M%p',
         '%b %dth %Y %I:%M%p',
         '%Y-%m-%d',
+        '%Y/%m/%d',
         '%d.%m.%Y',
         '%d/%m/%Y',
         '%Y/%m/%d %H:%M:%S',
@@ -845,6 +872,8 @@ def unified_strdate(date_str):
     return upload_date
 
 def determine_ext(url, default_ext=u'unknown_video'):
+    if url is None:
+        return default_ext
     guess = url.partition(u'?')[0].rpartition(u'.')[2]
     if re.match(r'^[A-Za-z0-9]+$', guess):
         return guess
@@ -1193,13 +1222,6 @@ def format_bytes(bytes):
     return u'%.2f%s' % (converted, suffix)
 
 
-def str_to_int(int_str):
-    if int_str is None:
-        return None
-    int_str = re.sub(r'[,\.]', u'', int_str)
-    return int(int_str)
-
-
 def get_term_width():
     columns = os.environ.get('COLUMNS', None)
     if columns:
@@ -1267,15 +1289,28 @@ class HEADRequest(compat_urllib_request.Request):
         return "HEAD"
 
 
-def int_or_none(v, scale=1, default=None, get_attr=None):
+def int_or_none(v, scale=1, default=None, get_attr=None, invscale=1):
     if get_attr:
         if v is not None:
             v = getattr(v, get_attr, None)
-    return default if v is None else (int(v) // scale)
+    if v == '':
+        v = None
+    return default if v is None else (int(v) * invscale // scale)
 
 
-def float_or_none(v, scale=1, default=None):
-    return default if v is None else (float(v) / scale)
+def str_or_none(v, default=None):
+    return default if v is None else compat_str(v)
+
+
+def str_to_int(int_str):
+    if int_str is None:
+        return None
+    int_str = re.sub(r'[,\.]', u'', int_str)
+    return int(int_str)
+
+
+def float_or_none(v, scale=1, invscale=1, default=None):
+    return default if v is None else (float(v) * invscale / scale)
 
 
 def parse_duration(s):

youtube_dl/version.py

@@ -1,2 +1,2 @@
 
-__version__ = '2014.07.20'
+__version__ = '2014.08.21.2'